Influence of the natural aluminium oxide layer on ... - ALU-WEB.DE

Ebner 

Special Edition 

European Aluminium Congress 2011 

22 to 23 November 2011, Maritim Hotel Düsseldorf 

TECHNOLOGIES FOR THE 

ALUMINIUM INDUSTRY 

Adressing market requirements 

in alumin- 

ium flat rolled products 

Heavy duty extrusion 

presses for large 

pr<strong>of</strong>ile applications 

Soldering and brazing <strong>of</strong> 

<strong>aluminium</strong> and its alloys

Trust... 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

SMS SIEMAG AG 

�� 

�� 

�� 

�� 

�� 

�� 

��

Christian Wellner 

Executive Director <strong>of</strong> GDA 

Gesamtverband der Aluminiumindustrie 

e.V. (German Aluminium 

Association) 

Aluminium industry 

continuing to grow 

in future 

Despite <strong>the</strong> current crisis in <strong>the</strong> banking 

and finance sector, <strong>the</strong> global <strong>aluminium</strong> 

industry is continuing to look optimistical- 

ly to <strong>the</strong> future. The industry is still on a 

path <strong>of</strong> solid growth despite a weakening 

in <strong>the</strong> growth dynamics and a dampener 

being put on business expectations. 

Forecasts for <strong>aluminium</strong> continue to be 

optimistic. The dynamic development is 

closely linked to <strong>the</strong> innovative capability 

<strong>of</strong> <strong>the</strong> sector and <strong>the</strong> metal’s beneficial 

properties. Aluminium has established itself 

in numerous user markets. Whe<strong>the</strong>r it 

be in <strong>the</strong> most important market, <strong>the</strong> transport 

sector, in mechanical engineering, 

in <strong>the</strong> electronics sector or in packaging, 

<strong>aluminium</strong> companies have continued to 

push technological development forward 

and opened up new uses and fields <strong>of</strong> application 

for <strong>the</strong> lightweight metal again 

and again. The worldwide consumption 

<strong>of</strong> <strong>aluminium</strong> will continue to grow in <strong>the</strong> 

next 10 to 15 years, with an annual global 

consumption <strong>of</strong> some 70 million tonnes 

expected by around 2020. Growing demand 

from Asia and <strong>the</strong> important user 

markets – automotive, mechanical engineering, 

building and construction, packaging 

and solar energy – will help <strong>the</strong> lightweight 

metal achieve continuous growth. 

Growing consumption <strong>of</strong> <strong>aluminium</strong> will 

result in enormous importance being given 

to <strong>the</strong> development <strong>of</strong> innovative process 

technologies and plant concepts for <strong>the</strong> 

production and processing <strong>of</strong> <strong>aluminium</strong>. 

More complex material properties demand 

a high degree <strong>of</strong> technological expertise 

FOREWORD 

and <strong>the</strong> development <strong>of</strong> special solutions. 

Future technological developments will 

also focus on resource efficiency and potential 

savings <strong>of</strong> CO 2 and energy in all 

processes. In future, too, we must repeatedly 

develop material properties fur<strong>the</strong>r, 

create new products and optimise production 

processes via investments in application-oriented 

research and development. 

What innovative technologies can 

equipment suppliers <strong>of</strong>fer <strong>the</strong> <strong>aluminium</strong> 

industry? Will new processes or material 

developments open up new markets for 

<strong>aluminium</strong>? What new trends are emerging? 

Experts from <strong>the</strong> <strong>aluminium</strong> industry 

will be discussing <strong>the</strong>se and o<strong>the</strong>r questions 

with <strong>the</strong>ir equipment partners and suppliers 

at <strong>the</strong> European Aluminium Congress 

‘Technologies for <strong>the</strong> Aluminium Industry’ 

on 22-23 November 2011 in Düsseldorf. 

It provides <strong>the</strong> equipment suppliers and 

technology partners <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> producing 

and processing industry with a comprehensive 

platform to present and discuss 

<strong>the</strong>ir latest developments. 

The congress programme <strong>of</strong>fers <strong>the</strong> opportunity 

to exchange information and 

have detailed discussions with representatives 

<strong>of</strong> <strong>the</strong> European <strong>aluminium</strong> industry, 

its equipment manufacturers and o<strong>the</strong>r 

suppliers. 

I wish all EAC congress participants an 

informative programme <strong>of</strong> presentations 

and I hope <strong>the</strong>y enjoy interesting discussions 

and make valuable contacts during 

both days <strong>of</strong> <strong>the</strong> congress in Düsseldorf. 

ALUMINIUM · XX/2011 EAC CONGRESS 2011 3 

�

FOREWORD 

Aluminium industry continuing to grow in future ............................... 3 

SESSION ROLLING INDUSTRY 

Foil slitting technology – an integrative approach ............................. 6 

Focus on rolling mill efficiency: design and control...........................10 

Batch-type and continuous floater furnace facilities for <strong>aluminium</strong> 

alloy strip ..................................................................................13 

Addressing market requirements in <strong>aluminium</strong> flat products ...............16 

SESSION MEASURING & CONTROL 

Longer campaigns with improved monolithics for lining 

<strong>aluminium</strong> melt-hold furnaces .......................................................19 

Optimization <strong>of</strong> an <strong>aluminium</strong> rolling mill with a high speed 

X-ray thickness gauge ..................................................................22 

State-<strong>of</strong>-<strong>the</strong>-art casthouse production management ..........................26 

Monitoring <strong>of</strong> <strong>aluminium</strong> extrusion dies cleaning process 

by an optic sensor .......................................................................28 

SESSION EXTRUSION 

<strong>Influence</strong> <strong>of</strong> raw material constitution on <strong>the</strong> quality <strong>of</strong> 

gloss alloy-type extrusion billets – a report from practice ................. 31 

Heavy duty extrusion presses for large pr<strong>of</strong>ile applications .................34 

Extrusion line from a single source for intelligent pr<strong>of</strong>iles ............. 36/38 

Integrated processing <strong>of</strong> <strong>aluminium</strong> pr<strong>of</strong>iles subsequent to 

hot extrusion ..............................................................................40 

Flexible automated material flow in extrusion operations ...................44 

SESSION APPLICATION-ORIENTED TECHNOLOGIES 

Laser cleaning and surface modification <strong>of</strong> <strong>aluminium</strong> – 

first step to a green plant ............................................................48 

Corrosion behaviour <strong>of</strong> <strong>aluminium</strong> materials in aqueous 

cleaning solutions .......................................................................50 

Soldering and brazing <strong>of</strong> <strong>aluminium</strong> and its alloys ............................54 

<strong>Influence</strong> <strong>of</strong> <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong> on brazeability ............57 

SESSION MELTING, RECYCLING & HEAT TREATMENT 

The new generation <strong>of</strong> <strong>aluminium</strong> heat treatment 

plants – a vanguard concept ........................................................60 

New standards in gas-fired heating <strong>of</strong> moulds, ladles, launders 

and melting furnaces through flameless burner technology ................63 

Effects <strong>of</strong> gaseous pyrolysis products on <strong>aluminium</strong> recycling yield ......66 

SESSION SOFTWARE & SIMULATION 

Optimizing <strong>the</strong> energy consumption <strong>of</strong> coil annealing 

furnaces by ma<strong>the</strong>matical modelling <strong>of</strong> <strong>the</strong> annealing process ............70 

Numerical analysis <strong>of</strong> <strong>the</strong> seam welds <strong>of</strong> industrial extrusion pr<strong>of</strong>iles ...72 

Imprint ......................................................................................53 

List <strong>of</strong> advertisers 

CONTENTS 

ALUMINIUM · XX/2011 EAC CONGRESS 2011 5 

6 

38 

50 

F. W. Brökelmann Aluminiumwerk 

GmbH & Co. KG 

BWG Bergwerk- und Walzwerk- 

61 

Maschinenbau GmbH 

Danieli Fröhling Josef Fröhling 

4 

GmbH & Co.KG 15 

Drache Umwelttechnik GmbH 71 

Ebner Industrie<strong>of</strong>enbau GmbH, Austria 18/19 

Eisenmann AG 

Kampf Schneid- und 

21 

Wickeltechnik GmbH & Co. KG 

Kind & Co. Edelstahlwerk 

25 

Kommanditgesellschaft 45 

LOI Thermprocess GmbH 9 

S+C Extrusion Tooling Solutions GmbH 35 

SMS Meer GmbH 76 

SMS Siemag AG 2

ROLLING INDUSTRY 

Foil slitting technology – an integrative approach 

Walter Brockhorst and Gabriele Barten, Achenbach Buschhütten GmbH 

System supplier Achenbach Buschhütten 

Achenbach sees itself as system supplier 

<strong>of</strong> machinery for producing non-ferrous 

metal rolling mills and foil slitting machines. 

Achenbach’s initial communication 

with <strong>the</strong>ir customers is threefold: 

‘We are <strong>the</strong> Specialists!’, ‘We are Leaders 

in technology and quality!’ and ‘We are 

your Partner!’ The brandname Achenbach 

provides orientation and security 

for <strong>the</strong> customers. Achenbach is a familyowned 

and family-run company having 

approx. 310 employees. Achenbach rolling 

mills and machinery are operated in 

nearly 60 countries all over <strong>the</strong> world. 

Company development and mission 

Our mission is ‘Technology for Future Concepts’, 

which controls and coordinates think- 

ing and acting at Achenbach. Achenbach has 

<strong>the</strong> technological know-how to realize <strong>the</strong> 

future ideas <strong>of</strong> <strong>the</strong>ir customers for producing 

strips, foils as well as foil products in <strong>the</strong> 

form <strong>of</strong> most modern machinery. The aim is 

to provide our customers with first-class and 

tailor-made rolling mills and machinery in 

order to open up success potentials for <strong>the</strong>ir 

company’s future. High machine performance 

and numerous references have ensured <strong>the</strong> 

excellent reputation <strong>of</strong> Achenbach: 

• Established as a hammer mill in 1452, 

Achenbach has been manufacturing metal 

rolling mills since 1888. After <strong>the</strong> Second 

World War, Achenbach became <strong>the</strong> specialist 

for non-ferrous metals, such as <strong>aluminium</strong> or 

copper and copper alloys. Achenbach is <strong>the</strong> 

world market leader for <strong>aluminium</strong> thin-strip 

and foil rolling mills for more than 20 years 

now. 

• On this basis, Achenbach extended its product 

range in 2006. First came <strong>aluminium</strong> foil 

doublers, separators and slitting machines 

for <strong>aluminium</strong> foils, <strong>the</strong>n slitting machines 

for <strong>the</strong> converting industry (slitting machines 

for mon<strong>of</strong>oils, laminates, non-woven as well 

as all adhesive and non-adhesive materials) 

were added. The convincing market entry was 

facilitated by <strong>the</strong> trend-setting new developments 

with regard to design and construction. 

Nearly 20 machines <strong>of</strong> this type have already 

been successfully commissioned. 

Integrative technology 

As a system supplier, Achenbach follows an 

integrative approach to <strong>of</strong>fer overall technical 

solutions. ‘Everything from one single source’ 

gives <strong>the</strong> chance to optimally design <strong>the</strong> pro- 

duction process as a whole, from <strong>the</strong> rolling 

process via <strong>the</strong> slitting process up to <strong>the</strong> fur<strong>the</strong>r 

processing. 

Achenbach develops and manufactures 

rolling mills, slitting machines as well as important 

auxiliary systems on <strong>the</strong>ir single premises 

in Buschhütten. Therefore, <strong>the</strong> knowledge regarding 

<strong>the</strong> rolling mill is decisive for developing 

trendsetting slitting machines on <strong>the</strong> one 

hand. Vice versa, experiences in winding and 

slitting <strong>of</strong> rolled and even converting materials 

give valuable information for designing <strong>the</strong> 

rolling mills. On <strong>the</strong> o<strong>the</strong>r hand, <strong>the</strong> knowledge 

from manufacture, assembly and commissioning 

<strong>of</strong> this machinery form <strong>the</strong> basis 

for future developments. 

By covering all four fields, Achenbach deduces 

synergy effects in two directions for <strong>the</strong> 

benefit <strong>of</strong> <strong>the</strong> customers: 

1. Synergy effects between <strong>the</strong>oretic design 

and practical realization: support and extension 

<strong>of</strong> technology and quality leadership. It is 

a question <strong>of</strong> trustworthiness to not only design 

machines, but also to manufacture <strong>the</strong>m. 

2. Synergy effects between processing and 

converting industry: continuously growing demands 

on <strong>the</strong> quality <strong>of</strong> <strong>the</strong> slitting products 

in combination with highly sensitive materials 

underline <strong>the</strong> importance <strong>of</strong> a mutual knowhow 

transfer. 

This is exactly what Achenbach means by 

an integrative approach in foil slitting technology. 

The reproducibility due to <strong>the</strong> overall 

data flow and <strong>the</strong> optimization <strong>of</strong> <strong>the</strong> material 

flow within <strong>the</strong> production is <strong>the</strong> customers’ 

benefit. 

By meeting <strong>the</strong>se demands, Achenbach is 

qualified as a supplier not only for first-class 

stand-alone systems but also for highly-complex 

machinery ‘on greenfield sites’ or <strong>the</strong>ir 

modernization. With more than 120 years 

<strong>of</strong> experience in mechanical engineering, 

Achenbach is a qualified partner for sophisticated 

special requirements and is traditionally 

always ready to face tricky modernization 

projects. 

Market requirements 

With respect to <strong>the</strong> different kinds <strong>of</strong> foil slitting 

machines its demand can be qualitatively 

characterized as follows: growing technical 

requirements toge<strong>the</strong>r with more specific 

desires and needs <strong>of</strong> <strong>the</strong> customers, define a 

highly-complex context. This is reflected in 

6 ALUMINIUM · EAC CONGRESS 2011 

Images: Achenbach

<strong>the</strong> following developments and desires <strong>of</strong> <strong>the</strong> 

customers: 

• continuously higher speeds <strong>of</strong> <strong>the</strong> single 

machines 

• fur<strong>the</strong>r reduction <strong>of</strong> non-productive times 

• highly sensitive materials, which require a 

more specialized design 

• trend to smaller batch sizes 

• increasing demand for integrated 

additional functions such as simultaneous 

oiling or perforation for fur<strong>the</strong>r processing 

• growing importance <strong>of</strong> resource efficiency, 

here reduction <strong>of</strong> <strong>the</strong> energy consumption 

• high operational safety as strict side 

condition. 

Considerable conflicts <strong>of</strong> objectives result in 

optimization problems which require innovative 

solutions. 

Achenbach Optifoil product line 

Achenbach’s response is a system <strong>of</strong> nine basic 

machine types: As ‘Optifoil’ product line, 

<strong>the</strong>se machines types are <strong>the</strong> basis to fulfil all 

slitting jobs for a wide variety <strong>of</strong> materials: 

tailored to <strong>the</strong> individual customer’s needs, 

highly-productive and in top quality, especially 

with regard to <strong>the</strong> straight-edged finished 

roll rewinding and absolutely clean cuts: 

• Optifoil Varioslit – <strong>the</strong> flexible all-rounder 

for all foil slitting jobs 

• Optifoil Varioslit Plus – <strong>the</strong> high-produc- 

tive all-rounder for all foil slitting jobs 

• Optifoil Fibreslit – <strong>the</strong> specialist for slitting 

tasks in <strong>the</strong> paper and textile industry 

• Optifoil Jumboslit – <strong>the</strong> specialist for all 

slitting jobs <strong>of</strong> jumbo rolls 

Optifoil Jumboslit 

• Optifoil Inspector – <strong>the</strong> specialist for 

surface control 

• Optifoil Lightslit – <strong>the</strong> slitting specialist for 

thin metal foils 

• Optifoil Heavyslit – <strong>the</strong> slitting specialist 

for thick metal foils and large roll diameters 

• Optifoil Separator and Doubler for 

producing even thinnest metal foils. 

The brand name Achenbach generally represents 

<strong>the</strong> following quality features: 

• complete reliability resulting from 

superior quality <strong>of</strong> all components 

• first-class measuring, drive and control 

components 

• energy-saving and emission-minimized 

engineering 

• maximum availability by online fault 

diagnosis as well as service and support 

during <strong>the</strong> whole operational period 

• effective operation and easy maintenance 

due to modular design. 

Success factors in foil slitting technology 

In principal five key factors are setting <strong>the</strong> 

trends for <strong>the</strong> quality <strong>of</strong> slitting technology: 

(1) Optimum slitting quality at highest speeds 

Robust design: It is one precondition and is 

mainly reflected in <strong>the</strong> mechanical design <strong>of</strong> 

<strong>the</strong> single components such as bottom knife 

shaft, bottom knife slitting tools and top knife 

selection as well as <strong>the</strong> definition <strong>of</strong> <strong>the</strong> slitting 

tool diameter; in both cases, a special design is 

inevitable. Regarding <strong>the</strong> design <strong>of</strong> <strong>the</strong> slitting 

tools, a too high moment <strong>of</strong> inertia is to be 

avoided on <strong>the</strong> one hand. On <strong>the</strong> o<strong>the</strong>r hand, 


<strong>the</strong> smooth run <strong>of</strong> <strong>the</strong> slitting unit in continuous 

operation at maximum speeds is to be 

secured. 

Manifold fine adjustment: The fine adjustment 

with a slitting geometry tailored to <strong>the</strong> 

respective material is decisive for <strong>the</strong> excellent 

slitting quality; for this purpose Achenbach 

<strong>of</strong>fers optionally an electronic setting device 

realized by a positioning device. By means <strong>of</strong> 

this or <strong>of</strong> <strong>the</strong> 4-axes-setting, <strong>the</strong> slitting tools 

can be adapted even to <strong>the</strong> most difficult material 

features in <strong>the</strong> setup time <strong>of</strong> <strong>the</strong> machine. 

This manifold fine adjustments are especially 

important for highly-sensitive materials such 

as s<strong>of</strong>t <strong>aluminium</strong> or multi-<strong>layer</strong> composite 

materials. 

Highest precision: Even if highest precision 

with slitting tolerances < 1/100 mm is required, 

<strong>the</strong> use <strong>of</strong> slitting cassettes is advantageous to 

guarantee <strong>the</strong> perfect cut at high speeds. This 

especially goes for slitting tasks with higher 

material thickness. 

(2) Highest reliability at 

maximum machine utilization 

This is guaranteed by <strong>the</strong> robust design <strong>of</strong> 

<strong>the</strong> machine frame as any kind <strong>of</strong> vibration 

is safely avoided. The principle <strong>of</strong> stability is 

also a must-have for purchase parts, i.e. only 

high-class quality drive components, slitting 

and winding tools, etc. are used. The highly 

efficient use <strong>of</strong> <strong>the</strong> installed energy performance 

as well as <strong>the</strong> optimum interaction <strong>of</strong> all 

mechanical, electric and hydraulic modules 

is ano<strong>the</strong>r advantage. The integrative quality 

control according to DIN ISO 9001, which 

continuously checks <strong>the</strong> in-house manufactured 

components as well as purchase parts, 

represents <strong>the</strong> first-class mechanical engineering 

Achenbach is standing for. The customers’ 

benefits are extremely high machine speeds 

and process stability. 

(3) Quick finished roll handling 

Highest process speeds are <strong>the</strong> first factor for 

high productivity, low non-productive times 

<strong>the</strong> second. With an increasing degree <strong>of</strong> technology 

and decreasing batch sizes, <strong>the</strong> aim <strong>of</strong> 

reducing <strong>the</strong> non-productive times becomes 

more and more important. That means to 

facilitate an optimum handling cycle for <strong>the</strong> 

customer. Achenbach <strong>the</strong>refore <strong>of</strong>fers a wide 

product range including roll handling, roll turret 

and roll stripping devices, which simplify 

<strong>the</strong> handling by combining automatic material 

clamping and integrated cross-cutting units. In 

order to quicken <strong>the</strong> finished roll handling 

<strong>the</strong>re are special handling devices for heavy 

weights and for small rolls. Perfectly combined, 

<strong>the</strong> result is a fully-automatic handling cycle 

ALUMINIUM · EAC CONGRESS 2011 7


Optifoil Heavyslit with rotary frame 

for unloading and immediately palletizing with 

a defined schedule. In any case, <strong>the</strong> reduction 

<strong>of</strong> handling times is <strong>the</strong> customer’s benefit. 

(4) Shortest times for material changes 

The customers <strong>of</strong> <strong>the</strong> slitting machine operators 

increasingly demand ‘Just-in-Time Deliveries’. 

Small batch sizes and short reaction times 

are becoming a competitive advantage: To be 

competitive it is not enough to be good – you 

must be flexible. Achenbach has accepted this 

challenge. With <strong>the</strong> tailor-made machines <strong>of</strong> 

<strong>the</strong> Optifoil product line, <strong>the</strong> customers can 

process smaller batches <strong>of</strong> various slitting 

products with highest productivity. While <strong>the</strong> 

slitter has to quickly adapt to various slitting 

widths, <strong>the</strong> <strong>aluminium</strong> foil separators focus 

on different material thicknesses and material 

widths. 

The automatic material threading device is 

one example for modules guaranteeing that,a 

splice table with adhesive tape unwinder is 

ano<strong>the</strong>r one. Fur<strong>the</strong>r examples are: recipe 

storage for various slitting widths and winding 

data or a combined drive technology with 

two motors on one winding shaft. By means 

<strong>of</strong> that, foil materials with different strip thicknesses 

can be separated with continuous winding 

tightness; throughout <strong>the</strong> entire diameter 

range as well as throughout a large strip width 

range. With <strong>the</strong> help <strong>of</strong> this combined drive 

concept, different and highly sensitive materials 

can be run on one single-slitter at continuous 

winding tightness. 

By using slitting cassettes or an automatic 

knife shaft control <strong>the</strong> handling times as well 

shortened. Auxiliary devices, which simplify 

and automate <strong>the</strong> core positioning, finally 

round <strong>of</strong>f <strong>the</strong> range <strong>of</strong> products for shortening 

<strong>the</strong> set-up times. 

(5) Custom-built machinery 

In order to optimally support <strong>the</strong> customers in 

meeting <strong>the</strong>ir special needs, it is important to 

respond to <strong>the</strong> customers’ desires. By <strong>the</strong> way, 

some custom-built components has been later 

taken as basic design: 

• Energy-recovering unwinder (hybrid winding): 

While <strong>the</strong> braking power was, for example, 

formerly dissipated via braking resistors, 

this power is today directly resupplied into 

<strong>the</strong> winding drives and <strong>the</strong>refore <strong>the</strong> energy 

consumption is reduced. 

• X-frames for material fine adjustment: Ano<strong>the</strong>r 

custom-built design is <strong>the</strong> fine adjustment 

<strong>of</strong> <strong>the</strong> material web by a X-frame additionally 

installed in front <strong>of</strong> <strong>the</strong> slitting unit. 

With its quick reaction to material waste, 

as <strong>the</strong> set-up time for material change are Optifoil Heavyslit with automatic slitting gap setting 

such a X-frame supports <strong>the</strong> normal side edge 

control by an absolutely precise material web 

alignment directly at <strong>the</strong> cutting knife. This 

auxiliary device is recommended especially 

for sensitive materials. 

• Oil applicator: The option <strong>of</strong> additionally 

installing an oil applicator is just one example 

for an ‘added value’ in slitting technology. 

During <strong>the</strong> slitting process, <strong>the</strong> material web is 

covered with oil, which is required when <strong>the</strong> 

slitting process is followed by a deep-drawing 

process. 

• Double-face strip inspection: The doubleface 

strip inspection is one fur<strong>the</strong>r custombuilt 

component. This system scans <strong>the</strong> surface 

<strong>of</strong> <strong>the</strong> coil and analyses its data. Thus, faults 

can be specifically and quickly eliminated for 

<strong>the</strong> inspected coil. As a preventative measure, 

<strong>the</strong> production process can be optimized on 

<strong>the</strong> basis <strong>of</strong> <strong>the</strong> inspection data. 

(6) Long-term partnership 

Traditionally, Achenbach considers itself to be 

<strong>the</strong> partner <strong>of</strong> its customers. Being a partner 

means establishing and maintaining long-term 

business relations. A product understanding 

going beyond <strong>the</strong> delivery <strong>of</strong> first-class machinery 

is part <strong>of</strong> this philosophy and comprises 

also <strong>the</strong> commitment for maximum 

availability throughout <strong>the</strong> whole operational 

period. This is guaranteed by <strong>the</strong> Achenbach 

Service & Support for maintaining <strong>the</strong> machine 

performance: as immediate measures in 

case <strong>of</strong> breakdown, as preventative measures 

or as optimization measures; trusted support 

staff is always available. 

Rolling mills and foil slitting machines are 

<strong>the</strong> core parts in our customers’ production. 

Their quality is a decisive factor for our customers’ 

competitiveness in <strong>the</strong>ir own markets. 

For many years, <strong>the</strong> Achenbach brand mark 

has already been known for its non-ferrous 

rolling mills all over <strong>the</strong> world. This is soon 

to follow for <strong>the</strong> Optifoil foil slitting machines 

– for <strong>the</strong> customers’ benefit. 

� 

8 ALUMINIUM · EAC CONGRESS 2011

�� 

�� 

�� 

� �� 

� �� 

� �� 

� �� 

� �� 

�� 

�� 

�� 

�� 

�� 

�� 

� �� 

� �� 

� �� 

� �� 

� ��


Focus on rolling mill efficiency: design and control 

Rainer Neukant, Achenbach Buschhütten GmbH 

Long time, <strong>the</strong> challenge in rolling mill 

building was to continuously increase <strong>the</strong> 

productivity on <strong>the</strong> one hand and on <strong>the</strong> 

o<strong>the</strong>r hand to improve <strong>the</strong> tolerances <strong>of</strong> 

all kind <strong>of</strong> flat rolled products. Today, 

setting <strong>the</strong> course with innovative solutions 

means more and more focusing on 

rolling mill efficiency, with respect to 

design and control. In this context it is <strong>of</strong> 

high importance to particularly aim <strong>the</strong> 

development work at both, machinery 

efficiency and resources efficiency. The 

respective fields <strong>of</strong> innovations are manifold: 

simulation, mechatronics, process 

models, use <strong>of</strong> resources and not least 

service and support, to mention only a 

few. All <strong>of</strong> <strong>the</strong>se innovative search fields 

are to be considered in an integrative approach, 

ensuring at <strong>the</strong> same time high 

productivity, best quality and <strong>the</strong> defined 

production flexibility. 

1 Machinery efficiency 

1.1 Computational Fluid Dynamics (CDF) 

In a rolling mill enormous quantities <strong>of</strong> media 

(hydraulic, cooling and lubrication fluids and 

exhaust air) must be handled. The fume hood 

as an example illustrates that it is possible to 

be innovative even with regard to a relatively 

simple part <strong>of</strong> machinery. A fume hood is 

needed to collect <strong>the</strong> exhaust air loaded with 

gaseous rolling oil. There are several reasons 

to design this part <strong>of</strong> <strong>the</strong> rolling mill exhaust air 

purification system as efficient as possible: 

• guaranteeing best conditions for <strong>the</strong> 

operators working closely to <strong>the</strong> rolling 

mill and to fulfil environmental, health 

and safety requirements 

• maximum recovery <strong>of</strong> rolling oil from <strong>the</strong> 

waste air to comply with economic targets. 

The classic fume hood design is based on both, 

experience values and relatively simple geometric 

aspects. Today, competitive simulation 

programs are available allowing more detailed 

views into fluid mechanics <strong>of</strong> <strong>the</strong>se exhaust air 

purification systems. The knowledge gained 

from <strong>the</strong> CFD simulations results in fluid-optimized 

design features to increase <strong>the</strong> exhaust 

efficiency and at <strong>the</strong> same time to reduce <strong>the</strong> 

energy consumption. 

1.2 Chemical CAD 

The application <strong>of</strong> physical / chemical simulation 

s<strong>of</strong>tware is a fur<strong>the</strong>r example for advanced 

engineering methods to improve <strong>the</strong> 

efficiency <strong>of</strong> rolling mills. Particularly for <strong>the</strong> 

field <strong>of</strong> exhaust air purification systems a 

physical and chemical process simulation tool 

forms <strong>the</strong> basis for <strong>the</strong> optimized design <strong>of</strong> today’s 

‘Airpure’ systems. 

With <strong>the</strong> help <strong>of</strong> <strong>the</strong>se CAD tools, new 

Airpure systems can be perfectly designed 

while existing systems can be modified for 

better performance on <strong>the</strong> shortest term. By 

means <strong>of</strong> <strong>the</strong> simulation s<strong>of</strong>tware, <strong>the</strong> effects 

<strong>of</strong> rolling oil components and additives on <strong>the</strong> 

exhaust air can be previously tested and recommendations 

to improve a time-consuming 

and expensive exhaust air purification can be 

made at <strong>the</strong> earliest stage. 

1.3 Mechatronics 

Setting <strong>the</strong> course for fur<strong>the</strong>r developments 

to <strong>the</strong> customers’ benefit, an integrative approach 

is required considering at <strong>the</strong> same 

time mechanical as well as electronic aspects 

<strong>of</strong> engineering. The ‘Win-SprayS’ rolling oil 

distribution system is just one example for it 

reflected by <strong>the</strong> following solutions: 

• Patented valve design: best reliability is 

achieved by eliminating pneumatic components, 

i.e. <strong>the</strong> valve is just electrically actuated. 

All moving parts that are subject to wear and 

tear are integrated in a cartridge that can be 

simply replaced from <strong>the</strong> front. Before introduction 

into <strong>the</strong> market, <strong>the</strong> functionality <strong>of</strong> 

this valve series was comprehensively verified 

in test runs <strong>of</strong> 10,000,000 switching cycles. 

• Different valve sizes: two sizes <strong>of</strong> spraying 

valves are available to cover <strong>the</strong> range <strong>of</strong> rolling 

oil spraying systems from cold strip to foil 

rolling mills. The smaller valve is optimized 

for narrow 26 mm spacing in <strong>the</strong> spraying bar 

and applies 44 l/min, while <strong>the</strong> big valve is suitable 

for 52 mm spacing applying 150 l/min 

• Valves meeting <strong>the</strong> control system requirements: 

with <strong>the</strong> aid <strong>of</strong> ‘Matlab’ models <strong>the</strong> 

spraying geometry is designed for a uniform 

cooling effect. The correctness <strong>of</strong> <strong>the</strong>se design 

models is verified measuring <strong>the</strong> oil distribution 

for individual nozzles as well as measuring 

<strong>the</strong> heat transfer. These measurements 

are done on a test stand with rolling oil. Both 


Images: Achenbach

Pulse and Volume (Level) Control methods 

are supported. 

• Advanced design tools: toge<strong>the</strong>r with careful 

tests in <strong>the</strong> laboratory, today a verification 

and optimization <strong>of</strong> existing installations in 

<strong>the</strong> rolling mill is possible. As an example, a 

pressure-sensitive film is used to visualize <strong>the</strong> 

footprint <strong>of</strong> <strong>the</strong> spraying system. This is mainly 

done to identify worn nozzles and to ei<strong>the</strong>r 

confirm or optimize spraying angles for best 

overlapping. 

1.4 Process automation 

In cold rolling mills, particularly in foil rolling 

mills, best results have been achieved 

with powerful closed-loop control systems for 

many years. Such systems have been steadily 

improved with respect to better measuring 

systems (e.g. <strong>the</strong> ‘Optiroll’ flatness measuring 

roll), better actuators (e.g. <strong>the</strong> Win-SprayS 

rolling oil distribution) and faster control systems. 

In that way product quality (e.g. thickness 

and flatness) as well as productivity (by 

higher rolling speeds) could be maximized. 

Fur<strong>the</strong>r optimization is now focused on 

meeting <strong>the</strong> high quality requirements already 

at <strong>the</strong> strip head. This requires a precise presetting 

<strong>of</strong> <strong>the</strong> mill including for example an 

optimum bending force to meet <strong>the</strong> flatness 

tolerances even before <strong>the</strong> feedback controller 

is able to compensate for a wrong flatness. 

This requires control methods, which are not 

based on measuring <strong>the</strong> parameters to be 

controlled. Here process models are necessary, 

which are able to approximate <strong>the</strong> rolling 

result. For a long time, such future-orientated 

physical models were available as <strong>of</strong>f- 

line models for rolling mill design and layout. 

In recent years, online-model applications 

have been developed. 

The integrative Optiroll automation system 

comprises both, online process models and 

model-based functions available throughout 

<strong>the</strong> different levels <strong>of</strong> automation hierarchy. 

The Roll Speed Compensation (RSC) is 

one example for a model-based function that 

is integrated in <strong>the</strong> Level 1 control system for 

highest dynamic behaviour, as it compensates 

for speed-related tribologic effects in cold 

rolling. With RSC, much closer thickness tolerances 

are achieved compared to <strong>the</strong> classic 

closed loop control which is depending on <strong>the</strong> 

slow feedback <strong>of</strong> measured thickness. Without 

RSC, <strong>the</strong> changing tribologic conditions in <strong>the</strong> 

roll gap would result in significant thickness 

deviations during acceleration <strong>of</strong> <strong>the</strong> rolling 

mill. 

Beyond integrated model-based functions 

<strong>the</strong> Optiroll automation system provides 

trendsetting solutions for dedicated Level 2 

process models to comprehensively optimize 

<strong>the</strong> rolling process. This is particularly required 

for both, presetting <strong>the</strong> closed loop controller 

(Level 1) and those controlling parameters, 

which cannot be measured during <strong>the</strong> rolling 

process. With regard to Level 2 process models, 

<strong>the</strong>re are two different approaches: 

• Before rolling, roll force and roll stack 

deflection are calculated in order to achieve 

minimum thickness deviation at <strong>the</strong> strip head 

and to reach target pr<strong>of</strong>ile /flatness as quick as 

possible in order to avoid scrap material. For 

this, <strong>the</strong> calculation <strong>of</strong> <strong>the</strong> roll force on Level 

2 provides two options: Tribologic models for 

<strong>the</strong> roll gap friction with long-term and shortterm 

adaption, as far as cold rolling is concerned. 

When it comes to hot rolling, adaption 

<strong>of</strong> flow stress strain curves is applied. 


• During rolling, two different types <strong>of</strong> modeling 

<strong>the</strong> rolling process are designed: 

On <strong>the</strong> one hand, <strong>the</strong> calculation <strong>of</strong> <strong>the</strong>rmal 

expansion <strong>of</strong> <strong>the</strong> work roll to correct <strong>the</strong> rolling 

force and <strong>the</strong> mechanical bending in order 

to avoid pr<strong>of</strong>ile and / or flatness errors and on 

<strong>the</strong> o<strong>the</strong>r hand, calculation <strong>of</strong> strip temperatures 

in order to keep <strong>the</strong> material properties 

in a certain range. 

While all o<strong>the</strong>r functions are focusing on 

precise pre-setting, <strong>the</strong> <strong>the</strong>rmal models as well 

are periodically triggered during rolling, so that 

online corrections with respect to pr<strong>of</strong>ile and 

flatness or with respect to <strong>the</strong> strip temperature 

are possible. Meeting <strong>the</strong> strip temperature 

during all passes is not only required for 

<strong>the</strong> metallurgical quality <strong>of</strong> hot-rolled aircraft 

alloys, but also for <strong>the</strong> exact finishing temperature 

<strong>of</strong> cold-rolled canstock qualities. 

For best performance, <strong>the</strong> integration <strong>of</strong> 

Level 1 and Level 2 functions is fundamentally 

important. Due to <strong>the</strong> modular design, <strong>the</strong> 

Optiroll automation system is predestined for 

modernizing existing rolling mills. Above that, 

especially <strong>the</strong> Level 2 options can be customized 

to fulfil <strong>the</strong> individual requirements for 

improving <strong>the</strong> rolling mill performance. 

2 Resources efficiency 

2.1 Electrical power 

Electrical power and gas are <strong>the</strong> major forms 

<strong>of</strong> energy that are to be considered for <strong>the</strong> efficiency 

<strong>of</strong> a rolled-products production plant. 

Gas is required for heating processes. For <strong>the</strong> 

rolling mill, however, electrical energy is <strong>the</strong> 

most important factor. 

The first approach to achieve an energy-efficient 

rolling mill is <strong>the</strong> identification <strong>of</strong> <strong>the</strong> 

major energy consumers. Electrical drives are 



<strong>the</strong> essential consumers <strong>of</strong> electrical power in 

a rolling mill. 60 to 80% <strong>of</strong> <strong>the</strong> electrical power 

installed in a foil rolling mill are required 

for <strong>the</strong> main drives, i.e. <strong>the</strong> mill motor and 

<strong>the</strong> coiler drives. Accordingly, top priority to 

achieve best energy efficiency is given to <strong>the</strong> 

design <strong>of</strong> drives, motors and control systems 

main drives <strong>of</strong> <strong>the</strong> rolling mill: 

• In cold rolling mills, decoiler and recoiler 

drives are connected to a common DC link, 

which is fed by <strong>the</strong> rectifier unit. So <strong>the</strong> generated 

power <strong>of</strong> <strong>the</strong> de-coiler, which is applying 

<strong>the</strong> tension to <strong>the</strong> mill, is kept in <strong>the</strong> system 

feeding <strong>the</strong> mill motor and re-coiler motor, 

which are operated in motoric mode. 

• Modern drive systems are additionally 

equipped with an active feeder, which (contrary 

to a simple rectifier) is able to feed back 

electrical energy into <strong>the</strong> mains supply. Usually, 

this is possible whenever <strong>the</strong> rolling mill 

is stopped and <strong>the</strong> rotational energy <strong>of</strong> <strong>the</strong> 

coils and rolls can be recovered by inverters 

that are working in generator mode. 

• Saving electrical energy is moreover possible 

with efficient motors that consume fur<strong>the</strong>r 

1 to 2% less electrical energy compared to 

conventional motors. Considering that 99% <strong>of</strong> 

<strong>the</strong> motor cost during its lifetime arises from 

energy cost, also such small effects are a valuable 

contribution to an efficient rolling mill. 

• Besides <strong>the</strong> installation <strong>of</strong> efficient components 

also an intelligent automation and control 

system is a decisive factor for an efficient 

rolling mill. Frequency-controlled pumps and 

fans are good examples for such intelligent 

and energy-saving control systems. Frequency 

control allows always and continuously adjusting 

<strong>the</strong> power as per <strong>the</strong> actual process 

requirements. An ‘Airpure’ fan for example 

can be reduced in its rotations per minute 

(and accordingly power consumption) when 

less air must be moved in case not all rolling 

mills <strong>of</strong> a rolling mill line are in operation. 

The same applies for pumps <strong>of</strong> a rolling oil 

circulation system. Compared to a bypass-controlled 

volume flow, <strong>the</strong> frequency-controlled 

pump is able to save up to 50% <strong>of</strong> electrical 

energy with a ‘Superstack’ rolling oil filtration 

system. 

Intelligent flow-rate control and efficient 

drives for pumps and fans are good examples 

for saving electrical energy with shortest return 

on investment. 

2.2 Rolling oil filtration 

Saving electrical energy for fluid circulation is 

not <strong>the</strong> only criterion for efficiency <strong>of</strong> <strong>the</strong> rolling 

oil purification system. The second aspect 

is maintaining <strong>the</strong> good quality <strong>of</strong> rolling oil 

by perfect filtration. In that way, <strong>the</strong> lifetime 

<strong>of</strong> rolling oil is maximized and fresh rolling oil 

must be added to <strong>the</strong> system less frequently. 

In o<strong>the</strong>r words: <strong>the</strong> SuperstackII micr<strong>of</strong>iltration 

system preserves a valuable resource – 

mineral-oil based rolling oil. 

A third aspect <strong>of</strong> efficiency is <strong>the</strong> continuously 

controlled dosage <strong>of</strong> filter aid. Depending 

on <strong>the</strong> differential pressure filter aid is 

added, preventing <strong>the</strong> filter cake from getting 

clogged. By means <strong>of</strong> this, <strong>the</strong> filtration cycle is 

extended compared to an uncontrolled forming 

<strong>of</strong> <strong>the</strong> filter cake. 

Final aspects <strong>of</strong> maintaining <strong>the</strong> rolling oil 

quality are in-house developed Achenbach 

recipes for filtration aids and additives. These 

recipes are designed specifically to <strong>the</strong> customer’s 

rolling processes. 

The result <strong>of</strong> <strong>the</strong>se features is perfectly 

clean rolling oil with extended filtration cycles, 

preserving valuable resources. 

2.3 Savings and recovery 

For <strong>the</strong> <strong>aluminium</strong> cold rolling process a huge 

amount <strong>of</strong> rolling oil is circulated for cooling 

and lubricating. As this oil is applied to <strong>the</strong> 

hot work rolls through sprays, a significant 

amount <strong>of</strong> rolling oil is taken by <strong>the</strong> exhaust 

air <strong>of</strong> <strong>the</strong> rolling mill in form <strong>of</strong> fume and mist. 

Recovering that oil from <strong>the</strong> exhaust air is not 

only recovering a valuable resource, but also 

recovering energy in form <strong>of</strong> mineral oil. That 

oil is fed back into <strong>the</strong> process without any 

restrictions. 

Today’s high degree <strong>of</strong> rolling oil recovery 

is based on <strong>the</strong> advanced design feasible by 

physical/chemical engineering. Important features 

<strong>of</strong> this optimization are: 

• recuperators for higher heat recovery 

• improved condensers for low-temperature 

distillation 

• reduced pressure losses in air ducts. 

Optimized fume hoods and speed-controlled 

fans are fur<strong>the</strong>r criteria <strong>of</strong> <strong>the</strong> highly efficient 

Airpure system. The maximized recovery <strong>of</strong> 

rolling oil and <strong>the</strong> minimum energy consumption 

<strong>of</strong> <strong>the</strong> Airpure exhaust air purification is 

not only remarkable with respect to energy 

and resources efficiency, but also for reasons 

<strong>of</strong> environmental protection: A rolling mill 

with an Airpure system is surpassing <strong>the</strong> strict 

European restrictions for VOC emissions. 

The efficiency <strong>of</strong> <strong>the</strong> Airpure system in recovering 

resources becomes visible comparing 

<strong>the</strong> amount <strong>of</strong> rolling oil which is recovered 

from <strong>the</strong> exhaust air with <strong>the</strong> electrical energy 

that is generated by a wind turbine: <strong>the</strong> typical 

electrical energy generated by one state<strong>of</strong>-<strong>the</strong>-art 

wind turbine for on-shore installation 

is comparable to <strong>the</strong> energy (mineral oil) 

recovered from one foil rolling mill. 

Saving energy and resources and saving costs 

here are in accordance with sustainability and 

ecology <strong>of</strong> a modern <strong>aluminium</strong> cold rolling 

mill. These benefits are multiplied in rolling 

mill lines with three or even more rolling 

mills. 

3 Service and support 

A crucial aspect for any pr<strong>of</strong>essional service 

and support package for rolling mill machinery 

is an integrative modular concept that 

aims at maintaining and even increasing <strong>the</strong> 

machinery performance by taking <strong>the</strong> following 

measures: 

• Restoring by immediate measures in <strong>the</strong> 

case <strong>of</strong> breakdown 

• Preserving by remote and on-site 

maintenance 

• Improving by optimization measures. 

Optimization measures in <strong>the</strong> sense <strong>of</strong> Achenbach 

Service and Support comprise: 

• inspection with ‘Quick Wins’ 

• analysis <strong>of</strong> <strong>the</strong> process data 

• options for system modernization 

• sustainability measures including tele- 

service. 

Offering three support packages – Optiroll 

check, resources and efficiency check, and 

productivity check – Achenbach act as a pr<strong>of</strong>essional 

partner to guarantee high efficiency 

and availability <strong>of</strong> <strong>the</strong>ir rolling mills all over 

<strong>the</strong> world. In order to support <strong>the</strong>ir customers 

in maintaining competitive advantages in 

<strong>the</strong>ir markets, Achenbach developed a concept 

to ascertain fields with potential for systems’ 

optimization and / or conversion respectively 

modernization by installing particular new 

individual components. In any case, this concept 

<strong>of</strong> supporting <strong>the</strong>ir customers include special 

training units. 

In concrete terms and in view <strong>of</strong> <strong>the</strong> important 

issue <strong>of</strong> resources efficiency, this can 

mean: 

• new vacuum pumps for better perform- 

ance <strong>of</strong> <strong>the</strong> Airpure system 

• new drives for higher speed and thus 

improved productivity 

• new automation systems for better 

product quality 

• new components for continuous availabil- 

ity and maintainability. 

All <strong>the</strong> aforementioned facts classify Achenbach 

as leaders in technology and quality 

delivering everything from a single source 

– mechanical engineering, machinery, and 

service and support – always keeping one eye 

on rolling mill efficiency in both, installation 

<strong>of</strong> new machinery, modernization and supporting 

high performance in <strong>the</strong> long run. � 


Batch-type and continuous floater 

furnace facilities for <strong>aluminium</strong> alloy strip 

Carl-August Preimesberger, Ebner Industrie<strong>of</strong>enbau GmbH 

The production <strong>of</strong> <strong>aluminium</strong> and <strong>aluminium</strong> 

alloy strips – starting at <strong>the</strong> hot 

rolling mill or a continuous casting plant 

through to sellable strips for <strong>the</strong> widest 

range <strong>of</strong> applications – involves <strong>the</strong> material 

passing through several stages <strong>of</strong> heat 

treatment. Depending on <strong>the</strong> starting 

material and <strong>the</strong> technological specifications 

<strong>of</strong> <strong>the</strong> final product, <strong>the</strong> heat treatment 

processes may involve annealing, 

homogenizing, solution heat treatment, 

recrystallizing, aging or temper annealing. 

Basically, all <strong>the</strong> heat treatment processes 

can be performed during continuous annealing 

with a floater furnace. However, 

cost effective operation <strong>of</strong> such a facility 

is only possible if it used predominately 

to solution heat-treat strips. On <strong>the</strong> o<strong>the</strong>r 

hand, it is not technically possible to 

solution heat-treat strips in a batch-type 

furnace. From an economical point <strong>of</strong> 

view, this kind <strong>of</strong> furnace is ideal for <strong>the</strong> 

remaining heat treatment processes. 

The Ebner product range includes proven 

designs for both types <strong>of</strong> furnace 

The Hicon floater furnace manufactured by 

Ebner Industrie<strong>of</strong>enbau in Leonding, Austria, 

features <strong>the</strong>ir proven design concept for <strong>the</strong> 

continuous heat treatment <strong>of</strong> cold-rolled <strong>aluminium</strong> 

strip, dramatically improving <strong>the</strong> performance 

and quality achieved in conventional 

continuous furnaces. Floater furnace facilities 

have been built to process strip with a thickness 

between 0.3 and 6.35 mm and a width up 

to 2,400 mm. Heating up times <strong>of</strong> 1 min/mm 

have been achieved at setpoint temperatures 

<strong>of</strong> 560°C and a Δt <strong>of</strong> +0/-3°C. 

The combination <strong>of</strong> <strong>the</strong> most precise temperature 

uniformity and a perfect strip surface 

finish specified by end users in <strong>the</strong> automotive 

and aerospace industries prompted Ebner to 

develop <strong>the</strong> concept <strong>of</strong> a floater furnace in 

cooperation with renowned manufacturers <strong>of</strong> 

<strong>aluminium</strong> products. It is possible to achieve 

<strong>the</strong> required surface properties because <strong>the</strong> 

<strong>aluminium</strong> strip is transported totally contactfree 

through <strong>the</strong> furnace facility during <strong>the</strong> 

entire heating up and cooling process. The 

properties apply to <strong>the</strong> deep drawing characteristics 

and corrosion resistance. 

The Ebner research and development department 

advanced nozzle systems which 

Photos: Ebner 

were <strong>the</strong>n continually improved in a test facility. 

Powerful Hicon high convection recirculation 

technology provides <strong>the</strong> support and 

stability required for both thick and thin <strong>aluminium</strong> 

strip, while achieving <strong>the</strong> best possible 

heat transfer rates. Each furnace zone has two 

opposing fan units controlled by frequency 

converters. These atmosphere recirculation 

fans were also developed by Ebner. The highperformance 

impellers, mounted directly on 

<strong>the</strong> fan motor shaft, have undergone extensive 

long-term testing in order to optimize <strong>the</strong>m 

for long service life. Modern automated welding 

technology is used to produce impellers <strong>of</strong> 

optimal and reproducible quality to operate 

under extreme <strong>the</strong>rmo-mechanical stress. 

The specially developed nozzle system 

delivers a uniform transfer <strong>of</strong> <strong>the</strong>rmal energy 

across <strong>the</strong> entire width <strong>of</strong> <strong>the</strong> strip during <strong>the</strong> 

heating up phase. This helps to even improve 

<strong>the</strong> temperature uniformity throughout <strong>the</strong> 

soaking phase, in order to meet <strong>the</strong> required 

technological specifications. This furnace section 

fulfils <strong>the</strong> strict AMS 2750 D aerospace 

standard, applicable worldwide, specifying 

that a furnace used for supplying plate to <strong>the</strong> 

aircraft industry (Furnace Class 1) must maintain 

a maximum deviation <strong>of</strong> ±3K from <strong>the</strong> 

temperature setpoint in <strong>the</strong> entire workload 

space during <strong>the</strong> soaking time. In most cases 

<strong>the</strong> final deviation form <strong>the</strong> setpoint is down 

to around 1.5°C. 

Although <strong>the</strong> Hicon floater furnaces are 

made up <strong>of</strong> individual furnace zones, a distinction 

is made between heating-up and soaking 

zones. The direct gas-fired heating-up zones 

Continuous floater furnace facility 


<strong>of</strong> <strong>the</strong> furnace are each equipped with four 

low-Nox, tow-stage, all-metall burners. The 

combustion air is preheated in a recuperator 

to approx. 400°C. The soaking zones are fitted 

with two burners which are turndown controlled 

continuously. This system, which has been 

tried and proven in many facilities, makes 

it possible to keep well below <strong>the</strong> threshold 

specified in <strong>the</strong> stringent TA-Luft standard. 

Once <strong>the</strong> strip leaves <strong>the</strong> furnace section, it 

passes through a combined air/water quench. 

This quench, developed by Ebner, makes 

possible to implement a special quenching 

program designed for each individual alloy, 

in order to achieve optimal material properties 

at minimal distortion <strong>of</strong> <strong>the</strong> strip. Typical 

cooling gradients specified for aerospace al- 

loy strip exceed 300°C/s. The automation 

systems s<strong>of</strong>tware developed by Ebner, automatically 

selects from a database <strong>the</strong> correct 

quenching program suitable for <strong>the</strong> alloy and 

<strong>the</strong> dimensions <strong>of</strong> <strong>the</strong> strip furnace facility. 

If quenching need not be that rapid, cooling 

can also be effected by air cooling only. The 

water cooling system is <strong>the</strong>n switched <strong>of</strong>f and 

<strong>the</strong> charge is merely cooled by air in <strong>the</strong> highperformance 

air cooling section. By virtue <strong>of</strong> 

this high degree <strong>of</strong> flexibility, Hicon floater 

furnace facilities fulfil all requirements placed 

on heat treatment and quenching <strong>of</strong> a wide 

variety <strong>of</strong> alloys. 

The Ebner automation system is based on 

a Simatic Step7 system with object-oriented 

programming incorporating WINCC and HMI 

Faceplates. These faceplates combine operation 

and supervision <strong>of</strong> <strong>the</strong> facility with <strong>the</strong> 

visualization <strong>of</strong> each 

device incorporated in 

that facility, while at 

<strong>the</strong> same time, keeping 

an online maintenance 

log book. 

Apart form this, an 

operating hours counter 

is provided for each 

peripheral device. The 

integrated diagnostics 

system simplifies maintenance 

and servicing 

<strong>of</strong> <strong>the</strong> facility, keeping 

downtimes to a minimum. 

Hardware and 

s<strong>of</strong>tware documentation 

with comprehen- 



Gas-fired radiant tube heated batch-type furnace facility 

sive annotations provides <strong>the</strong> user with excellent 

support for carrying out maintenance. 

By means <strong>of</strong> remote maintenance via modem 

Ebner specialists can provide support whenever 

needed. Apart from <strong>the</strong> process control 

system terminal, touch pads connected via 

WLAN are provided, for facilitating remote 

adjustments and service work for <strong>the</strong> entire 

facility. 

In addition to <strong>the</strong> technical and technological 

advantages, <strong>the</strong> durable design <strong>of</strong> <strong>the</strong> key 

components and service-friendly implementation 

<strong>of</strong> <strong>the</strong> facility has led to well-known 

companies investing in Hicon floater furnaces. 

Apart from floater furnace facilities that have 

already been handed over to AMAG in Austria, 

Aleris Aluminium in Duffel, Belgium, 

and Southwest Aluminium in China, <strong>the</strong> most 

recently-placed orders have also been received 

by Ebner. 

For more than 40 years Ebner has been 

building batch-type furnaces for a wide range 

<strong>of</strong> annealing processes in air or protective atmosphere. 

This kind <strong>of</strong> batch-type furnace can 

process coils with an outer diameter <strong>of</strong> up to 

2,650 mm, coil width <strong>of</strong> up to 2,800 mm and 

a net coil weight <strong>of</strong> 25 tonnes, accommodating 

up to 12 coils in each furnace with <strong>the</strong> charge 

configured in two lanes. 

The key requirements for a modern batchtype 

furnace are: 

• high rate <strong>of</strong> heat transfer 

• uniform temperature distribution 

• low specific energy consumption 

• low maintenance costs 

• high degree <strong>of</strong> operational safety 

• low environmental impact. 

The furnace atmosphere is recirculated using 

speed-controlled direct drive fan units. 

The process atmosphere is drawn in by fans 

mounted in <strong>the</strong> ro<strong>of</strong> <strong>of</strong> <strong>the</strong> furnace and <strong>the</strong>n 

passed through a baffle system over radiant 

tubes suspended in left and right-hand side 

walls <strong>of</strong> <strong>the</strong> furnace. The temperature inside 

<strong>the</strong> furnace is controlled by <strong>the</strong>rmocouples located 

in <strong>the</strong> baffle system in <strong>the</strong> furnace ro<strong>of</strong> 

as well as <strong>the</strong>rmocouples mounted in <strong>the</strong> side 

walls ahead <strong>of</strong> and beyond <strong>the</strong> charge. 

The latest design <strong>of</strong> single-lane batch-type 

furnaces feature a jet-flow system for directing 

<strong>the</strong> furnace wind at <strong>the</strong> coils. Ebner batch-type 

furnaces with dual-lane charge configurations 

use a combined jet/mass flow system. The 

advantage is that <strong>the</strong> temperature difference 

between <strong>the</strong> gas flowing out <strong>of</strong> <strong>the</strong> nozzles 

and <strong>the</strong> gas flowing past <strong>the</strong> centerline <strong>of</strong> <strong>the</strong> 

furnace is negligible so that all <strong>the</strong> coils in <strong>the</strong> 

furnace are heated through more evenly. Ano<strong>the</strong>r 

advantage is that <strong>the</strong> coils in <strong>the</strong> furnace 

can always be charged in <strong>the</strong> same place. This 

means that an application-specific jet nozzle 

system can be used to target each coil. If each 

coil can be assigned to a separate furnace zone, 

<strong>the</strong>n it is also possible to trim back <strong>the</strong> overshoot 

temperature for each coil individually. 

Hicon singlecoil 

overhead 

furnaces are a 

special type <strong>of</strong> 

batch furnace. 

The special 

feature <strong>of</strong> this 

kind <strong>of</strong> singlecoil 

overhead 

furnace facility 

is that each 

individual furnace 

chamber 

is supported on 

a steel structure 

and each 

furnace cham- 

Gas-fired radiant tube single-coil overhead furnace facility 

ber holds just one coil. The coil is placed on 

a charging frame integrated into <strong>the</strong> furnace 

plug, which is <strong>the</strong>n raised into <strong>the</strong> furnace 

chamber. 

The individual chambers are served by one 

charger that transports <strong>the</strong> furnace plugs plus 

coils between <strong>the</strong> charging area and <strong>the</strong> overhead 

furnace units. 

The charging area alongside <strong>the</strong> facility is 

used to charge and remove <strong>the</strong> coils. By defining 

a precise position for <strong>the</strong> charging area, it 

can be entered into a coordinates system to be 

served automatically by <strong>the</strong> overhead crane. 

Because even charging and discharging can be 

performed automatically with <strong>the</strong> charger, it 

is possible to implement <strong>the</strong>se furnaces as 

fully-automated systems. 

A Hicon single-coil overhead furnace facility 

delivers <strong>the</strong> shortest possible heating up 

time with a temperature difference as low as 

5°C throughout <strong>the</strong> whole coil at <strong>the</strong> end <strong>of</strong> 

heating. 

In addition to <strong>the</strong> high flexibility and 

throughput <strong>of</strong> <strong>the</strong>se facilities, <strong>the</strong> main advantages 

include <strong>the</strong> high as-annealed quality <strong>of</strong> 

<strong>the</strong> material since <strong>the</strong> anneal process can be 

matched to a specific coil. These facilities are 

also equipped with an annealing time calculation 

program featuring an <strong>of</strong>fline model that 

pre-calculates processing times in advance <strong>of</strong> 

<strong>the</strong> anneal and an online model that supervises 

<strong>the</strong> anneal process in real time and optimizes 

it if necessary. This calculation model is also 

available for standard multi-coil batch-type 

furnace facilities. 

Ebner has supplied more than 120 batchtype 

furnaces to numerous customers worldwide, 

each designed specifically for one or 

more <strong>of</strong> a wide range <strong>of</strong> applications. Among 

<strong>the</strong>se customers are <strong>the</strong> world‘s largest and 

most well-known producers <strong>of</strong> <strong>aluminium</strong> and 

<strong>aluminium</strong> alloy strip. 

� 



Addressing market requirements in <strong>aluminium</strong> flat products 

Sean Carter and Detlef Neumann, Danieli Fröhling 

In order to succeed in today’s very competitive 

<strong>aluminium</strong> plate, strip / sheet and 

foil markets, it is not sufficient to only 

fulfil <strong>the</strong> standard basic requirements 

when developing plant concepts for both 

rolling mills and downstream cutting lines 

but to define and deliver enhanced solutions 

for <strong>the</strong> future. Particularly, when 

considering developments in <strong>the</strong> processing 

industry with demands for increased 

production levels, larger coil densities, 

thinner finished products and increased 

material quality <strong>the</strong>se factors have to 

be taken into consideration. Danieli is a 

100% metals focused company with over 

8,700 employees, with in-house manufacturing 

in Italy, Germany, Thailand 

and China; along with its process partner 

Innoval Technology, Danieli supplies <strong>aluminium</strong> 

rolling and processing solutions 

as well as plate stretchers and turnkey 

(including FSTK) projects. 

General market consideration 

Generally, <strong>the</strong> world-wide market for technology 

to produce downstream <strong>aluminium</strong> products, 

for example rolled products, is quite different 

when looking for <strong>the</strong> expected investments 

in <strong>the</strong> near to mid future. Europe’s and 

North America’s strategy will be to maintain 

and build upon its leading position <strong>of</strong> efficien- 

cy, quality and service but without significant 

expansion into new capacities while realising 

maximum pay-back on <strong>the</strong> existing assets. 

The well-established production companies 

will concentrate on: 

• Process and yield increase 

• Relocation <strong>of</strong> facilities and /or consolidate 

production capacities at cen tral points 

• Investing in modernisations and revamps 

to maximise current potential as well as 

increasing coil density, processing speeds 

and product quality 

• Improving finished strip quality 

• Concentrating on high-end products such 

as foil, lithography, etc. 

• Customer service (reduced lead times, 

just-in-time production) 

Danieli Wean United Hot Finishing Mill Photos: Danieli Fröhling 

• Development <strong>of</strong> new niche high value 

products such as clad material, sub 6 μm 

foil, bright finished stock, etc. 

• Development into new uses for <strong>aluminium</strong> 

rolled products. 

These measures will be increasingly important 

when considering <strong>the</strong> competition that comes 

from <strong>the</strong> large new investments and production 

capacity currently being implemented in 

Asia and also partly in South America and 

Russia. In <strong>the</strong> near future Asia will emerge as 

<strong>the</strong> dominant market for <strong>aluminium</strong> flat rolled 

products in <strong>the</strong> world and Asian producers 

will try to supply into traditional markets that 

are still occupied by European and American 

suppliers. Here, both quality and quantity as 

well as supply lead time will be <strong>the</strong> key issues 

that purchasers <strong>of</strong> rolled materials will 

look for. 

An example <strong>of</strong> <strong>the</strong> development <strong>of</strong> <strong>aluminium</strong> 

products is in aerospace industry that 

is constantly pushing <strong>the</strong> design barriers with 

increasingly complex metallic or hybrid aircraft 

fuselage skin and structures. This includes 

<strong>the</strong> development <strong>of</strong> new alloys, for example 

damage tolerant Al-Cu-Li fuselage sheet and 

high strength alloys containing Scandium (Sc), 

Aluminium-Glare composite material and 

ever larger very heavy gauge plates for <strong>the</strong> 

manufacture <strong>of</strong> aircraft spars and ribs. 

With <strong>the</strong> increasing word-wide demand for 

automotive, aerospace, and canstock products 

this paper focuses on <strong>the</strong> latest developments 

in cold rolling technology and strip processing. 

Modern <strong>aluminium</strong> rolling 

technology and market trends 

In general, <strong>the</strong> <strong>aluminium</strong> rolling market can 

be considered in three broad sectors relating 

to <strong>the</strong> flat finished products and <strong>the</strong> associated 

equipment to produce <strong>the</strong> rolled material: 

• Hot rolling – plates, shate and coiled 

feedstock for cold and foil rolling 

• Cold rolling – sheet for automotive, aero- 

space, canstock, lithographic, foil feed- 

stock and general applications 

• Foil rolling – foil for food, cosmetics, 

tobacco and pharmaceutical packaging; 

technical applications such as heat 

exchangers and cable wrap; household 

applications. 

Danieli has expertise in all <strong>of</strong> <strong>the</strong> above-mentioned 

technologies. 

The majority <strong>of</strong> <strong>aluminium</strong> cold mills rolling 

sheet products today are a 4-high singlestand 

type, which in many cases produces a 

quality product meeting end-user requirements. 

However, such conventional mills are 

now approaching or have reached <strong>the</strong> technical 

limits <strong>of</strong> <strong>the</strong> latest market requirements 

that can include: 

1. Increasing coil dimensions including wide 

strip width ranges <strong>of</strong> less than 1 metre to 2.7 

metre or greater 

2. Achievement <strong>of</strong> excellent levels <strong>of</strong> strip 

quality including flatness, thickness, surface 

finish, dryness, coil build-up, etc across an 

ever increasing wide range <strong>of</strong> products with 


Danieli Fröhling 6-high Diamond Mill 

rolling speeds up to 1,800 m/min 

4. Wide entry to exit thickness range, typical- 

ly from 10 mm to less than 1 mm 

3. Ability to handle and roll coil weights up to 

or greater than 35 tonnes without a compromise 

in rolling dimensionally smaller coils 

4. Large rolling load range to cater for hard 

alloys to Electrical Discharge Texturing (EDT) 

and skin-passes on a single stand 

5. Operational efficiency by reduction in coilto-coil 

times and mill stop-time 

6. Greater production flexibility by a wide 

mill control range thus having <strong>the</strong> ability to 

react to future market demands for new strip 

products or requirements, i.e. a ‘future pro<strong>of</strong>’ 

mill 

7. Environmental considerations such as electrical 

power efficiency and fume recovery with 

coolant re-generation 

8. Improved working environment and practices 

for mill operators. 

Some existing cold mills can be modernised 

to include some <strong>of</strong> <strong>the</strong> above latest requirements 

and that a new 4-high single stand cold 

mill <strong>of</strong> an advanced design may exactly meet 

<strong>the</strong> specific market and commercial requirements 

<strong>of</strong> a client, for example in <strong>the</strong> case <strong>of</strong> 

<strong>the</strong> Danieli Fröhling a ‘4-high Diamond Mill’ 

being supplied to Nikkei Siam Aluminium 

(NSA), Thailand, to roll primarily fin-stock 

material. However, <strong>the</strong> majority <strong>of</strong> <strong>the</strong> latest 

market requirements require an advanced 

6-high mill stand solution ei<strong>the</strong>r as a single 

stand or tandem mill configuration. 

The Danieli Fröhling ‘6-High Diamond 

Mill’ in a single stand configuration being 

supplied to Aleris Europe is <strong>the</strong> latest generation 

<strong>of</strong> state-<strong>of</strong>-<strong>the</strong>-art cold rolling mills and is 

designed from inception to meet <strong>the</strong> current 

and future market requirements. The superior 

technology selected by Aleris and provided 

by <strong>the</strong> Diamond Mill includes <strong>the</strong> following: 

• 6-high roll stack configuration to ensure 

mill stack stability with optimum sized 

workrolls for <strong>the</strong> specified thickness range 

whilst providing <strong>the</strong> highest level <strong>of</strong> strip flatness 

performance across a wide strip range 

that has a max / min width ratio <strong>of</strong> greater than 

two. This configuration ensures flexibility <strong>of</strong> 

scheduling current rolling products as well as 

future products. The parallel roll design delivers 

a practical operational solution with a 

simple roll grind regime and has an effective 

actuator control range to guarantee high qual- 

Danieli Fröhling Advanced Slitter 


ity strip flatness. 

• Hydraulic roll-gap adjustment, utilising 

bottom mounted double-acting roll load cylinders 

designed for a roll separation force with 

an operational range that covers low load EDT 

passes as well as high rolling load capability 

for AA5xxx series hard alloys 

• The Danieli Fröhling ‘Hi-Res’ coolant spray 

concept with Hot Edge Sprays (HES) providing 

precise <strong>the</strong>rmal control <strong>of</strong> <strong>the</strong> workrolls that 

coupled with <strong>the</strong> dynamic work-roll and intermediate 

roll bending and intermediate roll 

side-shifting in order to meet Aleris’s stringent 

product performance criteria 

• Close attention to <strong>the</strong> design <strong>of</strong> all rolls in 

contact with <strong>the</strong> strip and <strong>the</strong> coil handling to 

maintain a defect free material surface finish 

• DAN-ECO 2 fume cleaning and rolling oil 

recovery system to meet <strong>the</strong> highest European 

emission standards for <strong>the</strong> separation <strong>of</strong> 

Volatile Organic Compounds (VOC) as well as 

recovering <strong>the</strong> rolling oil to be fully re-used for 

<strong>the</strong> rolling process 

• DANPurity rolling oil filtration system to 

maintain <strong>the</strong> original design specifications <strong>of</strong> 

<strong>the</strong> rolling oil by <strong>the</strong> filtration <strong>of</strong> containments 

thus ensuring high strip surface quality. 

When <strong>the</strong> Aleris Diamond Mill starts producing 

quality material at <strong>the</strong> end <strong>of</strong> 2012, it 

will be one <strong>of</strong> <strong>the</strong> most advanced cold mills in 

operation around <strong>the</strong> world. 

Modern slitting and trimming technology 

As <strong>the</strong> requirements for ‘totally’ flat material 

with tight tolerances exists not only for fin- 



ished rolled material but also for <strong>the</strong> slit strip, 

development <strong>of</strong> even more precise and rigid 

slitting machines is <strong>of</strong> highest importance. 

Shearing stresses have a big influence 

on subsequent workability <strong>of</strong> <strong>the</strong> strip. Single 

strips may curl or even jump out <strong>of</strong> <strong>the</strong> 

stamping die if stresses induced during slit- 

ting are too high. Width deviations could result 

in imperfect shapes due to <strong>the</strong> very narrow 

trim web at <strong>the</strong> edge <strong>of</strong> <strong>the</strong> strip before 

pressing. 

With highly accurate adjustment devices, 

modern slitting shears provide <strong>the</strong> best chance 

<strong>of</strong> an exact and repeatable knife shaft adjustment. 

It can be seen that <strong>the</strong> slitting shear must 

be able to be adjusted as closely as possible 

to certain parameters (immersion and cutting 

gap) according to <strong>the</strong> material conditions, this 

being essential for thinner strip gauges in particular 

Danieli Fröhling accommodates all <strong>of</strong> 

<strong>the</strong>se through a high standard <strong>of</strong> design and 

manufacturing: 

• Pre-tensioned elements giving backlash- 

free movement <strong>of</strong> machine parts. 

• High accuracy measuring systems to 

provide <strong>the</strong> adjustment system with 

necessary data. Rigid and solid design to 

prevent vibration. 

• Data bases store recipes for different 

slitting programmes 

• Tool clamping systems to ensure reliable 

clamping automatically actuated. 

Following <strong>the</strong>se demands, Danieli Fröhling 

developed a new generation <strong>of</strong> slitting heads. 

One <strong>of</strong> <strong>the</strong> main parameters for <strong>the</strong> strip edge 

quality is <strong>the</strong> knives’ immersion into <strong>the</strong> material 

to be cut. The optimal immersion depends 

on material grade, condition and strip 

thickness. The real immersion is a function <strong>of</strong> 

cutting force and machine rigidity. Part <strong>of</strong> <strong>the</strong> 

machine rigidity is <strong>the</strong> deflection <strong>of</strong> <strong>the</strong> knife 

shaft. 

By reducing <strong>the</strong> bearing distance and 

optimisation <strong>of</strong> <strong>the</strong> whole bearing system, 

a reduction <strong>of</strong> deflection by 44 percent is 

achieved. 

For subscribers 

www.alu-epaper.de 

Even faster – your 

added PLUS to 

<strong>the</strong> print medium 

The stiffness <strong>of</strong> a knife shaft could be increased 

by more than 130 percent. 

All <strong>of</strong> this results in even better, reproducible 

cutting edge quality and a lower transfer 

<strong>of</strong> cutting stress to <strong>the</strong> strip. 

In longitudinal slitting lines, a looping pit 

is applied for compensation <strong>of</strong> <strong>the</strong> different 

recoiler diameters <strong>of</strong> <strong>the</strong> single strips – caused 

by <strong>the</strong> differences <strong>of</strong> single strip cross section 

due to <strong>the</strong> rolling process. Due to <strong>the</strong> downgauging 

<strong>of</strong> finished products nowadays and 

<strong>the</strong> request <strong>of</strong> <strong>the</strong> finishing industry for larger 

recoiler diameters, tensioning <strong>of</strong> <strong>the</strong> strip with 

advanced braking units becomes ever more 

important. 

Danieli Fröhling has developed <strong>the</strong> vacuum 

braking roll in order to avoid or minimise <strong>the</strong> 

aforementioned disadvantages. Around 27 

<strong>of</strong> <strong>the</strong>se units have been supplied to industry 

to date, and experiences have been ga<strong>the</strong>red 

over <strong>the</strong> past about 17 years. 

Advantage compared to <strong>the</strong> strip surface: 

• Very low surface pressure during braking 

• No damage <strong>of</strong> <strong>the</strong> strip surface 

• No friction between upper surface <strong>of</strong> <strong>the</strong> 

strip and braking means 

• Lowest possible friction at <strong>the</strong> lower side 

<strong>of</strong> <strong>the</strong> strip 

• Same specific strip tension at all strips, 

independent on <strong>the</strong> rolling pr<strong>of</strong>ile over 

<strong>the</strong> strip width. 

The international top level <strong>of</strong> trimming lines 

in <strong>the</strong> <strong>aluminium</strong> industry process coils with 

weight <strong>of</strong> around 30 tonnes at speeds <strong>of</strong> more 

that 1,200 m/min. Danieli Fröhling trimming 

lines, having proven <strong>the</strong>ir reliability at stable 

operation speeds <strong>of</strong> up to 1,500 m/min since 

years, are equipped with special features to 

provide precise products at maximum pro- 

duction, such as: 

• High-speed cropping 

• High precision electrostatic oiler 

• Edge trimming shear with automatic 

adjustment 

• Scrap suction devices for high-speed 

applications 

• Dedicated design <strong>of</strong> exit guide unit. 

Product services and support 

Vital to any organisation is <strong>the</strong> services pro- 

vided with both a new project and future support. 

This can range from training, process 

studies right through to major FSTK projects. 

Danieli with its partner Innoval has <strong>the</strong> organisational 

structure to provide <strong>the</strong>se services to 

our clients. 

Summary 

It is evident that <strong>the</strong> end-users <strong>of</strong> high-end 

<strong>aluminium</strong> products in particular in <strong>the</strong> automotive, 

beverage and aerospace markets 

are continuously innovating to maintain <strong>the</strong>ir 

competitiveness, develop new products and 

increase <strong>the</strong>ir efficiency and added-value. 

Consequently, rolling and processing companies 

must respond to <strong>the</strong>se developments and 

also implement similar philosophies encompassing 

<strong>the</strong> efficient production <strong>of</strong> high quality 

end-products. 

The result <strong>of</strong> this market and product 

advancement requires companies such as 

Danieli to react and supply advanced high 

quality equipment and process solutions to 

meet <strong>the</strong>se evolving challenges by reacting to 

market requirements and providing a spectrum 

<strong>of</strong> enhanced solutions for <strong>the</strong> future. 

Danieli can do this! 

� 


Longer campaigns with improved monolithics 

for lining <strong>aluminium</strong> melt-hold furnaces 

Andy Wynn, John Coppack and Tom Steele, Thermal Ceramics UK Ltd. 

To remain competitive, <strong>aluminium</strong> producers 

continue to increase productivity 

through <strong>the</strong>ir melt-hold furnaces. Increasing 

heat input to <strong>the</strong> furnace using more 

powerful burners is common practice. 

But faster melting leads to increased 

metal losses from surface oxidation and 

to segregation from large heat gradients. 

These effects are countered by increased 

use <strong>of</strong> fluxes and increased stirring. Given 

<strong>the</strong> increasingly challenging environment 

within which <strong>the</strong> refractory lining has to 

work, traditional lining solutions can no 

longer be relied upon to provide <strong>the</strong> service 

lives that were previously achieved. 

Therefore, a new generation <strong>of</strong> furnace 

lining materials is required to cope with 

today’s <strong>aluminium</strong> furnace. This work 

reports on a new monolithic material 

with improved performance, compared to 

existing materials, designed for use in <strong>the</strong> 

ramp/hearth area <strong>of</strong> <strong>aluminium</strong> furnaces. 

Improved behaviour against <strong>the</strong> critical 

performance criteria in this furnace region 

are demonstrated in <strong>the</strong> laboratory 

using industry standard test methods. 

The refractory lining <strong>of</strong> a typical furnace 

used for holding and melting <strong>aluminium</strong> has 

to withstand a wide variety <strong>of</strong> physical and 

chemical environments. Each <strong>of</strong> <strong>the</strong> different 

areas within <strong>the</strong> furnace (Fig. 1) presents a different 

set <strong>of</strong> operating conditions, in terms <strong>of</strong> 

peak temperature, metal contact, salt contact, 

etc. Therefore, in order for a monolithic material 

to perform in a particular area <strong>of</strong> <strong>the</strong> 

furnace, it needs to cope with <strong>the</strong> specific 

environmental conditions in that region. This 

is why furnace linings are complex arrangements, 

with different materials installed in different 

locations [1]. 

Background: In <strong>the</strong> last 30 years, a group <strong>of</strong> 

monolithic technologies has emerged, specifically 

designed to perform within <strong>the</strong> unique 

environment <strong>of</strong> <strong>aluminium</strong> melt-hold fur- 

naces. These Al-resistant grades <strong>of</strong>ten contain 

‘non-wetting’ additives, particularly in <strong>the</strong> 

metal contact areas, to minimize interaction 

between <strong>the</strong> refractory and <strong>the</strong> melt to suppress 

damage from ‘corundum growth’ [2]. 

As <strong>aluminium</strong> producers strive to increase 

productivity, <strong>the</strong> environment within <strong>the</strong> furnace 

is becoming more arduous. Chamber 

temperatures are increasing and more aggressive 

fluxes are being used, necessitating more 

frequent and severe cleaning operations <strong>of</strong> 

<strong>the</strong> furnace walls. To maintain high productivity, 

it is necessary to minimize <strong>the</strong> frequency 

<strong>of</strong> furnace downtime. The more aggressive 

conditions today mean that lining materials 

developed in <strong>the</strong> past are now being used be- 

yond <strong>the</strong>ir original intended design boundaries 

and <strong>the</strong>ir service performance is under threat, 

leading to more frequent lining repairs. 

Aluminium producers take a furnace <strong>of</strong>fline 

for repair once a critical lining area has degraded 

to <strong>the</strong> point <strong>of</strong> affecting <strong>the</strong> efficiency 

and/or safety <strong>of</strong> <strong>the</strong> operation. At this stage, 

not all <strong>the</strong> lining will have degraded to <strong>the</strong> 

point that it is in need <strong>of</strong> replacement or repair. 

Therefore, <strong>the</strong> frequency <strong>of</strong> furnace down- 

time is determined by <strong>the</strong> area <strong>of</strong> <strong>the</strong> furnace 

most quickly and frequently degraded during 

operation. In order to increase campaign 

times and decrease frequency <strong>of</strong> stoppages, 

we need to improve <strong>the</strong> service life <strong>of</strong> this 

weak link in <strong>the</strong> lining. To identify <strong>the</strong> region 

most frequently and quickly degraded, we 

worked with several <strong>aluminium</strong> producers. 

Their feedback suggested that <strong>the</strong> most common 

area that was <strong>the</strong> cause <strong>of</strong> repair downtime 

was <strong>the</strong> ramp / hearth area. 

The failure mechanisms within <strong>the</strong> furnace 

environment, that limit refractory service life, 

are listed in [1]. Since our target is to improve 

refractory performance in <strong>the</strong> ramp / hearth 

region, we need to understand which <strong>of</strong> <strong>the</strong>se 

failure modes are most critical to lining performance 

in this region. 

Performance targets: A study <strong>of</strong> working 

practices and furnace operating conditions at 

MEASURING & CONTROL 

a number <strong>of</strong> <strong>aluminium</strong> producers revealed 

that <strong>the</strong> ramp / hearth region <strong>of</strong> an <strong>aluminium</strong> 

melt-hold furnace is subjected to severe mechanical 

and <strong>the</strong>rmal stress during <strong>the</strong> loading 

<strong>of</strong> large ingot down <strong>the</strong> ramp. Frequent loading 

<strong>of</strong> heavy ingot to feed <strong>the</strong> furnace, <strong>of</strong>ten 

by fork lift truck, subjects <strong>the</strong> ramp to severe 

abrasive forces. As <strong>the</strong> ingot is usually at room 

temperature, <strong>the</strong>re is also considerable <strong>the</strong>r- 

mal shock on <strong>the</strong> ramp / hearth refractory, 

which is at furnace operating temperature. 

As <strong>the</strong> bottom <strong>of</strong> <strong>the</strong> ramp and <strong>the</strong> complete 

hearth are in contact with molten metal, <strong>the</strong> 

refractory is also subject to chemical attack 

from <strong>the</strong> alloy, alloying elements and flux additions. 

A study <strong>of</strong> ramp / hearth degradation 

<strong>of</strong> Al-resistant materials containing ‘non-wetting’ 

additives suggested that damage leading 

to furnace downtime is mostly due to <strong>the</strong> 

mechanical action <strong>of</strong> <strong>the</strong> erosion and <strong>the</strong>rmal 

shock from ingot loading. We <strong>the</strong>refore focused 

our work on developing a new Al-resistant 

material with improved abrasion and 

<strong>the</strong>rmal shock resistance. To achieve significant 

improvements in performance we set out 

to increase abrasion and <strong>the</strong>rmal shock resistance 

by 20% compared to existing materials. 

As metal and alkali resistance are secondary 

performance parameters in this furnace region, 

we also had to ensure that any changes 

we made to <strong>the</strong> materials did not degrade 

chemical resistance. 

EXPERIMENTAL 

Fig. 1: Furnace 

lining zones in a 

typical <strong>aluminium</strong> 

melt-hold furnace 

Two existing, industry leading Al-resistant 

monolithic materials used by many <strong>aluminium</strong> 

producers in <strong>the</strong> ramp / hearth area <strong>of</strong> 

ALUMINIUM · EAC CONGRESS 2011 19 

Images: Thermal Ceramics


Standard 

1 

Standard 

2 

melt-hold furnaces were selected as baseline 

materials for <strong>the</strong> study. Detailed analysis was 

undertaken in order to identify those aspects 

<strong>of</strong> <strong>the</strong> materials technology that were considered 

to be constraining performance, leading 

to mechanical failure. The bond chemistry 

and aggregate granulometry were <strong>the</strong>n re-engineered 

to find <strong>the</strong> optimum balance <strong>of</strong> material 

types and grain size, shape and distribu- 

tion that produced <strong>the</strong> maximum improvement 

in abrasion and <strong>the</strong>rmal shock performance 

without negatively affecting o<strong>the</strong>r important 

properties. This paper presents <strong>the</strong> results 

<strong>of</strong> performance and property measurements 

<strong>of</strong> <strong>the</strong> final, optimized development composition 

compared to <strong>the</strong> baseline standards. All 

materials in <strong>the</strong> study were tested against <strong>the</strong> 

four key performance parameters. 

Primary performance parameters 

New 

material 

Water (%) 5.5-6.5 5.7 5.3 

Bulk 

density 

(kg/m 3 ) 

PLC 

(%) 

CCS 

(MPa) 

110°C 2840 2630 2640 

815°C 2800 2590 

1000°C 2790 2580 

1300°C 2570 2510 

815°C -0.29 -0.43 

1000°C -0.32 -0.26 

1300°C -0.35 0.38 0.95 

110°C 128 122 147 

815°C 163 99 

1000°C 129 95 

1300°C 138 119 144 

Table 1: Physical properties <strong>of</strong> materials studied 

Standard 

1 

Standard 

2 

New 

material 

% Al 2 O 3 80.6 65.8 66.6 

% SiO 2 11.2 26.7 25.6 

% CaO 1.8 3.6 3.2 

% TiO 2 2.0 2.1 2.2 

% Fe 2 O 3 1.2 1.0 1.0 

% MgO 0.2 0.1 0.2 

% Alkalis 0.2 0.2 0.2 

Table 2: Chemical analysis <strong>of</strong> materials studied 

1. Abrasion Resistance Test (ASTM C704); 

pre-fired samples are blasted with a stream <strong>of</strong> 

SiC grit <strong>of</strong> specified grain size for a set time. 

Samples are cross-sectioned and <strong>the</strong> amount 

<strong>of</strong> material abraded across <strong>the</strong> section is 

measured in cm 3 . 

2. Thermal Shock Resistance Test (ASTM 

C1100 – Ribbon Test); pre-fired samples are 

subjected to alternating heating and cooling 

cycles on one face using a ribbon burner. The 

modulus <strong>of</strong> elasticity (E-modulus) <strong>of</strong> samples 

is measured non-destructively by ultrasonics 

before and after testing. The percentage 

<strong>of</strong> retained E-modulus is used as a measure 

<strong>of</strong> retained strength. 

Secondary performance parameters 

1. Aluminium Resistance ‘Cup’ Test; 

Cup samples are prepared and filled 

with 7075 alloy. Samples are ramped up 

to 1,000°C and held for 100 hours. After 

cooling, <strong>the</strong> samples are sectioned vertically 

and visually assessed for <strong>the</strong> degree 

<strong>of</strong> metal penetration and corundum 

growth [3]. 

2. Alkali Resistance ‘Cup’ Test; sample 

preparation is <strong>the</strong> same as for <strong>the</strong> Al resistance 

cup test. Instead <strong>of</strong> Al, <strong>the</strong> samples 

are filled with mixtures <strong>of</strong> K 2CO 3 

and Na 2CO 3 and fired to 900, 1,000 or 

1,100°C for five hours. After sectioning, 

samples are analyzed by visual inspection 

for cracks, bulges, depth <strong>of</strong> penetration 

and colour change. 

RESULTS AND DISCUSSION 

The characteristics <strong>of</strong> <strong>the</strong> optimized new 

material compared to <strong>the</strong> two standard 

baseline materials are displayed in Tables 1 

and 2. Both baseline materials are low cement, 

vibrocast grades. The optimized new material 

in Table 1 could be cast at 5.3% water, lower 

than <strong>the</strong> baseline grades, and gave free flow 

<strong>of</strong> 125 mm and tapped flow <strong>of</strong> 160 mm. 

Primary performance parameters 

Abrasion resistance test results <strong>of</strong> <strong>the</strong> mate- 

rials are presented in Fig. 2. As dried, <strong>the</strong> new 

optimized material delivered 16% better resistance 

to abrasion compared to Standard 

1 and 20% compared to Standard 2. When 

pre-fired, <strong>the</strong> new material delivered 30% 

improvement on abrasion resistance compared 

to Standard 1 and 20% compared to 

Standard 2. 

Thermal shock resistance test results <strong>of</strong> 

<strong>the</strong> materials are presented in Fig. 3. After 5 

cycles, Standard 1 lost 42% <strong>of</strong> its E-modulus 

and Standard 2 lost 32%, compared to only 

20% loss for <strong>the</strong> new material. These results 

suggest <strong>the</strong> new material can deliver 52% improvement 

on <strong>the</strong>rmal shock resistance compared 

to Standard 1 and 38% compared to 

Standard 2. 

Secondary performance parameters 

As baseline materials 1 and 2 are commonly 

used in service, we fully expected <strong>the</strong>m to 

Fig. 2: Abrasion loss resistance <strong>of</strong> test materials 

Fig. 3: Thermal shock resistance <strong>of</strong> test materials 

pass <strong>the</strong> Al resistance ‘cup’ testing. Both materials, 

and all <strong>of</strong> our new development formulations, 

contain well proven ‘non-wetting’ 

additives. Our optimized new composition 

passed all Al contact testing and performed 

identically to Standards 1 and 2 in <strong>the</strong> visual 

assessment <strong>of</strong> Al ‘cup’ test samples (Fig. 4). 

Analysis <strong>of</strong> <strong>the</strong> alloy after testing revealed 

that although all materials pass <strong>the</strong> test (target 

pick up < 0.5% Si, < 0.1% Fe), Si pick up is 

much reduced in <strong>the</strong> new material compared 

to <strong>the</strong> two standards. Since ‘Cup’ test failures 

are normally accompanied by increased 

concentrations <strong>of</strong> Si and Fe in <strong>the</strong> alloy after 

testing, this result suggests a much reduced 

interaction between <strong>the</strong> new material and <strong>the</strong> 

test alloy compared to <strong>the</strong> standards, indicating 

superior ‘non-wetting’ behaviour. 

For <strong>the</strong> alkali resistance testing, we expected 

Standards 1 and 2 to possess good resistance 

to alkalis as <strong>the</strong>y are used commonly 

in service. Our final, optimized new composi- 

tion passed all Alkali contact testing with 

K 2CO 3 and Na 2CO 3 and performed identically 

to Standards 1 and 2 in <strong>the</strong> visual assessment 

<strong>of</strong> Alkali ‘cup’ test samples after testing, at all 

test temperatures. 

CONCLUSIONS 

1. By working closely with alumin- 


Standard 

1 

Standard 

2 

New 

material 

% Si pick up 0.314 0.093 0.011 

% Fe pick up 0.052 0.04 0.04 

Table 3: Alloy analysis after Al ‘Cup’ testing 

ium producers, <strong>the</strong> most frequent cause <strong>of</strong> 

melt-hold furnace downtime has been identified 

as mechanical damage in <strong>the</strong> ramp / hearth 

region <strong>of</strong> <strong>the</strong> refractory lining. 

2. The main factors leading to mechanical 

damage in this region have been identified as 

severe abrasion and <strong>the</strong>rmal shock from <strong>the</strong> 

frequent loading <strong>of</strong> heavy, cold ingot. 

3. Through re-engineering <strong>of</strong> <strong>the</strong> bond chemistry 

and aggregate granulometry, significant 

improvements have been achieved in abrasion 

and <strong>the</strong>rmal shock resistance for material in 

<strong>the</strong> ramp / hearth region <strong>of</strong> <strong>aluminium</strong> melthold 

furnaces. 

4. An optimized formulation has been shown 

to deliver 20-30% improvement in abrasion 

resistance and 40-50% improvement in <strong>the</strong>rmal 

shock resistance compared to existing 

materials. 

5. The new material has been shown to pass 


Fig. 4: Al ‘Cup’ testing – Standard 2 (left) and New material (right) – dried samples 

industry standard <strong>aluminium</strong> contact and 

alkali resistance tests. More detailed investigation 

indicates <strong>the</strong> new material possesses 

superior ‘non-wetting’ characteristics compared 

to existing materials. 

6. These results suggest that <strong>the</strong> new material 

should extend service life in <strong>the</strong> ramp / hearth 

area and is thus expected to reduce <strong>the</strong> frequency 

<strong>of</strong> furnace downtime and allow <strong>aluminium</strong> 

producers to run longer production 

campaigns, increasing productivity and minimizing 

<strong>the</strong> need for expensive repairs. 

7. This new material is now on trial in <strong>the</strong> 

ramp/hearth area <strong>of</strong> melt-hold furnaces at several 

<strong>aluminium</strong> producers around <strong>the</strong> world. 

Aluminium am laufenden Band 

�� 

�� 

REFERENCES 

1. A.M. Wynn, T.J. Coppack, T. Steele, K.J. Moody, 

and L. Caspersen, Monolithic Material Selection for 

<strong>the</strong> Lining <strong>of</strong> Aluminum Holding & Melting Furnaces, 

TMS 2010, Seattle, USA, Feb. 14-18, 2010. 

2. D. Jones, A.M. Wynn, and T.J. Coppack, The Development 

and Application <strong>of</strong> an Aluminium Resistant 

Castable, UNITECR ’93, Sao Paulo, Brasil, Oct 

31-Nov 3, 1993. 

3. A.M. Wynn, T.J. Coppack, and T. Steele, Methods 

<strong>of</strong> Assessing Monolithic Refractories for Material 

Selection in Aluminium Melt-Hold Furnaces, 53rd International Refractories Colloquium, Aachen, 

Germany, Sept. 8-9, 2010. 

�� 

�� 

�� 


�


Optimization <strong>of</strong> an <strong>aluminium</strong> rolling 

mill with a high speed X-ray thickness gauge 

Chr. Burnett, A. Quick and J. Olschewski; Thermo Fisher Scientific 

The high speed production achieved by 

modern <strong>aluminium</strong> rolling mills requires 

reliable and robust thickness sensors. 

The Automatic Gauge Control (AGC) 

systems and hydraulic actuators in place 

today are capable <strong>of</strong> reacting to strip 

changes within just a few milliseconds so 

accurate measurements must be supplied 

with comparable speed. In order to aide 

process control engineers as <strong>the</strong>y optimize 

<strong>the</strong> throughput <strong>of</strong> a given mill, <strong>the</strong> 

International Electrical Commission (IEC) 

has produced Standard 61336 defining 

specific terms associated with thickness 

measurement equipment and <strong>the</strong> testing 

protocol associated with verifying gauge 

performance. This paper describes a new 

Thermo Scientific sensor that provides 

high speed measurements allowing faster 

feedback loops and tighter control <strong>of</strong> 

flat sheet <strong>aluminium</strong> and its alloys. For 

direct feedback, various communication 

protocols are available such as analogue 

signals, Pr<strong>of</strong>ibus or E<strong>the</strong>rnet. In addition 

to using <strong>the</strong> measurement values for direct 

feedback, single 1 ms measurement 

values can be archived using iba analyzer 

allowing efficient post-rolling analysis for 

mill optimization and product improvement 

resulting in fur<strong>the</strong>r cost savings. 

Rolled <strong>aluminium</strong> strip is used in a wide variety 

<strong>of</strong> industrial and consumer applications 

around <strong>the</strong> world. With <strong>the</strong> ever-growing 

concern for energy efficiency and sustainability, 

<strong>aluminium</strong> producers strive to provide 

<strong>the</strong> world class quality strip on <strong>the</strong> first coil 

<strong>of</strong> a campaign, maximizing mill yield and 

minimizing scrap material. Aluminium sheet 

producers and <strong>the</strong>ir customers have agreed 

upon standards to describe various physical 

parameters for <strong>the</strong> material traded. Thickness, 

width, hardness, strength, are among <strong>the</strong> key 

variables defined in a simple product code. 

Both parties fully understand <strong>the</strong> standards 

and any disputes are governed by <strong>the</strong> scope 

<strong>of</strong> actual standard definition produced by 

one <strong>of</strong> a few <strong>aluminium</strong> industry associations. 

Thickness gauge manufactures are also expected 

to produce and test <strong>the</strong>ir equipment 

in accordance with an international standard 

known as IEC 61336. Unfortunately, <strong>the</strong>se 

standards are not as well known by <strong>the</strong>ir customers, 

<strong>the</strong> <strong>aluminium</strong> sheet producers. This 

can result in confusion during <strong>the</strong> gauge selection 

process, and unmet expectations for new 

mills and mill upgrade projects. This paper 

will review <strong>the</strong> role <strong>of</strong> <strong>the</strong> thickness gauge in 

a mill, present an overview <strong>of</strong> <strong>the</strong> standards to 

which all gauging systems should be held and 

present how a modern high speed X-ray thickness 

gauge can be used as a tool to optimize 

an <strong>aluminium</strong> rolling mill. 

Aluminium rolling mills 

The evolution <strong>of</strong> <strong>aluminium</strong> rolling mills 

has accelerated as <strong>the</strong> speed <strong>of</strong> processors 

and digital controls have grown by orders <strong>of</strong> 

magnitude. The capital cost in a rolling mill is 

substantial and investors understand that in 

order to achieve <strong>the</strong> maximum ROI and shortest 

payback time, <strong>the</strong> mill needs to produce 

high quality sheet at <strong>the</strong> fastest possible mill 

speeds. Diligent plant managers are always 

focused on safely maximizing mill output. To 

accomplish this, mills are operated 24 hours a 

day, seven days a week. When a mill is down 

for any reason, <strong>the</strong> accountants not only consider 

<strong>the</strong> energy and labour consumed while 

<strong>the</strong> mill is idle, but <strong>the</strong> value <strong>of</strong> <strong>the</strong> product 

that could have been made during that ‘lost 

time’. It is no surprise to hear that mills operate 

at <strong>the</strong> highest speeds allowed by <strong>the</strong>ir motors 

and drives. However, raw production in tonnes 

means nothing if <strong>the</strong> material produced does 

not meet quality standards. 

Many believe that strip quality begins in <strong>the</strong> 

meltshop, and that is not far from <strong>the</strong> truth. 

Controlling <strong>the</strong> chemistry <strong>of</strong> <strong>the</strong> molten <strong>aluminium</strong> 

not only assures <strong>the</strong> alloy produced 

will meet <strong>the</strong> mechanical properties desired <strong>of</strong> 

<strong>the</strong> final product, but that <strong>the</strong> strip will handle 

Hot roughing mill Hot finish mill Cold rolling mill Foil mill 

Max. thickness > 400 mm 20-30 mm 2-6 mm 0.6 mm 

Min. thickness 20-30 mm 2-10 mm 0.150 mm 0.006 (x2) mm 

Rolling speeds 

(metres per min) 

~ 100 ~1000 ~2000 ~2000 

Table 1: Overview <strong>of</strong> <strong>aluminium</strong> rolling mill parameters 

<strong>the</strong> tonnes <strong>of</strong> pressure and tensions <strong>of</strong> <strong>the</strong> high 

speed rolling process. Table 1 summarizes <strong>the</strong> 

typical thicknesses and rolling speeds for each 

mill type. Not surprisingly, as <strong>the</strong> material gets 

thinner, <strong>the</strong> speeds increase dramatically. 

Automatic Gauge Control (AGC) 

The average human response time is on <strong>the</strong> 

order <strong>of</strong> a quarter <strong>of</strong> a second, at <strong>the</strong> maximum 

rolling speed <strong>of</strong> a cold or foil mill, 8 

metres <strong>of</strong> strip is produced. It is easy to see 

why AGC is an essential component <strong>of</strong> modern 

rolling mills. Comprehensive AGC algorithms 

incorporate readings from dozens <strong>of</strong> sensors 

around <strong>the</strong> mill. 

Some <strong>of</strong> <strong>the</strong> key AGC input parameters 

are speed and tension (Fig. 1). Conservation <strong>of</strong> 

mass dictates that <strong>the</strong> mass per unit time entering 

<strong>the</strong> mill must equal <strong>the</strong> mass per unit time 

exiting <strong>the</strong> mill. So as <strong>the</strong> material is rolled 

thinner, <strong>the</strong> speeds must increase. If <strong>the</strong> drive 

motors are <strong>of</strong>f, even by a few centimetres per 

minute, <strong>the</strong> strip may break. There is a delicate 

balance between <strong>the</strong> reduction caused by 

<strong>the</strong> mill force, and <strong>the</strong> reduction caused by 

drawing (extruding) <strong>the</strong> material through <strong>the</strong> 

gap. 

The thickness gauges are primarily used 

to look for dramatic deviations in thickness, 

o<strong>the</strong>r than that, <strong>the</strong>ir feedback is used to check 

<strong>the</strong> predictions <strong>of</strong> <strong>the</strong> mass flow models. Small 

changes are corrected by slight adjustments in 

speed and tension, as <strong>the</strong> motor power can 

be controlled very quickly. Larger changes 

are compensated by <strong>the</strong> hydraulic cylinders in 

<strong>the</strong> mill stand, which can react in a matter <strong>of</strong> 

milliseconds. 

When a mill is operating at 1,800 m/min, 

<strong>the</strong> material moves 3 cm every millisecond. 

If <strong>the</strong> mill is using 0.5 metre diameter rolls, 

<strong>the</strong> circumference <strong>of</strong> <strong>the</strong> roll would be roughly 

equivalent to 50 milliseconds. In order to see 

any eccentricity or periodic event related to a 

roll <strong>of</strong> this diameter, one would need to have a 

thickness sensor not only capable <strong>of</strong> operating 

at 5 ms, but being able to provide measurements 

with manageable signal to noise values. 

For cable shielding applications where electrical 

signals are carried at 1 GHz or higher, <strong>the</strong> 

ability to see thickness variations over a distance 

<strong>of</strong> 30 cm is essential to avoid unwanted 

electrical harmonics by <strong>the</strong> end consumer. 


Fig. 1: Selected AGC input variables 

Thickness gauge selection 

While <strong>the</strong>re are several choices in thickness 

gauge technology, <strong>the</strong>re really is only one 

choice for <strong>the</strong> speed and accuracy requirements 

demanded in optimizing a rolling mill: 

<strong>the</strong> X-ray gauge. 

Direct contact gauges have <strong>the</strong> advantage 

<strong>of</strong> being insensitive to alloy, <strong>the</strong> measurement 

stylus marks <strong>the</strong> strip, and <strong>the</strong> mechanical tolerances 

<strong>of</strong> <strong>the</strong> frame prevent measurement 

near <strong>the</strong> centreline <strong>of</strong> <strong>the</strong> strip. Additionally, 

<strong>the</strong> small measurement spot size <strong>of</strong> <strong>the</strong> stylus 

translates microvariations in <strong>the</strong> strip surface 

into a noisy signal. While <strong>the</strong>se variations may 

actually be in <strong>the</strong> strip, <strong>the</strong> signal needs to be 

filtered to reduce <strong>the</strong> noise, and <strong>the</strong> filtering 

will delay responses to actual longer term 

changes. Therefore, high speed AGC is not 

practical with contact gauges. 

There are o<strong>the</strong>r non-contact, non-radiation 

based sensors available, some that use eddy 

current and o<strong>the</strong>rs that use laser, but each also 

has <strong>the</strong>ir drawbacks, <strong>the</strong> eddy current sensors 

have an extremely narrow gap between 

emitter and receiver which can turn this non- 

contact gauge into a contact gauge should <strong>the</strong> 

strip vary by as little as 12 mm. Additionally, 

like <strong>the</strong> contact gauge, <strong>the</strong> sensor frame 

mechanics limit <strong>the</strong> measurement location to 

only a few centimetres from <strong>the</strong> edge. La- 

Fig. 2: Simulated sensor responses to a 250 ms, 1.0% deviation from target 

ser gauges do 

benefit from a 

larger air gap, 

but <strong>the</strong>y can be 

sensitive to <strong>the</strong> 

high amount <strong>of</strong> 

steam and mist 

that occurs in 

a rolling mill. 

Additionally 

<strong>the</strong> laser camera 

technology 

limits <strong>the</strong> resolution 

<strong>of</strong> measurement 

to a few microns. While this may be 

acceptable in certain applications, it does not 

meet <strong>the</strong> needs <strong>of</strong> <strong>the</strong> thinner faster mills. 

Non-contact radiation based thickness 

gauges can use ei<strong>the</strong>r radioisotope or X-ray 

sources. However in <strong>the</strong> case where <strong>the</strong> gauge 

measurement is to be used in a closed loop 

control or AGC system, <strong>the</strong>re is really only 

one solution: X-rays. The number <strong>of</strong> photons 

emitted from an X-ray source is approx. 

1,000 times that <strong>of</strong> <strong>the</strong> commercially available 

isotopes. Due to <strong>the</strong> statistical nature <strong>of</strong> 

radiation detection, measurements made with 

more photons have a better signal to noise 

ratio, and consequently a more true measurement. 

In <strong>the</strong> case <strong>of</strong> X-ray versus isotope, <strong>the</strong> 

noise level for an isotope is on <strong>the</strong> order <strong>of</strong> 

20 to 30 times worse than that <strong>of</strong> an X-ray 

based sensor when <strong>the</strong> same averaging time 

is used. Statistical noise at that level creates 

a situation where small changes in thickness 

are lost in <strong>the</strong> noise <strong>of</strong> <strong>the</strong> signal. While multiple 

isotope pellets might be used in an effort 

to increase <strong>the</strong> signal, <strong>the</strong> regulatory and safety 

considerations make this option prohibitive. 

Ano<strong>the</strong>r approach to improve <strong>the</strong> noise on 

isotope based systems is to increase <strong>the</strong> averaging, 

or response time. However, when this 

is done, small, and instantaneous changes in 

product thickness are blurred to <strong>the</strong> point <strong>of</strong> 

not being seen (Fig. 2). 


Just as low signal to noise ratio is a serious 

factor in source selection, one must take care 

to select a source <strong>of</strong> a proper energy as to 

not have too much signal. While <strong>the</strong> density 

<strong>of</strong> <strong>aluminium</strong> and its alloys is a tremendous 

advantage in aerospace and o<strong>the</strong>r applications, 

it presents unique challenges in non-contact 

radiation gauging. At typical X-ray energies, 

it is nearly thirteen times less absorbing compared 

to steel. When compounded by <strong>the</strong> thin 

dimension <strong>of</strong> a foil mill, <strong>the</strong> X-ray energies 

must be reduced to <strong>the</strong> point where measuring 

and compensating for environmental factors 

such as air temperature and pressure become 

essential. If an X-ray gauge is operated at too 

high <strong>of</strong> an energy, <strong>the</strong> dynamic range <strong>of</strong> <strong>the</strong> 

detector output is reduced, limiting <strong>the</strong> measurement 

resolution and precision. In <strong>the</strong> case 

<strong>of</strong> <strong>aluminium</strong> can stock production at around 

250 microns, a 10 micron change in thickness 

results in a signal change <strong>of</strong> less than 0.075% 

at a photon energy <strong>of</strong> 60 keV, where as <strong>the</strong> 

same thickness change at a photon energy <strong>of</strong> 

10 keV will produce a signal change <strong>of</strong> over 

7% (Fig. 3). When <strong>the</strong> statistical noise on <strong>the</strong> 

measurement is ±0.1%, it is easy to see that 

<strong>the</strong> 60 keV source is just too much for <strong>the</strong> can 

stock application. 

X-ray gauge compensation 

for alloys and clads 

Commercially pure <strong>aluminium</strong> has almost no 

practical applications due to its weak mechanical 

properties. In order for <strong>aluminium</strong> to be 

useful to industry, small amounts <strong>of</strong> copper, 

manganese, magnesium and o<strong>the</strong>r elements 

are added. While <strong>the</strong>se elements add strength 

to <strong>the</strong> <strong>aluminium</strong> sheet, <strong>the</strong>ir radiation absorption 

properties vary dramatically making 

alloy compensation complex for X-ray gauges. 

For example, <strong>the</strong> AA5000 series <strong>of</strong> alloys 

with <strong>the</strong>ir small amount magnesium actually 

absorb less radiation than pure <strong>aluminium</strong>, but 

<strong>the</strong> AA 7000 series can absorb twice as much 

radiation, depending on <strong>the</strong> amount <strong>of</strong> zinc 

and copper present. 

While physics can used to predict a <strong>the</strong>oretical 

absorption for any alloy combination, 

<strong>the</strong> actual density can vary by several percent 

when compared to <strong>the</strong> <strong>the</strong>oretical density. It 

is this density difference that makes physical 

measurement <strong>of</strong> alloys necessary. More 

advanced alloy compensation algorithms can 

extend this ‘density correction’ from one alloy 

to similar alloys, but many <strong>aluminium</strong> 

gauge manufacturers require <strong>the</strong> producer 

to provide a library <strong>of</strong> samples across a variety 

<strong>of</strong> thicknesses and alloys to cover <strong>the</strong> 

whole production range. This is not possible 



Fig. 3: Sensor response as a function <strong>of</strong> <strong>aluminium</strong> thickness for different photon energies 

for some producers so <strong>the</strong>y must resort to <strong>the</strong> 

‘stop-and-measure’ technique. A single correction 

factor is calculated by comparing <strong>the</strong> 

uncorrected gauge measurement to a physical 

contact measurement made while <strong>the</strong> strip is 

stopped. While simple, it limits <strong>the</strong> accuracy 

<strong>of</strong> <strong>the</strong> gauge to <strong>the</strong> accuracy <strong>of</strong> <strong>the</strong> contact 

measurements which not only slow production 

down, but are known to be operator dependent. 

The challenges <strong>of</strong> alloy compensation are 

fur<strong>the</strong>r complicated when <strong>the</strong> <strong>aluminium</strong> producer 

rolls clad products. The alloy <strong>layer</strong>s have 

different equivalent absorptions which by 

means <strong>of</strong> proprietary s<strong>of</strong>tware can be converted 

to a single correction factor for that product. 

Without an alloy/clad compensation algorithm, 

<strong>the</strong> compensation approach based on sample 

sets for each alloy becomes a logistical headache 

for storage and quality assurance checks. 

International standards 

International organizations like <strong>the</strong> Aluminum 

Association, ASTM, Japanese Standards 

Association and o<strong>the</strong>rs provide guidance on 

not only alloy chemistry tolerances, but sheet 

dimensional tolerances as well. For instrument 

suppliers, <strong>the</strong> International Electrotechnical 

Commission (IEC) has produced standards to 

provide guidance and definition for specific 

terms and tests used. The standards act as a 

consistent scale to compare one instrument to 

ano<strong>the</strong>r without confusing nomenclature obscuring 

<strong>the</strong> true performance <strong>of</strong> each. 

The Nuclear Instrumentation Technical 

Committee (IEC Technical Committee 45) 

produced IEC 61336 ‘Thickness measurement 

systems utilizing ionizing radiation – Definitions 

and test methods’. The first committee 

release was in 1983, and an update was 

drafted in 1996. The document is available 

for purchase from <strong>the</strong> IEC at http://webstore. 

iec.ch/webstore/webstore.nsf/ArtNum_PK/ 

21703?OpenDocument, as such we cannot 

reproduce it here. The standard first defines 

common terms in order to clarify what specific 

words mean. For example <strong>the</strong> ‘mean response 

time’ may be called <strong>the</strong> ‘first time constant’ by 

some, and something else by o<strong>the</strong>rs. (Fig. 4) 

The fixed definitions avoid any ambiguity in 

interpretation. They additionally provide guidance 

in setting up and carrying out <strong>the</strong> tests 

defined in <strong>the</strong> second part <strong>of</strong> <strong>the</strong> standard. 

The standard dedicates no less than nine 

terms to clarify time based parameters. Some 

are related to defining <strong>the</strong> time associated with 

<strong>the</strong> time taken to respond to a change, while 

o<strong>the</strong>rs are related to <strong>the</strong> digital processing <strong>of</strong> 

signals. This is particularly critical with <strong>the</strong> 

advent <strong>of</strong> high speed data processors. Incoming 

data can be manipulated and processed by 

advanced filters to mask or hide true statistical 

variations. As depicted in Fig. 2 above, radiation 

measurements are statistical by <strong>the</strong>ir very 

nature. All noise figures should be quoted with 

a reference to <strong>the</strong> number <strong>of</strong> Sigmas, or confidence 

levels (CL). Most gauge manufactures 

present 2 sigma (95% CL) noise figures, but 

not all. Straightforward data processing provides 

for predictable results and a better representation 

<strong>of</strong> <strong>the</strong> process dynamics. If a change 

occurs in <strong>the</strong> process, advanced filtering may 

portray a portion <strong>of</strong> <strong>the</strong> change, but not <strong>the</strong> 

full change. Process engineers and <strong>the</strong>ir AGC 

algorithms may over react, or under correct 

thanks to <strong>the</strong> manipulated 

data. 

In order 

to ethically 

improve <strong>the</strong> 

speed <strong>of</strong> <strong>the</strong> sensor response 

to change, without 

increasing <strong>the</strong> statistical 

noise on <strong>the</strong> measurement, 

<strong>the</strong> physical characteristics 

<strong>of</strong> <strong>the</strong> sensor 

need to be optimized for 

<strong>the</strong> application. The raw 

signal; must be shielded to remove as much 

electrical noise as possible. The ideal approach 

for this is represented in <strong>the</strong> Thermo Scientific 

RM 210 AS X-ray thickness gauge. In this versatile 

instrument, <strong>the</strong> radiation detector output 

is digitized right away. The analog detector signal 

is converted to a digital number within a 

few millimeters <strong>of</strong> its origin. This practically 

eliminates <strong>the</strong> possibility <strong>of</strong> electrical noise 

impinging on <strong>the</strong> signal. In comparison tests, 

following <strong>the</strong> IEC 61336 guide, <strong>the</strong> noise on 

<strong>the</strong> new designed detector improved by a factor 

<strong>of</strong> 30%. Additionally, users <strong>of</strong> this system 

can benefit fur<strong>the</strong>r by taking advantage <strong>of</strong> <strong>the</strong> 

detector’s ability to operate at a 1 ms mean 

response time. At this speed, and with <strong>the</strong> reduced 

noise, process engineers have <strong>the</strong> tools 

to analyze data at high speeds, revealing mill 

chatter and o<strong>the</strong>r higher frequency anomalies. 

In a typical rolling mill that produces can 

stock <strong>aluminium</strong> at 250 um, <strong>the</strong> noise at a 10 

ms mean response time might be on <strong>the</strong> order 

<strong>of</strong> ±0.20% (2 sigma). With <strong>the</strong> improved 

signal processing <strong>of</strong> this system <strong>the</strong> noise will 

drop to ±0.15% (2-sigma). For a mill that produces 

200,000 tonnes a year, that translates 

to a savings <strong>of</strong> almost $200,000 in <strong>aluminium</strong> 

material alone (using <strong>aluminium</strong> price at $1 

per pound). 

An additional benefit to digitizing <strong>the</strong> signal 

so early in its journey to <strong>the</strong> AGC system, 

is speed. Once digitized, <strong>the</strong> data can be processed 

with out <strong>the</strong> time consuming ADC/DAC 

conversions. The reduction <strong>of</strong> a few milliseconds 

<strong>of</strong> process delay time can assure <strong>the</strong> AGC 

has time to fully correct any strip thickness 

deviations. This can result in higher quality 

product. 

Data archiving 

A final benefit is realized in <strong>the</strong> powerful tool 

<strong>of</strong> data archiving. This ideal system is available 

with a s<strong>of</strong>tware feature that stores any 

gauge data stream in <strong>the</strong> iba ‘.dat’ format. This 

format is gaining popularity as <strong>the</strong> iba PDA 

data analysis tool also grows in popularity. 

Fig. 4: Graphical representation <strong>of</strong> gauge response 

to an instantaneous thickness change 


�� 

The flexibility <strong>of</strong> <strong>the</strong> iba visualization tool 

is its real strength. As simple example <strong>of</strong> <strong>the</strong> 

flexibility is depicted in Fig. 5 showing a coil 

report with <strong>the</strong> thickness data presented as a 

function <strong>of</strong> length, with tolerances and coil 

statistics. Easy to use features allow for time 

based, and length based data analysis. Built in 

ma<strong>the</strong>matical tools such FFT can point process 

engineers to mill components that might need 

maintenance. In this situation, mill downtime 

can be best managed, and unplanned downtime 

dramatically reduced. 

The data archiving feature can also be con- 

figured to accept and record data from o<strong>the</strong>r 

sensors with in <strong>the</strong> mill. Any data point that 

is available to <strong>the</strong> mill computer through an 

E<strong>the</strong>rnet connection can be collected by <strong>the</strong> 

system to allow for comprehensive data analysis. 

Thus permitting <strong>the</strong> pairing <strong>of</strong> thickness 

measurement output to mill tensions and 

speeds in such a way that complies with <strong>the</strong> 

IEC 61336 testing standards. The IEC 61336 

Annex B defines appropriate test points for 

data collection and analysis. While traditional 

analog outputs are typically used for gauge 

validation, it is equally acceptable to use <strong>the</strong> 

data transferred via E<strong>the</strong>rnet, or 

o<strong>the</strong>r means to a data archive. 

Summary 

Advances in online measurement <strong>of</strong> 

flat sheet have culminated in a state 

<strong>of</strong> <strong>the</strong> art X-ray based sensor system 

that provides high speed/low noise 

measurements permitting <strong>aluminium</strong> 

producers to realize material 

savings and quality improvements 

not previously achievable. This sen- 


sor incorporates alloy and clad compensation 

based on <strong>the</strong> fundamental principles <strong>of</strong> radiation 

physics. It is housed in a robust frame designed 

for <strong>the</strong> rigors <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> rolling 

mill. The complete package is manufactured 

and tested following IEC 61336 definitions 

which assure process engineers receive clear 

thickness data with out ambiguity. Thus allowing 

for mill optimization to achieve world 

class quality and strip uniformity at <strong>the</strong> highest 

rolling speeds. This results in maximizing <strong>the</strong> 

return for <strong>the</strong> <strong>aluminium</strong> rolling mill owners 

and investors. 

Acknowledgement 

The author thanks <strong>the</strong> International Electrotechnical 

Commission (IEC) for permission to reproduce 

Information from its International Standard IEC 

61336 ed.1.0 (1996). All such extracts are copyright 

<strong>of</strong> IEC, Geneva, Switzerland. All rights reserved. 

Fur<strong>the</strong>r information on <strong>the</strong> IEC is available 

from www.iec.ch. IEC has no responsibility for <strong>the</strong> 

placement and context in which <strong>the</strong> extracts and 

contents are reproduced by <strong>the</strong> author, nor is IEC 

in any way responsible for <strong>the</strong> o<strong>the</strong>r content or accuracy 

<strong>the</strong>rein. 

Fig. 5: Typical coil report using iba data archiving tools � 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

��


Optimized raw material usage driven by dynamic alloying 

State-<strong>of</strong>-<strong>the</strong>-art casthouse production management 

Dieter Deutz, PSI Metals Non Ferrous GmbH 

Increasing raw material and scrap prices 

in <strong>the</strong> non-ferrous industry are leading 

to a continually rising portion <strong>of</strong> material 

costs in overall production costs. The 

effect <strong>of</strong> optimized raw material usage 

is getting more and more important to 

manage production costs and to ensure 

competitive advantages. PSImetals, <strong>the</strong> 

production management solution specialized 

for <strong>the</strong> needs <strong>of</strong> non ferrous metals 

producers, takes this into account by recalculating 

<strong>the</strong> charging for all scheduled 

melt batches giving an overall optimized 

decision support for <strong>the</strong> responsible persons. 

The PSI system supports all aspects <strong>of</strong> production 

management. This includes a knowledge 

based order dressing component to manage 

<strong>the</strong> highest value <strong>of</strong> production know-how. 

The sales and operational planning, due date 

quoting and order related production planning 

processes are covered as well as scheduling 

and execution. All components are directly 

integrated in one common factory model. This 

makes it possible to combine <strong>the</strong> as-is-situation 

<strong>the</strong> short term and even <strong>the</strong> long term 

decision model. 

On executing level starting with <strong>the</strong> goods 

receipt, all material flows are not only tracked 

but controlled. Excavators, wheel loaders 

and fork-lift trucks and its weighing systems 

are integrated by mobile computers. Melting, 

holding, casting and all o<strong>the</strong>r equipments are 

directly conducted by level 2 integration and 

pdc terminals. So all production processes are 

guided and every time online transparent. 

PSImetals with its simple and intuitive human 

interface is leading to short term implementation 

and direct acceptance. The real time connection 

<strong>of</strong> <strong>the</strong> actual analysis measurement 

during melting for <strong>the</strong> optimization <strong>of</strong> <strong>the</strong> 

batch composition and <strong>the</strong> subsequent alloy 

takes into account <strong>the</strong> furnace sump, lumpiness 

and many o<strong>the</strong>r aspects leads to highest 

quality with lowest material costs. Each material 

unit is tracked from <strong>the</strong> ingredients to <strong>the</strong> 

final product. 

The PSI System delivers <strong>the</strong> current stocks, 

and knows <strong>the</strong> planned scrap inflow. The 

proactive control <strong>of</strong> <strong>the</strong> composition optimization 

uses this knowledge to form follow-up 

charges, and thus achieve significantly lower 

usage <strong>of</strong> new material and at <strong>the</strong> same time 

lower scrap stocks at lower costs. 

The tracking and optimizing <strong>of</strong> energy consumption 

as basis for smarter energy procurement 

is also already possible. 

This text gives an overview <strong>of</strong> how to optimize 

material flows within <strong>aluminium</strong> casthouses 

by using a holistic production management 

system. 

Casthouse production management 

Main cost drivers <strong>of</strong> <strong>the</strong> production costs in 

an <strong>aluminium</strong> casthouse are raw material and 

energy costs. Both topics are addressed by 

PSImetals Production Management System 

(PMS). 

To reach targets like 

Minimize material and energy costs 

• Optimize scrap usage by integrated 

alloying 

• Control number <strong>of</strong> alloying steps 

• Track and plan energy consumption 

Secure and optimize production processes, 

especially alloying 

• Integrated testing device and alloying 

recalculation 

• Control deviation handling by customer 

managed workflows 

• Tool management 

Real time transparency & material genealogy 

• Tracking <strong>of</strong> all relevant events 

• Tracking <strong>of</strong> all material items 

• Tracking <strong>of</strong> material genealogy 

first <strong>of</strong> all real time transparency is needed by 

driving and supporting <strong>the</strong> production processes. 

Step 1 – material management 

and material flow control 

The fur<strong>the</strong>r approaches are explained based 

on a casthouse example which is shown in 

Fig. 1: 

Fig. 1: Areas <strong>of</strong> production within a casthouse 

Beginning with a scrap and metal stockyard 

<strong>the</strong> casthouse examples consist <strong>of</strong> melting 

furnaces, casting, sawing, homogenization and 

scalping devices. PSImetals covers all types 

<strong>of</strong> casthouses. The example should point out 

high complexity in different alloys and dimensions 

especially when casting ingots and billets 

is not limited. 

Tracking <strong>of</strong> all types <strong>of</strong> materials is essential 

for <strong>the</strong> fur<strong>the</strong>r steps. The starting point is 

<strong>the</strong> material input in <strong>the</strong> factory. On basis <strong>of</strong> 

notifications and purchase orders <strong>the</strong> input is 

booked and <strong>the</strong> material is ready for fur<strong>the</strong>r 

qualification and/or preparation steps. Each 

movement is driven by PSImetals to ensure 

that every material is a <strong>the</strong> right position in 

accordance to <strong>the</strong> schedules. At each workplace 

<strong>the</strong> corresponding schedule gives <strong>the</strong> 

workers <strong>the</strong>ir program for <strong>the</strong> shift. By activating 

<strong>the</strong> next task related transport orders 

for forklifts, cranes or fully automatic handling 

devices are generated. 

This principle <strong>of</strong> controlling and driving 

<strong>the</strong> material movements, tracking <strong>the</strong> location 

and <strong>the</strong> as-is values <strong>of</strong> each material unit is 

not only valid for raw materials (RM), but 

also for work in process (WIP), semi-finished 

goods (SFG) and finished goods (FG). 

PSImetals provides all needed customizing 

functionality to make all adoptions to <strong>the</strong> customer 

situation as easy as possible. As an example 

<strong>the</strong> built in editor for <strong>the</strong> yard topology 

is mentioned. With this editor <strong>the</strong> customer 

can change his structure directly without any 

programming. This is in <strong>the</strong> casthouse area 

used to change <strong>the</strong> structure <strong>of</strong> boxes in according 

to <strong>the</strong> product mix and <strong>the</strong> procurement 

strategy. 

Step 2 – integrated dynamic alloying 

Based on <strong>the</strong> sufficient transparency <strong>the</strong> real 

benefits are achieved by using <strong>the</strong> built-in al- 


Fig. 2: Optimized raw material calculation using a LP solver always considering given restrictions and 

optimization targets 

loying module which can be used by <strong>the</strong> production 

planning, scheduling or execution. For 

calculating <strong>the</strong> optimized raw material supply 

<strong>the</strong> actual situation, all scheduled batches or 

production orders are taking into account. 

The calculating is down by feeding a LP (linear 

programming) solver. Fig. 2 shows <strong>the</strong> schedule 

generation inclusive material supply. In 

<strong>the</strong> first stage <strong>the</strong> melting and casting batches 

are built up and sequenced for each device. 

With <strong>the</strong> second stage <strong>the</strong> material supply for 

every scheduled batch is calculated. 

Managing energy 

The energy costs are getting more and more 

into <strong>the</strong> focus. The complexity <strong>of</strong> <strong>the</strong> regionally 

different influences and restrictions makes 

it difficult to manage. The forecasting and procurement 

<strong>of</strong> energy and even <strong>the</strong> reactivity in 

adopting <strong>the</strong> production on a short term notice 

to a different energy situation are <strong>the</strong> challenges 

for <strong>the</strong> future. 

PSImetals provides <strong>the</strong> functionality for 

consider <strong>the</strong> various aspects <strong>of</strong> increasing 

energy efficiency, forecasting and production 

control: 

Tracking and short term control 

• High level <strong>of</strong> transparency in energy 

Fig. 3: Energy management to track and control 

<strong>the</strong> energy consumption in production 

consumption 

• Early warning system to avoid violations 

<strong>of</strong> contractual limited peak loads 

• Predicted overall energy consumption for 

operational decision support 

Energy consumption in planning 

and scheduling 

• Energy consumption values – part <strong>of</strong> 

production order routings 

• Consumption forecast for different 

time-frames 

• Support for contract negotiation with 

energy supplier 

Summary 

Potential benefits are summarized in <strong>the</strong> following 

overview: 

Transparency and online material genealogy 

• Decrease <strong>of</strong> reaction time and efforts 

• Integrated alloying 

• Optimized material supply <strong>of</strong> all 

scheduled batches 

• Decrease in production costs 

• Secured alloying process 

• Reduced alloying adjustment steps 

(~ 1,1 steps in average) 

• Support in deviation handling 

• Decrease in material costs by over 10% 

Fig. 4: Managing <strong>the</strong> load peaks <strong>of</strong> your energy 

demand to avoid limit violation costs 


• Additional benefits by adopted 

procurement strategies 

Integrated energy management 

• Minimizing energy costs 

• Basis for new business model ‘selling 

control power’ 

Next steps 

For fur<strong>the</strong>r improvements in adopting <strong>the</strong> 

ma<strong>the</strong>matical models to <strong>the</strong> real world ‘Qualicision’ 

will be used. Qualicision is an extended 

fuzzy logic based optimizing module which 

leads to more human way <strong>of</strong> decision by recalculating 

<strong>the</strong> different parameters <strong>of</strong> <strong>the</strong> linear 

programming model. With <strong>the</strong> Qualicision 

module shows how this level as a qualitative 

decision level will be integrated into <strong>the</strong> system 

structure: 

About PSI Metals 

PSI is <strong>the</strong> leading IT supplier for production 

management solutions in <strong>the</strong> metals industry 

combining supply chain management, 

advanced planning and scheduling, production 

execution and logistics optimization. 

For more than 40 years, we have been delivering 

added-value solutions to maximize <strong>the</strong> 

plant performances <strong>of</strong> numerous metals producers 

around <strong>the</strong> globe. PSImetals s<strong>of</strong>tware 

solutions enable producers <strong>of</strong> <strong>aluminium</strong>, 

steel and copper products to ensure <strong>the</strong>ir competitive 

edge by delivering products as agreed 

in quantity, quality and time whilst considering 

inventory, productivity and performance 

targets. 

The PSImetals solution line is an end-toend 

approach for <strong>the</strong> overall supply chain caring 

for all <strong>the</strong> needs <strong>of</strong> <strong>the</strong> metals industry. 

From your supplier to your customer, PSImetals 

<strong>of</strong>fers powerful and tailor-made products 

to support all processes from planning to 

execution within your supply chain always 

considering <strong>the</strong> complexity <strong>of</strong> metals production: 

• Planning level to support all planning processes 

from Business Planning via Production 

Planning to Detailed Scheduling 

• Execution level to monitor and control production 

activities as well as to assure quality 

• Level <strong>of</strong> material- and transport logistic to 

optimise all transports requested to keep production 

running 

• Energy management level 

• Cross-application KPI and production 

monitoring functions. 

All information is based on an integrated 

factory model for consistent real time plant 

status information. 

� 



Monitoring <strong>of</strong> <strong>aluminium</strong> extrusion dies 

cleaning process by an optic sensor 

A. Pascual Formoso 1 ; E. Piñeiro Ben 1 ; L. Herrero Castilla 1 ; C. Domínguez 2 , A. Llobera 2 

1 2 Aimen Technology Center, Pontevedra, Centro Nacional de Microelectrónica (IMB-CNM, CSIC), 

Barcelona, Universidad Autónoma de Barcelona, Bellaterra (Barcelona) 

The aim <strong>of</strong> this work has been <strong>the</strong> assessment 

<strong>of</strong> <strong>the</strong> use <strong>of</strong> a multiple internal 

reflection sensor in order to monitor <strong>the</strong> 

advance <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> extrusion dies 

cleaning process. The main goal <strong>of</strong> <strong>the</strong> 

sensor is that it is able to measure different 

initial concentrations <strong>of</strong> NaOH and it 

shows <strong>the</strong> point in which <strong>the</strong> aluminate 

starts to precipitate allowing to distinguish 

states <strong>of</strong> <strong>the</strong> bath on concentration 

ratios <strong>of</strong> [Al] / [NaOH] = 0.1. 

The <strong>aluminium</strong> extrusion industry has a significant 

economic importance all over Europe. 

The process is carried out using extrusion dies, 

steel disks with an opening cut through <strong>the</strong>m, 

with <strong>the</strong> size and shape <strong>of</strong> <strong>the</strong> intended crosssection 

<strong>of</strong> <strong>the</strong> final extruded product. An <strong>aluminium</strong> 

billet pre-heated (500ºC) is forced 

to flow through <strong>the</strong> die in order to get <strong>the</strong> 

required pr<strong>of</strong>ile. At <strong>the</strong> end <strong>of</strong> <strong>the</strong> cycle, it is 

necessary to remove all <strong>the</strong> remaining metal 

inside <strong>the</strong> cavities <strong>of</strong> <strong>the</strong> die to reuse it in new 

pieces production. In fact, <strong>the</strong> extrusion dies 

are responsible for quality and process performance 

and <strong>the</strong>ir cost is between 35-50% <strong>of</strong> 

<strong>the</strong> cost <strong>of</strong> manufacture (Guía Tecnológica). 

The ordinary cleaning method consists <strong>of</strong> 

loading <strong>the</strong> dies into open tanks, containing a 

solution <strong>of</strong> NaOH (caustic soda) at high concentration 

and temperature near <strong>the</strong> boiling 

point. The cleaning process may even exceed 

10 hours, depending on die size and <strong>aluminium</strong> 

quantity adhered to its surface. Generally, 

<strong>the</strong> baths are changed after each use without 

making a preliminary analysis to determine 

if <strong>the</strong> bath still retains its cleaning properties. 

Daily bath emptying (which can exceed 3,000 

litre per day) causes a high cost due to <strong>the</strong> 

new formulation reagents (water and sodium 

hydr<strong>oxide</strong>) and energy. In addition, it generates 

a large amount <strong>of</strong> hazardous waste (with 

high basicity and high content <strong>of</strong> dissolved 

<strong>aluminium</strong>) (Guía Tecnológica). 

Several studies (Li et al, 2005a) have shown 

that <strong>aluminium</strong> in a caustic solution at a temperature 

approaching to boiling point leads 

to sodium aluminate (NaAlO 2), resulting in 

hydrogen gas (H 2 (g)). Sodium aluminate so- 

lutions begin to run out, since a mesoscopic 

point <strong>of</strong> view, due to presence <strong>of</strong> <strong>aluminium</strong> 

hydr<strong>oxide</strong>s like gibbsite and bayerite (Harris 

et al, 1999). Optical scattering techniques, like 

Dynamic Light Scattering (DLS), have been 

proven its feasibility to characterize <strong>the</strong> nucleation 

and growth <strong>of</strong> <strong>the</strong>se species. 

The growing interest <strong>of</strong> industry for control 

and monitoring <strong>of</strong> production process 

motivates research in <strong>the</strong> development <strong>of</strong> new 

optical devices that allow <strong>the</strong>ir integration into 

production lines (Harris et al, 1999). These 

systems provide real-time information about 

a process. One <strong>of</strong> <strong>the</strong>se labs on a chip device 

are <strong>the</strong> multiple internal reflection (MIR) systems. 

(Llobera et al, 2007). 

The aim <strong>of</strong> this work has been <strong>the</strong> assessment 

<strong>of</strong> <strong>the</strong> use <strong>of</strong> a low cost MIR sensor 

fabricated in PDMS in order to monitor 

<strong>the</strong> advance <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> extrusion dies 

cleaning process. 

Method 

Solution preparation: Due to <strong>the</strong> impossibility 

<strong>of</strong> working with real extrusion dies at laboratory 

scale, small pieces <strong>of</strong> <strong>aluminium</strong> have 

been used for <strong>the</strong> simulations. Cleaning baths 

are prepared using NaOH solution in a benchscale. 

This bath is heated to <strong>the</strong> temperature 

<strong>of</strong> work and <strong>the</strong> fragment <strong>of</strong> <strong>aluminium</strong> is put 

into <strong>the</strong> bath. Once treatment time is completed, 

<strong>the</strong> attacked <strong>aluminium</strong> piece is removed. 

When <strong>the</strong> bath is cool 

a dark gelatinous compounds 

start to precipitate 

(<strong>aluminium</strong> 

hydr<strong>oxide</strong> and aluminates). 

The <strong>aluminium</strong> 

concentration in solution 

and free sodium 

hydr<strong>oxide</strong> concentration 

are determined in 

each test <strong>of</strong> experimental 

set-up. The ratio <strong>of</strong> 

free sodium hydr<strong>oxide</strong> 

and <strong>aluminium</strong> in so- 

lution gives an idea <strong>of</strong> 

bath exhaustion. 

Optical characterization: This work has been 

carried out to characterize <strong>the</strong> optical properties 

<strong>of</strong> samples such as refractive index, transmittance 

spectral response and scattering. 

The size distribution has been analyzed by 

dynamic light scattering, from samples prepared 

according to <strong>the</strong> procedure mentioned 

previously (to 95ºC from a solution <strong>of</strong> initial 

NaOH concentration <strong>of</strong> 19%). The size <strong>of</strong> particles 

suspended in <strong>the</strong> bath has been determined 

using a Spectrometer Autosizer 4800 

<strong>of</strong> Malvern Instruments (PCS technique). The 

refractive index has been measured at different 

times (with a refractometer ABBEMAT- 

HP) for so what‘s <strong>the</strong> bath depletion point. 

Reference measurement system: Experiments 

with laboratory samples have been 

conducted to evaluate <strong>the</strong> behaviour <strong>of</strong> <strong>the</strong> 

samples measured by <strong>the</strong> sensor and verify 

<strong>the</strong> feasibility <strong>of</strong> <strong>the</strong> proposed detection 

methods, as well as <strong>the</strong> needs and tolerances 

<strong>of</strong> <strong>the</strong> coupling and sealing systems. For <strong>the</strong>se 

tests, a setup similar to <strong>the</strong> proposed sensor 

has been used, but discrete optical elements 

were used instead <strong>of</strong> waveguides, as a reference 

measurement system. This assembly allows 

to develop transmission and dispersion 

measurements at 90° simultaneously. The light 

is injected through a multimode optical fibre 

and collected with plastic optical fibres <strong>of</strong> 1 

mm in diameter. 

In order to improve <strong>the</strong> signal to noise ratio 

and eliminate <strong>the</strong> contribution <strong>of</strong> light which is 

Fig. 1: Attenuation and dispersion at 90° for solutions caustic soda with different 

concentrations <strong>of</strong> <strong>aluminium</strong> Images: Aimen 


necessary in <strong>the</strong> specific case <strong>of</strong> <strong>the</strong> measurement 

<strong>of</strong> dispersion, <strong>the</strong> laser is modulated to 

low frequencies (721 Hz) and <strong>the</strong> response <strong>of</strong> 

both detectors is processed via a synchronic 

detection algorithm or lock-in. 

To perform <strong>the</strong>se tests, aliquots <strong>of</strong> soda/ 

<strong>aluminium</strong> solution are sampled at different 

reaction times. The measurements made 

with <strong>the</strong> reference system allow observing <strong>the</strong> 

dependence <strong>of</strong> <strong>the</strong> transmission and optical 

dispersion <strong>of</strong> <strong>the</strong> sample with reaction time. 

Fig. 1 show <strong>the</strong> measurements obtained in dB 

relative to <strong>the</strong> power transmitted to <strong>the</strong> sample 

<strong>of</strong> NaOH (blank) and correlation coefficients 

<strong>of</strong> a linear fit. 

MIR based sensor: From a fluidic point <strong>of</strong> 

view, <strong>the</strong> proposed system consist <strong>of</strong> a MIR 

structure directly connected to fluidic input/ 

output reservoirs. The sensor operation is 

based on circulating fluid through <strong>the</strong> guide 

MIR. The light is injected through optical fibres 

and light is collected after travelling part 

way through <strong>the</strong> fluid (Fig. 2) (Llobera et al, 

2007). The measured magnitude is <strong>the</strong> loss 

<strong>of</strong> irradiance associated with <strong>the</strong> presence <strong>of</strong> 

particles relevant process in question, which 

in this particular case are compounds <strong>of</strong> <strong>aluminium</strong> 

(<strong>aluminium</strong> particles and <strong>aluminium</strong> 

hydr<strong>oxide</strong>). 

In order to study <strong>the</strong> process that takes 

place in <strong>the</strong> cleaning bath, <strong>the</strong> alkaline solution 

should be circulated through <strong>the</strong> interior 

<strong>of</strong> <strong>the</strong> guide MIR. 

Fig. 2: Experimental setup 

Results 

The experimental tests performed to validate 

<strong>the</strong> optical sensor at <strong>the</strong> laboratory, demonstrate 

<strong>the</strong> capacity <strong>of</strong> <strong>the</strong> device to determine 

optimal initial conditions and <strong>the</strong> point <strong>of</strong> depletion 

<strong>of</strong> <strong>the</strong> bath, allowing <strong>the</strong> optimization 

process. 

Resistance testing: Some experiments have 

been conducted in order to test <strong>the</strong> resistance 

<strong>of</strong> materials <strong>of</strong> MIR guide. The device is immersed 

in a bath (19% [NaOH], 95ºC) during 

a time <strong>of</strong> 8 hours. Fig. 3, shows <strong>the</strong> response 

<strong>of</strong> MIR guide after <strong>the</strong> resistance test on a several 

samples. The curve for <strong>the</strong> soda solution is 

similar to <strong>the</strong> curve produced by <strong>the</strong> transmission 

<strong>of</strong> water, while <strong>the</strong> signal associated with 

<strong>the</strong> sample with <strong>aluminium</strong> is an 

order <strong>of</strong> magnitude lower than <strong>the</strong> 

previous one. The presence <strong>of</strong> <strong>aluminium</strong> 

particles lead to losses to 

reduce light scattering coupled to 

<strong>the</strong> output fibre. The sensor is not 

damaged in its functional capacity. 

The materials used for monitoring 

purposes are feasible for at least 

one cycle <strong>of</strong> industrial process. 

Process monitoring: The aim 

<strong>of</strong> this study has been <strong>the</strong> assessment 

<strong>of</strong> <strong>the</strong> use <strong>of</strong> an optic sensor 

in order to monitor <strong>the</strong> advance 

<strong>of</strong> <strong>the</strong> <strong>aluminium</strong> extrusion dies 

cleaning process, and estimate <strong>the</strong> 

end point <strong>of</strong> <strong>the</strong> bath. Aliquots <strong>of</strong> 

caustic soda (NaOH 14%, 95°C) / 

<strong>aluminium</strong> solution are sampled 

at different reaction times have 

been injected into <strong>the</strong> guide MIR. 

Fig. 4 illustrates <strong>the</strong> signal decay 

with increasing time <strong>of</strong> reaction 

and <strong>the</strong>refore decreases as <strong>the</strong> 

ratio [NaOH] / [Al]. These variations 

are due to <strong>the</strong> formation and 

growth <strong>of</strong> particles in <strong>the</strong> sample. 

In Fig. 5 (see next 

page) two characteristic 

regions can be distinguished. 

In <strong>the</strong> region 

between 10 and 60 min 

a rapid decrease <strong>of</strong> <strong>the</strong> 

ratio [NaOH]/[Al] is observed toge<strong>the</strong>r 

with a low sensitivity <strong>of</strong> <strong>the</strong> 

signal <strong>of</strong> <strong>the</strong> device (0.4 ±0.06 dB/ 

(M/M)). The second region is located 

above 60 min. Here <strong>the</strong> change is 

less pronounced but device sensitivity 

<strong>of</strong> <strong>the</strong> device is higher (3.7 ±0.5 

dB/(M/M)) for ratios <strong>of</strong> [NaOH]/ 

[Al] inferior to 3. Starting from this 

point <strong>the</strong> values converge to bath 

saturation ones. In <strong>the</strong> first region, <strong>the</strong> concentration 

<strong>of</strong> <strong>the</strong> dissolved <strong>aluminium</strong> increases 

but <strong>the</strong> particle size is below <strong>the</strong> wavelength 

<strong>of</strong> <strong>the</strong> emitted light so that no signal <strong>of</strong> scattering 

is recorded. This explains <strong>the</strong> low intensity 

measured by <strong>the</strong> device. However, in <strong>the</strong> second 

region <strong>the</strong> particle size is superior to <strong>the</strong> 

selected wavelength and <strong>the</strong> sensitivity <strong>of</strong> <strong>the</strong> 

device increases. These results are in accordance 

with <strong>the</strong> observations made by Harris et 


al, 1999, regarding <strong>the</strong> measurement <strong>of</strong> <strong>the</strong> 

induction period, from where a linear growth 

<strong>of</strong> <strong>the</strong> particles present in <strong>the</strong> saturated solution 

<strong>of</strong> Al in NaOH at high temperature is 

observed. 

The maximum sensitivity <strong>of</strong> <strong>the</strong> sensor for 

process monitoring is achieved with a wave- 

Fig. 3: Device response for a sample <strong>of</strong> distilled water, [NaOH] 

5.94 M and [NaOH]/[Al] 0.78 

Fig. 4: Spectral response <strong>of</strong> <strong>the</strong> measurand for each time <strong>of</strong> <strong>the</strong> 

bath 

length <strong>of</strong> 577 nm with a linear slope <strong>of</strong> 0.041 

±0.004 dB / min. This allows to distinguish 

states <strong>of</strong> <strong>the</strong> bath on concentration ratios <strong>of</strong> 

[Al] / [NaOH] = 0.1. 

Initial conditions: Laboratory scale assays to 

measure <strong>the</strong> transducer response for different 

initial conditions <strong>of</strong> <strong>the</strong> baths show that <strong>the</strong> device 

has <strong>the</strong> capacity to distinguish different 

initial concentrations <strong>of</strong> caustic soda. Different 

samples <strong>of</strong> NaOH solution (10%, 14% and 

19% concentration) have been injected into 

<strong>the</strong> device. Considering <strong>the</strong> dissolution <strong>of</strong> <strong>the</strong> 

10% as blank, an increasing to 19% NaOH 

concentration represents an increase in <strong>the</strong> signal 

<strong>of</strong> 1.04 dB (Fig. 6). These increases in signal 

are due to changes in <strong>the</strong> refractive index <strong>of</strong> 

<strong>the</strong> different solutions: 1.360 (NaOH 10%) 

1.367 (NaOH 14%) 1.379 (19% NaOH). 

The sensitivity <strong>of</strong> <strong>the</strong> device to scattering 

losses and refractive index changes enable to 



Fig. 5: Evolution <strong>of</strong> intensity losses and ratio [NaOH]/[Al] in function 

<strong>of</strong> time at laboratory conditions 

determine as <strong>the</strong> optimal initial conditions as 

<strong>the</strong> exhaustion point <strong>of</strong> <strong>the</strong> reaction. 

Conclusions 

This work aims <strong>the</strong> control and analysis <strong>of</strong> <strong>the</strong> 

industrial process <strong>of</strong> <strong>aluminium</strong> extrusion die 

www 

<strong>aluminium</strong>ePaper.com 

Please be our guest 

and discover <strong>the</strong> benefits <strong>of</strong> <strong>the</strong> 

ALUMINIUM-ePaper yourself in a 

free three-month trial: 

� accessible at least a week 

before <strong>the</strong> printed edition 

� available from any location 

� simple download 

� keyword researches 

� linked list <strong>of</strong> contents 

� direct contact with advertisers 

cleaning using optical sensors. An integrated 

optical technology, with capacity for monitoring 

industrial processes in conditions <strong>of</strong> very 

extreme temperature and pH has been developed. 

The results show that <strong>the</strong> prototype exceeds 

all critical requirements. 

The data obtained after <strong>the</strong> validation 

demonstrate <strong>the</strong> applicability 

<strong>of</strong> optical 

sensors for monitoring 

bath dedicated 

to <strong>aluminium</strong> dissolution. 

Absorption 

and scattering losses, 

related to <strong>aluminium</strong> 

particles present in 

<strong>the</strong> cleaning solutions, 

have been correlated 

with initial 

caustic solution conditions. 

Fur<strong>the</strong>rmore, 

<strong>the</strong> degradation study 

<strong>of</strong> <strong>the</strong> sensor reflects 

after use for an industrial 

bath, <strong>the</strong> device 

does not suffer 

damage affecting <strong>the</strong> 

proper operation. 

The results <strong>of</strong> this 

research indicate that 

<strong>the</strong> application <strong>of</strong> 

this optic sensor as a 

monitoring technique 

in <strong>aluminium</strong> caustic 

solutions is feasible. 

The industrial application 

<strong>of</strong> this sensor 

would help <strong>aluminium</strong> 

companies to 

select <strong>the</strong> most suitable 

cleaning opera- 

Fig. 6: Spectral response <strong>of</strong> different solutions <strong>of</strong> NaOH 

according to its requirements. By stating <strong>the</strong> 

amount <strong>of</strong> dissolved <strong>aluminium</strong>, <strong>aluminium</strong> 

industry could work under different conditions 

<strong>of</strong> sodium hydr<strong>oxide</strong> concen tration, temperature 

and time. This would mean a decrease 

in consumption <strong>of</strong> raw materials (caustic soda 

and water), energy and a reduction in effluents 

and waste management. Monitoring is a useful 

investment with wide practical benefits. 

Acknowledgements 

This work was supported by project <strong>of</strong> <strong>the</strong> 

Ministerio de Industria, Turismo y Comercio. 

Secretaría de Estado de Telecomunicaciones y 

para la Sociedad de la Información Dirección 

General para el Desarrollo de la Sociedad de 

la Información, TSI-020301-2008-11. The assistance 

<strong>of</strong> Iñigo Salinas and David Izquierdo 

<strong>of</strong> University <strong>of</strong> Zaragoza for design and 

measurement implementation with reference 

system is gratefully acknowledged. 

Bibliography 

Guía Tecnológica. Directiva 96/61 relativa a la prevención 

y control integrados a la prevención y control 

integrados de la contaminación. Epígrafe 2.5. 

Metalurgia de aluminio. 

Harris D.R., Keir R.I., Prestidge C.A., Thomas J.C., 

1999. A dynamic light scattering investigation <strong>of</strong> 

nucleation and growth in supersaturated alkaline 

sodium aluminate solutions (syn<strong>the</strong>tic Bayer liquors), 

Colloids and Surfaces A, p. 154 343-352. 

Li H.; Addai-Mensah J.; Thomasb J. C.; Gersona A. 

R. 2005a. The influence <strong>of</strong> Al(III) supersaturation 

and NaOH concentration on <strong>the</strong> rate <strong>of</strong> crystallization 

<strong>of</strong> Al(OH)3 precursor particles from sodium 

aluminate solutions. Journal <strong>of</strong> Colloid and Interface 

Science 286, p. 511-519. 

Llobera A. Demming S.; Wilkea R. and Bu¨ttgenbacha 

S.; 2007. Multiple internal reflection poly (dimethylsiloxane) 

systems for optical sensing. Advance Article 

on <strong>the</strong> www.rsc.org/loc | Lab on a Chip. 

tion conditions and � 


<strong>Influence</strong> <strong>of</strong> raw material constitution on <strong>the</strong> quality <strong>of</strong> 

gloss alloy-type extrusion billets – a report from practice 

Klaus H<strong>of</strong>fmann, Ralf Munk, Marcel Rosefort and Dietmar Bramh<strong>of</strong>f, Trimet Aluminium AG 

The Trimet group provides <strong>the</strong> <strong>aluminium</strong> 

processing industry with primary metal, 

alloys, semi-finished products, castings 

and recycling. The competences <strong>of</strong> <strong>the</strong> 

business unit Primary Products are electrolysis 

and continuous casting. Products 

are rolling slabs, extrusion and forging 

billets, hot dip-coating alloys, initial products 

for safety parts and liquid <strong>aluminium</strong>. 

Due to <strong>the</strong> demand for high quality 

and services regarding <strong>the</strong>se products in 

particular <strong>of</strong> <strong>the</strong> automotive industry it 

is essential to enforce and secure quality 

control. All <strong>the</strong>se high quality products 

from Trimet, especially <strong>the</strong> safety parts, 

<strong>the</strong> gloss alloys and parts with functional 

surfaces for automotive applications, are 

characterized by an extremely homogenous 

microstructure without 

internal defects or inclusions. 

Within our permanent quality 

improvement programme an examination 

has been carried out to 

measure <strong>the</strong> influence <strong>of</strong> raw material 

constitution on <strong>the</strong> product 

quality. Therefore <strong>the</strong> use <strong>of</strong> liquid 

metal, ingot metal and mixtures <strong>of</strong> 

<strong>the</strong>se has been tested. The paper 

covers <strong>the</strong> essential production 

steps for such challenging high 

quality billets. It gives a report 

on <strong>the</strong> quality requirements, <strong>the</strong> 

qualifying <strong>of</strong> <strong>the</strong> alloying material and <strong>the</strong> 

process parameters. Especially <strong>the</strong> methods 

<strong>of</strong> analysis, like PoDFa and inclusion 

identification, and <strong>the</strong> correlations between 

<strong>the</strong> process parameters, <strong>the</strong> feedstock 

and metal quality are described. 

The market for extrusion billets for bright surface 

products in Europe is bigger than 15,000 

tonnes a year. The highest demand for this 

special material comes from <strong>the</strong> automotive 

industry. There it is used for trim parts, ro<strong>of</strong> 

racks and cover plates. But <strong>the</strong>re is also a 

market for this material in <strong>the</strong> lighting and audio 

industry. One can say that <strong>the</strong> time <strong>of</strong> <strong>the</strong> 

black coated trims is over – for <strong>the</strong> moment. 

To fulfil <strong>the</strong> very challenging specifications 

for <strong>the</strong> final parts, <strong>the</strong> whole process from <strong>the</strong> 

alumina for <strong>the</strong> electrolyses up to <strong>the</strong> anodising 

<strong>of</strong> <strong>the</strong> final part has to be fully tuned and 

straight designed. All changes in this process 

chain have to be scrutinised, and all changes 

have to be discussed and well co-ordinated 

between <strong>the</strong> partners <strong>of</strong> this sensitive process 

chain. 

So, why change such a sensitive process? 

If we want to improve <strong>the</strong> quality <strong>of</strong> <strong>the</strong> end 

product, we will have to change <strong>the</strong> process 

from time to time, or <strong>the</strong>re are technical needs 

to change a step in <strong>the</strong> process. This paper discusses 

<strong>the</strong> influence <strong>of</strong> <strong>the</strong> solid/liquid metal 

ratio. 

The trimal BQ process 

Trimet developed a special process named BQ 

(Brilliant Quality) to fulfil all <strong>the</strong>se requirements 

for a brilliant end product. 

The trimal BQ process is a special process 

for producing extrusion billets, to fulfil high 

requirements and to cover all <strong>the</strong> production 

steps and parameters that are adjusted for <strong>the</strong> 

end product: <strong>the</strong> alumina, melting, casting, homogenizing, 

testing. 

This process includes all parts <strong>of</strong> <strong>the</strong> pro- 

Production flow in <strong>the</strong> casthouse 

Images: Trimet 

EXTRUSION 

duction chain starting with <strong>the</strong> alumina and 

anodes for <strong>the</strong> electrolysis at <strong>the</strong> Essen or 

Hamburg smelter to produce <strong>the</strong> liquid and 

solid raw material. Followed by alloying, metal 

treatment, grain refining and casting, finished 

up with <strong>the</strong> customised homogenizing process 

unique for <strong>the</strong>se billets. 

The electrolysis process 

At <strong>the</strong> electrolysis <strong>the</strong>re are several pots operated 

as high purity pots. For <strong>the</strong>se pots <strong>the</strong> 

alumina is not used in <strong>the</strong> waste gas purification 

process to avoid <strong>the</strong> enrichment <strong>of</strong> o<strong>the</strong>r 

elements. The anodes are hand selected according 

to <strong>the</strong>ir metal content and <strong>the</strong>ir operating 

life in <strong>the</strong> cells is shorter than in <strong>the</strong> 

standard pots. 

The control <strong>of</strong> <strong>the</strong> pots is done by 

<strong>the</strong> Trimet 9-box matrix model [1]. This 

is increasing current efficiency and reducing 

energy consumption. After <strong>the</strong> 

transport, crucible with <strong>the</strong> pot room 

metal is arriving at <strong>the</strong> casthouse <strong>the</strong> 

dross is accurately removed from <strong>the</strong> 

metal surface. Now <strong>the</strong> metal is transferred 

to one <strong>of</strong> <strong>the</strong> melting furnaces or 

casted into sow forms. 

Melting and casting 

The process in <strong>the</strong> casthouse (see flow 

chart) starts with <strong>the</strong> loading <strong>of</strong> solid metal into 

<strong>the</strong> gas fired melting furnace. When <strong>the</strong> solid 

metal is heated up to <strong>the</strong> required temperature 

<strong>the</strong> hot pot room metal (approx. 930°C) 

is added to <strong>the</strong> furnace. In this step we use <strong>the</strong> 

energy which is stored in <strong>the</strong> pot room metal 

to bring <strong>the</strong> sows into <strong>the</strong> liquid state. The liq- 


EXTRUSION 

Grain refiner rod (AlTixBx ) 

• Chemical analysis 

• Efficiency 

• Metallography 

Oxide, <strong>oxide</strong> lines 

TiB2 agglomerate 

Al3Ti - phases 

Grain refiner input [2] 

uid metal will <strong>the</strong>n be mixed thoroughly. 

For <strong>the</strong> quality <strong>of</strong> <strong>the</strong> end product <strong>the</strong> quality 

<strong>of</strong> <strong>the</strong> metal input is essentially important. If 

low quality base material is used, <strong>the</strong> production 

<strong>of</strong> a clean melt will be impossible. At Trimet 

<strong>the</strong> hot metal is analysed and melt-dross 

is skimmed <strong>of</strong>f; when cold metal is used, only 

ingots from in-house production are added to 

<strong>the</strong> pot room metal. 

After skimming <strong>the</strong> surface <strong>of</strong> <strong>the</strong> bath, 

a sample will be taken and <strong>the</strong> required elements 

will be brought into <strong>the</strong> batch. This 

addition will be done in <strong>the</strong> appropriate sequence 

and at <strong>the</strong> needed temperature. For <strong>the</strong> 

alloying we use master alloys from certified 

suppliers. The melt will be stirred, skimmed 

and a sample will be taken. If <strong>the</strong> analysis is 

Filter box for <strong>the</strong> continuous casting [3] 

Unacceptable <strong>oxide</strong> lines and TiB 2 

agglomerates in a grain refiner rod 

within <strong>the</strong> range given by <strong>the</strong> customer and 

<strong>the</strong> temperature at <strong>the</strong> demanded level, <strong>the</strong> 

melt will be transferred into <strong>the</strong> inductive 

heated casting furnace. From this point on 

<strong>the</strong>re is no fur<strong>the</strong>r addition <strong>of</strong> any alloying 

agent. The only exception is <strong>the</strong> grain refiner 

rod, which is added process-related during <strong>the</strong> 

casting. All delivered batches <strong>of</strong> grain refiner 

are tested for chemical analysis, <strong>oxide</strong> lines, 

TiB 2 agglomerates and Al 3Ti particle size. 

In <strong>the</strong> following it is very important to prevent 

any turbulence to avoid <strong>the</strong> forming <strong>of</strong> 

dross in <strong>the</strong> melt. The melt cleaning process is 

divided into three separate steps. 

In <strong>the</strong> casting furnace <strong>the</strong> first step <strong>of</strong> <strong>the</strong> 

metal cleaning is 

done. The melt is 

cleaned by a mixture 

<strong>of</strong> Ar and Cl, 

this is dispersed 

by a rotary gas 

injector into <strong>the</strong> 

melt. During this 

step sodium, calcium, 

hydrogen 

and non-metallic 

inclusions are removed. 

After <strong>the</strong> 

second step, a suf- 

Automated casting using <strong>the</strong> spout and floater technique at <strong>the</strong> Trimet plant in Essen [4] 

ficient holding time, <strong>the</strong> casting is started and 

during casting <strong>the</strong> third step <strong>of</strong> <strong>the</strong> metal cleaning 

is operated in-line by a CFF (see Fig. 3). 

Usually we use 40 ppi CCFs for <strong>the</strong> trimal BQ 

process. For <strong>the</strong> automated casting we use <strong>the</strong> 

spout and floater technique, which has some 

advantages concerning <strong>the</strong> inclusion content. 

With this technique we are able to cast all our 

alloys in <strong>the</strong> desired quality. 

Quality control 

First a short overview <strong>of</strong> <strong>the</strong> inspections and 

tests which have to be done before, during and 

after <strong>the</strong> BQ process to secure <strong>the</strong> expected 

brilliant quality. 

Melting: analysis <strong>of</strong> hot and cold metal and <strong>of</strong> 

master alloys. 

Metal treatment: analysis, quantities, times, 

filter. 

Casting: H 2 measurement, analysis at casting 

start, casting parameter, analysis at end casting, 

PoDFA samples. 

First inspection: ultrasonic test, length, shape, 

surface defects. 

Homogenization: recipe, temperature, time, 

cooling. 

Structure: cracks, inclusions, porosity, coarse 

grain, surface defects. 

Of particular importance during quality 

control in <strong>the</strong> cast shop are <strong>the</strong> melt analysis 

concerning H 2 and inclusions (PoDFA-measurements), 

<strong>the</strong> continuous control <strong>of</strong> all casting 

parameters and <strong>the</strong> ultrasonic testing after 

casting. 

These checks are supplemented by metallographic 

quality control in <strong>the</strong> Trimet laboratory. 

This means detailed controlling <strong>of</strong> possible 

cracks, inclusions, porosity, grain structure, 

etc. 

Trials about <strong>the</strong> solid / liquid ratio 

So why should we change a stable and successful 

process? One <strong>of</strong> <strong>the</strong> reasons was <strong>the</strong> 

crisis that started in 2009. Trimet had to reduce 

its output <strong>of</strong> <strong>the</strong> casthouse and to switch 

<strong>of</strong>f around two thirds <strong>of</strong> <strong>the</strong> electrolyses capacity. 

This implied that <strong>the</strong> number <strong>of</strong> high 

purity pots was decreased, too. This resulted 

in a lack <strong>of</strong> high purity metal for a trimal BQ 

campaign. 

The trials were supposed to answer <strong>the</strong> 

question whe<strong>the</strong>r <strong>the</strong>re is a difference in <strong>the</strong> 

quality <strong>of</strong> <strong>the</strong> billets when <strong>the</strong>y are produced 

from 100% solid metal or 100% liquid metal 

or 50/50%. 

The melt quality was checked with <strong>the</strong> 

PoDFA method. The Porous Disc Filtration 

Apparatus is a method by which liquid metal 


Principle <strong>of</strong> taking PoDFA-samples [5] 

from <strong>the</strong> melt is pressed by pressed air or 

sucked by vacuum through a ceramic filter 

disc (see Fig. 5). Inclusions are concentrated 

by filtration in front <strong>of</strong> <strong>the</strong> filter disc. This area 

has to be prepared with metallographic techniques 

for <strong>the</strong> microscopic examination <strong>of</strong> <strong>the</strong> 

sample and <strong>the</strong> identification <strong>of</strong> <strong>the</strong> inclusion 

types. The evaluation has to be done by an 

experienced metallographer who is counting 

<strong>the</strong> covered area <strong>of</strong> <strong>the</strong> single inclusions. The 

concentration is given for each type <strong>of</strong> inclusion 

by <strong>the</strong> covered area in mm 2 /kg filtered 

<strong>aluminium</strong>. The inclusions are identified by 

<strong>the</strong>ir colour and shape with <strong>the</strong> assistance <strong>of</strong> 

<strong>the</strong> PoDFA-catalogue. 

For <strong>the</strong> identification <strong>of</strong> unknown inclu- 

Typical inclusions in <strong>aluminium</strong> melt [5] 

sions a scanning electron 

microscope with 

energy dispersive Xray 

is helpful. 

The advantage <strong>of</strong> 

<strong>the</strong> PoDFA method 

is <strong>the</strong> possibility to 

identify <strong>the</strong> nature <strong>of</strong> 

<strong>the</strong> inclusions and to 

count <strong>the</strong>ir frequency. 

A handicap <strong>of</strong> this 

technique is that <strong>the</strong> 

result <strong>of</strong> <strong>the</strong> test is a 

snap-shot <strong>of</strong> just one 

moment. This is <strong>the</strong> reason why <strong>the</strong> sampling 

has to be done very carefully and well planned 

to get repeatable results. It has to be scheduled 

where <strong>the</strong> sampling will be done, when and 

how <strong>of</strong>ten it has to be done. 

As metal input for <strong>the</strong>se tests we used liquid 

pot room metal and cold pot room metal as <strong>the</strong> 

solid raw material. For <strong>the</strong> tests we produced 

a 6,000 series with approx. 0,5% silicon and 

magnesium, which is usually used for automotive 

applications. The PoDFA samples were 

taken at <strong>the</strong> beginning and at <strong>the</strong> end <strong>of</strong> <strong>the</strong> 

castings. One sample was taken in front <strong>of</strong> <strong>the</strong> 

filter and one after <strong>the</strong> filter. 

These tests were also used to verify <strong>the</strong> performance 

<strong>of</strong> <strong>the</strong> ceramic foam filter system. 

The results <strong>of</strong> both parameters for six casting 

heats are shown in <strong>the</strong> figure. There is no significant 

difference in <strong>the</strong> inclusion concentration 

in <strong>the</strong> product whe<strong>the</strong>r 100% liquid pot 

room metal is used as metal input or 100% 

solid potroom metal or 50% <strong>of</strong> each is used as 

metal input. The comparison <strong>of</strong> <strong>the</strong> PoDFA results 

at <strong>the</strong> beginning <strong>of</strong> <strong>the</strong> casting and at <strong>the</strong> 

end <strong>of</strong> <strong>the</strong> casting shows that <strong>the</strong> filter system 

is capable to reduce all kind <strong>of</strong> inclusions. 

The comparison <strong>of</strong> <strong>the</strong> PoDFA results 

shows that <strong>the</strong> filter system is working effec- 

PoDFA measurements to compare inclusion amount using hot metal and cold metal (50%), before and after <strong>the</strong> ceramic foam filter-system 

SESSION EXTRUSION 

tively and is reducing <strong>the</strong> content <strong>of</strong> all types 

<strong>of</strong> inclusions in <strong>the</strong> melt. The filter system is 

working solid during <strong>the</strong> whole casting run. 

Conclusion 

The exceedingly sophisticated standard <strong>of</strong> <strong>the</strong> 

production process <strong>of</strong> Trimet Aluminium as 

well as <strong>the</strong> intensive fur<strong>the</strong>r development <strong>of</strong> 

a product will also satisfy <strong>the</strong> growing quality 

requirements in <strong>the</strong> future. The sum <strong>of</strong> process 

steps, quality control and quality management 

is <strong>the</strong> basis <strong>of</strong> <strong>the</strong> high standard <strong>of</strong> Trimet 

quality. 

In addition <strong>the</strong> metal input is essential for 

<strong>the</strong> product quality. The advantage <strong>of</strong> Trimet 

is <strong>the</strong> use <strong>of</strong> own liquid and solid metal for 

<strong>the</strong> production. Trials have shown that <strong>the</strong>re 

is no significant difference between 100% 

liquid pot room metal and 100% solid pot 

room metal or any mixture as metal input for 

<strong>the</strong> product quality <strong>of</strong> <strong>the</strong> trimal BQ process. 

Due to this great variability concerning <strong>the</strong> 

metal input, economic fluctuations, such as <strong>the</strong> 

above mentioned crisis, induce no quality variations 

<strong>of</strong> <strong>aluminium</strong> production at Trimet. 

References 

[1] Light Metals 2003: Increased Current Efficiency 

And Reduced Energy Consumtion At The Trimet 

Smelter Essen Using 9 Box Matrix Control, [pp. 

449ff] T.Rieck, M. Iffert, P. White, R. Rodrigo and 

P. Kelchtermans 

[2] Light Metals 2008: Producing Bright Shining 

Products For Special Applications – Experience 

from Practice”, J. Dressler, D. Bramh<strong>of</strong>f, C. Deiters, 

H. Koch 

[3] Fa. Drache Umwelttechnik 

[4] C. Kammer: Aluminium Taschenbuch, Aluminium 

Verlag Düsseldorf 

[5] ABB: PoDFA – Inclusion Identification And 

Quantification Analysis. � 


EXTRUSION 

Heavy duty extrusion presses for large pr<strong>of</strong>ile applications 

Axel Bauer, SMS Meer GmbH 

SMS Meer, a member <strong>of</strong> <strong>the</strong> SMS group, 

is currently building several heavy extrusion 

presses for large pr<strong>of</strong>ile applications. 

Two <strong>of</strong> <strong>the</strong>se presses have a nominal 

pressing force <strong>of</strong> 150 MN and are considered 

as <strong>the</strong> largest state-<strong>of</strong>-<strong>the</strong>-art frontloading 

presses in <strong>the</strong> world. Typical applications 

for those presses for large pr<strong>of</strong>iles 

are among o<strong>the</strong>rs railway pr<strong>of</strong>iles. 

Looking at <strong>the</strong> market for railway pr<strong>of</strong>iles, it is 

clearly seen that it is a tight market with only 

few global p<strong>layer</strong>s, e. g. Constellium (formerly 

Alcan Engineered Products) or Aleris, but also 

some growing competitors from Far East like 

<strong>the</strong> Jilin Midas Group. 

The global distribution <strong>of</strong> <strong>aluminium</strong> railway 

applications shows regional differences: 

in North America <strong>the</strong> passenger transportation 

system is more or less limited to commuter 

service within <strong>the</strong> great metropolitan areas 

only, while in Europe in general <strong>the</strong> passenger 

volume is very high. On <strong>the</strong> o<strong>the</strong>r hand freight 

transportation by rail is very common in North 

America, whereas in Europe we have mainly 

freight transportation by road. For China, 

<strong>the</strong> most populous territorial state, passenger 

transportation by train is getting more and 

more important. 

All <strong>the</strong> fast developing countries like China, 

Russia, India or Brazil <strong>the</strong>se days do have 

infrastructural programmes which include 

new railway lines. These projects comprise 

high-speed lines for <strong>the</strong> long distance connections 

as well as investments in Intercity 

connections, commuter trains or metros. China’s 

aim is to build 20.000 km <strong>of</strong> high-speed 

tracks until 2020 while in Russia an investment 

programme <strong>of</strong> 7,5 billion euros for <strong>the</strong> erection 

<strong>of</strong> 11.000 km <strong>of</strong> high-speed rail tracks 

has started recently. Also India, which has <strong>the</strong> 

world’s largest railway network, is looking for 

fundamental modernization. 

Even in those projects where normally <strong>the</strong> 

big p<strong>layer</strong>s like Siemens, Bombardier or Alstom 

are in involved, <strong>the</strong>re is always a certain 

percentage <strong>of</strong> local content in <strong>the</strong> supplies, 

i. e. <strong>the</strong> <strong>aluminium</strong> extrusions are produced 

locally. 

This is one explanation for <strong>the</strong> big invest- 

View <strong>of</strong> <strong>the</strong> manipulator for loose dummy block handling and <strong>the</strong> 52,000-litre hydraulic tank 

ments in large extrusion presses in China. 

The International Union <strong>of</strong> Railways (UIC) 

is expecting that <strong>the</strong> global high-speed network 

will triple within <strong>the</strong> next 15 years. 

Taking into account that this is only an outlook 

on <strong>the</strong> high-speed network, a huge demand for 

railway extrusions is expected. 

Aluminium is not <strong>the</strong> only material railway 

cars can be built with. Steel is a permanent 

competitor. There are even some ideas to use 

composite material for certain areas <strong>of</strong> <strong>the</strong> car 

body. But <strong>the</strong> comparison between <strong>the</strong> integrated 

<strong>aluminium</strong> design and <strong>the</strong> steel structure 

design clearly shows a weight reduction 

<strong>of</strong> more than 20 percent on <strong>aluminium</strong> design. 

This is resulting in an estimated energy saving 

<strong>of</strong> five percent for every ten percent <strong>of</strong> 

weight reduction. The limited axle load <strong>of</strong> <strong>the</strong> 

European railway network only allows for 

higher pay loads if <strong>the</strong> net weight <strong>of</strong> <strong>the</strong> train 

is reduced. 

Additionally <strong>the</strong>re are o<strong>the</strong>r known advantages 

<strong>of</strong> <strong>aluminium</strong> like nearly 100 percent 

recyclability, good corrosion resistance, high 

strength to weight ratio or <strong>the</strong> decorative as- 


Photos: SMS Meer

Counter platen with fixed tie rods 

EXTRUSION 

pect <strong>of</strong> <strong>aluminium</strong>. Looking at <strong>the</strong> global greenhouse gas emissions, 

27 percent <strong>of</strong> <strong>the</strong>se emissions are caused by transportation 

and 73 percent are caused by road transportation, while 

only 2 percent arise due to rail transportation. Rail transportation 

is a ‘green solution’ to move people or goods and a low 

carbon technology. A slogan <strong>of</strong> <strong>the</strong> railway manufacturer Bombardier 

is pointing out, that ‘<strong>the</strong> climate is right for trains!’ 

Eventually all this leads to an increasing market for large 

pr<strong>of</strong>iles and since <strong>the</strong> numbers <strong>of</strong> operative heavy presses were 

limited, now big investments in extrusion plants for heavy duty 

and large sections are being made, especially in China. 

But it is not only <strong>the</strong> number <strong>of</strong> presses; it is also <strong>the</strong> size 

<strong>of</strong> <strong>the</strong> presses that is constantly increasing over <strong>the</strong> last years. 

While a 65 MN press was considered as a huge press a few years 

ago, today some people would call it ‘medium size’ only. 

Considering <strong>the</strong> recent orders <strong>of</strong> SMS… 

• 70 MN frontloading press for Nanshan Aluminium 


• 82 MN frontloading press for Yankuang Aluminium 

• 82 MN frontloading press for Nanshan USA 

• 100 MN frontloading press for Jilin Lijuan 

• 150 MN frontloading tube press for Yankuang Aluminium 


…we see a high potential on this market regarding also o<strong>the</strong>r 

press manufacturers from Europe or Asia who are getting significant 

orders on big presses, too. 

� 

ETS 

Extrusion Tooling Solutions 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

��

EXTRUSION 

Gerhardi Alutechnik and GIA Clecim – a European partnership 

Extrusion line from a single source for intelligent pr<strong>of</strong>iles (I) 

Christoph Deiters, Gerhardi AluTechnik GmbH 

The German Gerhardi Alutechnik, founded 

in 1796 and even today owned by <strong>the</strong> 

founder family, consequently transfers 

values like reliability, long term thinking 

and tradition, but also innovation 

and willingness to invest. The latter has 

carried <strong>the</strong> company through more than 

two centuries and was once more proved 

during <strong>the</strong> last three years: management 

and owners decided for a second extrusion 

line with <strong>the</strong> necessary mechanical 

periphery and a considerable expansion 

<strong>of</strong> <strong>the</strong> factory. This investment lead to an 

augmentation <strong>of</strong> capacity from 6, 000 to 

16.000 tonnes a year. 

Today <strong>the</strong>re are 125 employees who develop, 

produce and market 10,000 to 11,000 tonnes 

<strong>of</strong> <strong>aluminium</strong> pr<strong>of</strong>iles. These are sold to <strong>the</strong> 

building industry (‘Geraconstruct’ including 

‘Gerasolar’ – shutters pr<strong>of</strong>iles and guide rails, 

shading technique, room-dividing elements, 

sanitary equipment, skirting boards, solar 

industry), <strong>the</strong> automotive industry (‘Geramotive’ 

– window frames, guide rails, decorative 

parts, ro<strong>of</strong> rail systems), <strong>the</strong> electronic industry 

(‘Geratronics’ – pr<strong>of</strong>iles for cooling elements) 

and industrial technique (‘Geratechnics’ – machine 

building, lighting industry, furniture). 

Gerhardi is a state<strong>of</strong>-<strong>the</strong>-art 

manufacturer 

<strong>of</strong> semi-finished <strong>aluminium</strong> 

products and – also due to 

<strong>the</strong> new 33 MN extrusion line – 

in a position to meet <strong>the</strong> market’s high 

requirements as to technical and organizational 

quality. 

The Spanish partner, GIA Clecim Press, 

has become one <strong>of</strong> <strong>the</strong> leaders in <strong>the</strong> metal 

extrusion market during <strong>the</strong> past years. The 

company was established in <strong>the</strong> early eighties 

by Gaspar Fernández in Albacete, Spain. This 

is nowadays <strong>the</strong> only manufacturing plant with 

300 employees exclusively working for <strong>the</strong> 

field <strong>of</strong> extrusion <strong>of</strong> metals and special alloys. 

GIA took its position as a complete plant 

supplier throughout <strong>the</strong> years. The company 

got <strong>the</strong> vast majority <strong>of</strong> <strong>the</strong> Spanish market as 

well as presence in many o<strong>the</strong>r countries. In 

2006, GIA Clecim Press purchased <strong>the</strong> knowhow 

and all activities <strong>of</strong> <strong>the</strong> Presses division 

from Siemens. After this acquisition, GIA 

Clecim Press became <strong>the</strong> press manufacturer 

with <strong>the</strong> largest number <strong>of</strong> references in Europe. 

Why did Gerhardi invest 

in a new extrusion press? 

Gerhardi installed <strong>the</strong> first extrusion line in 

1945. Up to 2009 <strong>the</strong> company produced with 

a 2,000-tonne extrusion press <strong>of</strong> 1979. From 

<strong>the</strong> management’s viewpoint this meant a 

risk – for <strong>the</strong> customers as well as for <strong>the</strong> enterprise, 

which bears <strong>the</strong> responsibility for <strong>the</strong> 

employees and for which <strong>the</strong> subject ‘future’ 

traditionally was and still is <strong>of</strong> considerable 

importance according to <strong>the</strong> company’s philosophy 

‘Common Future in Pr<strong>of</strong>ile’. 

As a consequence – after thorough consideration 

<strong>of</strong> all strategic alternatives – owners 

and management decided to invest in a second 

extrusion line in order to be a reliable supplier 

<strong>of</strong> semi-finished <strong>aluminium</strong> products. 

As <strong>the</strong> market forecast for semi-finished 

products in <strong>the</strong> alu-minium sector is positive 

and even shows considerable 

growth rates, <strong>the</strong> company 

acts in a 

developable 

and growing 

market. 

This could 

already be 

observed 

and had 

major influence 

at <strong>the</strong> 

time <strong>of</strong> <strong>the</strong> 

investment 

decision – a very courageous decision: 

Regarding <strong>the</strong> market potential <strong>of</strong> about 

800.000 tpy in Germany, <strong>the</strong> decision makers, 

however, came to <strong>the</strong> conclusion that <strong>the</strong> 

accumulated maximum production capacity 

<strong>of</strong> 16.000 tpy <strong>of</strong> two Gerhardi presses 

might be a reasonable and realizable target. 

The investment would be suitable to secure 

<strong>the</strong> existence <strong>of</strong> <strong>the</strong> company by providing 

<strong>the</strong> technical prerequisites for an expanded 

product range and adapting <strong>the</strong>m 

to <strong>the</strong> company’s high innovation potential. 

Installation <strong>of</strong> <strong>the</strong> extrusion line 

New products like pr<strong>of</strong>iles <strong>of</strong> glossy and 

crash-relevant <strong>aluminium</strong> alloys for automotive 

applications, resulting from <strong>the</strong> close 

cooperation especially with Gerhardi’s raw 

material supplier, required more flexibility <strong>of</strong> 

<strong>the</strong> production process. 

What did Gerhardi buy? 

Gerhardi has invested in a second extrusion 

line with many technical options. It was important 

to run <strong>the</strong> press with <strong>the</strong> same billet 

diameter (8 inch) as <strong>the</strong> existing press. This 

guarantees that <strong>the</strong> matrices <strong>of</strong> <strong>the</strong> older device 

can also be used in <strong>the</strong> new press. The 

dialog with GIA resulted in <strong>the</strong> decision for a 

press with a power <strong>of</strong> 33 MN in order to be 

later able to apply a 9 inch or maybe also a 

10 inch billet. 

In order to reach a constant quality, which 

is especially important for <strong>the</strong> decorative parts 

and <strong>the</strong> functional parts, Gerhardi bought a 

combined billet heating: pre-heating with gas 

and complementary induction heating, which 

allows an even temperature for <strong>the</strong> pressing 

process at <strong>the</strong> point <strong>of</strong> matrix. Special attention 

is given to <strong>the</strong> cooling process. 

GIA installed a pronounced cooling tunnel 

with air- / air-water- and water-cooling. 

These cooling mechanisms allow <strong>the</strong> extru- 

sion <strong>of</strong> higher strength material – also an 


Gerhardi

option for <strong>the</strong> future. 

As Gerhardi wanted to use its building’s 

full size <strong>of</strong> 120 metres for <strong>the</strong> line, <strong>the</strong> length 

<strong>of</strong> <strong>the</strong> run-out-system is considerable. The 

60-metre pr<strong>of</strong>iles are stacked and <strong>the</strong>n sewed 

to customized size from 2,000 mm upwards. 

Press, stacker and saw are followed by a fully 

automatic assembly line which guarantees that 

no worker touches <strong>the</strong> pr<strong>of</strong>iles until reaching 

<strong>the</strong> packing unit. 

The building concept over two floors prescribes 

a stringent material flow. Oven storage, 

packaging and processing <strong>of</strong> <strong>the</strong> pr<strong>of</strong>iles are 

effected on <strong>the</strong> upper floor. 

All <strong>the</strong>se requirements were laid down in a 

specification sheet. The Gerhardi project team 

travelled a lot, visited various extrusion companies 

and as many suppliers as possible. The 

managing director and his team went through 

all <strong>of</strong> Europe to get to know <strong>the</strong> relevant companies, 

to talk to <strong>the</strong>m and to see reference 

lines. 

Whom did Gerhardi choose? 

After an intensive analysis <strong>of</strong> all <strong>of</strong>fers as to 

economic aspects, but also as to <strong>the</strong> product 

availability and <strong>the</strong> vertical range <strong>of</strong> manufacture, 

Gerhardi decided to buy an extrusion 

line made by <strong>the</strong> Spanish company GIA Clecim. 

The core business for GIA is <strong>the</strong> manufacturing 

<strong>of</strong> extrusion plants and 95% <strong>of</strong> jobs 

are turnkey projects. The strategies <strong>of</strong> nonsubcontracting 

lets GIA get <strong>the</strong> highest quality 

for such equipment. GIA holds a manufacturing 

ratio <strong>of</strong> 80%, which means that all equipment 

is manufactured, assembled and tested 

in its facilities. 

For Gerhardi <strong>the</strong> most important criteria 

in favour <strong>of</strong> GIA were <strong>the</strong> 

following: 

• Under an economic 

point <strong>of</strong> view <strong>the</strong> GIA 

Clecim <strong>of</strong>fer was persuading 

in comparison 

to all <strong>the</strong> o<strong>the</strong>r suppliers 

• Gerhardi got a good 

impression <strong>of</strong> <strong>the</strong> technical 

aspects <strong>of</strong> a GIA press 

– especially <strong>of</strong> <strong>the</strong> solid 

machine construction, 

<strong>the</strong> extremely effective 

and good hydraulic system, 

<strong>the</strong> interaction in a 

completely coordinated 

system 

• All from a single 

source – GIA is <strong>the</strong> only 

producer <strong>of</strong> complete extrusion 

systems from <strong>the</strong> 

billet store over <strong>the</strong> billet heating to <strong>the</strong> press, 

<strong>the</strong> saw and <strong>the</strong> assembling device 

• GIA has an enormous in-house production 

depth. All mechanical parts are manufactured 

by <strong>the</strong> company itself and even <strong>the</strong> hydraulic 

system is developed in Albacete and produced 

by GIA 

• GIA is a family enterprise like Gerhardi and 

<strong>the</strong> customer negotiates with <strong>the</strong> owner himself 

• GIA very much wanted to install a reference 

extrusion line in Germany so that Gerhardi 

could take it for granted that <strong>the</strong> Spanish 

partner would make considerable efforts in 

this project. 

How was <strong>the</strong> installation 

process effected? 

33 MN GIA extrusion press for Gerhardi 

After signing <strong>the</strong> contract both parties formed 

project teams who communicated directly 

with each o<strong>the</strong>r. GIA employed a German 

speaking coordinator for <strong>the</strong> German market, 

so that <strong>the</strong> communication could take place 

in German. Innumerable plans were exchanged. 

In interaction between architects, 

construction companies and o<strong>the</strong>r project 

partners <strong>the</strong> prerequisites for <strong>the</strong> installation 

<strong>of</strong> <strong>the</strong> extrusion line were created. Various 

journeys from Germany to Spain and vice 

versa were necessary to realize <strong>the</strong> planned 

targets. 

After <strong>the</strong> preparing activities <strong>the</strong> actual installation 

was effected by a competent team <strong>of</strong> 

technical assemblers and s<strong>of</strong>tware engineers. 

Highlights <strong>of</strong> this phase were certainly <strong>the</strong> socalled 

marriage, <strong>the</strong> merging <strong>of</strong> <strong>the</strong> first part 

<strong>of</strong> <strong>the</strong> press with <strong>the</strong> building, and <strong>the</strong> first 

warm billet. 

Summary 

EXTRUSION 

The installation <strong>of</strong> a technologically sophisticated 

extrusion line is always a great challenge 

– especially if it is a European line like in <strong>the</strong> 

case <strong>of</strong> Gerhardi, where <strong>the</strong> devices came primarily 

from Spain, but also from <strong>the</strong> Ne<strong>the</strong>rlands, 

from Italy and Germany. 

Gerhardi – after a careful choice <strong>of</strong> <strong>the</strong> 

suppliers – was very satisfied that all partners 

were eager to constructively accompany <strong>the</strong> 

investment. All companies involved learnt a 

lot in <strong>the</strong> project – like <strong>the</strong> adequate estimation 

<strong>of</strong> a permanent dialogue, <strong>of</strong> <strong>the</strong> quality 

<strong>of</strong> exchanged information, <strong>the</strong> presence and 

otivation <strong>of</strong> <strong>the</strong> own team within <strong>the</strong> planning 

and installation process and <strong>the</strong> anticipatory 

understanding for <strong>the</strong> partner’s different 

mentality. Gerhardi is aware <strong>of</strong> <strong>the</strong> special 

challenge <strong>of</strong> a European extrusion line, but 

is at <strong>the</strong> same time convinced to have placed 

with <strong>the</strong> GIA Clecim press <strong>the</strong> right investment 

following its philosophy ‘Common future 

in Pr<strong>of</strong>ile’. ‘Common’ not only with customers 

and employees, but also, as it was laid 

down in <strong>the</strong> corporate presentation many 

years ago, also with <strong>the</strong> suppliers. 

Especially with <strong>the</strong> Spanish partners Gerhardi 

developed a business relationship, which 

also lead to common private activities. Visits 

<strong>of</strong> Mr Gaspar Roldan or his fa<strong>the</strong>r Gaspar 

Fernandez were remarkable events as well as 

visits <strong>of</strong> <strong>the</strong> Gerhardi owners in Spain. 

Gerhardi and GIA – this is not only a partnership 

resulting from <strong>the</strong> purchase <strong>of</strong> an 

extrusion press for intelligent Gerhardi <strong>aluminium</strong> 

pr<strong>of</strong>iles, but also a European friendship. 

� 


EXTRUSION 

Gerhardi and GIA Clecim – an European partnership 

Extrusion line from a single source for intelligent pr<strong>of</strong>iles (II) 

Gaspar Fernandez Roldán, GIA Clecim Press 

GIA Clecim Press would like use this 

unique opportunity at <strong>the</strong> GDA conference 

to introduce itself as a company able 

to attend <strong>the</strong> existing <strong>aluminium</strong> extrusion 

demands. GIA Clecim Press is a dynamic 

company which has become, during 

<strong>the</strong> past years, one <strong>of</strong> <strong>the</strong> leaders in 

<strong>the</strong> metal extrusion market in <strong>the</strong> world. 

At <strong>the</strong> beginning <strong>of</strong> <strong>the</strong> 1980s, when GIA in 

Albacete, Spain, began producing extrusion 

dies, <strong>the</strong>re was a boom in Spain’s building 

industry which provided huge market opportunities 

for <strong>aluminium</strong> window and structural 

sections. Good starting conditions, which encouraged 

<strong>the</strong> young company to extend 

its engineering performance besides 

die manufacturing to o<strong>the</strong>r applications 

in extrusion plants. So GIA became an 

equipment manufacturer which <strong>of</strong>fered 

handling and processing equipment for 

extruded sections to <strong>the</strong> <strong>the</strong>n growing 

Spanish market. 

A substantial gap was filled when, 

towards <strong>the</strong> end <strong>of</strong> <strong>the</strong> 1980s, <strong>the</strong> technical 

staff at GIA ventured to attempt 

<strong>the</strong> centrepiece <strong>of</strong> <strong>the</strong> plants: <strong>the</strong> first 

extrusion press designed by <strong>the</strong> company 

itself and made by it in 1990 was 

a 14 MN press <strong>of</strong> conventional, backloading 

configuration, which began 

operating as part <strong>of</strong> a complete plant 

supplied by GIA to a Spanish extruder 

in Albacete. 

Following <strong>the</strong> company policy to 

become a supplier <strong>of</strong> turn-key projects 

in <strong>the</strong> extrusion business, GIA developed 

and introduced to <strong>the</strong> market <strong>the</strong> 

short-stroke back loading design <strong>of</strong> 18 

to 22 MN. Due to <strong>the</strong> factory location 

and <strong>the</strong> great growth <strong>of</strong> extruded <strong>aluminium</strong> 

consumption, <strong>the</strong> main approached market at 

<strong>the</strong> time was Spain where <strong>the</strong> equipment had 

a huge demand and customer acceptance. 

At <strong>the</strong> end <strong>of</strong> <strong>the</strong> 1990s <strong>the</strong> leap to trading 

overseas took place. After orders from <strong>the</strong> Dominican 

Republic o<strong>the</strong>rs followed from Mexico 

and <strong>the</strong> Sou<strong>the</strong>rn states <strong>of</strong> <strong>the</strong> USA. By 

<strong>the</strong> end <strong>of</strong> <strong>the</strong> millennium, that is within about 

ten years, GIA had supplied 45 new extrusion 

plants equipped with 18 and 35 MN presses 

<strong>of</strong> its own. Changed conditions in <strong>the</strong> extru- 

sion market and also <strong>the</strong> advance into new 

markets such as <strong>the</strong> gulf region demanded <strong>the</strong> 

development <strong>of</strong> larger presses from <strong>the</strong> beginning 

<strong>of</strong> <strong>the</strong> new millennium. Thus, <strong>the</strong> range 

<strong>of</strong> short stroke back-loading presses was extended 

up to 45 MN. As before, <strong>the</strong>se formed 

<strong>the</strong> centrepiece <strong>of</strong> complete GIA plants. 

The decision by Siemens VAI in 2006 to 

break away from <strong>the</strong> extrusion press construction 

activities <strong>of</strong> its French traditional trademark 

Clecim opened up new perspectives for 

GIA. It was soon decided to acquire <strong>the</strong> rights 

to and designs <strong>of</strong> Clecim presses and, toge<strong>the</strong>r 

with <strong>the</strong> company’s own press programme, to 

market ‘GIA Clecim’ presses. For GIA, Clecim’s 

technology was a milestone for <strong>the</strong> company’s 

fur<strong>the</strong>r development. At <strong>the</strong> time <strong>of</strong> <strong>the</strong> 

sale Clecim still had a proud reference list <strong>of</strong> 

more than 100 extrusion presses built since 

<strong>the</strong> early 1970s. 

GIA Clecim Press became <strong>the</strong>n <strong>the</strong> press 

manufacturer with <strong>the</strong> largest number <strong>of</strong> references 

in Europe. This transfer <strong>of</strong> property 

congregated technology and know-how from 

all companies which have merged to Clecim 

along <strong>the</strong> time. GIA Clecim Press also took 

advantage <strong>of</strong> this transfer by hiring expert engineers 

from former Clecim Presses Division. 

Therefore, GIA Clecim Press became <strong>the</strong> 

successor <strong>of</strong> <strong>the</strong> companies Loewy, Davy, Morane, 

Somu, Loire, Secim, Clesid and finally 

Clecim, since <strong>the</strong> foundation <strong>of</strong> Chavanne 

Brun in Montbrison (France) in 1857. All <strong>the</strong>se 

company were extraordinary press builders 

along <strong>the</strong> time, and have demonstrated outstanding 

designs for <strong>the</strong> extrusion field. 

GIA Clecim Press has in its archive all <strong>the</strong> 

information, calculations and drawings <strong>of</strong> <strong>the</strong> 

presses designed by <strong>the</strong>se manufacturers during 

<strong>the</strong>ir activity. Our company maintains <strong>the</strong> 

willing <strong>of</strong> keeping contact with all users <strong>of</strong> 

<strong>the</strong>se equipments as well as continues increasing 

this reference list. 

Particularly important in <strong>the</strong> <strong>aluminium</strong> extrusion 

field, GIA acquired an established <strong>the</strong> 

range <strong>of</strong> modern short-stroke presses. Since 

<strong>the</strong>n, along with its own back loading 

presses <strong>the</strong> company has been able 

to react flexibly to <strong>the</strong> various needs 

<strong>of</strong> <strong>the</strong> market and <strong>of</strong>fer <strong>the</strong> optimum 

press configuration for each individual 

case as <strong>the</strong> market de-mands. Ano<strong>the</strong>r 

benefit for GIA is <strong>the</strong> fact that Clecim 

extrusion presses were not built for 

<strong>aluminium</strong> alone. Through Clecim 

GIA also acquired technologies and 

references for <strong>the</strong> extrusion <strong>of</strong> o<strong>the</strong>r 

materials, such as stainless steel, copper, 

special alloys as well as particular 

projects. In that connection it is worth 

mentioning that GIA Clecim Press has 

just commissioned 22 MN press for a 

stainless steel plant in Austria and has 

been awarded for o<strong>the</strong>r interesting 

projects as <strong>the</strong> 42 MN extrusion press 

for stainless also in Sweden and a 35 

MN complete extrusion plant for copper 

alloys in China. 

The meanwhile consolidated engineering 

programme that originated 

from GIA and Clecim today includes 

presses <strong>of</strong> any load and type including direct 

and indirect extrusion with or without piercer, 

optionally in front-loading or back-loading 

versions. In <strong>the</strong> recent past several complete 

plants have been commissioned all over <strong>the</strong> 

world such as a 14 MN line in Bosnia, 18 MN 

in Ireland, a 30 MN front loading press plant 

in Bahrain, a 22 MN press line in Saudi Arabia, 

a 65 MN front loading press to Japan. 

All <strong>the</strong> plants are developed, designed, produced, 

assembled and tested at <strong>the</strong> company’s 

own works in Spain. In <strong>the</strong> mechanical workshop 

parts weighing up to 100 tonnes can be 

machined. 


Photos: GIA 

In autumn 2009 Gerhardi Alutechnik in 

Lüdenscheid, Germany, began operating its 

second extrusion line which was supplied by 

GIA. Its centrepiece is a 35 MN short-stroke, 

front-loading press <strong>of</strong> GIA Clecim Press design. 

The press enables short extrusion cycle 

times and so achieves correspondingly high 

productivity. 

To understand and appreciate <strong>the</strong> big step 

this project was for GIA Clecim Press, it is 

worth mentioning <strong>the</strong> challenges and <strong>the</strong> importance 

<strong>of</strong> this contract. It should be known 

that Gerhardi’s production programme is not 

concerned with mass goods but contains many 

special products for niche markets, some <strong>of</strong> 

<strong>the</strong>m made in relatively small quantities to 

<strong>the</strong> highest quality standards, e. g. stretch-bent 

decorative trim strips <strong>of</strong> a high-gloss alloy can 

be mentioned here. The Gerhardi contract is 

one <strong>of</strong> <strong>the</strong> projects where GIA engineering 

shows state-<strong>of</strong>-<strong>the</strong>-art technological developments, 

which <strong>the</strong> extrusion industry in highlydeveloped 

industrialised countries now demands 

from extrusion plant manufacturers. 

Gerhardi is one <strong>of</strong> <strong>the</strong> extrusion companies 

with <strong>the</strong> highest level <strong>of</strong> flexibility without giving 

away any productivity ratios. This extrusion 

plant meant a great step forward for Gerhardi 

but for GIA also as it shows <strong>the</strong> Spanish 

company is at <strong>the</strong> level <strong>of</strong> <strong>the</strong> most demanding 

extruders. 

The more than 100 extrusion presses built 

so far by GIA are almost without exception 

used in extrusion lines which, from <strong>the</strong> log 

store to <strong>the</strong> finishing <strong>of</strong> <strong>the</strong> sections, were 

designed and constructed by <strong>the</strong> company. 

GIA group is divided in several business units 

coordinated to provide complete solutions to 

extruders. GIA Matriceria, as ever, produces 

extrusion dies and all kind <strong>of</strong> tooling for extrusion. 

Log stores, shears, billet saws, billet loader, 

extrusion presses, handling and puller systems 

are all designed and made by <strong>the</strong> core 

unit GIA. Special positioning devices for <strong>the</strong> 

section saws ensure precise lengths and rapid 

material flow. The fully automatic stackers 

developed by GIA not only ensure a rapid 

and reliable production sequence, but are so 

constructed as to protect <strong>the</strong> finished sections 

from damage. 

Besides pure extrusion lines, under <strong>the</strong> 

trade name ‘GIA Aplicaciones’ GIA supplies 

units for heating and heat treatment within 

<strong>the</strong> extrusion line which include log furnaces, 

die heaters, ageing ovens and die nitriding 

furnaces. In addition, o<strong>the</strong>r units developed, 

designed and manufactured by <strong>the</strong> company 

itself are <strong>of</strong>fered, which are important for finishing 

<strong>the</strong> sections, namely units for powder 

coating and anodising. Thus, <strong>the</strong> company can 

supply complete extrusion plants, from <strong>the</strong> 

EXTRUSION 

log store right up to <strong>the</strong> surface finishing <strong>of</strong> 

<strong>the</strong> ready sections. 

Ano<strong>the</strong>r company <strong>of</strong> <strong>the</strong> group, GIA Hidraulica, 

not only builds hydraulic power units 

for various applications in general mechanical 

engineering, especially in combination with 

GIA’s own extrusion machines. This unit also 

specialises in electrical and electronic equipment 

for <strong>the</strong> control and automation <strong>of</strong> extrusion 

plants and is, moreover, responsible for 

commissioning <strong>the</strong> plants supplied by GIA. 

There is a fifth business unit, GIA Suministros, 

in charge <strong>of</strong> <strong>the</strong> service and supply <strong>of</strong> 

spare parts for GIA Clecim Press extrusion 

equipment. This unit keeps a great number <strong>of</strong> 

parts in stock in order to provide a perfect 

service to its customers. 

GIA Clecim Press is located, as ever, in 

Albacete, Spain, which is nowadays <strong>the</strong> only 

manufacturing plant with 35.000 m2 <strong>of</strong> bays 

and where counts with 275 employees exclusively 

working in <strong>the</strong> field <strong>of</strong> extrusion. 

Core business for GIA Clecim Press is <strong>the</strong> 

manufacturing <strong>of</strong> extrusion and surface treatment 

plants and 90% <strong>of</strong> <strong>the</strong> jobs are turnkey 

projects. Its strategy <strong>of</strong> non-subcontracting let 

<strong>the</strong>m get <strong>the</strong> highest quality for such equipment. 

GIA Clecim Press holds a manufacturing 

ratio <strong>of</strong> 80%, which means that all equipment 

is manufactured, assembled and tested 

in <strong>the</strong>ir own facilities. 

� 


EXTRUSION 

Integrated processing <strong>of</strong> <strong>aluminium</strong> 

pr<strong>of</strong>iles subsequent to hot extrusion 

A. Jäger, N. Ben Khalifa, A.E. Tekkaya, Institute <strong>of</strong> Forming Technology 

and Lightweight Construction, TU Dortmund University, Dortmund, Germany 

Strategies <strong>of</strong> <strong>the</strong>rmo-mechanical processing 

<strong>of</strong> <strong>aluminium</strong> pr<strong>of</strong>iles subsequent to 

hot extrusion were developed and tested. 

By using <strong>the</strong> process heat <strong>of</strong> extrusion for 

subsequent forming and heat treatment 

steps, process chains for <strong>the</strong> production 

<strong>of</strong> graded <strong>aluminium</strong> pr<strong>of</strong>iles were developed 

and realized on an experimental 

scale. Due to <strong>the</strong> geometric limitations, 

<strong>the</strong> technology <strong>of</strong> realizing a grading in 

<strong>the</strong> geometry <strong>of</strong> open and hollow pr<strong>of</strong>iles 

is different. Therefore, two innovative 

concepts for <strong>the</strong> <strong>the</strong>rmo-mechanical 

processing <strong>of</strong> open and hollow sections 

were developed, allowing electromagnetic 

compression and rolling to become 

integrated within <strong>the</strong> process chains. Possible 

applications are seen in producing 

functionally graded products with locally 

adapted properties. 

Conventional hot metal extrusion is used to 

produce straight, semi-finished products in 

mass production (Laue and Stenger, 1976) with 

a constant cross section over <strong>the</strong> length and 

homogeneous mechanical and micro-structural 

properties (Hall and Mudawar, 1996). 

Aluminium sections are frequently used as 

construction elements. Generally, <strong>the</strong> loading 

conditions <strong>of</strong> structure elements in technical 

constructions differ locally. The constant cross 

Fig. 2: Concept <strong>of</strong> extrusion and subsequent forming by rollers (schematic) 

Fig. 1: Concept <strong>of</strong> extrusion with in-line electromagnetic compression (longitudinal section, schematic) 

section, which is attributed to <strong>the</strong> peculiarity 

<strong>of</strong> <strong>the</strong> extrusion process itself, mostly represents 

a compromise between <strong>the</strong> functionality 

and <strong>the</strong> component design, which e. g. involves 

a local oversizing and thus an excessive use 

<strong>of</strong> resources. Here, a local adaption <strong>of</strong> <strong>the</strong> 

geometry could be suitable in order to meet 

<strong>the</strong> locally specific demands on <strong>the</strong> structure 

properties. For this reason, <strong>the</strong> development 

<strong>of</strong> innovative forming technologies is indispensable 

in order to manufacture products 

with graded, locally adapted structural properties 

as well as varying geometric shapes. 

The increase <strong>of</strong> <strong>the</strong> formability <strong>of</strong> <strong>aluminium</strong> 

can be enhanced by heat treatment 

between <strong>the</strong> single forming steps. The integration 

<strong>of</strong> <strong>the</strong>rmo-mechanical processing after 

extrusion is a desirable 

substitute for additional 

heat treatment processes 

for several reasons: 

additional production 

steps lead to increased 

production time, energy 

consumption and production 

costs. Incorporating<strong>the</strong>rmo-mechanical 

processing immediately 

after extrusion 

will save <strong>the</strong> time and 

energy needed to reheat 

<strong>the</strong> pr<strong>of</strong>iles for o<strong>the</strong>r hot 

forming steps. 

In this paper, strategies 

for <strong>the</strong> <strong>the</strong>rmo-mechanical 

processing <strong>of</strong> 

<strong>aluminium</strong> pr<strong>of</strong>iles by 

hot forming subsequent to hot extrusion for 

<strong>the</strong> manufacturing <strong>of</strong> geometrically graded 

<strong>aluminium</strong> pr<strong>of</strong>iles are given. Due to <strong>the</strong> geometric 

limitations, <strong>the</strong> technology <strong>of</strong> subsequent 

processing <strong>the</strong> geometry <strong>of</strong> open and 

hollow pr<strong>of</strong>iles is different. Two concepts for 

<strong>the</strong> processing <strong>of</strong> <strong>aluminium</strong> pr<strong>of</strong>iles by applying 

hot extrusion combined with subsequent 

hot electromagnetic compression and alternatively 

combined with an integrated hot rolling 

concept were developed and tested on an 

experimental scale. By process integration <strong>the</strong> 

process heat <strong>of</strong> extrusion is used for <strong>the</strong> successive 

forming operation. 

Description <strong>of</strong> <strong>the</strong> new concepts 

For <strong>the</strong> processing <strong>of</strong> hollow pr<strong>of</strong>iles a combination 

<strong>of</strong> hot extrusion and electromagnetic 

compression was developed. Hot metal extrusion 

is used to produce tubular semi-finished 

products continuously by pushing a billet 

through a die (DIN 8583-6, 2003). In typical 

industrial hot <strong>aluminium</strong> extrusion, pr<strong>of</strong>ile 

exiting speeds <strong>of</strong> up to 50 m/min are common 

(Ostermann, 2007). In contrast to this, 

electromagnetic compression is used to reduce 

<strong>the</strong> cross section <strong>of</strong> a workpiece locally (Harvey 

and Brower, 1961). The applied magnetic 

pressure can achieve values <strong>of</strong> up to several 

hundred MPa and accelerates <strong>the</strong> workpiece 

to typical strain rates <strong>of</strong> about 10 4 s -1 completing 

<strong>the</strong> entire forming process in approximately 

50 to 200 μs. Due to <strong>the</strong> favourable 

relation between resulting exit speed in extrusion 

and processing time in electromagnetic 

compression, a longitudinal translation be- 


Images: IUL

Fig. 3: Experimental setup for extrusion with subsequent electromagnetic compression 

tween <strong>the</strong> workpiece and <strong>the</strong> tool coil during 

<strong>the</strong> compression process can be neglected. For 

example, with an exiting speed <strong>of</strong> 50 m/min, 

and 100 μs for <strong>the</strong> electromagnetic forming 

process, a relative movement between <strong>the</strong> 

extrudate pr<strong>of</strong>ile and <strong>the</strong> tool coil <strong>of</strong> about 

160 μm results, which meets <strong>the</strong> geometrical 

tolerance field <strong>of</strong> macroscopic forming processes. 

Therefore, <strong>the</strong> application <strong>of</strong> electromagnetic 

compression subsequent to extrusion 

is possible without a compensation <strong>of</strong> <strong>the</strong> 

relative speed between <strong>the</strong> workpiece and <strong>the</strong> 

tooling. 

To integrate both processes, a tool coil for 

compression was positioned behind <strong>the</strong> die 

exit and coaxial to <strong>the</strong> extrudate in order to 

reduce <strong>the</strong> workpiece cross section locally 

(Fig. 1). Additionally, a counter die in <strong>the</strong> shape 

<strong>of</strong> a mandrel can be mounted to <strong>the</strong> mandrel 

<strong>of</strong> a porthole extrusion die, which extended 

into <strong>the</strong> tool coil (Jäger et al., 2009). By this, 

besides achieving a more defined geometry in 

comparison to a free forming operation, an 

increase <strong>of</strong> <strong>the</strong> geometrical complexity <strong>of</strong> locally 

compressed areas can be achieved. 

For processing open sections, like L-, T 

and double-T-shaped pr<strong>of</strong>iles, a rolling pro- 

cess was developed. By executing <strong>the</strong> subsequent 

forming operation as a rolling process, 

<strong>the</strong> continuous run out <strong>of</strong> <strong>the</strong> pr<strong>of</strong>ile can be 

taken into account. 

Fig. 4: Dimensions <strong>of</strong> <strong>the</strong> mandrel for electromagnetic forming 

A special rolling stand was developed and 

mounted behind <strong>the</strong> extrusion platen <strong>of</strong> an 

extrusion press, allowing <strong>the</strong> continuous forming 

<strong>of</strong> <strong>the</strong> outgoing extrusion pr<strong>of</strong>iles. The 

rollers mesh toge<strong>the</strong>r with sinusoidal teeth, 

creating a corrugated contour on <strong>the</strong> web <strong>of</strong> 

a continuously fed I-beam-shaped pr<strong>of</strong>ile. The 

contour <strong>of</strong> <strong>the</strong> forming dies stores <strong>the</strong> desired 

geometry <strong>of</strong> <strong>the</strong> workpiece. As <strong>the</strong> exit speed <strong>of</strong> 

<strong>the</strong> extrudate varies, compensation strategies 

have to be provided. For this, <strong>the</strong> rolling stand 

was swivel-mounted and <strong>the</strong> rotational speed 

<strong>of</strong> <strong>the</strong> rollers was adapted permanently (Fig. 2). 

Extrusion and electromagnetic forming 

A test rig for evaluating <strong>the</strong> concept <strong>of</strong> <strong>the</strong> 

integration <strong>of</strong> an electromagnetic compression 

step subsequent to extrusion was built (Fig. 3). 

A tool coil was positioned behind <strong>the</strong> die exit 

in order to reduce <strong>the</strong> workpiece cross section 

locally. To adapt <strong>the</strong> tool coil and <strong>the</strong> extrudate 

geometry, a field shaper made out <strong>of</strong> a 

CuCrZr-alloy, electrically insulated with a 

polyimide foil, is used. Beyond shaping and 

concentrating <strong>of</strong> <strong>the</strong> electromagnetic field, <strong>the</strong> 

field shaper also prevents an overheating <strong>of</strong> 

<strong>the</strong> tool coil by <strong>the</strong>rmal radiation <strong>of</strong> <strong>the</strong> processed 

hot extrudate. To assure a uniform air 

gap between <strong>the</strong> field shaper and <strong>the</strong> workpiece 

guiding rollers made out <strong>of</strong> brass are ar- 

EXTRUSION 

ranged pairwise at <strong>the</strong> inlet and run-out side 

<strong>of</strong> <strong>the</strong> tool coil. For heat treating <strong>of</strong> <strong>the</strong> extruded 

and subsequently hot deformed workpieces 

made <strong>of</strong> heat treatable alloys, an additional 

quenching setup was positioned behind 

<strong>the</strong> tool coil. Atomizing nozzles are located 

around <strong>the</strong> press axis providing air atomized 

water mist streams. To use <strong>the</strong> press heat <strong>of</strong> 

extrusion for <strong>the</strong> subsequent <strong>the</strong>rmo-mechanical 

processing steps efficiently, <strong>the</strong> whole 

setup is designed compactly. From <strong>the</strong> die exit 

to <strong>the</strong> quenching, <strong>the</strong> whole setup has a length 

<strong>of</strong> less than one meter. 

Experimental trials were performed using 

a 250-ton direct extrusion press (Collin 

PLA250t) and a pulse power generator (Maxwell-Magneform 

series 7000). Solenoid coils 

fabricated from copper and embedded in a 

fibre-reinforced resin were used as tool coils 

(Poynting GmbH). Using an EN AW-6082 <strong>aluminium</strong> 

alloy, a porthole extrusion die and a 

squared field shaper, a squared hollow pr<strong>of</strong>ile 

(30 x 30 x 1.8 mm) was extruded with an exiting 

speed <strong>of</strong> about 16 mm/s while compression 

by electromagnetic forming was applied 

periodically in intervals <strong>of</strong> approximately 

10s. Semi-finished products with local bulges 

arranged around <strong>the</strong> circumference <strong>of</strong> <strong>the</strong> 

squared tube can be produced. 

In order to expand <strong>the</strong> range <strong>of</strong> geometrical 

shapes with cross-sections o<strong>the</strong>r than that 

<strong>of</strong> <strong>the</strong> extruded <strong>aluminium</strong>, a mandrel can be 

used to define <strong>the</strong> cross section geometry. A 

squared mandrel for <strong>the</strong> processing <strong>of</strong> a round 

tube was chosen and specially designed to prevent 

force fit between <strong>the</strong> workpiece and <strong>the</strong> 

die (Fig. 4). By using a squared field shaper, 

with an aperture <strong>of</strong> 43.4 x 43.4 mm and a 

length <strong>of</strong> 27 mm <strong>of</strong> <strong>the</strong> concentrator, in combination 

with <strong>the</strong> extrusion <strong>of</strong> a round tube 

(Ø 40 x 2 mm) fur<strong>the</strong>r trials were performed. 

In this combination <strong>of</strong> field shaper, workpiece 

and mandrel an adapted air gap and 

pressure distribution along <strong>the</strong> workpiece 

circumference can be achieved which helps 

to prevent a force fit between <strong>the</strong> extrudate 

and <strong>the</strong> mandrel. In order to character- 

ize <strong>the</strong> forming results when a mandrel was 

used as a form defining tool, an experimental 

study was performed at charging energies 

<strong>of</strong> 6.4 kJ, 8 kJ and 8.8 kJ. To measure <strong>the</strong> 

high speed current pulse in <strong>the</strong> tool coil a 

Rogowski coil was used. The best forming result/geometrical 

accuracy could be achieved at 

a charging energy <strong>of</strong> 8 kJ. For this a maximum 

coil current amplitude <strong>of</strong> 62 kA and frequency 

<strong>of</strong> <strong>the</strong> discharging current <strong>of</strong> 4.5 kHz were detected. 

To prevent damage by overheating <strong>of</strong> 

<strong>the</strong> tool coil and <strong>the</strong> electrical isolation, <strong>the</strong> 

field shaper was cooled by compressed air 


EXTRUSION 

Fig. 5: (a) tubular product, (b) position <strong>of</strong> <strong>the</strong> mandrel (longitudinal section), (c) application ‘crashbox’ 

during stops, e. g. due to billet loading. In an 

industrial application a water-cooling <strong>of</strong> <strong>the</strong> 

field shaper by integrated cooling channels 

would be useful. Strategies for cooling <strong>of</strong> <strong>the</strong> 

coil or <strong>the</strong> field shaper in electromagnetic 

forming are suggested e. g. by Golovashchenko 

et al. (2006) or Hahn (2006). 

Fig. 5 a shows a locally geometrically graded 

pr<strong>of</strong>ile manufactured by <strong>the</strong> introduced 

process chain processing <strong>of</strong> a round tube in 

combination with a squared mandrel (Fig. 5 b). 

The distance between <strong>the</strong> local bulges is determined 

by <strong>the</strong> pr<strong>of</strong>ile’s exit speed and <strong>the</strong> 

discharging frequency <strong>of</strong> <strong>the</strong> electromagnetic 

forming machine, which is limited by <strong>the</strong> capacitor 

charging time. The general feasibility 

<strong>of</strong> this technological concept could be proven. 

Details about <strong>the</strong> geometric accuracy, <strong>the</strong> process 

limits and defects are provided in Jäger et 

al. (2011a). 

A possible application might be <strong>the</strong> use as 

a ‘crashbox’, as a crash absorbing element in 

a car bumper system, with an adapted forcedisplacement 

characteristic (Fig. 5 c), e. g. designed 

as a telescoping tube or inversion tube. 

The inversion process involves <strong>the</strong> turning 

inside out <strong>of</strong> a tube under axial compressive 

load, characterized by a constant load level 

(Guist, L.R. and Marble, D.P., 1966.). 

Extrusion and rolling 

A roll stand for <strong>the</strong> corrugating <strong>of</strong> <strong>aluminium</strong> 

pr<strong>of</strong>iles subsequent to hot extrusion was developed 

and aligned to <strong>the</strong> press axis <strong>of</strong> an 

extrusion press (Fig. 7). The roll stand consists 

<strong>of</strong> two, e. g. sinusoidal shaped, counter 

rotating rollers driven by a gear motor. For 

adjusting <strong>of</strong> <strong>the</strong> rolling gap, one <strong>of</strong> <strong>the</strong> rollers 

can be moved laterally by a manually driven 

threaded spindle. 

As <strong>the</strong> exit speed <strong>of</strong> <strong>the</strong> extrudate varies, 

in particular when extrusion is initiated 

or restarted after billet loading, <strong>the</strong> roll 

stand is swivelling-mounted at <strong>the</strong> extrusion 

platen <strong>of</strong> <strong>the</strong> press. By this, a clearance for 

<strong>the</strong> compensation <strong>of</strong> a varying exit speed <strong>of</strong> 

<strong>the</strong> extrudate is given. Additionally, excessive 

reaction forces between <strong>the</strong> rolling on 

<strong>the</strong> extrusion process are prevented. To adapt 

<strong>the</strong> circumferential speed <strong>of</strong> <strong>the</strong> rollers to 

<strong>the</strong> extrusion speed, <strong>the</strong> deflection <strong>of</strong> <strong>the</strong> roll 

stand is measured by a potentiometer which is 

mounted in <strong>the</strong> swivelling axis <strong>of</strong> <strong>the</strong> rolling 

stand. The obtained signal is used for a closed 

Fig. 6: Experimental setup for extrusion and rolling 

loop process control <strong>of</strong> <strong>the</strong> rolling speed. 

This allows <strong>the</strong> rolling stand to be kept parallel 

to <strong>the</strong> extrusion platen. Fur<strong>the</strong>r details 

<strong>of</strong> <strong>the</strong> process control are given in Jäger et 

al. (2011b). For damping <strong>of</strong> possible vibrations, 

a hydraulic shock absorber is mounted 

between <strong>the</strong> extrusion press and <strong>the</strong> free end 

<strong>of</strong> <strong>the</strong> roll stand. 

Experimental trials were performed extruding 

a double-T-shaped pr<strong>of</strong>ile <strong>of</strong> an EN AW- 

6082 <strong>aluminium</strong> alloy using a 1,000-tonne 

direct extrusion press (SMS Meer) and two 

sets <strong>of</strong> sinusoidal shaped rollers, pairwise. 

Corrugated web beams with partial corrugation 

in <strong>the</strong> web <strong>of</strong> <strong>the</strong> beam section could be 

produced quasi continuously (Figs. 7 a & 7 b). 

As an alternative to produce regularly 

deformed pr<strong>of</strong>iles, irregularly shaped rollers 

with e. g. varying amplitude and phase and 

even with sections <strong>of</strong> a constant radius can 

be used to manufacture geometrical graded 

products with e. g. locally adapted torsion rigidity 

or bending stiffness. 

For <strong>the</strong> first trials two sets composed <strong>of</strong> 

two sinusoidally shaped rollers were manufac- 

tured out <strong>of</strong> sheet metal blanks with a thickness 

<strong>of</strong> 10 mm by laser cutting. Geometrical 

details <strong>of</strong> <strong>the</strong> corrugation and <strong>the</strong> positioning 

<strong>of</strong> <strong>the</strong> rollers are given in Figs. 7 a) and 7 b). 

To prevent shear cutting in <strong>the</strong> transition 

area between <strong>the</strong> flange and <strong>the</strong> web, <strong>the</strong> 

rollers were positioned at a distance <strong>of</strong> 5 mm 

parallel to <strong>the</strong> flanges. The metal forming operation 

processes comprise elements <strong>of</strong> bending 

and stretch bending, partially performed 

as a free forming and a tool-bounded operation. 

Microscopic analysis revealed a crack 


initiation at <strong>the</strong> pole <strong>of</strong> <strong>the</strong> corrugation in <strong>the</strong> 

transition between <strong>the</strong> flange and <strong>the</strong> web for 

<strong>the</strong> tested configuration (Fig. 7 c). To prevent 

excessive shear, especially in <strong>the</strong> transition 

area between <strong>the</strong> flange and <strong>the</strong> web, a redesign 

<strong>of</strong> geometry and <strong>the</strong> rollers is necessary. 

Possible applications are seen in <strong>the</strong> production 

<strong>of</strong> lightweight construction pr<strong>of</strong>iles 

(Fig. 8 a) with reduced web thickness. Beyond 

that, pr<strong>of</strong>iles with a varying or even a partial 

corrugation geometry are conceivable (Fig. 

8 b), which would result in graded bending 

rigidity or torsional stiffness properties. 

Conclusion 

Two strategies for <strong>the</strong> manufacturing <strong>of</strong> products, 

or ra<strong>the</strong>r semi-finished products, with 

locally adapted geometrical properties were 

developed. Due to geometric limitations and 

accessibilities <strong>of</strong> <strong>the</strong> cross section <strong>the</strong> technology 

<strong>of</strong> integrated processing <strong>of</strong> open and hollow 

pr<strong>of</strong>iles by forming is different. 

For open sections a rolling setup for corrugating 

<strong>the</strong> web <strong>of</strong> an I-beam section subsequent 

to hot extrusion was developed and 

tested. Future work will focus on <strong>the</strong> extension 

<strong>of</strong> <strong>the</strong> geometrical complexity <strong>of</strong> <strong>the</strong> products 

and on <strong>the</strong> analysis <strong>of</strong> <strong>the</strong> geometrical, mechanical 

and resulting structural properties. 

Experiments were conducted on hollow 

pr<strong>of</strong>iles which were compressed by electromagnetic 

forming subsequent to extrusion. 

Due to an extremely short processing time <strong>of</strong> 

<strong>the</strong> high speed forming process, a compensation 

<strong>of</strong> <strong>the</strong> relative speed between <strong>the</strong> workpiece 

and <strong>the</strong> tooling can be ignored. By <strong>the</strong> 

contactless force induction, an additional benefit 

is given for <strong>the</strong> heat balance. To fur<strong>the</strong>r 

understand this process, analysis <strong>of</strong> process 

dependencies in <strong>the</strong> system workpiece, field 

shaper and mandrel is necessary. 

Fig. 7: (a) Longitudinal cut, (b) cross section, (c) crack 


This work was carried out in <strong>the</strong> subproject 

TPA2 within <strong>the</strong> Transregional Collaborative 

Research Centre SFB/TR30, funded by <strong>the</strong> 

German Research Foundation (DFG). 

References 

Fig. 8: Corrugated web beam, a) continuously corrugate, b) partly corrugated 

DIN 8583-6, 2003. Manufacturing processes forming 

under compressive conditions – Part 6: Extrusion; 

Classification, subdivision, terms and definitions, 

Beuth, Berlin. 

Guist, L.R., Marble, D.P., 1966. Prediction <strong>of</strong> <strong>the</strong> 

inversion load. NASA 

Technical Note 3622. 

Golovashchenko, S., 

Dmitriev, V., Canfield, 

P., Krause, A., Maranville, 

C., 2006. Patent 

application US 

2006/0086165A1. 

Apparatus for electromagnetic 

forming with 

durability and efficiency 

enhancement. 

Hahn, R., 2004. 

Werkzeuge zum impulsmagnetischenWarmfügen 

von Pr<strong>of</strong>ilen aus 

Aluminium- und Magnesiumlegierungen,doctoral 

<strong>the</strong>sis, Technische 

Universität Berlin. 

Hall, D. D., Mudawar, I., 

EXTRUSION 

1996. Optimization <strong>of</strong> quench history <strong>of</strong> <strong>aluminium</strong> 

parts for superior mechanical properties, International 

Journal <strong>of</strong> Heat and Mass Transfer 39 (1), 

p. 81-95. 

Harvey, G. W., Brower, D. F., 1961. US patent 2 976 

907. Metal forming device and method, 28 March. 

Jäger, A., Risch, D., Tekkaya, A. E., 2009. Patent 

application DE 10 2009 039 759 A1. Verfahren und 

Vorrichtung zum Strangpressen und nachfolgender 

elektromagnetischer Umformung, 31 August. 

Jäger, A., Risch, D., Tekkaya, A. E., 2011a. Thermomechanical 

processing <strong>of</strong> <strong>aluminium</strong> pr<strong>of</strong>iles by 

integrated electromagnetic compression subsequent 

to hot extrusion, Journal <strong>of</strong> Materials Processing 

Technology 211 (5), Special Issue: Impulse Forming, 

p. 936-943. 

Jäger, A., Ben Khalifa, N., Psyk, V., Tekkaya, A. 

E. 2011b. Thermo-mechanical processing <strong>of</strong> <strong>aluminium</strong> 

pr<strong>of</strong>iles subsequent to hot extrusion, steel 

research international, Special edition: International 

Conference on Technology <strong>of</strong> Plasticity, 

ICTP, 2011, Aachen, Germany 

Laue, K. Stenger, H., 1976. Strangpressen, Aluminium 

Verlag, Düsseldorf, p. 1. Ostermann, F., 

2007. Anwendungstechnologie Aluminium, 2nd 

Edition. Springer, Berlin, p. 440-441. 

Authors 

Dipl.-Ing. Andreas Jäger is head <strong>of</strong> <strong>the</strong> department 

<strong>of</strong> bulk metal forming at <strong>the</strong> Institute <strong>of</strong> Forming 

Technology and Lightweight Construction (IUL). 

Dipl.-Ing. Nooman Ben Khalifa is chief engineer for 

research at <strong>the</strong> IUL. 

Pr<strong>of</strong>. Dr.-Ing. A. Erman Tekkaya is head <strong>of</strong> <strong>the</strong> IUL. 


EXTRUSION 

Flexible automated material flow in extrusion operations 

Georg Papadopoulos, H+H Herrmann + Hieber GmbH 

The logistics specialist H+H Herrmann + 

Hieber based in Denkendorf near Stuttgart 

has worked for more than 20 years 

in <strong>the</strong> special field <strong>of</strong> in-house transport 

in extrusion plants. Since 1989 H+H has 

planned and supplied machinery and 

equipment for almost every major extrusion 

plants in Germany and for many 

elsewhere in Europe. 

The fact that material flow automation in this 

sector has become so successfully established, 

can be attributed on <strong>the</strong> one hand to <strong>the</strong> continual 

growth <strong>of</strong> extrusion production in Germany, 

but in addition economic conditions in 

Europe after <strong>the</strong> fall <strong>of</strong> <strong>the</strong> Iron Curtain have 

played a part not to be underestimated. The 

economic opening has created opportunities, 

but has at <strong>the</strong> same time brought Germany, as 

an established industrial location, under strong 

pressure. Faced with <strong>the</strong> choice between relocating 

to regions with lower labour costs 

and embarking upon rationalisation efforts, 

<strong>the</strong> extrusion industry solved <strong>the</strong> dilemma by 

opting for <strong>the</strong> second alternative. This set in 

motion unprecedented rationalisation initiatives, 

which have brought extrusion plants in 

Germany to a worldwide peak position from 

<strong>the</strong> technological standpoint. 

Numerous technical and organisational innovations 

have contributed to this successful 

development. And in this it is certain that, not 

least, <strong>the</strong> step by step automation <strong>of</strong> material 

flow through <strong>the</strong> extrusion plant has been an 

important factor. It can now be stated with 

some confidence that <strong>the</strong> extrusion plants 

which operate most successfully are those 

which paid attention to that aspect in good 

time. Since <strong>the</strong>re is unanimous conviction 

in <strong>the</strong> branch that in times to come competitive 

pressure will not get any less severe, <strong>the</strong> 

compulsion to rationalise and automate is becoming 

still more urgent – and thus too, <strong>the</strong> 

demand for automated internal logistics. 

Analysis and planning – <strong>the</strong> 

starting point <strong>of</strong> any project 

At H+H Herrmann + Hieber every project 

begins with analysis and comprehensive, forward-looking 

planning. This demands specialists 

who not only have mastery over <strong>the</strong> tools 

for material flow planning, but in addition 

possess comprehensive knowledge about <strong>the</strong> 

processes involved in extrusion operations. 

Automated material flow at Pandolfo Alluminio 

In every case <strong>the</strong> planning is based on <strong>the</strong> 

individual circumstances <strong>of</strong> a project and on 

<strong>the</strong> planning figures <strong>of</strong> <strong>the</strong> production operation. 

As demonstrated for H+H in well over 

50 successfully completed extrusion projects, 

<strong>the</strong> specifications and requirements to be met 

by <strong>the</strong> planning are always different. In practice 

<strong>the</strong>re is no standard solution. 

Planning starts with observation <strong>of</strong> <strong>the</strong> 

current situation, for which all <strong>the</strong> data are 

collected. The quantitative framework (press 

capacity, ageing capacity, etc.) is specified by 

<strong>the</strong> production department. For <strong>the</strong> definition 

<strong>of</strong> <strong>the</strong> target situation, <strong>the</strong> planning must generally 

take into account that what is involved 

is not static values, but a dynamic, variable 

operating situation. This means that <strong>the</strong> extrusion 

plant must by capable <strong>of</strong> responding flexibly 

to changing market needs. The associated 

material flow variations necessarily have to be 

based on forward-looking planning considerations. 

In this context overall planning takes into 

account not just <strong>the</strong> transport <strong>of</strong> extruded sections, 

but also that <strong>of</strong> all auxiliary materials 

required for <strong>the</strong> production process. This includes 

in particular <strong>the</strong> transport <strong>of</strong> spacers 

and packing materials, as well as <strong>the</strong> transport 

<strong>of</strong> scrap. 

When <strong>the</strong> individual material flows have 

been defined and <strong>the</strong> necessary areas calculated, 

<strong>the</strong> next step is to implement <strong>the</strong> solution 

in design and structural terms. For this, 

in <strong>the</strong> first instance tried and tested basic 

modules are available – manually controlled 

trolleys, automatic cranes and o<strong>the</strong>rs – whose 

dimensions and capacities can be adapted to 

suit <strong>the</strong> requirements in each case. During this 

planning phase it is found in many cases that 

<strong>the</strong> planned material flow cannot be achieved 

with <strong>the</strong> standard components available – or 

only insufficiently so. At that point <strong>the</strong> innovative 

ability <strong>of</strong> <strong>the</strong> planning team is called 

into play. 

Generally, in projects <strong>of</strong> this type <strong>the</strong> 

changes to be made have to be carried out 

while interfering as little as possible with 

on-going extrusion operations. If this is not 

achieved <strong>the</strong> investment project is doomed to 

failure already from <strong>the</strong> start, since it is <strong>the</strong>n 

no longer viable. 

Pandolfo Alluminio – an 

example <strong>of</strong> a successful project 

How to set about implementing such a project, 

and what its effects can be, is shown here by 

considering <strong>the</strong> example <strong>of</strong> <strong>the</strong> Italian extrusion 

plant Pandolfo Alluminio in Padua in 

nor<strong>the</strong>rn Italy. 

“Please supply us with a detailed material 

flow plan, to include all structural changes for 

automating <strong>the</strong> pallet transport with a view 

to increasing output in <strong>the</strong> medium term by 

about a third, up to 32,000 tonnes per year!” 

With this request, made by Gianfranco Pandolfo, 

chairman <strong>of</strong> <strong>the</strong> board and CEO <strong>of</strong> <strong>the</strong> 

Italian extrusion plant Pandolfo Alluminio to 

H+H, a comprehensive modernisation and 

extension programme worth some 15 million 

euros began in 2003. The result sets new standards 

in <strong>the</strong> sector. 


Images: H + H

The medium-sized, family-run company was 

at that time producing up to 23,000 tpy at its 

Plant 1, with around 400 employees and four 

extrusion lines. In Plant 2, about five kilometres 

from <strong>the</strong> extrusion plant, <strong>the</strong> Pandolfo 

group also operates efficient section machining 

and surface finishing facilities. 

At <strong>the</strong> start <strong>of</strong> <strong>the</strong> project, toge<strong>the</strong>r with 

Pandolfo H+H worked out a general plan 

which – after approval by <strong>the</strong> customer – was 

implemented in several steps, and this in such 

a way that section production was affected as 

little as possible. The fourth and for <strong>the</strong> time 

being <strong>the</strong> last step <strong>of</strong> this comprehensive 

project was completed according to plan in 

2008. 

The project starts at <strong>the</strong> point where <strong>the</strong> 

sawn section lengths, stacked in racks, are 

available. At this point <strong>the</strong> previously fixed 

material flow has now been replaced by flexible 

handling. This flexible automation makes 

special demands. At Pandolfo <strong>the</strong> number <strong>of</strong> 

logistical options available was drastically increased 

by free allocation <strong>of</strong> <strong>the</strong> material to 

furnaces, storage positions and packing stations 

and by partial sorting out for fur<strong>the</strong>r 

processing. 

At <strong>the</strong> start <strong>of</strong> <strong>the</strong> project, according to plan 

<strong>the</strong> heat treatment 

capacities were extended 

and modernised. 

Two older heat 

treatment furnaces 

were dismantled and 

replaced by new units. 

In addition <strong>the</strong> heat 

treatment capacity 

<strong>of</strong> <strong>the</strong> plant was also 

massively increased 

by <strong>the</strong> commissioning 

<strong>of</strong> a fur<strong>the</strong>r, large 

furnace unit. The final 

result is that <strong>the</strong> plant 

now has a total <strong>of</strong> seven 

furnace units. 

In accordance with works specifications <strong>the</strong> 

total production from all four presses can be 

allocated arbitrarily to any <strong>of</strong> <strong>the</strong> total <strong>of</strong> seven 

heat treatment furnaces. Operation is <strong>the</strong>refore 

free and individual charges, made up in 

accordance with various criteria, can be distributed 

between <strong>the</strong> existing furnaces without 

any restriction. Besides, <strong>the</strong>re is <strong>of</strong> course 

<strong>the</strong> option to omit heat treatment entirely. 

All <strong>the</strong> furnaces are charged by a central 

automatic crane which moves over <strong>the</strong> full 

EXTRUSION 

Interim storage <strong>of</strong> racks at <strong>the</strong> transfer station between two automatic cranes 

width <strong>of</strong> <strong>the</strong> shed. In this case <strong>the</strong> automatic 

crane is, as it were, <strong>the</strong> link between <strong>the</strong> presses 

and <strong>the</strong> heat treatment. The racks, filled by 

stacking machines, are stored intermediately 

with <strong>the</strong> help <strong>of</strong> <strong>the</strong> crane at deposition points 

underneath <strong>the</strong> crane track and are <strong>the</strong>n taken 

to <strong>the</strong> furnace concerned as necessary. 

At <strong>the</strong> outlet <strong>of</strong> each furnace <strong>the</strong> heat 

treated goods are checked. Special inspection 

positions have been installed for this purpose. 

All <strong>the</strong> racks that leave <strong>the</strong> heat treatment fur- 

TOOL STEEL AND TOOLING 

COMPETENCE 

FROM 

ONE SOURCE 

Kind & Co., Edelstahlwerk, KG 

Bielsteiner Str. 124 – 130 

D-51674 Wiehl 

Tel +49 (0) 22 62 / 84-0 

Fax +49 (0) 22 62 / 84-175 

info@kind-co.de · www.kind-co.de

EXTRUSION 

naces pass through this inspection 

stage. The heat treatment process 

data are recorded by <strong>the</strong> transport 

management system (TMS), archived, 

and sent on to <strong>the</strong> superordinated 

operational computer level 

<strong>of</strong> <strong>the</strong> plant for charge tracking 

in <strong>the</strong> context <strong>of</strong> <strong>the</strong> QA system. 

The checked racks are again 

stored intermediately and are <strong>the</strong>n 

taken to individual packing positions 

as necessary. In this area too 

<strong>the</strong>re has to be unrestricted flexibility: 

any rack can be assigned to 

any packing position. To be able 

to achieve this a hi<strong>the</strong>rto unique 

transport and storage system was 

developed. Owing to <strong>the</strong> numerous transport 

options and <strong>the</strong> consequently large storage 

area involved, a system comprising several 

automatic cranes was provided. The cranes operate 

redundantly, i. e. <strong>the</strong>y match <strong>the</strong>ir actions 

to one ano<strong>the</strong>r in a flexible manner. Fur<strong>the</strong>rmore, 

as a special feature automatic optimisation 

<strong>of</strong> all transport and storage movements 

is provided and this covers all <strong>the</strong> transport 

units. In practice this means that for every task 

<strong>the</strong> optimum procedure (shortest path, smallest 

number <strong>of</strong> individual movements, etc.) is 

sought. This takes place having regard to <strong>the</strong> 

system as a whole, i.e. <strong>the</strong> rack movements are 

co-ordinated automatically in accordance with 

<strong>the</strong> specified optimum. 

A fur<strong>the</strong>r automatic crane is provided for 

goods to be separated out, which are taken in 

special racks to Plant 2 for fur<strong>the</strong>r processing. 

These sections have to be re-stacked in <strong>the</strong> 

special racks. The same crane provides transport 

to <strong>the</strong> re-stacking station. 

The detailed planning <strong>of</strong> this area had 

to ensure that <strong>the</strong> floor <strong>of</strong> <strong>the</strong> shed remains 

freely accessible and not obstructed by transport 

units. This was achieved by locating <strong>the</strong> 

rack transport so far as possible around <strong>the</strong> 

edges. Full racks are transported at floor level 

whereas in contrast empty racks are generally 

moved three metres above floor level. 

The packing positions are located after <strong>the</strong> 

store areas. Here too, <strong>the</strong> packing positions 

can receive material automatically from any 

part <strong>of</strong> <strong>the</strong> store. For this, <strong>the</strong> racks are placed 

by one <strong>of</strong> <strong>the</strong> automatic cranes onto <strong>the</strong> appropriate 

roller track and taken automatically 

to <strong>the</strong> packing position intended. At present a 

total <strong>of</strong> seven packing positions are available. 

Each <strong>of</strong> <strong>the</strong>m is equipped with accessories 

which facilitate <strong>the</strong> work <strong>of</strong> <strong>the</strong> packers and 

help to increase packing output. 

The system as a whole is controlled by a 

central transport management system. This 

Automatic crane No 4 with overhead roller track for empty racks at <strong>the</strong> back 

system, which also assists production planning, 

was developed by H+H toge<strong>the</strong>r with <strong>the</strong> 

company Aberle Steuerungstechnik in Leingarten. 

At Pandolfo <strong>the</strong> new system was installed 

already during <strong>the</strong> first phase <strong>of</strong> project 

implementation, so that <strong>the</strong> areas completed 

one after ano<strong>the</strong>r could be integrated into 

<strong>the</strong> system step by step. The central transport 

management system records all <strong>the</strong> section 

racks by means <strong>of</strong> barcodes and controls <strong>the</strong>ir 

movements up to <strong>the</strong> storage <strong>of</strong> <strong>the</strong> packed 

goods. It is consequently in a position at anytime 

to provide information about <strong>the</strong> current 

position and processing status <strong>of</strong> each and 

every rack. This ability makes <strong>the</strong> transport 

management system a valuable aid to production 

planning. 

Moreover, it was decided at Pandolfo 

to integrate scrap transport as well into <strong>the</strong> 

automatic system. Any scrap produced in or 

arriving at <strong>the</strong> packing area, which has to be 

returned to <strong>the</strong> foundry, is taken <strong>the</strong>re fully 

automatically by a central transport system. 

To maximise value recovery, this even takes 

place after sorting into distinct alloys. 

Success factors: flexibility 

and low operating costs 

Already at <strong>the</strong> beginning it was stated that automated 

section transport through <strong>the</strong> plant 

– under <strong>the</strong> conditions prevailing in <strong>the</strong> work 

<strong>of</strong> central-European extrusion plants – has 

economic advantages and substantially improves 

<strong>the</strong> competitiveness <strong>of</strong> an appropriately 

equipped extrusion plant. 

The prerequisite for this is, firstly, that operations 

must take place absolutely faultlessly. 

For that reason particular attention is paid to 

<strong>the</strong> aim <strong>of</strong> ‘maximum availability’. At Pandolfo, 

for example, <strong>the</strong> design <strong>of</strong> a new rack 

type adapted to <strong>the</strong> system has proved to be a 

decisive factor. Before being fed into <strong>the</strong> trans- 

port system each rack is checked 

for damage. In this way and with 

a series <strong>of</strong> o<strong>the</strong>r measures, a total 

availability <strong>of</strong> 98.5 percent over 

<strong>the</strong> year, and for some parts <strong>of</strong> <strong>the</strong> 

plant even more than 99 percent, 

is achieved. 

The end result is that <strong>the</strong> logistical 

objectives <strong>of</strong> <strong>the</strong> project have 

been fully implemented: 

• automated material transport 

• absolute flexibility, i. e. all aggregates 

can freely be called into 

action 

• linking <strong>of</strong> <strong>the</strong> material flow to 

and from Plant 2 

• automatic return transport to <strong>the</strong> 

central scrap collection point and <strong>the</strong> foundry, 

etc. 

The conditions set by <strong>the</strong> customer: 

• maximum availability <strong>of</strong> all units 

• no damage during transport 

• implementation without interfering with 

operations 

• high safety standards 

• unobstructed shed floor surface 

have also been met in full. Fur<strong>the</strong>rmore, at 

<strong>the</strong> conclusion <strong>of</strong> <strong>the</strong> project it emerged that 

<strong>the</strong> forecast planning figures are in some cases 

being substantially exceeded. 

The extrusion enterprise is benefiting from 

<strong>the</strong> automation <strong>of</strong> its mat flow in many ways. 

The rationalisation effect is directly visible in 

<strong>the</strong> reduced personnel needs and <strong>the</strong> troublefree 

operation. Transport-generated scrap, a 

not negligible cost factor during <strong>the</strong> operation 

<strong>of</strong> a conventionally equipped extrusion plant, 

has never yet been produced so far! 

Above all, however, <strong>the</strong> company is in a 

position to respond, with <strong>the</strong> same staff, to 

changing market circumstances. In <strong>the</strong> case <strong>of</strong> 

Pandolfo Alluminio it has proved possible in 

this way to increase <strong>the</strong> plant’s capacity by 

more than <strong>the</strong> promised 35 percent. If necessary 

<strong>the</strong> company is now capable <strong>of</strong> producing 

35,000 tonnes a year. Should <strong>the</strong> market 

weaken, however, as it did unexpectedly for 

extrusion plants at <strong>the</strong> end <strong>of</strong> 2008, production 

can be adapted flexibly thanks to <strong>the</strong> 

lower staffing level. 

To what extent economy and flexibility 

can be increased in specific, individual cases 

depends on <strong>the</strong> initial situation and on <strong>the</strong> 

circumstances <strong>of</strong> <strong>the</strong> case. No generally valid 

figure can be given, for that reason. However, 

<strong>the</strong> projects implemented until now show that 

<strong>the</strong> return <strong>of</strong> investment (ROI) for <strong>the</strong> resources 

devoted to extrusion plant automation is 

not longer than four to five years. 

� 


�� 

�� 

�� 

Please be our guest 

and discover <strong>the</strong> benefits <strong>of</strong> <strong>the</strong> 

Aluminium ePaper yourself in a 

free three-month trial: 

�� 

�� 

�� 

�� 

�� 

�� 

��

APPLICATION-ORIENTED TECHNOLOGIES 

Laser cleaning and surface modification 

<strong>of</strong> <strong>aluminium</strong> – first step to a green plant 

Edwin Büchter, Clean-Lasersysteme GmbH 

Advanced industrial lasers have evolved 

well beyond simple cutting and welding 

applications. Laser technology now <strong>of</strong>fers 

an industrial de-coating and surface 

cleaning solution that is cost-effective 

as well as responsive to environmental 

concerns. From <strong>the</strong> automated cleaning 

<strong>of</strong> moulds to precise de-coating to 

<strong>oxide</strong> removal, laser surface treatments 

are proving to be an attractive option to 

traditional labour intensive methods. Besides 

simple cleaning application <strong>the</strong> laser 

technology is also capable to apply modifications 

to <strong>the</strong> substrate surface such as 

roughening or changing <strong>the</strong> metallurgical 

structure. Using a laser for surface modification 

means substituting critical conventional 

chemical media blasting process 

by cleaning with laser light. 

In <strong>the</strong> past decade laser cleaning and surface 

treatment systems have generated significant 

interest as a viable alternative to conventional 

cleaning and paint removal technologies. 

Research <strong>of</strong> mobile, reliable, and powerful 

laser systems for cleaning and paint removal 

operations began in <strong>the</strong> late eighties with <strong>the</strong> 

modification <strong>of</strong> weld or cutting lasers into laser 

systems for surface preparation. This approach 

did not meet <strong>the</strong> requirements for a surface 

preparation laser, which are significantly different 

than for cutting and welding. In <strong>the</strong> 

early nineties, research took place around <strong>the</strong> 

world for more efficient, reliable laser systems 

for surface preparation work. It took ano<strong>the</strong>r 

few years to develop <strong>the</strong> technology from a 

laboratory system to a system capable for use 

in day-to-day industrial operations. Today, laser 

systems are widely used for various surface 

preparation tasks in many industries. Those 

tasks include: mould cleaning, paint removal, 

joining pretreatment, oil and grease removal, 

and many more. 

Operating principle 

The laser generates a directed and monochromatic 

beam <strong>of</strong> light. In order to create a high 

power density, <strong>the</strong> laser beam is also typically 

tightly focused. In <strong>the</strong> focal point, <strong>the</strong> intense 

laser beam is absorbed by <strong>the</strong> contamination 

or paint and <strong>the</strong>rmally incinerates or sublimates 

<strong>the</strong> target material, i. e. paint or con- 

Operating principle (Nd:YAG): pulsed laser vaporizes <strong>the</strong> absorbing ‘dirt’ <strong>layer</strong>, scanned beam covers and 

cleans <strong>the</strong> whole surface until <strong>the</strong> blank substrate reflects <strong>the</strong> laser radiation 

taminations. This incineration or vaporization 

will, in combination with <strong>the</strong> resulting micro 

<strong>the</strong>rmal shockwave, remove <strong>the</strong> target material 

as long as <strong>the</strong> target material is able to 

absorb <strong>the</strong> laser energy. The better <strong>the</strong> target 

material absorbs <strong>the</strong> energy, <strong>the</strong> faster it can 

be removed. 

Colour, chemical composition and thickness 

<strong>of</strong> <strong>the</strong> target <strong>layer</strong> all have a direct impact on 

<strong>the</strong> effectiveness <strong>of</strong> <strong>the</strong> process. The removal 

process automatically stops once a metal substrate 

is reached since metal surfaces reflect 

<strong>the</strong> laser beam and do not generally absorb 

<strong>the</strong> laser energy. 

The heat transfer into <strong>the</strong> substrate material 

can be a critical factor. To minimize this 

effect, many laser equipment manufacturers 

use pulsed laser sources. 

The laser intensity, known as <strong>the</strong> laser 

power per beam spot, is a critical parameter 

for <strong>the</strong> heat transfer into <strong>the</strong> substrate material. 

Very short laser pulses with a pulse duration 

<strong>of</strong> only a few nanoseconds (ns) in combination 

with a very small focus diameter (less 

<strong>the</strong>n 500 μm) result in a minimal heat transfer 

into <strong>the</strong> substrate material. Under normal operating 

conditions and with <strong>the</strong> right process 

parameters, damage to <strong>the</strong> substrate material 

can be eliminated. The heat transfer factor <strong>of</strong> 

continuous wave laser systems is much higher 

and might result in substrate temperatures that 

will damage <strong>the</strong> substrate. Test results with a 

Precise automated paint stripping / manual paint stripping for repair applications 


Surface modification and de-coating <strong>of</strong> anodized <strong>layer</strong>s for electrical bonding 

Laser modification on <strong>aluminium</strong> surfaces for adhesive bonding and improved wettability prior to painting 

handheld pulsed Nd:YAG laser with an average 

laser power <strong>of</strong> 500 W (peak power <strong>of</strong> over 

400 kW) on an <strong>aluminium</strong> sheet resulted in 

maximum substrate temperatures <strong>of</strong> 80°C. 

Pulsed laser systems generate laser power 

levels well beyond <strong>the</strong> average power <strong>of</strong> <strong>the</strong> 

laser source. A pulsed 150 W solid-state laser 

will generate a peak pulse power <strong>of</strong> over 

160 kW. This high peak power and <strong>the</strong> above 

mentioned beam parameter results in a power 

intensity removing many target materials with 

acceptable production rates. 

Currently, <strong>the</strong>re are three different kinds 

<strong>of</strong> laser sources available for surface preparation 

works. The main difference is <strong>the</strong> laser 

generation and <strong>the</strong> resulting beam delivery 

configuration. Respecting 

<strong>the</strong> automation and 

integration possibilities, 

solid-state Nd:YAG laser 

<strong>of</strong>fer special advantages 

in industrial, automated 

applications such adhesive 

bonding preparation, pint 

stripping, mould cleaning 

and pre- / post treatment 

for welding and brazing. 

The operative wavelength 

<strong>of</strong> 1,064 nanometres 

(nm) lies within <strong>the</strong> 

transmission bandwidth <strong>of</strong> 

common optical glass and 

enables <strong>the</strong> use <strong>of</strong> fibre 

optic cables for beam delivery. Fibre optics 

dramatically increases <strong>the</strong> flexibility <strong>of</strong> <strong>the</strong> 

laser system. Nd:YAG lasers can be used for 

work on hard to reach areas. Currently, <strong>the</strong> 

maximum fibre optic length is 150 ft. Nd:YAG 

lasers are nearly maintenance free and very 

simple to operate. Pulsed systems for surface 

preparation use reach average laser power 

levels <strong>of</strong> up to 1,000 W. 

Sustainable technology 

Due to <strong>the</strong> media free cleaning with light, 

cleanLaser <strong>of</strong>fer several advantages. These 

advantages, which are leading to sustainable 

savings, are: 

Laser cleaning <strong>of</strong> <strong>aluminium</strong> car body parts for welding preparation 


• No media and chemical materials required 

• Energy consumption far low 

(typically 5 kW or less) 

• Up to 80% lower running costs and waste 

<strong>of</strong> resources significant reduced 

• No noise and dirt allocation leading to 

better and safer workplaces. 

This combination <strong>of</strong> saving resources an reduction 

<strong>of</strong> costs will have a sustainable long 

term effect in industrial cleaning and surface 

applications. 

In October 2010, <strong>the</strong> founders <strong>of</strong> cleanLA- 

SER have been awarded with Europe’s highest 

doped environmental award, <strong>the</strong> German environmental 

award (Deutscher Umweltpreis). 

Clean Laser has also been nominated for <strong>the</strong> 

Clean Tech Media Award 2011. 

Current application fields 

for cleanLaser technology 

Currently, a wide range <strong>of</strong> industrial application 

fields are applicable for cleanLaser surface 

treatment and cleaning technology: 

• Cleaning <strong>of</strong> metal parts, especially alumin- 

ium for following production steps 

• Pre-treatment prior to painting 

• Adhesive bonding and pre-treatment <strong>of</strong> 

<strong>aluminium</strong> for long term stability 

• Welding and brazing preparations 

• (Precise) de-coating and ablation <strong>of</strong> <strong>layer</strong>s 

for joining or electrical bonding <strong>of</strong> plated 

<strong>aluminium</strong> 

• Modification and structuring <strong>of</strong> (metal) 

surfaces for enhanced performance. 

cleanLaser Systems 

Clean-Lasersysteme GmbH, an ISO 9001: 

2008 certified laser system manufacturer <strong>of</strong>fers 

a wide range <strong>of</strong> powerful laser systems for 

surface applications. These systems are available 

with different end effectors. Laser optics 

are available for manual use and as well as for 

automated applications. Due to cleanLaser’s 

fibre coupling technology, <strong>the</strong> compact basic 

laser unit can be connected to a laser optics by 

using an up to 50 m long flexible fibre cable. 

The flexibility <strong>of</strong>fers easy integration as 

well as suitable power ranges for applications 

in an industrial environment. Advantages such 

as in-line cleaning are evident. 

High efficient laser systems mainly based 

on Q-switched solid state laser sources guaranty 

short laser pulse impact times for sensitive 

cleaning by nothing else <strong>the</strong>n laser light. 

Current laser systems are available in a 

power range between 20 to 1,000 Watts, ready 

for serving nearly every speed and cycle time 

demand. 

� 


Audi AG


Corrosion behaviour <strong>of</strong> <strong>aluminium</strong> 

materials in aqueous cleaning solutions 

Silvio Koehler, Georg Reinhard, Excor Korrosionsforschung GmbH 

Especially in <strong>the</strong> car and aviation industry 

<strong>aluminium</strong> and its alloys are <strong>the</strong> most important 

materials. One advantage <strong>of</strong> <strong>aluminium</strong> 

is <strong>the</strong> generally passive and corrosion resistance 

in aqueous solutions within a defined 

pH range except for pitting corrosion due to 

some reactive species, such as chloride. While 

pure <strong>aluminium</strong> is covered by a homogenous 

<strong>oxide</strong> <strong>layer</strong>, <strong>aluminium</strong> alloys <strong>of</strong>ten consist <strong>of</strong> 

several phases (aggregates and mixed phases 

<strong>of</strong> alloying elements) leading to an inhomogeneous 

<strong>oxide</strong> <strong>layer</strong>. The effect <strong>of</strong> alloying elements 

on <strong>the</strong> breakdown <strong>of</strong> <strong>the</strong> passive film 

was extensively studied using various grades 

<strong>of</strong> <strong>aluminium</strong> and different metals [1,2]. 

At <strong>the</strong> cleaning <strong>of</strong> <strong>aluminium</strong> materials 

most <strong>of</strong> <strong>the</strong> problems arise because <strong>of</strong> <strong>the</strong> 

wrong pH-value <strong>of</strong> <strong>the</strong> respective cleaning 

bath. The Pourbaix diagram <strong>of</strong> <strong>aluminium</strong> can 

be used for a rough estimation, but it is not valid 

by 100% for <strong>aluminium</strong> alloys. The second 

problem is <strong>the</strong> present <strong>of</strong> corrosion promoting 

anions, not only chloride. In combination with 

an unsuitable pH-value <strong>the</strong> follows for <strong>the</strong> material 

to clean can be dramatically, also in <strong>the</strong> 

present <strong>of</strong> small quantities. 

Before <strong>the</strong> corrosion ability <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> 

materials Al1050 and 2014 was characterized 

in <strong>the</strong> aqueous solutions <strong>of</strong> commercial 

available cleaning concentrates, <strong>the</strong> effect <strong>of</strong> 

<strong>the</strong> pH and <strong>the</strong> chloride concentration were 

investigated in model electrolytes with pH 

values between 3 and 10 at room temperature. 

In ano<strong>the</strong>r set <strong>of</strong> experiments sodium chloride 

was added to solutions <strong>of</strong> pH8. In all solutions 

<strong>the</strong> free corrosion potential as well as time 

depending impedance spectra were recorded 

using <strong>the</strong> <strong>aluminium</strong> materials as working 

electrode in a top part measuring cell. 

For pure <strong>aluminium</strong> (Al 1050) <strong>the</strong> results 

measured in <strong>the</strong> model electrolytes with 

<strong>the</strong> different pH values were in accordance 

to <strong>the</strong> Pourbaix diagram. Fur<strong>the</strong>r, passive 

<strong>aluminium</strong> is ra<strong>the</strong>r stable up to high chloride 

contents. But in <strong>the</strong> industrial aqueous 

cleaning solutions Al1050 corrosion attacks 

were observed, although <strong>the</strong> pH value <strong>of</strong> <strong>the</strong> 

cleaning solution was in <strong>the</strong> passive region 

<strong>of</strong> <strong>aluminium</strong>. Surfactants and complexing 

agents included in <strong>the</strong> cleaners seem to have 

a corrosion promoting effect. In <strong>the</strong> case <strong>of</strong> 

<strong>the</strong> <strong>aluminium</strong> alloy Al 2014 in all solutions 

(model electrolytes as well as cleaning baths) 

a corrosion attack was observed. It could be 

shown by means <strong>of</strong> SEM that in <strong>the</strong> case <strong>of</strong> 

Al 2014 <strong>the</strong> selective corrosion around <strong>the</strong> 

cathodic copper-rich inter-metallic dominates 

<strong>the</strong> corrosion performance. 

The next step was to find suitable inhibitors 

for <strong>the</strong> cleaning solution, which are able 

to protect <strong>the</strong> <strong>aluminium</strong> materials, but <strong>the</strong>y 

should not lower <strong>the</strong> pH-value to uphold <strong>the</strong> 

cleaning effect. Promising results were observed 

at combinations <strong>of</strong> substances creating 

a converting <strong>layer</strong> on <strong>the</strong> <strong>aluminium</strong> basis 

material with substances inhibiting <strong>the</strong> copper 

rich phases selectively. 

1. Introduction 

Aluminium materials are widely used in <strong>the</strong> 

automotive, aerospace and railways industries 

[3]. Therefore, protection concepts for semifinished 

as well as final products made <strong>of</strong> <strong>aluminium</strong> 

and its alloys are requested. 

One way to protect <strong>aluminium</strong> surfaces 

during <strong>the</strong>ir storage into moist air is <strong>the</strong> use 

<strong>of</strong> volatile corrosion inhibitors. These are inhibiting 

substances with a sufficient vapour 

pressure, which can be incorporated in <strong>the</strong> 

matrix <strong>of</strong> <strong>the</strong> packaging materials like paper 

and films. The VCI can be released by sublimation 

and/or be transporting by water vapour 

into <strong>the</strong> package. In <strong>the</strong> case <strong>of</strong> densely closed 

packages <strong>the</strong> VCI enrich and saturate <strong>the</strong> inner 

atmosphere. Through a reversible adsorption 

on <strong>the</strong> metal surface <strong>the</strong>y inhibit corrosion as 

far as <strong>the</strong> packaging is closed [4,5]. Besides <strong>the</strong> 

densely closed packages <strong>the</strong> right combination 

<strong>of</strong> inhibitors has to be used depending on <strong>the</strong> 

composition <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> alloy [6]. 

The surfaces to protect have to be accessible 

and clean. To install an effective cleaning 

process a plenty <strong>of</strong> parameters and requirements 

have to take into account especially 

<strong>the</strong> compatibility <strong>of</strong> cleaner and material [7]. 

At <strong>the</strong> cleaning <strong>of</strong> <strong>aluminium</strong> materials most 

<strong>of</strong> <strong>the</strong> problems arise because <strong>of</strong> <strong>the</strong> wrong 

pH-value <strong>of</strong> <strong>the</strong> respective cleaning bath. The 

second problem is <strong>the</strong> present <strong>of</strong> corrosion 

promoting anions like chloride, because <strong>of</strong> 

<strong>the</strong> long-term use <strong>of</strong> <strong>the</strong> respective cleaning 

baths. In combination with an unsuitable pHvalue 

<strong>the</strong> follows for <strong>the</strong> material to clean can 

be dramatically, also in <strong>the</strong> present <strong>of</strong> small 

quantities. It is known that most <strong>of</strong> <strong>the</strong> cleaning 

processes take place at temperatures around 

60°C. Here, <strong>the</strong> pH value is shifted to more 

alkaline values compared to room temperature. 

Fur<strong>the</strong>r, <strong>the</strong> solubility <strong>of</strong> solids (e. g. surface 

<strong>oxide</strong> <strong>layer</strong>s, inorganic particles) as well 

as <strong>the</strong> velocity <strong>of</strong> <strong>the</strong>ir dissolution is increased 

and intended for good cleaning results with- 

in short contact times [8]. It is not <strong>the</strong> purpose 

<strong>of</strong> this paper to describe <strong>the</strong> mechanism during 

<strong>the</strong> cleaning process itself. Especially <strong>the</strong> 

electrochemical measurements should be presented 

as tools to answer questions about <strong>the</strong> 

compatibility <strong>of</strong> cleaning bath and material to 

clean and to evaluate <strong>the</strong> current ‘corrosiveness’ 

<strong>of</strong> a cleaning bath similar to <strong>the</strong> chip/ 

filter paper test according to DIN 51360 part 

2 finally with <strong>the</strong> aim to obtain conditions <strong>of</strong> 

<strong>the</strong> metal surface that <strong>the</strong> application <strong>of</strong> VCI 

is successful. 

2. Experimental 

Test panels (50 x 100 mm) <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> 

alloys mentioned in Table 1 were used for <strong>the</strong> 

experiments. At first <strong>the</strong> panels were cleaned 

with methanol and acetone assisted by an ultrasonic 

bath. After this <strong>the</strong>y have been stored 

in a desiccator for at least 12 h. So a reproducible 

formation <strong>of</strong> a native <strong>oxide</strong> <strong>layer</strong> was assured. 

For <strong>the</strong> electrochemical measurements 

<strong>the</strong> panels were put in a cylindrical top cell 

realizing a circular electrode area <strong>of</strong> 1.76 cm² 

acting as working electrode. As counter and 

reference electrode a platinum net and saturated 

calomel electrode (SCE) were used. All 

electrical measurements were performed using 

<strong>the</strong> impedance measurement system IM6 

(Zahner Instruments, Kronach, Germany) at 

room temperature (22°C). In a first set <strong>of</strong> experiments 

<strong>the</strong> influence <strong>of</strong> <strong>the</strong> pH value on <strong>the</strong> 

corrosion were investigated. Therefore, aqueous 

solutions (de-ionized water) including 

0.01M KNO 3 as supporting electrolyte <strong>of</strong> <strong>the</strong> 

Material Si Fe Cu Mn Mg Cr Zn Ti Zr 

1050 0.09 0.25 0.001 0.01 0.01 0 0.01 0.01 0 

2014 0.88 0.11 4.85 0.85 0.73 0.004 0.05 0.02 0.01 

Table 1: Composition <strong>of</strong> <strong>the</strong> investigated <strong>aluminium</strong> alloys 


following pH values were prepared: pH 3, pH 

5, pH 8 and pH 10. The pH value was adjusted 

with HNO 3 and NaOH, respectively. After 

<strong>the</strong>se experiments 1-3 wt% aqueous solutions 

(de-ionized water) including 0.01M KNO 3 as 

supporting electrolyte <strong>of</strong> <strong>the</strong> commercial available 

cleaner concentrates mentioned in Table 

2 were prepared. In all aerated solutions <strong>the</strong> 

free corrosion potential has been measured for 

30 min. After this electrochemical impedance 

spectra (EIS) were recorded at <strong>the</strong> respective 

open circuit potentials in a frequency range <strong>of</strong> 

100 kHz to 50 MHz after 5, 10, 15 and 30min 

<strong>of</strong> immersion time. Thereby, <strong>the</strong> impedance 

spectra were taken at o<strong>the</strong>r surface areas than 

<strong>the</strong>se used for <strong>the</strong> recording <strong>of</strong> <strong>the</strong> free corrosion 

potentials. The EIS curves were fitted using 

<strong>the</strong> Thales s<strong>of</strong>tware (Zahner Instruments, 

Kronach, Germany). 

The samples <strong>of</strong> Al 2014 were investigated 

by SEM (Scanning Electron Microscopy) 

No. kind <strong>of</strong> cleaner suitable for* pH value at 25°C* 

1 demulsifying Fe, (Al) 9.1 ± 0.3 (1%) 

2 demulsifying Fe, Al, Zn 8.0 ± 0.2 (2%) 

3 emulsifying Fe, Al, Zn 8.6 ± 0.3 (1%) 

* according to product data sheet 

Table 2: List <strong>of</strong> investigated cleaners 

coupled with EDX (Energy-dispersive X-ray 

spectroscopy). Here, surface regions affected 

and not affected by cleaning solution 2 were 

investigated and compared concerning <strong>the</strong> ratio 

<strong>of</strong> <strong>the</strong> elements Al, Cu and O. The SEM 

images were recorded with a LEO 440 (STS, 

North Billerica, USA) at activation energies 

<strong>of</strong> 5 keV and 20 keV, respectively. The EDX 

spectra were recorded by <strong>the</strong> coupled SDD 

(Silicon Drift Detector) AXAS (Ketek GmbH, 

München, Germany). 

In ano<strong>the</strong>r set <strong>of</strong> experiments 0.01M benzotriazole 

and/or 0.01M sodium di-hydrogen 

phosphate was added to a 2 wt% solution <strong>of</strong> 

cleaner 2. The free corrosion potential as well 

as electrochemical impedance spectra were 

recorded as described above. 

3. Results and discussion 

At first <strong>the</strong> influence <strong>of</strong> <strong>the</strong> pH value and <strong>the</strong> 

chloride concentration were investigated on 

<strong>the</strong> alloy 1050, which consists <strong>of</strong> pure <strong>aluminium</strong> 

by almost 99.5%. As expected from 

<strong>the</strong> Pourbaix-diagram <strong>of</strong> <strong>aluminium</strong> [9] <strong>the</strong> 

free corrosion potential E corr measured in <strong>the</strong> 

pH range between 3 and 5 remains in a region, 

where <strong>the</strong> <strong>aluminium</strong> surface is in a passive 

state, while at pH 10 <strong>the</strong> free corrosion potential 

corresponds to an active dissolution (see 

Fig. 1). 


In alkaline solutions <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> dissolves 

under <strong>the</strong> formation <strong>of</strong> [Al(OH) 4 ]- complexes. 

The impedance spectra recorded at pH10 (Fig. 

2) show an increase <strong>of</strong> <strong>the</strong> C ox combined with 

a decreasing resistance in <strong>the</strong> region <strong>of</strong> lower 

frequencies. This behaviour is typical for a 

uniform dissolution <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> [10]. 

Strong deviations from <strong>the</strong> Pourbaix-diagram 

mentioned above can be observed at <strong>the</strong> 

investigation <strong>of</strong> <strong>aluminium</strong> alloys. The inter- 

Fig. 1: Free corrosion potential E corr <strong>of</strong> Al 1050 in 

solution <strong>of</strong> different pH value 

Fig.2: Impedance spectra <strong>of</strong> Al 1050 at several pH 

values after 30 min immersion time 

pretation <strong>of</strong> <strong>the</strong> free corrosion potentials <strong>of</strong> 

<strong>the</strong> <strong>aluminium</strong> alloys is ra<strong>the</strong>r complicated, 

because <strong>the</strong>y are mixed potentials with portions 

<strong>of</strong> bulk and/or selective corrosion processes. 

In <strong>the</strong> case <strong>of</strong> <strong>the</strong> copper-rich Al2014 

(Fig. 3) <strong>the</strong> curve in <strong>the</strong> beginning can be 

interpreted by a selective corrosion around 

<strong>the</strong> cathodic copper-rich inter-metallic phases 

followed by a stabilization / passivation at pHvalues 

between 3 and 10. In more acid as well 

as more alkaline solutions <strong>the</strong> free corrosion 

potentials are typical for <strong>the</strong> dissolution <strong>of</strong> 

<strong>the</strong> bulk <strong>aluminium</strong> [11,12]. More details are 

available from <strong>the</strong> impedance data. As already 

mentioned <strong>the</strong> alloy Al2014 is characterized 

by a higher content <strong>of</strong> copper, which forms inter-metallic 

phases. In <strong>the</strong> impedance spectra 

(Fig. 4) one can observe lower resistances in 

<strong>the</strong> low-frequency region compared to pure 

<strong>aluminium</strong>. Taking a look on <strong>the</strong> transient behaviour 

<strong>the</strong>se resistances keep constant on a 

low level at pH values below 3 and above 10. 

The capacities belonging to <strong>the</strong>m keep con- 

stant, too. Such a behaviour can be explained 

by a selective corrosion around <strong>the</strong> copper rich 

inter-metallic phases. 

Transferring <strong>the</strong> findings to industrial 

cleaning baths: Here, solutions <strong>of</strong> 3 commercial 

available cleaning concentrates were pre- 

Fig. 3: Free corrosion potential E corr <strong>of</strong> Al 2014 in 

solution <strong>of</strong> different pH value 

Fig. 4: Impedance spectra <strong>of</strong> Al 2014 at several pH 

values after 30min immersion time 

pared. All cleaners should be suitable to clean 

<strong>aluminium</strong> according to <strong>the</strong>ir product data 

sheets (see Table 2), however, at higher temperatures 

as room temperature. A strong uniform 

corrosion was observed already at room 

temperature in <strong>the</strong> baths prepared <strong>of</strong> cleaner 

1 and 3. This is explainable by <strong>the</strong> pH-value 

situated in <strong>the</strong> active region <strong>of</strong> <strong>aluminium</strong>. In 

<strong>the</strong> cleaner 2 with pH8 only a moderate attack 

was observed (Fig. 5). Maybe <strong>the</strong> bath ingredients 

accelerate <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> dissolution 

at local defects by complexion <strong>of</strong> <strong>aluminium</strong> 

[13]. For <strong>the</strong> <strong>aluminium</strong> alloy Al2014 <strong>the</strong> 

results found in <strong>the</strong> cleaning solutions were 

identical to this in <strong>the</strong> respective pH-model 

solutions. A black-colouring <strong>of</strong> <strong>the</strong> immerged 

alloy surfaces as shown in Fig. 6 was observed 

every time. 

Surface areas <strong>of</strong> Al 2014 affected (Fig. 6, 

black coloured area in <strong>the</strong> small picture) and 

not affected (Fig. 6, grey coloured area in <strong>the</strong> 

small picture) by cleaner 2 were investigated 

by SEM and EDX. According to <strong>the</strong> measured 

element ratios (Table 3) <strong>the</strong> selective corrosion 

and dissolution <strong>of</strong> <strong>aluminium</strong> around <strong>the</strong> copper 

rich phases could be proved. The black 



Fig. 5: Impedance spectra <strong>of</strong> Al1050 at pH8 and in 

solutions <strong>of</strong> <strong>the</strong> cleaner 1-3 (Table 2) 

Fig. 6: SEM image (5 keV, 200x): Al2014 after immersion 

in a solution <strong>of</strong> cleaner 2 

corrosion product is most likely Al(OH) 3, 

which only appears black because <strong>of</strong> <strong>the</strong> small 

<strong>layer</strong> thickness. 

The next step was to find suitable inhibitors 

for <strong>the</strong> cleaning solution, which are able to protect 

<strong>the</strong> <strong>aluminium</strong> materials, but <strong>the</strong>y should 

not lower <strong>the</strong> pH-value to uphold <strong>the</strong> cleaning 

effect. First experiments were done with sodium 

di-hydrogen phosphate and benzotriazole. 

The substances were added to a 2% solution 

<strong>of</strong> cleaner 2 toge<strong>the</strong>r and each alone. In Fig. 7 

<strong>the</strong> resistances R ox extracted from <strong>the</strong> impedance 

spectra at low frequency regions (around 

100 mHz) are presented. They correspond to 

<strong>the</strong> quality <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> and <strong>the</strong> corrosion 

resistance, consequently. Here, <strong>the</strong> inhibition 

effect <strong>of</strong> both substances is obvious, 

whereas <strong>the</strong> effect <strong>of</strong> each single substance is 

limited. The following mechanism is assumed: 

Phosphate is forming insoluble <strong>aluminium</strong> 

phosphate, inhibiting <strong>the</strong> dissolution <strong>of</strong> <strong>the</strong> 

<strong>aluminium</strong> bulk material. Benzotriazole is 

well known as inhibitor for copper. It forms 

conversion <strong>layer</strong>s on <strong>the</strong> copper rich phases, 

blocking <strong>the</strong> reduction <strong>of</strong> oxygen. In comparison 

<strong>the</strong> dissolution <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> bulk material 

dominates in <strong>the</strong> solution <strong>of</strong> cleaner 2. 

Therefore, <strong>the</strong> single addition <strong>of</strong> benzotriazole 

Cu Al O 

Non affected area 5 80 5 

Affected area 78 7 10 

Table 3: Results <strong>of</strong> EDX measurements, amounts <strong>of</strong> 

<strong>the</strong> elements Cu, Al and O in wt% 

Fig. 7: R OX extracted from <strong>the</strong> impedance spectra 

recorded in solutions <strong>of</strong> cleaner 2 with and without 

benzotriazole and sodium di-hydrogen phosphate 

to inhibit <strong>the</strong> inter-crystalline corrosion is insufficient 

to stop <strong>the</strong> bulk corrosion, whereas 

<strong>the</strong> single addition <strong>of</strong> phosphate can inhibit 

<strong>the</strong> bulk corrosion, but not <strong>the</strong> inter-crystalline 

corrosion. 

Finally, Fig. 8 shows panels <strong>of</strong> Al 2014 

completely cleaned (dip-cleaning assisted by 

ultrasound for 10 min) with cleaner 2 with and 

without <strong>the</strong> additionally added benzotriazole 

and sodium di-hydrogen phosphate. The effect 

is obvious [14]. 

4. Conclusions 

Using <strong>the</strong> electrochemical methods <strong>the</strong> different 

corrosion mechanisms <strong>of</strong> pure <strong>aluminium</strong> 

and Al 2014 could be detected. However, <strong>the</strong> 

behaviour <strong>of</strong> <strong>the</strong> free corrosion potentials is 

difficult to interpret especially for <strong>aluminium</strong> 

alloys. Here, <strong>the</strong> resulting potential is always 

a mixture <strong>of</strong> <strong>the</strong> response <strong>of</strong> <strong>the</strong> several inter 

metallic phases. 

Pure <strong>aluminium</strong> behaved passively in a 

pH range between 5 and 8. In <strong>the</strong> recorded 

impedance spectra only <strong>the</strong> resistance <strong>of</strong> <strong>the</strong> 

electrolyte and <strong>the</strong> capacity <strong>of</strong> <strong>the</strong> barrier <strong>layer</strong> 

were visible. At pH values above 10 a uniform 

dissolution <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> was observed 

characterized by a decreasing <strong>oxide</strong> <strong>layer</strong> resistance 

and an increasing surface capacity. 

The investigated <strong>aluminium</strong> alloy Al 2014 

is an inhomogeneous alloy and has a relatively 

high copper content. The surface <strong>oxide</strong> <strong>layer</strong> 

<strong>of</strong> <strong>the</strong> alloy is also inhomogeneous and <strong>the</strong> 

bulk material consists <strong>of</strong> several phases. For 

this alloy a uniform and/or selective corro- 

sion was found depending on <strong>the</strong> pH value 

already at room temperature. Below pH 3 

<strong>the</strong> alloy showed a uniform corrosion. In <strong>the</strong> 

pH range between 3 and 8 an increase <strong>of</strong> <strong>the</strong> 

resistance in <strong>the</strong> lower frequency range combined 

with an unchanged capacity at medium 

frequencies were detected. This response can 

be interpreted as selective corrosion around 

<strong>the</strong> cathodic copper rich phases. This (maybe) 

leads to a formation <strong>of</strong> corrosion products with 

a black visual appearance. They precipitated 

on <strong>the</strong> whole immersed surface. Additionally 

measurements by SEM/EDX proved <strong>the</strong> selective 

corrosion and dissolution <strong>of</strong> <strong>aluminium</strong> 

around <strong>the</strong> copper rich phases. At pH values 

above 10 a uniform dissolution was observed 

for Al 2014. 

Transferred to <strong>the</strong> industrial cleaning <strong>the</strong>se 

results reveal <strong>the</strong> importance <strong>of</strong> <strong>the</strong> ‘right 

choice’ <strong>of</strong> method and cleaner concentrate (pH 

value!) as well as an experienced bath monitoring 

(at least pH value and chloride content). 

Fur<strong>the</strong>rmore, <strong>the</strong> experiments showed that 

always <strong>the</strong> compatibility <strong>of</strong> <strong>the</strong> cleaner and 

<strong>the</strong> material has to be tested before <strong>the</strong> application. 

Therefore, <strong>the</strong> recording <strong>of</strong> electrochemical 

impedance spectra at <strong>the</strong> free corrosion 

potential is predestinated. A comparison 

<strong>of</strong> three commercial available cleaners (all 

should be suitable for <strong>aluminium</strong> according 

to <strong>the</strong>ir product data sheet) showed, that in 

cleaners with <strong>the</strong> pH value above 8.5 pure 

<strong>aluminium</strong> corroded heavily. But also in cleaners 

with a suitable pH-value a moderate at- 

tack was observed, caused by a complexion <strong>of</strong> 

<strong>aluminium</strong> by cleaner ingredients. 

The addition <strong>of</strong> both, sodium di-hydrogen 

phosphate and benzotriazole, can prevent 

Al2014 in near neutral cleaning solutions. But 

<strong>the</strong> single addition <strong>of</strong> benzotriazole to inhibit 

<strong>the</strong> inter-crystalline corrosion is insufficient to 

stop <strong>the</strong> bulk corrosion, whereas <strong>the</strong> single addition 

<strong>of</strong> phosphate can inhibit <strong>the</strong> bulk corrosion, 

but not <strong>the</strong> inter-crystalline corrosion. 

References 

[1] H. Zhan, J. M. C. Mol, F. Hannour, L. Zhuang, 

H. Terryn and J. H. W. de Wit; “The influence <strong>of</strong> 

copper content on intergranular corrosion <strong>of</strong> model 

Fig. 8: Al 2014 test panels after dip cleaning in a 2 wt% solution <strong>of</strong> cleaner 2; left: without benzotriazole 

and sodium di-hydrogen phosphate, right: with benzotriazole and sodium di-hydrogen phosphate 


AlMgSi(Cu) alloys”; Mat. Corr., 2008, 59, 670-675. 

[2] B. Zaid, D. Saidi, A. Benzaid, S. Hadji;“ Effects 

<strong>of</strong> pH and chloride concentration on pitting 

corrosion <strong>of</strong> AA6061 aluminum alloy“; Corros. 

Sci., 2008, 50, 1841-1847. 

[3] www.aluinfo.de, 2011. 

[4] U. Rammelt, S. Koehler, G. Reinhard; “Use <strong>of</strong> 

vapour phase corrosion inhibitors in packages for 

protecting mild steel against corrosion”; Corros. 

Sci., 2009, 51, 921-925. 

[5] S. Koehler, G. Reinhard; “Temporärer Korrosionsschutz”, 

Maschinenbau, 2008, 7, 34-37. 

[6] G. Reinhard, S. Lautner; “Temporärer Korro- 

sionsschutz von Aluminiumwerkst<strong>of</strong>fen mit flüch- 

tigen Korrosionsinhibitoren”; “GfKORR-Jahres- 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 


tagung 1998”, 1998, 97-108. 

[7] H. Thompsen, F.H. Rögner; “Reinigen und Vorbehandeln 

– Keine triviale Angelegenheit”; Metalloberfläche, 

2007, 61, 44-47. 

[8] H. Kollek „Reinigen und Vorbehandeln“, Curt 

R. Vincentz Verlag, Hannover, 1996, 13-20. 

[9] A. Bart, M.Stratmann (ed.); “Encyclopedia <strong>of</strong> 

Electrochemistry“ vol. 4 „Corrosion and Oxide 

films“; Vol. Editors: M. Stratmann, G.S. Frankel; 

Wiley VCH Weinheim 2003; Chapter 2. 

[10] Kyung-Keun Lee, Kwang-Bum Kim; “Electrochemical 

impedance characteristics <strong>of</strong> pure Al 

and Al–Sn alloys in NaOH solution”; Corros. Sci., 

2001, 43, 561-575. 

[11] Y. L. Cheng, Z. Zhang, F. H. Cao, J. F. Li, J. Q. 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

Zhang, J. M. Wang and C. N. Cao; “Study <strong>of</strong> <strong>the</strong> 

potential electrochemical noise during corrosion 

process <strong>of</strong> aluminum alloys 2024, 7075 and pure 

aluminum (pages 601–608)”; Mat. Corr., 2003, 54, 

601-608. 

[12] F.M. Queiroz, M. Magnani, I. Costa, H.G. de 

Melo; “Investigation <strong>of</strong> <strong>the</strong> corrosion behaviour <strong>of</strong> 

AA 2024-T3 in low concentrated chloride media“; 

Corros. Sci., 2008, 50, 2646-2657. 

[13] D. Mercier, M.-G. Barthés-Labrousse; “The role 

<strong>of</strong> chelating agents on <strong>the</strong> corrosion mechanisms 

<strong>of</strong> <strong>aluminium</strong> in alkaline aqueous solutions”; Corros. 

Sci., 2009, 51, 339-348. 

[14] C.M. von Klingspor, Bachelor-Thesis, HTW- 

University <strong>of</strong> Applied Science Dresden, 2011. � 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 



Soldering and brazing <strong>of</strong> <strong>aluminium</strong> and its alloys 

Christian Eisenbeis, SLV Duisburg 

The joining <strong>of</strong> <strong>aluminium</strong> always requires 

<strong>the</strong> use <strong>of</strong> special technologies. In particular, 

soldering and brazing techniques 

need special experience in order to obtain 

joints <strong>of</strong> good quality and sufficient 

strength. In addition, <strong>the</strong>re is <strong>of</strong>ten a 

good chance to use <strong>the</strong> brazing technologies 

for an efficient and highly mechanized 

fabrication process. Compared with 

welding, soldering and brazing allow 

appropriate higher or lower process-temperatures, 

depending on <strong>the</strong> type <strong>of</strong> <strong>the</strong> 

<strong>aluminium</strong> base metal. The market <strong>of</strong>fers 

new and improved filler metals and fluxes, 

and – at <strong>the</strong> same time – <strong>the</strong> demands 

on quality and strength <strong>of</strong> <strong>the</strong> brazed <strong>aluminium</strong> 

joints have increased. Traditional 

and highly developed procedures like 

flame brazing, resistant brazing, or controlled-arc-brazing 

are available ei<strong>the</strong>r 

for simple or sophisticated brazed components 

<strong>of</strong> <strong>aluminium</strong>. In this presentation 

<strong>the</strong> basics and <strong>the</strong> technical solutions will 

be explained and illustrated by figures 

and practical examples. 

1. The brazing process 

For soldering and brazing unlike welding an 

additional brazing filler metal is used whose 

melting temperature (or melting range) is 

below <strong>the</strong> melting temperature <strong>of</strong> <strong>the</strong> parent 

metal. The joining surfaces <strong>of</strong> <strong>the</strong> parent metals 

are wetted by <strong>the</strong> liquid braze without being 

molten <strong>the</strong>mselves. 

As a result, <strong>the</strong> following aspects are <strong>of</strong> 

importance: 

• It is possible to join dissimilar metals and 

metal alloys, respectively 

• Due to <strong>the</strong> relatively low process tempera- 

ture <strong>the</strong> component is less <strong>the</strong>rmally 

influenced 

• The amount <strong>of</strong> filler metal is relatively 

small 

• The rapid sequence <strong>of</strong> <strong>the</strong> brazing process 

• Low distortion compared with welding 

• Brazing/soldering <strong>of</strong> a variety <strong>of</strong> areas to 

be joined at <strong>the</strong> same time. 

The joint is built up by <strong>the</strong> exchange <strong>of</strong> diffusion 

<strong>of</strong> <strong>the</strong> metals within a small zone (Fig. 

1). To enable this diffusion-mechanism, <strong>the</strong> 

surface <strong>of</strong> <strong>the</strong> joining area must be free from 

<strong>oxide</strong>s and be protected against <strong>the</strong> renewed 

formation <strong>of</strong> <strong>oxide</strong>s during soldering/brazing. 

Therefore, <strong>the</strong>re are two preconditions to be 

fulfilled when brazing, using a flux: 

• The melting temperature <strong>of</strong> <strong>the</strong> braze 

must be below <strong>the</strong> melting temperature 

<strong>of</strong> <strong>the</strong> parent metal 

• The surface <strong>of</strong> <strong>the</strong> material must be 

completely free from <strong>oxide</strong>s 

• <strong>the</strong> flux must be intensive enough having 

<strong>the</strong> adequate effective temperature. 

2. The process ranges 

Brazing <strong>of</strong> <strong>aluminium</strong> is classified into brazing 

and soldering. At brazing, filler metals with 

melting temperatures above 450°C are used. 

Solders, having a melting point below 450°C, 

are called solder filler metals and <strong>the</strong> process 

is called soldering or s<strong>of</strong>t-soldering. 

The process temperature can be freely 

chosen, but it must be within <strong>the</strong> solidus temperature 

<strong>of</strong> <strong>the</strong> filler metal and <strong>the</strong> melting 

temperature <strong>of</strong> <strong>the</strong> parent metal. The flux will 

be chosen matching <strong>the</strong> soldering or brazing 

temperature; <strong>the</strong> effective temperature and 

<strong>the</strong> chemical intensity must be coordinated 

with <strong>the</strong> soldering/brazing process. 

More generally, <strong>the</strong> higher <strong>the</strong> brazing 

temperature, <strong>the</strong> higher is <strong>the</strong> mechanical 

strength. 

3. Structural requirements on brazing 

During brazing usually a brazing gap is produced 

by a braze drawn into <strong>the</strong> gap due to 

<strong>the</strong> capillary forces (capillary brazing). The 

braze displaces <strong>the</strong> flux as soon as <strong>the</strong> <strong>oxide</strong>s 

are dissolved on <strong>the</strong> joining surface. A relatively 

large-surface joint is produced and <strong>the</strong> 

braze <strong>layer</strong> located in <strong>the</strong> soldering gap will 

be mechanically streng<strong>the</strong>ned if a load is exerted 

on it. By this means considerable forces 

may be exerted that are able to exceed <strong>the</strong> 

strength <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> parent metal. 

4. Properties <strong>of</strong> <strong>aluminium</strong> brazed joints 

4.1 Brazed joints: The strength <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> 

brazed joint <strong>of</strong>ten is <strong>the</strong> same as a 

welded joint if <strong>the</strong> ‘joint is suitable for brazing’. 

During brazing <strong>of</strong> Al-materials <strong>of</strong> higher 

strength as well as <strong>of</strong> Al-cast alloys <strong>the</strong>re is 

<strong>the</strong> risk <strong>of</strong> a melt down <strong>of</strong> <strong>the</strong> surfaces to be 

joined since <strong>the</strong> melting temperature <strong>of</strong> such 

type <strong>of</strong> alloys can be near <strong>the</strong> melting point <strong>of</strong> 

<strong>the</strong> brazing filler metal. The chemical resistance 

<strong>of</strong> a brazed joint <strong>of</strong> <strong>the</strong> commonly used 

braze material Al88Si is not worse than that 

<strong>of</strong> a welded joint. It is true that brazed joints 

can be anodized, but in <strong>the</strong> joining area (braze 

<strong>layer</strong>) <strong>the</strong>y will get a darker colour. 

4.2 Soldered joints: Both, strength and corrosion 

resistance <strong>of</strong> soldered joints are much 

lower. A dry environment or corrosion protection 

(lacquering, greasing) will <strong>the</strong>n become 

necessary. It is nei<strong>the</strong>r possible to 

perform anodic oxidation nor to have oper- 

ating temperatures considerably above 

100°C. 

Fig. 1: Built-up <strong>of</strong> a brazed joint by diffusion (Source: Hydro Aluminium) 


Fig. 2: Al-base metal-alloys and AlSi3-brazing filler metal 

5. Aluminium brazes and solders 

5.1 Brazes: Due to <strong>the</strong> oxidic protective <strong>layer</strong> 

<strong>aluminium</strong> <strong>of</strong>fers good corrosion resistance. 

Therefore Al-brazes with 5 to 12% silicon 

have proven successful; hence <strong>the</strong> commonly 

used braze is Al 112 (B-Al88Si) ISO 17672 

(former AL 104 – EN1044) with a melting 

range <strong>of</strong> 575 to 585°C. Hence, at a working 

temperature <strong>of</strong> 585°C both, pure <strong>aluminium</strong> 

and all <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> alloys can be brazed 

that have a melting interval over 640°C (Fig. 

2). In practical terms this means that only pure 

<strong>aluminium</strong> and AlMn-alloys can be brazed 

well. 

5.2 Solders: Common solders are tin-zinc 

solders such as SnZn40 or ZnAl5. O<strong>the</strong>r sol- 

ders such as Sn96Ag can be used for <strong>aluminium</strong>, 

too. In general, <strong>the</strong> relatively poor suit- 

ability <strong>of</strong> <strong>aluminium</strong> to be soldered (stable <strong>oxide</strong> 

skin, poor corrosion resistance <strong>of</strong> <strong>the</strong> soldered 

joint) prevent from a more widespread 

soldering <strong>of</strong> <strong>aluminium</strong>. On <strong>the</strong> o<strong>the</strong>r hand, 

<strong>the</strong>re is an increased demand on soldering using 

a process temperature as low as possible 

due to <strong>the</strong> <strong>aluminium</strong> alloys applied. 

6. Fluxes for brazing/soldering <strong>of</strong> Al alloys 

Fluxes serve for dissolving <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> and 

to avoid a renewed contamination by oxygen 

on <strong>the</strong> fresh metal surface. 

6.1 Fluxes for brazing: Fluxes for light metals 

(FL) are stated in DIN EN 1045. The fluxes 

actuate above 500°C. A distinction is made 

between corrosive and non-corrosive fluxes. 


6.2 Fluxes for soldering: These are included 

in DIN EN ISO 9453 and <strong>of</strong>ten are based on 

zinc or tin chloride, possibly added by alkaline 

chlorides or organic substances. 

Destroying <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> and a sufficient 

wetting <strong>of</strong> <strong>the</strong> solder is not always successful 

by <strong>the</strong> use <strong>of</strong> <strong>the</strong> flux. New fluxes such 

as those alloyed with caesium under certain 

conditions may enhance wetting with solder 

(like with a slightly higher Mg content, too), 

which may be very important. 

7. Processes for soldering / 

brazing <strong>of</strong> <strong>aluminium</strong> 

The possibilities <strong>of</strong> soldering or brazing <strong>of</strong> <strong>aluminium</strong> 

(alloys) are considerably determined 

by <strong>the</strong> special properties <strong>of</strong> <strong>aluminium</strong> (problem 

<strong>of</strong> <strong>the</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong>, low melting 

temperature, very different wetting properties, 

high affinity <strong>of</strong> oxygen, susceptibility to 

corrosion if heavy metals are present). 

Some <strong>of</strong> <strong>the</strong> frequently used processes are: 

7.1 Flame brazing: During flame brazing <strong>the</strong> 

parts are heated to <strong>the</strong> soldering or brazing 

temperature using a torch (Fig. 4); depending 

on <strong>the</strong> temperature required propane, <strong>natural</strong> 

gas or acetylene are used (toge<strong>the</strong>r with air 

or oxygen). When heating up it must be observed 

that <strong>the</strong> flame is not always permitted 

to have contact with <strong>the</strong> soldering point, depending 

on <strong>the</strong> type <strong>of</strong> solder/braze. 

After reaching <strong>the</strong> working temperature <strong>the</strong> 

flux is already chemically active. The solder 

will become liquid thus wetting <strong>the</strong> surfaces 

and being drawn into <strong>the</strong> soldering gap, respec- 

tively. At <strong>the</strong> same time <strong>the</strong> flux will 

be displaced by <strong>the</strong> solder. 

7.2 Induction brazing: For <strong>the</strong> inductive 

heating <strong>of</strong> <strong>aluminium</strong> considerable 

induction power is required 

(<strong>aluminium</strong> is not ferromagnetic 

and has a high electric conductivity). 

Therefore, <strong>aluminium</strong> is induction 

brazed using a deep frequency and a 

high power. The main problem is still 

<strong>the</strong> control <strong>of</strong> <strong>the</strong> power, in order to 

avoid melting up <strong>of</strong> <strong>the</strong> material. 

7.3 Salt bath brazing: Here, <strong>the</strong> assembled 

parts applied with solder/ 

braze are immersed into a bath <strong>of</strong> 

molten flux which serves as a medium 

<strong>of</strong> heat transfer at <strong>the</strong> same time. 

This process has several advantages 

(no application <strong>of</strong> flux, good <strong>the</strong>rmal 

transfer, no air contact, tighter 

tolerances can be kept and brazing 

<strong>of</strong> several spots at <strong>the</strong> same time is 

possible). Parts that are particularly 

thin and <strong>of</strong> a complicated shape can 

be securely brazed using this process. 

On <strong>the</strong> o<strong>the</strong>r hand, a large consumption <strong>of</strong> 

fluxes is needed that adhere to <strong>the</strong> part and 

have to be removed, and disposed <strong>of</strong> using 

large amounts <strong>of</strong> rinsing water. Due to this 

reason salt bath brazing is used less and less. 

7.4 Furnace brazing: 

Controlled atmosphere brazing (CAB): brazing 

is performed in a continuous furnace 

using a shielding gas (nitrogen, sometimes 

nitrogen/hydrogen). Therefore, <strong>the</strong> application 

<strong>of</strong> a non-corrosive and thus milder flux 

is sufficient. Portions <strong>of</strong> hydrogen are used 

if <strong>aluminium</strong> is brazed with CrNi. The flux 

can be applied ei<strong>the</strong>r as a suspension, paste 

or dust (electrostatically), whereas <strong>the</strong> braze 

is increasingly directly applied to <strong>the</strong> surface 

by <strong>the</strong> manufacturer (Fig. 3). Advantages: 

little consumption <strong>of</strong> flux, only slight residuals 

on <strong>the</strong> part, cleaning not necessary. 

Fluxes from potassium and <strong>aluminium</strong> fluorides 

(KAlF4) have proven worldwide that 

have a melting range in <strong>the</strong> defined composition 

<strong>of</strong> 565 to 572°C. Complex parts such as 

<strong>aluminium</strong> radiators are manufactured by this. 

Vacuum brazing (VB): due to heating up in 

<strong>the</strong> furnace <strong>the</strong> <strong>aluminium</strong> material expands 

more than <strong>the</strong> adhering <strong>oxide</strong> <strong>layer</strong>; by this <strong>the</strong> 

<strong>oxide</strong> will be separated or solved, respectively. 

As this is performed under a vacuum, <strong>the</strong> surface 

laid bare remains metallically blank; <strong>the</strong> 

braze can wet <strong>the</strong> parent metal without <strong>the</strong> 

addition <strong>of</strong> a flux. Previously, <strong>the</strong> residual oxygen 

has been bound in <strong>the</strong> furnace by glowing 

magnesium chips (‘getter material’). Requirements 

<strong>of</strong> <strong>the</strong> braze gap: only 0.05 to 1.0 mm. 



Fig. 3: Controlled atmosphere brazing (CAB) <strong>of</strong> an <strong>aluminium</strong> radiator (Source: Solvay) 

7.5 Arc and laser beam brazing (braze weld- 

ing) 

Gas-metal arc brazing: <strong>the</strong> arc is used to break 

up <strong>the</strong> <strong>oxide</strong>s due to local heating. The braze 

melting in <strong>the</strong> arc will flow under <strong>the</strong> <strong>oxide</strong>s 

thus wetting <strong>the</strong> <strong>aluminium</strong> surface. Newly 

developed brazes with a slightly reduced melting 

temperature, partly on <strong>the</strong> basis <strong>of</strong> Zn, acting 

with arcs-very-poor-<strong>of</strong>-energy are used. 

Laser beam brazing: (e. g. Al- plates in <strong>the</strong> 

construction <strong>of</strong> car bodies) also makes use <strong>of</strong> 

a tool (e. g. a separate second laser beam or 

arc), in order to separate <strong>the</strong> <strong>oxide</strong>s. During 

laser beam brazing it is intended to use as little 

energy as possible and a defocused beam. 

There is <strong>the</strong> problem, however, to separate <strong>the</strong> 

<strong>oxide</strong>s on <strong>the</strong> one hand and to securely avoid 

melting up on <strong>the</strong> o<strong>the</strong>r. Advantage: a flux free 

brazing process <strong>of</strong> high process speed. 

7.6 Resistance brazing 

Resistance spot brazing: <strong>the</strong> high electric conductivity, 

<strong>the</strong> highly melting <strong>oxide</strong> <strong>layer</strong> as 

well as <strong>the</strong> good diffusion behaviour <strong>of</strong> Al-Cu 

makes this process considerably more complicated. 

Good results, however, can be obtained 

with braze cladded surfaces (e. g. Al88Si, <strong>layer</strong> 

thickness approx. 60 to 100 μm, plate thicknesses 

0.7 to 2 mm) using conventional spot 

welding units (alternating current, without 

current and power program). 

The advantages <strong>of</strong> this process: almost no deformation 

<strong>of</strong> <strong>the</strong> surface, short welding times, 

high process stability, application <strong>of</strong> simple 

welding units. 

8. Processes for soldering <strong>of</strong> <strong>aluminium</strong> 

alloys with reduced melting points 

The solders used for soldering <strong>of</strong> <strong>aluminium</strong> (alloys) 

were mentioned before. The commonly 

used processes are aimed at a <strong>the</strong>rmal energy 

as low as possible, or at joining temperatures 

as low as possible. This is necessary with <strong>aluminium</strong> 

alloys with melting temperatures that 

do not permit <strong>the</strong> 

use <strong>of</strong> brazes (e. g. 

Al88Si) any more 

if too much process 

heat could lead 

to s<strong>of</strong>tening <strong>of</strong> <strong>the</strong> 

material. In general, 

1xxx (Al-alloyed), 

2xxx (Cualloyed), 

3xxx 

(Mn-alloyed), 

4xxx (Si-alloyed) 

and 7xxx (Zn-alloyed) 

are more 

suitable for soldering, 

whereas 

<strong>the</strong> 5xxx (Mg-alloyed) 

and 6xxx 

(Si-/Mg-alloyed) 

are less suitable for soldering. It is ra<strong>the</strong>r <strong>the</strong> 

latter ones, however, that are <strong>of</strong>ten used in <strong>the</strong> 

automotive industry. 

9. Brazed <strong>aluminium</strong> dissimilar joints 

During brazing <strong>of</strong> <strong>aluminium</strong> <strong>the</strong> brazing temperatures 

are limited to approx. 600°C. For 

brazing <strong>of</strong> <strong>aluminium</strong> to titanium soldering 

using special zinc based solders is suitable. 

For joining <strong>aluminium</strong> to magnesium solders 

based on zinc and magnesium are available 

with good experience made in ultrasound 

based flux free soldering. Aluminium is <strong>of</strong>ten 

brazed with Cr-Steel or CrNi steel in a continuous 

furnace under controlled atmosphere. 

10. Summary 

In general, <strong>aluminium</strong> is suitable for soldering 

and brazing. There are, however limits due to 

<strong>the</strong> relatively low melting point and <strong>the</strong> influence 

<strong>of</strong> <strong>the</strong> wettability <strong>of</strong> <strong>the</strong> surface depending 

on <strong>the</strong> type <strong>of</strong> alloy. A variety <strong>of</strong> systems 

for soldering and brazing as well as types <strong>of</strong> 

fluxes is available. The most important process 

today is controlled atmosphere furnace 

brazing (CAB) with braze-cladded materials 

enhancing this very effective processing. 

Soldering processes are supported by metal 

alloyed fluxes with improved wettability. The 

application by ultrasound partly enables brazing 

without using a flux. Current developments 

in soldering and brazing and on processes 

in particular in <strong>the</strong> field <strong>of</strong> soldering will develop 

new technical possibilities <strong>of</strong> processing 

<strong>of</strong> <strong>aluminium</strong> and <strong>aluminium</strong> alloys and secure 

<strong>the</strong>m by new fields <strong>of</strong> manufacture. 

11. Standards and regulation 

DIN EN ISO 9453 (2006) • S<strong>of</strong>t solder alloys – 

Chemical compositions and forms; supersedes 

DIN EN 29454 (1994) 

DIN EN ISO 17672 (2010) • Brazing – Filler 

metals; supersedes DIN EN 1044 (2006) 

DIN 1707-100 (2011) • S<strong>of</strong>t solder alloys – 

Chemical composition and forms 

DIN EN ISO 3677 (1995) • Filler metals for soldering, 

brazing and braze welding – designations 

DIN EN 1045 (1997) • Fluxes for brazing 

DIN EN 13133 (2000) • Brazing / Brazer approval 

DIN EN 13134 (2000) • Brazing / Procedure 

approval 

DIN EN 12799 (2000) • Brazing / Non-destructive 

examination <strong>of</strong> brazed joints 

DIN EN 12797 (2000) • Brazing / Destructive 

examination <strong>of</strong> brazed joints 

DIN EN ISO 18279 (2004) • Imperfections 

in brazed joints 

Fig. 4: Flame brazing <strong>of</strong> <strong>aluminium</strong>: dissimilar metals (Source: Everwand & Fell) � 



<strong>Influence</strong> <strong>of</strong> <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong> on brazeability 

J. Zähr 1 , S. Oswald 2 , M. Türpe 3 , H.-J. Ullrich 1 , U. Füssel 1 

1 TU Dresden, Institute <strong>of</strong> Surface and Manufacturing Technology, Joining Technology and Assembly, 

Germany • 2 IFW Dresden, Germany • 3 Behr GmbH & Co. KG, Stuttgart, Germany 

For brazing <strong>of</strong> <strong>aluminium</strong>, <strong>the</strong> <strong>natural</strong> 

<strong>oxide</strong> <strong>layer</strong> has a significant influence. 

However, <strong>the</strong> behaviour <strong>of</strong> <strong>the</strong> <strong>oxide</strong> 

<strong>layer</strong> before brazing, e. g. due to changing 

climate conditions during <strong>the</strong> storage or 

<strong>the</strong> transport and during <strong>the</strong> heating process, 

are not analysed comprehensively 

yet. In this paper, two analysing methods 

are explained, which enable <strong>the</strong> analysis 

<strong>of</strong> <strong>the</strong> influence <strong>of</strong> increased humidities 

and temperatures on <strong>the</strong> composition and 

thickness <strong>of</strong> <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> <strong>oxide</strong> 

<strong>layer</strong>. Additionally, brazing tests are done 

to get a correlation between <strong>the</strong> <strong>oxide</strong> 

before <strong>the</strong> brazing process and <strong>the</strong> brazeability 

<strong>of</strong> <strong>the</strong> material. The brazing tests 

are done in a shielding gas lab furnace 

without using flux. The tests have shown 

that especially condensation leads to a 

growth <strong>of</strong> <strong>the</strong> oxidic <strong>layer</strong> and to a significant 

decrease <strong>of</strong> <strong>the</strong> brazeability. So 

<strong>the</strong> understanding about <strong>the</strong> influence <strong>of</strong> 

<strong>the</strong> atmospheric conditions on <strong>the</strong> oxidic 

<strong>layer</strong> and also <strong>the</strong> brazeability <strong>of</strong> <strong>the</strong> material 

is enhanced. 

Aluminium is a material widely used in industry, 

due to its good relation <strong>of</strong> weight to 

strength, good corrosion resistance and formability 

[1]. For assembly <strong>of</strong> several parts, a material 

joining is <strong>of</strong>ten necessary. For material 

joints between <strong>aluminium</strong> materials, a removal 

<strong>of</strong> <strong>the</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong> is necessary, 

<strong>the</strong> so called <strong>aluminium</strong> surface activation. 

The surface activation method used depends 

on <strong>the</strong> brazing technique. For flame brazing 

as well as brazing in a shielding gas furnace, 

flux is <strong>of</strong>ten used for <strong>the</strong> surface activation [2- 

[5]. The task <strong>of</strong> <strong>the</strong> flux is widely discussed in 

<strong>the</strong> literature. On <strong>the</strong> one hand side, <strong>the</strong> flux 

shall chemically dissolve <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> [5- 

7]. Thereby, <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> is cracked by <strong>the</strong> 

flux, so <strong>the</strong> flux can flow under <strong>the</strong> <strong>oxide</strong> particles 

and blast <strong>the</strong>m. Additionally, <strong>the</strong> flux 

decreases <strong>the</strong> surface tension <strong>of</strong> <strong>the</strong> solder. 

Therefore, <strong>the</strong> flow ability <strong>of</strong> <strong>the</strong> solder is increased 

[6]. 

Ano<strong>the</strong>r common brazing technique is <strong>the</strong> 

brazing inside a vacuum chamber. The surface 

activation is realized due to <strong>the</strong> Mg-gettering 

effect as well as <strong>the</strong> different <strong>the</strong>rmal expansion 

coefficients <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> base ma- 

terial and <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong> 

[8-9]. Most <strong>of</strong> <strong>the</strong> common research analyses 

for <strong>aluminium</strong> brazing are discussing <strong>the</strong> fur<strong>the</strong>r 

development <strong>of</strong> brazing techniques [10- 

11], solder systems [12-13] or <strong>the</strong> corrosion 

behaviour <strong>of</strong> <strong>the</strong> brazed parts [14]. However, 

<strong>the</strong> influence <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> before brazing 

is not characterized yet. There is only few 

information available in literature concerning 

<strong>the</strong> correlation between <strong>the</strong> <strong>oxide</strong> conditions 

before brazing and <strong>the</strong> brazeability <strong>of</strong> <strong>the</strong> 

material. Gray et. al. [15] mentions, that <strong>the</strong> 

<strong>oxide</strong> thickness influences <strong>the</strong> flow ability <strong>of</strong> 

<strong>the</strong> solder. Swiderksy [16] specifies this correlation: 

with increase <strong>of</strong> <strong>the</strong> <strong>oxide</strong> thickness 

from 4 to 22 nm an increase <strong>of</strong> <strong>the</strong> amount 

<strong>of</strong> <strong>the</strong> flux from 2 to 5 g/m 2 is necessary. But 

it is not clear, how this change <strong>of</strong> <strong>the</strong> <strong>oxide</strong> 

<strong>layer</strong> can occur. This paper explains <strong>the</strong> composition 

<strong>of</strong> <strong>the</strong> <strong>natural</strong> <strong>oxide</strong> <strong>layer</strong> as well as 

two possibilities for <strong>the</strong> analysis <strong>of</strong> this nm 

thick <strong>layer</strong>. Additionally, <strong>the</strong> influence <strong>of</strong> typical 

environmental conditions during storage 

and brazing are described. Afterwards, brazing 

tests are done to get a correlation between 

<strong>the</strong> oxidic <strong>layer</strong> thickness and composition to 

<strong>the</strong> brazeability. 

Natural <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong> – 

structure, analyzing methods 

The <strong>natural</strong> <strong>oxide</strong> <strong>layer</strong> consists <strong>of</strong> exactly two 

sheets: an <strong>oxide</strong> <strong>layer</strong> directly on <strong>the</strong> material 

(= barrier <strong>layer</strong>) and a hydroxidic <strong>layer</strong> 

at <strong>the</strong> transition to <strong>the</strong> atmosphere (= surface 

<strong>layer</strong>) [17]. In normal climate conditions (NC, 

20°C, 50% rel. humidity), <strong>the</strong> barrier <strong>layer</strong> 

has a thickness <strong>of</strong> ca. 2 nm and consists <strong>of</strong> 

amorphous Al 2 O 3 . The electrical conductivity 

<strong>of</strong> this compact <strong>layer</strong> is very low [17]. Therefore, 

<strong>the</strong> ion diffusion is minimized at room 

temperature, so this <strong>layer</strong> acts as corrosion 

protection for <strong>the</strong> <strong>aluminium</strong> material. Rising 

temperatures cause an increased ion diffusion, 

which leads to a growth <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong>. 

Additionally, crystalline Al 2 O 3 -structures are 

formed at temperatures above 400°C. 

The surface <strong>layer</strong> also has a thickness <strong>of</strong> 

2-3 nm under normal climate conditions (NC). 

This <strong>layer</strong> consists <strong>of</strong> <strong>aluminium</strong> hydr<strong>oxide</strong> 

(Al(OH) 3 ) as well as pores and crystalline parts 

<strong>of</strong> heterogeneous phases <strong>of</strong> <strong>the</strong> base material. 

This <strong>layer</strong> is formed by <strong>the</strong> reaction <strong>of</strong> <strong>the</strong> barrier 

<strong>layer</strong> (Al 2 O 3 ) or metallic <strong>aluminium</strong> with 

water. Therefore, this <strong>layer</strong> can grow significantly 

due to an increase <strong>of</strong> <strong>the</strong> relative humidity 

in <strong>the</strong> atmosphere or if condensed water is 

available at <strong>the</strong> surface. 

During transport or storage <strong>of</strong> <strong>aluminium</strong> 

materials before brazing, relative humidities 

above 90% or even condensation can occur at 

<strong>the</strong> surface <strong>of</strong> <strong>the</strong> base materials. These conditions 

influence <strong>the</strong> thickness <strong>of</strong> <strong>the</strong> surface 

<strong>layer</strong>. In contrast to this, <strong>the</strong> heating process 

during brazing leads to a change <strong>of</strong> <strong>the</strong> composition 

and <strong>the</strong> thickness <strong>of</strong> <strong>the</strong> barrier <strong>layer</strong>. 

In literature, <strong>the</strong>re are a lot <strong>of</strong> studies available 

concerning <strong>the</strong> influence <strong>of</strong> different ambient 

conditions on <strong>the</strong> <strong>aluminium</strong> <strong>oxide</strong> and 

hydr<strong>oxide</strong> <strong>layer</strong>. Never<strong>the</strong>less, <strong>the</strong>se analyses 

do not correlate with <strong>the</strong> real conditions for 

brazing materials and processes regarding <strong>the</strong> 

composition <strong>of</strong> <strong>the</strong> materials, <strong>the</strong> temperatures, 

humidities, durations and ambient at- 

Fig. 1: XPS-spectra <strong>of</strong> surface <strong>of</strong> Al-material (EN AW-Al Mn1Cu with solder EN AW-Al Si10) 


SESSION APPLICATION-ORIENTED TECHNOLOGIES 

Fig. 2: FTIR-Spectra while heating from room temperature to 605°C (left), summary <strong>of</strong> <strong>the</strong> peak area 

depending on heating temperature for <strong>the</strong> three phases existing at <strong>the</strong> material surface (right) 

mospheres. In <strong>the</strong> subsequently described 

analyses <strong>the</strong> following materials and atmospheric 

conditions were used: 

1. Material: Solder plated (both sides) Alsheets 

a. Base material: EN AW-Al Mn1Cu, 

thickness: 0.32 mm 

b. Solder: EN AW-Al Si10, thickness: 

0.04 mm (on both sides) 

2. Atmospheric conditions for <strong>the</strong> analysis 

<strong>of</strong> <strong>the</strong> influence <strong>of</strong> <strong>the</strong> storage and transport 

conditions 

a. Normal climate (NC) – 23°C/ 50% 

rel. humidity 

b. Humid climate (HC) – 40°C/ 92% 

rel. humidity 

c. Condensation (C) – 23°C/ > 100% 

rel. humidity 

3. Heating tests for <strong>the</strong> analysis <strong>of</strong> <strong>the</strong> influence 

<strong>of</strong> increased temperatures 

a. Heating temperature: 605°C 

b. Ambient atmosphere: Nitrogen 

The challenge for <strong>the</strong> analysis <strong>of</strong> <strong>the</strong> influence 

<strong>of</strong> real conditions on brazing materials is <strong>the</strong> 

choice <strong>of</strong> suitable methods which are able to 

analyse nm-thick <strong>layer</strong>s and to distinguish between 

<strong>aluminium</strong> <strong>oxide</strong> and hydr<strong>oxide</strong> phases. 

With XPS-measurements, <strong>the</strong> composition 

<strong>of</strong> <strong>the</strong> surface can be measured. The spectra 

shows <strong>the</strong> intensity depending on <strong>the</strong> binding 

energy, see Fig. 1. For <strong>natural</strong> <strong>aluminium</strong> surfaces 

oxygen, carbon, magnesium as well as 

<strong>aluminium</strong> can be measured. 

The important phases and <strong>the</strong>ir binding energies 

for <strong>the</strong> present analyses are [18]: 

Metallic Al 72,5 – 73 eV 

γ-Al 2O 3 73,7 – 74,5 eV 

Al(OH) 3 73,4 eV. 

Obviously it is not possible to distinguish between 

oxidic and hydroxidic phases on <strong>the</strong> 

base <strong>of</strong> <strong>the</strong> binding energy. Ano<strong>the</strong>r possibility 

to distinguish between <strong>the</strong>se two phases is 

described by Wittenberg et. al. [19]. In this 

study, <strong>the</strong> oxidic and hydroxidic phases are 

classified by <strong>the</strong> formation <strong>of</strong> <strong>the</strong> relation <strong>of</strong> 

<strong>the</strong> atomic concentration <strong>of</strong> <strong>the</strong> <strong>aluminium</strong> 

peak to <strong>the</strong> oxygen peak (Al/O). However, 

with <strong>the</strong> industrial material used in <strong>the</strong> present 

analysis, <strong>the</strong> measured <strong>aluminium</strong> intensity 

consists always <strong>of</strong> <strong>the</strong> oxidic bound <strong>aluminium</strong> 

and <strong>the</strong> metallic <strong>aluminium</strong>. The signal <strong>of</strong> 

<strong>the</strong> metallic bound <strong>aluminium</strong> originates from 

<strong>the</strong> metallic <strong>aluminium</strong> <strong>of</strong> <strong>the</strong> base material directly 

under <strong>the</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong>. Therefore, 

this metallic <strong>aluminium</strong> signal cannot be 

taken into account for <strong>the</strong> determination <strong>of</strong> 

<strong>the</strong> <strong>aluminium</strong> surface phases. Hence, only <strong>the</strong> 

oxidic bound <strong>aluminium</strong> is set into relation to 

<strong>the</strong> oxygen intensity in <strong>the</strong> present analysis. 

For Al 2O 3 <strong>the</strong> ratio <strong>of</strong> Al <strong>oxide</strong> to O is about 2/3, 

for <strong>the</strong> hydroxidic bound <strong>aluminium</strong> Al(OH) 3 

this ratio is about 1/3. Additionally, <strong>the</strong> <strong>oxide</strong> 

<strong>layer</strong> thickness can be estimated using <strong>the</strong> 

ratio <strong>of</strong> <strong>the</strong> intensity <strong>of</strong> <strong>the</strong> oxidic to metallic 

bound <strong>aluminium</strong> [20]. 

The XPS-measurement <strong>of</strong> <strong>natural</strong> <strong>aluminium</strong> 

surfaces as well as <strong>of</strong> materials stored 

under normal climate, humid climate and condensation 

show, that <strong>the</strong> hydroxidic <strong>layer</strong> is 

existent on all surfaces, even on <strong>the</strong> surface 

<strong>of</strong> <strong>natural</strong> (not stored) <strong>aluminium</strong> materials. In 

contrast to this, <strong>the</strong> storage conditions have a 

measurable influence on <strong>the</strong> <strong>oxide</strong> thickness. 

Especially, <strong>the</strong> influence <strong>of</strong> condensed water 

on <strong>the</strong> surface causes a significant increase <strong>of</strong> 

<strong>the</strong> <strong>oxide</strong> <strong>layer</strong>, see Fig. 4. 

The temperature increase during <strong>the</strong> brazing 

process has ano<strong>the</strong>r significant influence 

on <strong>the</strong> thickness and composition <strong>of</strong> <strong>the</strong> <strong>layer</strong>s 

at <strong>the</strong> surface <strong>of</strong> <strong>aluminium</strong> materials. The 

behaviour <strong>of</strong> <strong>the</strong>se <strong>layer</strong>s during heating from 

room temperature till 605°C (= brazing temperature) 

can be analyzed with <strong>the</strong> help <strong>of</strong> 

FTIR measurements under a N 2-atmosphere 

(= brazing atmosphere). All tests were done 

at Innoval Technology Ltd. Banbury, where a 

FTIR-spectrometer with an additional heating 

device is available.. A spectrum is taken <strong>of</strong> 

<strong>the</strong> surface every minute during <strong>the</strong> heating 

process. All spectra for one heating cycle are 

displayed in Fig. 2. 

It is obvious, that <strong>the</strong>re are three phases 

existing at <strong>the</strong> surface during <strong>the</strong> heating 

process. At room temperature <strong>the</strong> 

amorphous Al 2O 3-phase dominates 

<strong>the</strong> surface. With rising temperatures, 

especially at temperatures above 

400°C, <strong>the</strong> Al/O-signal moves to 

lower wave numbers. Unfortunately, 

<strong>the</strong> phase could not be defined until 

now. This phase can consist <strong>of</strong> ei<strong>the</strong>r 

crystalline Al 2O 3 or MgAl 2O 4. For a 

precise determination <strong>of</strong> this phase, 

TEM analyses are planned. At temperatures 

above 577°C, <strong>the</strong> solidus 

temperature <strong>of</strong> <strong>the</strong> solder, MgO is 

growing at <strong>the</strong> surface. With <strong>the</strong> help <strong>of</strong> this 

analyzing method, <strong>the</strong> behaviour <strong>of</strong> materials, 

which were stored under condensed water, 

was also analyzed. These materials have 

a major growth <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> during <strong>the</strong> 

heating process. 

Correlation between surface 

condition and brazeability 

Brazing method for analyzing influence <strong>of</strong> 

<strong>oxide</strong> <strong>layer</strong> on brazeability: The brazing tests 

were done in a laboratory shielding gas furnace. 

To see only <strong>the</strong> influence <strong>of</strong> <strong>the</strong> <strong>oxide</strong> 

<strong>layer</strong> on <strong>the</strong> brazeability, a mechanical surface 

activation method, see Fig. 3, was used instead 

<strong>of</strong> a chemical surface activation. To realize <strong>the</strong> 

cracking <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong>, <strong>the</strong> upper joining 

partner is pressed inside <strong>the</strong> liquid solder by 

applying a weight. Therefore, tensions are induced 

inside <strong>the</strong> <strong>oxide</strong> <strong>layer</strong>, which cause <strong>the</strong> 

cracking <strong>of</strong> this <strong>layer</strong>. Afterwards <strong>the</strong> liquid 

solder can flow to <strong>the</strong> upper joining partner 

and wet it. 

Brazing results for different surface conditions: 

A storage under humid climate as well 

as under condensation for a short period 

<strong>of</strong> time (3 days) causes a growth <strong>of</strong> <strong>the</strong> <strong>oxide</strong> 

<strong>layer</strong>, but does not influence <strong>the</strong> brazing 

Fig. 3: Mechanism <strong>of</strong> <strong>the</strong> mechanical surface 

activation 


level significantly, see Fig. 4. A longer storage 

under condensation suppresses <strong>the</strong> brazing 

process completely. However, storage in a normal 

climate after <strong>the</strong> condensation leads to a 

significant improvement <strong>of</strong> <strong>the</strong> brazeability <strong>of</strong> 

<strong>the</strong> material. 

It has to be followed, that <strong>the</strong> water, which 

is stored inside <strong>the</strong> pores <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> 

due to <strong>the</strong> condensation, influences <strong>the</strong> brazing 

level. This water leads to an increased 

growth <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> during <strong>the</strong> heating. 

This thickened <strong>oxide</strong> <strong>layer</strong> cannot be cracked 

by <strong>the</strong> weight <strong>of</strong> <strong>the</strong> upper joining partner. 

Hence, <strong>the</strong> flow <strong>of</strong> <strong>the</strong> solder is suppressed 

completely. In contrast to this, <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> 

thickness has a minor influence on <strong>the</strong> brazeability. 

Conclusions 

The present analysis characterizes <strong>the</strong> influence 

<strong>of</strong> <strong>the</strong> humidity <strong>of</strong> <strong>the</strong> atmosphere as 

well as <strong>of</strong> condensed water at <strong>the</strong> surface <strong>of</strong> 

<strong>aluminium</strong> materials on <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> 

<strong>oxide</strong> <strong>layer</strong>. With <strong>the</strong> help <strong>of</strong> XPS-measurements, 

it has been shown, that storage under 

normal or humid climate does not influence 

<strong>the</strong> <strong>oxide</strong> <strong>layer</strong> thickness significantly. In contrast 

to this, <strong>the</strong> influence <strong>of</strong> condensed water 

on <strong>the</strong> surface causes an intense growth <strong>of</strong> 

<strong>the</strong> <strong>oxide</strong> <strong>layer</strong>. FTIR measurements with an 

heating device have also shown that materials, 

which were stored under <strong>the</strong> influence <strong>of</strong> condensed 

water, have a significant thicker <strong>oxide</strong> 

<strong>layer</strong> at brazing temperature. 

Subsequent brazing tests have been done 

in a shielding gas furnace with a mechanical 

surface activation to see <strong>the</strong> influence <strong>of</strong> <strong>the</strong> 

<strong>oxide</strong> <strong>layer</strong> on <strong>the</strong> brazeability <strong>of</strong> <strong>the</strong> material. 

These tests have shown that <strong>the</strong> <strong>oxide</strong> 

thickness before brazing is not <strong>the</strong> critical 

parameter for <strong>the</strong> brazing result. In fact, <strong>the</strong> 

water inside <strong>the</strong> pores <strong>of</strong> <strong>the</strong> hydroxidic <strong>layer</strong> 

causes an increased growth <strong>of</strong> <strong>the</strong> <strong>oxide</strong> <strong>layer</strong> 

during <strong>the</strong> heating process. This effect <strong>of</strong> <strong>the</strong> 

condensed water can be avoided, if <strong>the</strong> materials 

are stored under normal climate after having 

water present at <strong>the</strong> surface <strong>of</strong> <strong>aluminium</strong> 

materials. 

The analyses showed that not only <strong>the</strong> solder 

or <strong>the</strong> brazing conditions are important for 

<strong>the</strong> brazing result <strong>of</strong> <strong>aluminium</strong> materials, but 

also <strong>the</strong> <strong>natural</strong> <strong>aluminium</strong> <strong>oxide</strong> <strong>layer</strong>. The 

<strong>natural</strong> <strong>oxide</strong> <strong>layer</strong> can be influenced significantly 

by <strong>the</strong> atmospheric conditions during 

<strong>the</strong> transport and <strong>the</strong> storage <strong>of</strong> <strong>the</strong> materials. 

Fur<strong>the</strong>rmore, it has to be assumed that also 

<strong>the</strong> surface <strong>of</strong> different brazing partners, e. g. 

<strong>aluminium</strong> <strong>oxide</strong> or stainless steels, can influence 

<strong>the</strong> brazeability. 

References 


[1] Ostermann, F.: Anwendungstechnologie Aluminium 

– ein Werkst<strong>of</strong>fhandbuch. Springer Berlin 

Heidelberg; Auflage: 2., 2007. 

[2] Kohlweiler, A.: Hartlöten von Aluminium – so 

geht es auch. In: Der Praktiker.(1996) Heft 2, S. 

40-45. 

[3] Meurer, C.; Belt, H.-J.; König, H.: Das Nocolok- 

Flux-Hartlötverfahren. In: Die Kälte und Klima- 

technik. (1997) Heft 10, S. 802-808. 

[4] Neitz, G.; Wielage, B.; Trommer; F.: Schutzgas- 

löten von Al-Wärmetauschern. In: Hart- und 

Hochtemperaturlöten und Diffusionsschweißen 

DVS-Berichte Band 212. Düsseldorf: DVS-Verlag, 

2001, S. 240-244. 

[5] Senaneuch, J.; Nylén, M.; Hutchinson, B.: Metallurgy 

<strong>of</strong> brazed <strong>aluminium</strong> heat-exchangers, Part I. 

In: Aluminium. 77 (2001) Heft 11, S. 896-899. 

[6] Müller, W.; Müller, J.-U.: Löttechnik – Leitfaden 

für die Praxis. Düsseldorf: Deutscher Verlag für 

Schweißtechnik DVS-Verlag GmbH, 1995. 

[7] Terrill, J. R.; Cochran, C. N.; Stokes, J. J.; Haupin; 

W. E.: Understanding <strong>the</strong> Mechanisms <strong>of</strong> Aluminium 

Brazing. In: Welding Journal. (1971) S. 833-839. 

[8] Ashburn, L. L.: Fluxless Vacuum Furnace Brazing 

<strong>of</strong> Aluminum particularly Advantageous for 

more critical Applications: I. In: Industrial Heating. 

(1994) S. 47-51. 

[9] Byrnes, E. R.: Vacuum Fluxless Brazing <strong>of</strong> Aluminium. 

In: Welding Journal. (1971) S. 712-716. 

[10] Füssel, U.; Six, S.: Entwicklung eines NIR- 

Lötverfahrens für die Fertigung von Solarabsorbern 

aus Aluminium. LÖT, 9. International Conference 

Brazing, High Temperature Brazing and Diffusion 

Welding. In: DVS-Berichte: Band 263 (2010) S. 

358-360 

[11] Möller, F.; Thomy, C.; Vollertsen, F.: Flussmittelfreies 

Hartlöten von Aluminiumlegierungen mit 

einem koaxialen Laser-Plasma-Hybridlötprozess. 

In: Schweißen und Schneiden 62/5 (2010) 294. 

[12] Bach, F.W.; Möhwald, K.; Holländer, U.; Langohr, 

A.: Niedrig schmelzende Aluminiumhartlote 

aus dem System Al-Si-Zn. 9. International Conference 

Brazing, High Temperature Brazing and 

Diffusion Welding. In: DVS-Berichte. Düsseldorf: 

DVS-Verlag. Band 263 (2010) Seite 117-121. 

[13] Tillmann, W.; LIU, C.; Wojarski, L.: Dotierung 

von Aluminiumbasisloten zum flussmittelfreien 

Hartlöten von Aluminiumlegierungen. 9. International 

Conference Brazing, High Temperature 

Brazing and Diffusion Welding. In: DVS-Berichte. 

Düsseldorf: DVS-Verlag.: Band 263 (2010) Seite 

106-112. 

[14] Melander, M.; Woods, R.: A corrosion study <strong>of</strong> 

brazed <strong>aluminium</strong> heat exchangers after field service. 

In: International Aluminium Journal. Band 86 

(2010) Heft 7/8, Seite 56-61. 

[15] Gray, A.; Flemming, A. J. E.; Evans, J. M.: Optimising 

<strong>the</strong> Properties <strong>of</strong> Long-life Brazing Sheet 

Alloys for Vacuum and Nocolok Brazed Components. 

VTMS 4 Conference Proceedings, London, 

May 1999. 

[16] Swidersky, H.-W.: Aluminium brazing with 

non-corrosive fluxes – state <strong>of</strong> <strong>the</strong> art and trends in 

Nocolok flux technology. In: Hart- und Hochtem- 

peraturlöten und Diffusionsschweißen DVS-Berichte 

Band 212. Düsseldorf: DVS-Verlag, DVS-Berichte 

Band 212 (2001), S. 164-169. 

[17] Altenpohl, D.: Aluminium und Aluminium- 

legierungen. Berlin: Springer Verlag, 1965. 

[18] Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; 

Bomben, K. D.: Handbook <strong>of</strong> X-Ray Photoelectron 

Spectroscopy: A Reference Book <strong>of</strong> Standard 

Spectra for Identification and Interpretation <strong>of</strong> 

XPS-Data. Physical Electronics, 1995. – ISBN-10: 

096481241X 

[19] Wittenberg, T. N.; Douglas, W. J.; Wang, P. S.: 

Aluminium hydr<strong>oxide</strong> growth on <strong>aluminium</strong> surfaces 

exposed to an air/1% NO 2 mixture. In: Journal 

<strong>of</strong> Materials Science, 23 (1988), S. 1745-1747. 

[20] Kozlowska, M.; Reiche, R.; Oswald, S.; Vinzelberg, 

H.; Hübner, R.; Wetzig, K.: Quantitative 

ARXPS investigation <strong>of</strong> systems at ultrathin <strong>aluminium</strong> 

<strong>oxide</strong> <strong>layer</strong>s. Surface Interface Anal. 36 

(2004) 1600. 

� 

Fig. 4: <strong>Influence</strong> <strong>of</strong> storage conditions and duration on <strong>oxide</strong> thickness and brazing level 


MELTING, RECYCLING & HEAT TREATMENT 

The new generation <strong>of</strong> <strong>aluminium</strong> 

heat treatment plants – a vanguard concept 

Markus Belte and Dan Dragulin, Belte AG 

The vigorous discussion in <strong>the</strong> <strong>aluminium</strong> 

heat treatment community about <strong>the</strong> development 

<strong>of</strong> <strong>the</strong> appropriate technique 

becomes within <strong>the</strong> framework <strong>of</strong> <strong>the</strong> 

present work a genuine concrete answer. 

This paper is completely dedicated to <strong>the</strong> 

presentation <strong>of</strong> a new generation <strong>of</strong> heat 

treatment plants; it first introduces <strong>the</strong> 

fundamentals <strong>of</strong> high speed heat transfer 

processes and presents practical results 

and <strong>the</strong>irs <strong>the</strong>rmodynamical analysis and 

<strong>the</strong>n <strong>of</strong>fers a detailed perspective <strong>of</strong> <strong>the</strong> 

vanguard concept <strong>of</strong> <strong>the</strong> new generation 

<strong>of</strong> <strong>aluminium</strong> heat treatment plants. The 

present work is based on original experiments 

and investigations obtained by <strong>the</strong> 

authors and will present techniques never 

used before in <strong>the</strong> industrial heat treatment 

<strong>of</strong> <strong>aluminium</strong>. 

Infrared radiation – <strong>the</strong>oretical aspects 

The radiation heat transfer process is described 

by <strong>the</strong> law <strong>of</strong> Stefan-Boltzmann: 

• ∂Q 

Q = ⎯⎯ = εσST 4 

∂t (1) 

• 4 4 

Q = εσS [T O - T U ] (2) 

• 

Where: Q = heat flow rate; ε = emissivity 1 ; 

σ = 5.67 x 10 -8 W/m 2 K 4 Stefan-Boltzmannconstant; 

S = surface <strong>of</strong> <strong>the</strong> emitting body; 

T = temperature <strong>of</strong> <strong>the</strong> emitting body (Kelvin); 

T O = surface temperature; T U = environment 

temperature 2 . 

The heating rate can be calculated using <strong>the</strong> 

following formula: 

mc 100 TWB Ta 

τ = ⎯ ⎯⎯ [ ξ´(⎯⎯)- ξ´´ (⎯⎯)] 

α T 3 

U 

TU TU (⎯⎯) 

100 

(3) 3 

Where: m = mass/surface unity <strong>of</strong> <strong>the</strong> product 

to be heated; c =specific heat; TWB = heat 

treatment temperature; TU = environment temperature; 

ξ = f(TWB/TU) A special case <strong>of</strong> <strong>the</strong> heat transfer process 

is <strong>the</strong> infrared heat transfer process. 

“Infrared radiation, that portion <strong>of</strong> <strong>the</strong> 

electromagnetic spectrum that extends from 

1 ε Є (0,1] ; for <strong>the</strong> ideal black body: ε = 1 

2 The environment could be <strong>the</strong> interior <strong>of</strong> a furnace. 

3 According to [3] 

<strong>the</strong> long wavelength, or red, end <strong>of</strong> <strong>the</strong> visiblelight 

range to <strong>the</strong> microwave range. Invisible 

to <strong>the</strong> eye, it can be detected as a sensation 

<strong>of</strong> warmth on <strong>the</strong> skin. The infrared range 

is usually divided into three regions: near 

infrared (nearest <strong>the</strong> visible spectrum), with 

wavelengths 0.78 to about 2.5 micrometres (a 

micrometre, or micron, is 10 -6 metre); middle 

infrared, with wavelengths 2.5 to about 50 micrometres; 

and far infrared, with wavelengths 

50 to 1,000 micrometres. Most <strong>of</strong> <strong>the</strong> radiation 

emitted by a moderately heated surface is infrared; 

it forms a continuous spectrum. Molecular 

excitation also produces copious infrared 

radiation but in a discrete spectrum <strong>of</strong> lines or 

bands.” [Encyclopaedia Britannica [1]] 

“Radiant heating has been used in <strong>the</strong> industry 

since <strong>the</strong> 1930s, where it was introduced to 

bake finishes on cars. It took time for <strong>the</strong> new 

technology to gain acceptance but today <strong>the</strong> 

technology is used in most industrialised countries 

especially for drying and curing <strong>of</strong> paints 

and drying <strong>of</strong> textile, pulp and paper. Infrared 

energy is unique because it can heat materials 

or objects without heating <strong>the</strong> air around <strong>the</strong>m. 

That allows infrared heat to be concentrated 

exactly where it is wanted without much loss 

<strong>of</strong> energy.” [2] 

In <strong>the</strong> case <strong>of</strong> an infrared heating <strong>the</strong> maxi- 

mum temperature which can be reached depends 

on <strong>the</strong> environment temperature: 

T - T U = (T max - T U)(1-e -B . τ ) (4) [3] 

Where: T = body temperature; τ = duration; 

T U = environment temperature; T max = maximum 

temperature; B = coefficient which takes 

into account <strong>the</strong> process conditions 

In <strong>the</strong> case <strong>of</strong> a classic heat transfer process 

performed exclusively through air convection, 

<strong>the</strong> body temperature can not exceed <strong>the</strong> air 

temperature. 

Gas radiation 

In <strong>the</strong> case <strong>of</strong> flame radiation <strong>the</strong>re are two 

types radiation: visible and invisible/infrared 

radiation. The last one is <strong>the</strong> invisible infrared 

radiation emitted by carbon di<strong>oxide</strong> and 

steam 4 . 

The quantity <strong>of</strong> heat transferred through 

radiation from a <strong>layer</strong> <strong>of</strong> CO 2 respectively H 2O 

to a grey surface are: 

3.2 3.2 

Tg Tw 

qCO2 = ε . 10.35(p . 

CO2 s) 0.4 [(⎯⎯) - (⎯⎯) 

100 100 

Tg 

0.65 

. (⎯) ] Tw 4 After J. H. Brunklaus 

Fig. 1: Bilateral (green) and unilateral (red) exposure to radiation (not coated, h = 10 cm) 

(5)[[7]→[6]] 


Abbildungen: Belte

q H2O = ε(46.52 - 94.9p H2Os)(p2 . s) 0.6 

3 ⎯⎯ 

3 ⎯⎯ 

2.32 + 1.37 √p 2 s 2.32 + 1.37 √p 2 s 

Tg Tw [(⎯⎯) -(⎯⎯) 

100 100 ] 

(6)[[7]→[6]] 

Where: q = quantity <strong>of</strong> heat [W/m²]; p = partial 

pressure [daN/cm²]; s = <strong>layer</strong> thickness [m]; 

T g = gas temperature [K]; T w = temperature 

<strong>of</strong> <strong>the</strong> grey surface [K]; ε = emissivity rapport: 

A λ = absorbed radiation and A λs = total radiated 

energy 

Aλ 

ε = ⎯ 

A λs 

Infrared heating <strong>of</strong> massive 

<strong>aluminium</strong> castings 

Electric radiant burners 

(7)[[7]→[6] 

Experiment purpose – determination <strong>of</strong>: 

• heating rate 

• maximal reached temperature 

• influence <strong>of</strong> 

surface quality 

�� 


coated or not 

as cast 

100% graphite coated 

• distance between <strong>the</strong> infrared source and 

<strong>the</strong> irradiated surface (h); 

Experimental conditions: 

• experimental infrared facility 

industrial IR installation 

power : 108 KW/m² 

nominal power: 37.8 kW 

(active surface <strong>of</strong> 0.35 m²) 

air cooling system 

• test specimen 

die casting: cylinder head 

Al-Si-Mg-Cu alloy 

weight: 26 kg 

wall thickness: 120 mm 

• exposure to radiation 

bilateral and unilateral 

In <strong>the</strong> case <strong>of</strong> a unilateral exposure to radiation 

(one part <strong>of</strong> <strong>the</strong> infrared module is switched 

<strong>of</strong>f) we obtain a heating rate <strong>of</strong> 0.11 °C/s (between 

50 and 425°C 5 ). In <strong>the</strong> case <strong>of</strong> a bilateral 

exposure to radiation we obtain a heating 

rate <strong>of</strong> 0.43 °C/s. Enhancing <strong>the</strong> installation 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

�� 

��@�� 

power by 100% (in <strong>the</strong> case <strong>of</strong> a bilateral exposure 

to radiation) quadruped <strong>the</strong> heating 

rate by a coeval reduction <strong>of</strong> <strong>the</strong> energy consumption. 

The unequal temperature maximum 

is conspicuous (fig.1). 

Energy consumption: 

• unilateral exposure to radiation: ~ 55 min 

55 

• E = ⎯ . 18.31 = 16.78kWh 

60 

• bilateral exposure to radiation: ~ 15 min 

15 

• E = ⎯ . 18.31 = 4.58kWh 

60 

Where: 0.35 m²………………………37.8 kW 

0.169 m² 6 ………….……………x kW 

X = 18.31 kW/side → X = 36.63 kW for a 

bilateral exposure 

The coating leads to a considerable enhancement 

<strong>of</strong> <strong>the</strong> surface heating rate and simultaneously 

to <strong>the</strong> increase <strong>of</strong> <strong>the</strong> temperature 

difference between <strong>the</strong> surface and <strong>the</strong> 

5 One has to pay maximum attention to <strong>the</strong> solidus temperature 

(especially for Al alloys containing Cu). 

6 casting surface exposed to radiation


center <strong>of</strong> <strong>the</strong> casting. 

Energy consumption and process efficiency 

between 55°C and 535°C 

• not coated: heating time = 0.27 h 

• E = 36.63 x 0.27 = 9.89 kWh 

• η = 3.579 7 /9.89 = 0.361 

• coated: heating time = 0.2 h 

• E = 36.63 x 0.2 = 7.3 kWh 

• η = 3.579 8 /7.3 = 0.49 

surface heating rate [°C/s] 

coated 0.59 

not coated 0.68 

Tab. 1: <strong>Influence</strong> <strong>of</strong> surface 

Gas fired burners 

The following experiment was strongly supported 

by LOI Italimpianti and performed by 

Elster Kromschröder. 

Experiment purpose – determination <strong>of</strong>: 

• heating rate (furnace preheated 

at ~ 500°C) 

Experimental conditions: 

• experimental heat treatment furnace 

using a ceramic flat flame burner as 

infrared radiation source 

power : 450 kW 

• test specimen 

die casting: cylinder head 

Al-Si-Mg-Cu alloy 

weight: 26 kg 

wall thickness: 120 mm 

• exposure to radiation 

unilateral 

Fig. 2: Flat flame burner: 450 kW 

Conclusion and prospects 

The industrial use <strong>of</strong> <strong>the</strong> infrared radiation for 

<strong>the</strong> heat treatment <strong>of</strong> aluminum castings in Europe 

is only a question <strong>of</strong> time. 

808 

7 Q = ⎯⎯ 

26 

x ∫(4.8 + 3.2 . 10-3T)dT = 3075.44kcal = 3.579kWh 

26.9 328 

808 

8 Q = ⎯⎯ 

26 

x ∫(4.8 + 3.2 . 10-3T)dT = 3075.44kcal = 3.579kWh 

26.9 328 

Fig. 3: Heating rate: 450 kW 

Belte’s industrial concept is illustrated in Fig. 

4. 

Our concept proves that <strong>the</strong> next generation 

<strong>of</strong> <strong>aluminium</strong> heat treatment plants will 

be: 

• extreme mobile 

commissioning/decommissioning 

duration: ~ three weeks 

• extreme flexible 

without base and ancillary frames 

• extreme short processing time 

• extreme environmentally friendly 

without gas emission (in <strong>the</strong> case <strong>of</strong> 

<strong>the</strong> use <strong>of</strong> electric radiant burners) 

References 

[1] http://www.britannica.com/ 

[2] M. Wagner, P. Buchet, W. Kesteleyn, A. Molin – 

Radiant heating in industrial processes, 22nd World 

Gas Conference June 1-5, 2003 Tokyo, Japan 

Fig. 4: Industrial infrared heating facility for <strong>aluminium</strong> alloys 

[3] C. Samoila, L. Druga, L.Stan – Cuptoare si Instalatii 

de Incalzire, EDP, Bukarest, 1983 

[4] D. Dragulin, M. Belte, M. Dragulin – Thermodynamische 

Aspekte der Wärmebehandlung von 

Metallen, pro Business Verlag, Berlin, 2010 

[5] M. Orfeuil – Electric Process Heating, Battelle 

Press, 1987 

[6] J. H. Brunklaus – Cuptoare Industriale, Editura 

Technica, Bukarest, 1977 

[7] A. Schack – der industrielle Wärmeübergang, 5. 

Auflage, Stahl-Eisen Verlag, 1957 

Authors 

Markus Belte is CEO <strong>of</strong> <strong>the</strong> Belte AG, located in 

Delbrück, Germany. 

Dr. Dan Dragulin is head <strong>of</strong> research and development 

at Belte AG. 


Abbildungen: promeos 

Resource efficiency at increased process quality 

becomes <strong>the</strong> dominating driver for investments 

in industry, and so in <strong>the</strong> world <strong>of</strong> casting. 

An optimized melting and material logis- 

tics process requires optimum temperature 

control <strong>of</strong> material and casting equipment, including 

melting furnace, transportation ladles 

or launders and casting moulds. 


New standards in gas-fired heating <strong>of</strong> moulds, ladles, launders 

and melting furnaces through flameless burner technology 

Jochen Volkert, promeos GmbH 

Fig.1: Porous burner 

Fig. 2: Different designs <strong>of</strong> <strong>the</strong> porous burner 

Based on its flameless high performance industry 

burner portfolio, promeos <strong>of</strong>fers new 

preheating and tempering solutions all along 

<strong>the</strong> process line. Energy savings <strong>of</strong> up to 75% 

and increased process stability due to enhanced 

control measures have been approved 

at numerous foundries all across <strong>the</strong> metal 

processing market. 

The clou is as easy as impressing: <strong>the</strong> 

absence <strong>of</strong> a free flame as spot-type heating 

source and <strong>the</strong> infinitely adjustable heat flux 

(similar to a dimming lamp) are core features 

to avoid inhomogeneity and inefficiency in 

heating ladles, launders or moulds. 

Porous burner technology 

In a porous burner (Fig. 1) <strong>the</strong> combustion takes 

place in a porous high temperature ceramic, 

<strong>the</strong> combustion reactor, instead <strong>of</strong> in an open 

flame. This results in a flameless, volumetric 

combustion in <strong>the</strong> form <strong>of</strong> glowing ceramic 

foam which can be used as a radiating surface 

as well as a homogenous source <strong>of</strong> heat. Any 

shapes are possible such as round or square 

burners, lines, cylinders, rings, rhombuses and 

o<strong>the</strong>r specially tailored shapes (Fig. 2). 

Special applications require special solutions, 

for example line shaped burner heads 

Table 1: Heating systems for casting moulds: benefits / value proposition 

which concentrate <strong>the</strong> heat from convection 

and radiation exactly where it is needed. 

The length <strong>of</strong> a reactor can thus be extended 

if desired. Dimensions <strong>of</strong> several meters in 

length with a reactor width from 15 to 200 mm 

allow specific burning performances from 2 

to 600 kW/m in length. 

Based on <strong>the</strong> unique features <strong>of</strong> <strong>the</strong> flameless 

porous burner combustion technology, 

promeos develops and manufactures tailored 

heating systems for process applications. 

Furnace engineering is required for both 

process technology and manufacturing to ensure 

optimum product quality. The outstanding 

design features <strong>of</strong> promeos burners and its 

ease <strong>of</strong> integration into any kind <strong>of</strong> apparatus 

allow <strong>the</strong> heat to be distributed over / into <strong>the</strong> 

required area / volume and applied where <strong>the</strong> 

process needs it, as opposed to being directed 

to only one spot. promeos supplies ‘tailor 

made suits’ for your individual needs, powerful 

and efficient. 

The unique combination <strong>of</strong> features like <strong>the</strong> 

power density, a rapid and precise power adjustment 

(similar to electric heating devices), 

a homogeneous heat flux and its compact and 

modular design result in <strong>the</strong> following benefits: 

• enhanced quality <strong>of</strong> your products 

+ reduced preheating times The high energy density and heat transfer via 

infrared radiation and controlled convention allow 

an unmatched quick heating <strong>of</strong> <strong>the</strong> moulds. 

Generally <strong>the</strong> preheating times are cut down 

50% and more. 

+ homogeneous temperature distribution So called ‘hot spots’ that are due to local fire or 

flames are completely eliminated because <strong>of</strong> <strong>the</strong> 

homogeneity <strong>of</strong> <strong>the</strong> heat input. In combination 

with <strong>the</strong> continuously variable power control <strong>of</strong> 

<strong>the</strong> burners <strong>the</strong> preheating <strong>of</strong> <strong>the</strong> mould can be 

controlled optimally. 

+ production time Production time is gained due to <strong>the</strong> shortened 

heating-up time and corresponding process control. 

+ less scrap production The ideal preheating <strong>of</strong> <strong>the</strong> mould reduces <strong>the</strong> 

number <strong>of</strong> required ‘warm castings’ – waste reduction 

is <strong>the</strong> consequence. 

+ durability <strong>of</strong> <strong>the</strong> tools The homogeneous temperature distribution reduces 

<strong>the</strong>rmal stress and <strong>the</strong>reby automatically 

increases <strong>the</strong> lifetime <strong>of</strong> <strong>the</strong> tools. 

Σ ROI < 24 months, guaranteed 

< 12 months, mostly 

Usually your investment is recovered within a 

period <strong>of</strong> max. 12 months. We will compile you 

with a detailed amortization calculation with 

every <strong>of</strong>fer you receive from us. 



• increased performance 

• reduced warm-up time or start-up time <strong>of</strong> 

your equipment 

• energy saving potential <strong>of</strong> up to 70% for 

your processes 

• operational cost saving <strong>of</strong> 50% by 

substituting electrical heating systems. 

Heating systems for casting moulds: 

higher production rates, quicker 

and more precise preheating 

The final step in <strong>the</strong> casting process is <strong>the</strong> 

most sensible – <strong>the</strong> temperature pr<strong>of</strong>ile <strong>of</strong> <strong>the</strong> 

casting tools is essential for <strong>the</strong> quality <strong>of</strong> <strong>the</strong> 

products. Since every individual mould requires 

its optimum heating device, promeos 

<strong>of</strong>fers customized heating tools, individually 

adapted to shape and process (Fig. 3). 

The modular ‘Lego-like’ design <strong>of</strong> <strong>the</strong> flamefree 

porous burner units enables promeos in 

a unique way to design ‘custom-made suits’ 

for different tools at reasonable cost. Opti- 

mum mould preheating, leading to significantly 

reduced scrap production (‘initial cast- 

Fig. 3: Drying curve refractory concrete 

+ reduced preheating time The homogeneously distributed heat input at a 

high energy density allows an unmatched quick 

heating <strong>of</strong> <strong>the</strong> ladle. Generally <strong>the</strong> heating-up 

times are cut down 50% and more. 

+ increased durability(exact heat-up curves, 

homogeneous temperature distribution with 

functions for drying, sintering and warming) 

So called ‘hot spots’ that are due to local fire or 

flames as well as cold areas (‘temperature holes’) 

are completely eliminated. A minimal loss <strong>of</strong> 

material and an increased durability <strong>of</strong> <strong>the</strong> ladle 

lining are <strong>the</strong> positive consequences. The temperature 

control device allows continuously variable 

temperature regulation and even a precise 

realization <strong>of</strong> given temperature curves. 

+ 65% reduction <strong>of</strong> operating costs* Energy savings <strong>of</strong> 65% are not unusual. This is 

due to <strong>the</strong> extremely improved heat transfer and 

<strong>the</strong> continuously variable power control, required 

for temperature holding during stand-by time. 

+ 80% CO 2 emission savings* 70% less CO 2 emissions and significantly lower 

levels <strong>of</strong> pollutants (CxHy and NOx) mean > 80% 

reduction <strong>of</strong> CO 2 equivalents. 



* Maximum value proven by customer data 

Table 2: Ladle heating systems: benefits / value proposition 

Fig. 4: Heating system for casting moulds 

ing’) is not longer a miracle. The required investments 

pay back in less than 24 months – 

a must in times <strong>of</strong> increasing resource efficiency 

constraints (see Table 1). 

Ladle heating systems: preheating 

allows energy savings <strong>of</strong> up to 65% 

The transport ladles lined with refractory material 

are heated to a target temperature in 

<strong>the</strong> empty state. A well-adapted insulated cap, 

which includes <strong>the</strong> burner, is placed on top <strong>of</strong> 





<strong>the</strong> ladle. The burner transmits <strong>the</strong> heat onto 

a radiating body which transfers <strong>the</strong> combustion 

energy as infrared radiation to <strong>the</strong> lining 

<strong>of</strong> <strong>the</strong> ladle. 

The heat is homogeneously and effectively 

transmitted all over <strong>the</strong> ladle wall. The hot 

gas flow is directed to <strong>the</strong> bottom <strong>of</strong> <strong>the</strong> ladle 

and afterwards through a slit between <strong>the</strong> 

radiating body and <strong>the</strong> lining <strong>of</strong> <strong>the</strong> ladle to 

<strong>the</strong> hot gas outlet, whereby <strong>the</strong> heat transfer 

is additionally improved. The promeos ladle 

heating substitutes existing ‘fires’ to closed 

systems with energy savings <strong>of</strong> up to 65%. It 

<strong>the</strong>refore combines maximum energy efficiency 

with ideal industrial safety and maximum 

process reliability. 

Reliable Drying and Sintering: The variable 

power control enables <strong>the</strong> realization <strong>of</strong> ex- 

Fig. 5: 

Ladle heating 

system 

act given drying and sintering pr<strong>of</strong>iles and in 

addition an optimum stand-by functionality 

with minimum energy consumption. promeos’ 

ladle heating systems is a reliable solution for 

an optimum casting process (see Fig. 3 and 

Table 2). 

Heating <strong>of</strong> launders and filter 

boxes: for a better process control 

The above explained advantages <strong>of</strong> homogeneously 

and efficiently heated ladles are also 

valid for launders and filter boxes – <strong>the</strong> better 

<strong>the</strong> material transportation systems, <strong>the</strong> 

better <strong>the</strong> casting quality. 

promeos burner systems, integrated into 

<strong>the</strong> lid <strong>of</strong> filter or into <strong>the</strong> covers <strong>of</strong> launders 

provide a homogeneous heating <strong>of</strong> <strong>the</strong> filter 

and <strong>the</strong> refractory lining. While hot spots are 

avoided and <strong>the</strong> lifetime <strong>of</strong> <strong>the</strong> refractory is 

enhanced, <strong>the</strong> temperature loss <strong>of</strong> <strong>the</strong> liquid 

metal can be reduced tremendously. Increased 

energy efficiency by reduced melting temperatures 

and improved material quality are evident 

(see Fig. 6 and 7 and Table 3 next page). 


New: Hot air heating system 

for industrial applications 

promeos <strong>of</strong>fers a compact and mobile, robust 

and modular hot air heating system for industrial 

applications – to be used for drying, preheating, 

tempering or whatsoever. 

Fig. 6: Heating system for filterboxes 

Operating mode: The exhaust gases <strong>of</strong> <strong>the</strong> 

central burner are mixed with secondary air in 

a compact mixing chamber inside <strong>of</strong> <strong>the</strong> heating 

system. While <strong>the</strong> burner control defines 

<strong>the</strong> heat capacity, <strong>the</strong> air (heating) temperaure 

can be controlled by <strong>the</strong> ratio <strong>of</strong> exhaust 

gas and secondary air mass flow (see Fig. 8). 

The hot air is <strong>of</strong>fered as point source, line 

source, area source or 3D source using different 

nozzles, almost unlimited applications 

flexibility, which was only known from electrically 

heated blowers up to now. 

Fig. 7: Heating system for launders 


promeos <strong>of</strong>fers ‘gas heat in electric quality’ 

without <strong>the</strong> existing shortcomings <strong>of</strong> electrical 

hot air guns such as limited temperature, 

limited power density or limited lifetime due 

to unsatisfying robustness. 

The hot air heating system from promeos 

is designed even for rough boundary conditions, 

such as steel plants and foundries etc.. 

It is both robust and compact and can be used 

stationary and as mobile heat source. A crane 

Fig. 8: Operating range <strong>of</strong> <strong>the</strong> hot air heating system for industrial applications 

hook and transportation wheels enables <strong>the</strong> 

user to move it quickly where it is needed. 

Benefits: While <strong>the</strong> cost saving potential 

<strong>of</strong> 50% through substituting electrical energy 

by gas is self-arguing, a CO 2-reduction <strong>of</strong> more 

+ no loss <strong>of</strong> temperature / no freezing <strong>of</strong> <strong>the</strong> 

melt 

The extremely uniform heating <strong>of</strong> <strong>the</strong> refractory 

surface and <strong>of</strong> <strong>the</strong> filters prevents an uncontrol- 

led cooling <strong>of</strong> <strong>the</strong> melt. 

+ no overheating <strong>of</strong> <strong>the</strong> furnaces The heat loss from <strong>the</strong> furnace to <strong>the</strong> casting station 

can be reduced / prevented so far, that an 

unwanted overheating <strong>of</strong> <strong>the</strong> furnaces can be 

avoided. 

+ increased durability <strong>of</strong> <strong>the</strong> refractory materials So called ‘hot spots’ that are due to local fire 

or flames as well as <strong>the</strong> non-uniform heating <strong>of</strong> 

<strong>the</strong> launders and filters are completely avoided. 

Thereby <strong>the</strong> <strong>the</strong>rmo-mechanical loading is significantly 

reduced and <strong>the</strong> lifetime <strong>of</strong> <strong>the</strong> material 

is increased. 

+ reduced energy consumption An overheating <strong>of</strong> <strong>the</strong> furnaces is not necessary. 

+ increased product quality The material is melted, prepared, transported and 

casted at an optimum metallurgical temperature. 

+ lower costs The saving <strong>of</strong> energy costs and <strong>the</strong> avoidance <strong>of</strong> 

‘warm casting / tapping’ lead to lower operating 

costs without plant down-times and thus to higher 

productivity. 



Table 3: Heating <strong>of</strong> launders and filterboxes: benefits / value proposition 

than 60% in Germany and most world markets 

is worth to be mentioned. 

promeos has set new standards in pre- 

heating solutions for <strong>the</strong> casting industry. 

Based on its unique combustion technology, 

promeos developed from a technology supplier 

to a reliable system manufacturer and 

service supplier. Energy efficiency, product 

quality and process reliability were and are 

<strong>the</strong> driving force. Future subjects in product 

development focus on a new type <strong>of</strong> melting 

and tempering furnace and fur<strong>the</strong>r energy 

efficiency improvements including internal 

energy recovery by combustion air preheating. 

� 







Effects <strong>of</strong> gaseous pyrolysis 

products on <strong>aluminium</strong> recycling yield 

1 Jaroni, B.; 1 Gisbertz, K.; 2 Rombach, G.; 1 Friedrich, B. 

1 RWTH Aachen University, IME Process Metallurgy and Metals Recycling, Germany 

2 Hydro Aluminium Rolled Products GmbH, Germany 

After a coated aluminum product has 

reached <strong>the</strong> end <strong>of</strong> life cycle it needs to 

be recycled in an economical, effective 

and ecological way. State-<strong>of</strong>-<strong>the</strong>-art is <strong>the</strong> 

<strong>the</strong>rmal removal <strong>of</strong> coatings and o<strong>the</strong>r organic 

fractions via treatment by pyrolysis 

prior to melting. In modern twin chamber 

furnaces this step takes place inside <strong>the</strong> 

scrap melting chamber. Through different 

construction methods, it is possible 

to use <strong>the</strong> gaseous pyrolysis products for 

an internal combustion in order to safe 

<strong>natural</strong> gas. Lab-scale experiments have 

shown that <strong>the</strong> average residence time is 

too short to complete <strong>the</strong> pyrolysis. It has 

to be considered that <strong>the</strong> pyrolysis continuous 

while <strong>the</strong> scrap block is pushed 

under <strong>the</strong> aluminum melt. By reason <strong>of</strong> 

this production procedure, gaseous reaction 

products, e. g. CO, CO 2 and C xH y, 

have a long contact time inside <strong>the</strong> melt. 

It is assumed that <strong>the</strong> dross formation 

increases rapidly as a result <strong>of</strong> various 

gas-melt reactions. Based on this assumption 

<strong>the</strong> aim <strong>of</strong> this work is <strong>the</strong> evaluation 

<strong>of</strong> <strong>the</strong> influence <strong>of</strong> gaseous pyrolysis 

reaction products on <strong>the</strong> dross formation 

by <strong>the</strong> simulation in lab-scale experiments. 

Therefore, typical pyrolysis gases 

are injected into <strong>the</strong> melt via a gas purging 

device. Fur<strong>the</strong>rmore, <strong>the</strong> magnesium 

content <strong>of</strong> <strong>the</strong> aluminum melt is varied to 

show its impact in addition to <strong>the</strong> gases. 

1 Introduction 

The pyrolytic process in chamber furnaces is 

located on <strong>the</strong> bridge inside <strong>the</strong> scrap melting 

chamber. In modern twin chamber furnaces 

<strong>the</strong> average charging time is 25 minutes. The 

time interval depends on bath temperature, 

bath level, atmosphere composition and security 

door lock time. This time interval simultaneously 

limits <strong>the</strong> duration <strong>of</strong> scrap heating, 

organic scission and combustion <strong>of</strong> carbon released 

from <strong>the</strong> scrap surface. With <strong>the</strong> next 

scrap charge <strong>the</strong> de-coated scrap is pushed into 

<strong>the</strong> aluminum melt which means that charging 

time is equal to <strong>the</strong> duration <strong>of</strong> <strong>the</strong> <strong>the</strong>rmal 

pretreatment. The industrial de-coating time 

Fig. 1: Balance <strong>of</strong> forces 

seems to be too short for a complete pyrolysis 

step especially inside compacted scrap, e.g. 

coil-leftovers and packages. The de-coating 

grade is a function <strong>of</strong> time, temperature, heat 

transfer and package density. In comparison 

with aluminum blocks <strong>the</strong> heat transfer <strong>of</strong> 

scrap packages is reduced. As a result <strong>of</strong> <strong>the</strong> 

package structure <strong>the</strong> pyrolysis products cannot 

leave <strong>the</strong> compacted material with a typical 

density is well below 2 g/cm 3 . 

2 Gas-liquid reactions 

The reaction between gaseous reaction products 

and liquid aluminum follows a certain 

pattern. The bubble formation depends on 

<strong>the</strong> porosity <strong>of</strong> <strong>the</strong> medium. In <strong>the</strong> case <strong>of</strong> a 

scrap package <strong>the</strong>re is a high variation <strong>of</strong> <strong>the</strong> 

hole diameter. Generally speaking, <strong>the</strong> bubble 

diameter is a function <strong>of</strong> surface tension, <strong>the</strong> 

pore diameter and <strong>the</strong> amount <strong>of</strong> formed gas. 

With a porous medium as gas distributor only 

<strong>the</strong> larges pores are outgassing at low gas flow 

rates. An increasing gas flow rate also enables 

<strong>the</strong> gas evolution on / at / <strong>of</strong> smaller openings, 

so that <strong>the</strong> gas flow is distributed more on <strong>the</strong> 

entire surface <strong>of</strong> <strong>the</strong> porous medium. The most 

common example <strong>of</strong> viewing bubble formation 

as a result <strong>of</strong> <strong>the</strong> bubble size goes back to <strong>the</strong> 

balancing forces <strong>of</strong> a single bubble (Fig. 1). 

Depending on <strong>the</strong> predominant /prevailing 

bubble formation mechanism, some forces are 

negligible. The direction <strong>of</strong> <strong>the</strong> resistance force 

is a result <strong>of</strong> <strong>the</strong> liquid stream. In quiescent 

fluids <strong>the</strong> breakaway point <strong>of</strong> bubbles is later 

than in turbulent fluids. The bubble size can be 

calculated with equation 1. [1] (1) 

d[(p] gV Bv) 

⎯⎯⎯⎯⎯ = F inertia = F Resistance = F Flotation 

dt 

= F Pressure = F σ = F virtual Mass 

In <strong>the</strong> scrap melting chamber <strong>the</strong> conditions 

are turbulent because <strong>of</strong> <strong>the</strong> electromagnetic 

pumps (EMP) which are used to homogenize 

<strong>the</strong> melt temperature as well as <strong>the</strong> alloying 

element pr<strong>of</strong>ile in <strong>the</strong> furnace and consequently 

for <strong>the</strong> increase <strong>of</strong> <strong>the</strong> melting rate. In this 

paper we will focus on <strong>the</strong> <strong>the</strong>rmochemical 

modelling and <strong>the</strong> experimental pro<strong>of</strong> <strong>of</strong> our 

results. 

3 Thermochemical modelling 

‘FactSage’ is used for <strong>the</strong> <strong>the</strong>rmochemical 

modelling <strong>of</strong> <strong>the</strong> melting experiments to provide 

an overview <strong>of</strong> <strong>the</strong> possible reactions 

and <strong>the</strong>ir reaction products. It is taken into 

account in all <strong>the</strong> representations shown in 

this chapter that <strong>the</strong> calculations are always 

based on reaction equilibrium. The modelling 

s<strong>of</strong>tware FactSage can access various materials 

databases (here Fact53, SGPS and SGTE). 

The equilibrium phases, which represent <strong>the</strong> 

system with <strong>the</strong> lowest total free enthalpy as 

a function <strong>of</strong> predetermined variables is based 

on Steady state conditions. The temperature 

is set to 750°C while <strong>the</strong> pressure <strong>of</strong> <strong>the</strong> melt 

and <strong>the</strong> added gas are set to 1 bar. The high 

metal volume used in <strong>the</strong> calculations has 

<strong>the</strong> consequence that <strong>the</strong> gases CO 2, CH 4 and 

CO are completely converted into solid reac- 


Fig. 2: Equilibrium phases for <strong>the</strong> reaction <strong>of</strong> 2350 g AlMg3 + CO 2 with 

up to 15 mol and under consideration <strong>of</strong> oxycarbide phases 

Fig. 3: Al and Mg losses in relation <strong>of</strong> Mg content and injected gas 

Fig. 4: Experimental setup with used gas analyzing system 


tion products except for H 2. 

In industrial scale <strong>the</strong> furnace atmosphere including <strong>the</strong> 

injected gas is constantly replaced and <strong>the</strong> remaining gaseous 

products are continuously removed from <strong>the</strong> reaction chamber. 

Accordingly <strong>the</strong> real system is an open one. 

An example for a typical calculation done in this project 

can be seen in Fig. 2. In this calculation one liter <strong>of</strong> AlMg3 alloy 

melt reacts with a variable amount <strong>of</strong> CO 2. The main products 

are MgO, Al 2O 3, Al 4C 3, MgAl 2O 4 and Al 2CO. 

It is easy to recognize that in <strong>the</strong> range up to 1.5 mol CO 2 

magnesium is preferably oxidized from <strong>the</strong> alloy melt, while 

<strong>the</strong> carbon content <strong>of</strong> <strong>the</strong> gas is fully converted into Al 4C 3. 

Magnesium is no longer present in <strong>the</strong> melt at at an addition <strong>of</strong> 

1.5 mol CO 2. At that point aluminum consumption increases. 

When <strong>the</strong> concentration <strong>of</strong> CO 2 reaches three mol, <strong>the</strong> balance 

shifts increasingly to <strong>the</strong> oxycarbide phase Al 2CO, while <strong>the</strong> 

amount <strong>of</strong> MgO phase remains constant. Above three moles 

<strong>of</strong> CO 2 <strong>the</strong>re is no more aluminum carbide in <strong>the</strong> system. At 

<strong>the</strong> same point magnesium <strong>oxide</strong> is transformed into <strong>the</strong> spinel 

MgAl 2O 4, so that <strong>the</strong> amount <strong>of</strong> this equilibrium phase starts 

to increase. The magnesium <strong>oxide</strong> is completely transformed 

into <strong>the</strong> spinel at a CO 2 concentration <strong>of</strong> 11.5 mol. Alumina 

appears as an additional oxidation product <strong>of</strong> <strong>the</strong> reaction between 

aluminum and carbon di<strong>oxide</strong> as a result <strong>of</strong> <strong>the</strong> complete 

consumption <strong>of</strong> MgO. 

In Fig. 3 <strong>the</strong> specific metal losses are presented in a bar 

chart diagram. The diagram shows <strong>the</strong> simulated aluminum and 

magnesium losses <strong>of</strong> <strong>the</strong> three investigated alloys. The results 

are grouped by <strong>the</strong> intended alloy and <strong>the</strong> gas species in different 

colours. The aluminum losses are shown in <strong>the</strong> lower 

bar part, while <strong>the</strong> magnesium losses <strong>of</strong> <strong>the</strong> AlMg alloys are 

added above in a weaker colour. The calculations revealed 

<strong>the</strong> metal losses for circa 87 mol aluminum or aluminum alloy 

(2,350 g) by addition <strong>of</strong> 1 mol <strong>of</strong> gas. The parameters have 

been pre-estimated to provide <strong>the</strong> opportunity to compare <strong>the</strong> 

calculated with <strong>the</strong> experimental results. As a consequence <strong>of</strong> 

this one mol <strong>of</strong> injected gas is used in <strong>the</strong> calculations, leading 

to a lack <strong>of</strong> oxygen in <strong>the</strong> system for full magnesium oxida- 

tion. As a result <strong>the</strong>re is no observable difference between <strong>the</strong> 

AlMg3 and <strong>the</strong> AlMg5 alloy in <strong>the</strong> calculation. 

4 Experiment 

4.1 Experimental setup: For <strong>the</strong> experimental campaign <strong>of</strong> this 

work three different aluminum alloys AlMg3, AlMg5 and pure 

Al were used. The feeding material <strong>of</strong> all experiments has always 

<strong>the</strong> same size and surface in order to minimize a failure 

<strong>of</strong> different aluminum <strong>oxide</strong> amounts inherent to <strong>the</strong> charge 

materials. CO, CO 2 and CH 4 as typical de-coating products 

are injected into <strong>the</strong> melt. The following sketch (Fig. 4) shows 

<strong>the</strong> experimental setup. 

All experiments are realized in an induction furnace. The 

used Al 2O 3 crucible has a volume <strong>of</strong> 1.3 litres. The induction 

coil is installed into a vacuum chamber which allows controlling 

<strong>the</strong> atmosphere inside <strong>the</strong> furnace. In all experiments <strong>the</strong> 

furnace atmosphere is 100% N 2. To realize a constant N 2 flow 

<strong>the</strong> gas is injected at <strong>the</strong> furnace bottom while <strong>the</strong> pressure is 

kept at 1 bar by an over pressure valve at <strong>the</strong> upper furnace. For 

injection <strong>of</strong> different gasses into <strong>the</strong> melt a glass lance is used. 



Fig. 5: a) Gas injection system; b) Induction coil with hood 

An exhaust hood above <strong>the</strong> crucible is used to 

collect <strong>the</strong> gaseous reaction products and <strong>the</strong> 

unconsumed gasses. An exhaust fan supplies 

<strong>the</strong> online quality gas measurement with <strong>the</strong> 

collected gasses. To make sure that no furnace 

atmosphere is sucked into <strong>the</strong> hood during <strong>the</strong> 

experiment more gas is injected than removed 

by <strong>the</strong> fan. Due to that <strong>the</strong> system has a slight 

over pressure under <strong>the</strong> hood, too. 

Fig. 5a) shows a drawing <strong>of</strong> <strong>the</strong> gas injection 

system. The induction coil and <strong>the</strong> used hood 

to collect <strong>the</strong> gaseous products are shown in 

Fig. 5b). 

In <strong>the</strong> first experimental campaign 12 tests 

are performed (Table 1). The gas injection time 

is limited by <strong>the</strong> crucible volume. The injection 

<strong>of</strong> gas was stopped when <strong>the</strong> crucible is 

fully filled by dross. The gas composition is 

limited by <strong>the</strong> range <strong>of</strong> <strong>the</strong> installed gas analyzing 

system and also for security reasons. 

4.2 Results 

a) b) 

In Fig. 6 <strong>the</strong> metal losses <strong>of</strong> <strong>the</strong> tests are presented. 

They are grouped according to <strong>the</strong> 

used feedstock materials. The chart shows <strong>the</strong> 

gross-remelting losses, which are calculated 

by <strong>the</strong> ratio <strong>of</strong> final weight <strong>of</strong> <strong>the</strong> cast to initial 

weight <strong>of</strong> <strong>the</strong> feedstock, see equation 2. 

For <strong>the</strong> gross-remelting losses <strong>the</strong> metal content 

<strong>of</strong> <strong>the</strong> dross is not considered. The column 

chart compares <strong>the</strong> influence <strong>of</strong> different gases 

as well as <strong>the</strong> influence <strong>of</strong> <strong>the</strong> main alloying 

element (magnesium) on <strong>the</strong> gross-remelting 

losses. 

The metal losses decrease for all three materials 

in <strong>the</strong> order <strong>of</strong> CO2 , CO and CH4 . The 

control experiments already indicated significant 

losses <strong>of</strong> <strong>the</strong> melt phase. In presence <strong>of</strong> 

magnesium <strong>the</strong> effect <strong>of</strong> carbon di<strong>oxide</strong> on <strong>the</strong> 

metal losses are doubled. The maximum loss 

<strong>of</strong> more than 45% is represent by <strong>the</strong> combination 

<strong>of</strong> <strong>the</strong> alloy AlMg3 and CO2 as injected 

gas. A linear dependence on <strong>the</strong> magnesium 

content <strong>of</strong> <strong>the</strong> alloy to <strong>the</strong> gross metal loss is 

not evident. 

To divide <strong>the</strong> metallic part from <strong>the</strong> nonmetallic 

dross components a subsequent salt 

melt step is conducted. A typical salt flux (55% 

Experiment No. Injected gas composition Mg-Contentin Mass.-% Treatment in min 

NV1 N2 (Chamber atmosphere) 0 120 

NV2 N2 (Chamber atmosphere) 2.82 20 

NV3 N2 (Chamber atmosphere) 4.86 20 

HV1KD Ar + 15 Vol.-% CO 2 0 32 

HV2KD Ar + 15 Vol.-% CO 2 2.82 21 

HV3KD Ar + 15 Vol.-% CO 2 4.86 14 

HV4KM Ar + 7 Vol.-% CO 0 32 

HV5KM Ar + 7 Vol.-% CO 2.82 20 

HV6KM Ar + 7 Vol.-% CO 4.86 20 

HV7KW Ar + 5 Vol.-% CH 4 0 19 

HV8KW Ar + 5 Vol.-% CH 4 2.82 19 

HV9KW Ar + 5 Vol.-% CH 4 4.86 19 

Table 1: Experimental plan 

Fig. 6: Gross-remelting losses 

mcast 

remelting loss gross in % = ⎯⎯⎯⎯ (2) 

m feedstock 

NaCl, 45% KCl and ≤ 1% CaF 2 ) is used. To 

guarantee a complete separation <strong>of</strong> metal and 

dross a high salt factor is adjusted. 

The net-remelting losses <strong>of</strong> <strong>the</strong> melting experiments 

can be determined from <strong>the</strong> measured 

metal yields <strong>of</strong> <strong>the</strong> salt experiments, see 

equation 3: (3) 

mcast + m recovered from dross 

remelting loss net in % = ⎯⎯⎯⎯⎯⎯⎯⎯⎯ 

m feedstock 

The results are shown in Fig. 7. The metal 

contents <strong>of</strong> <strong>the</strong> dross vary between 30 and 

97% (average 70%). The net metal losses decrease 

in <strong>the</strong> same way like <strong>the</strong> gross metal 

losses according to <strong>the</strong> oxidation effect <strong>of</strong> <strong>the</strong> 

treatment gas, in <strong>the</strong> order <strong>of</strong> CO 2 , CO and 

CH 4 . The net-remelting losses <strong>of</strong> <strong>the</strong> control 

experiments increase from 0.1% via 0.4% 

through to 1.3% with rising magnesium content 

<strong>of</strong> <strong>the</strong> melt. In contrast to <strong>the</strong> gross metal 

losses <strong>the</strong> losses after metal separation by 

salt treatment rise for all three injected gases 

with increasing magnesium content. Thus, <strong>the</strong> 

metal loss values for CO and CH 4 treatment 

based on <strong>the</strong> three points show a linear dependence 

on magnesium concentration. 

Comparing <strong>the</strong> gross and net metal losses it 

is good to see that <strong>the</strong> dross has a high metal 

content. This can be a result <strong>of</strong> dross handling 

and manual stripping but it can also be a result 

<strong>of</strong> <strong>the</strong> dross structure. It could be observed 

that especially in experiments with CO 2 injection 

<strong>the</strong> dross structure had a lot <strong>of</strong> <strong>layer</strong>s each 

with metal film. 

Before <strong>the</strong> dross is remelted under salt 

some samples are taken to analyze parts <strong>of</strong> 

<strong>the</strong> dross. The sampling is accomplished by 

optical criteria. XRD analyses have proven 

nei<strong>the</strong>r carbide nor oxycarbide phase while 

EDX analyses have shown <strong>the</strong> appropriate 

concentrations <strong>of</strong> Al, O and C, so that <strong>the</strong>re is 


Fig. 7: Net-remelting losses 

<strong>the</strong> possibility <strong>of</strong> an existing Aluminum-oxycarbide 

phase. Fig. 8 illustrates a typical SEM 

picture <strong>of</strong> one sample which is taken from <strong>the</strong> 

red marked area (see macroscopic photography 

on <strong>the</strong> upper right). 

The average element level <strong>of</strong> <strong>the</strong> two red 

marked areas consists <strong>of</strong> 17,10 At.-% C, 18,70 

At.-% O, 8,63 At.-% Mg and 55,58 At.-% Al 

(Fig. 8). The atomic components <strong>of</strong> aluminum 

and carbon are strongly represented. In contrast 

to <strong>the</strong> XRD analysis includes <strong>the</strong> EDX 

analysis <strong>the</strong> possibility <strong>of</strong> <strong>the</strong> presence <strong>of</strong> <strong>the</strong> 

carbide phase (Al 4 C 3 ) and <strong>the</strong> oxycarbid phase 

(Al 2 OC). 

Fig. 8: SEM picture <strong>of</strong> <strong>the</strong> analyzed area (Experiment HV6KM) 

5 Conclusions 


It should be underlined that all identified dependencies 

are related to <strong>the</strong> experimental 

scale <strong>of</strong> one litre <strong>of</strong> melt. The behaviour <strong>of</strong> 

<strong>the</strong> reaction dynamics in a multiple chamber 

furnace can only carried out partially based 

on <strong>the</strong> performed experiments. The interactions 

<strong>of</strong> pyrolysis product gases and <strong>the</strong> aluminum 

melt and <strong>the</strong>refore <strong>the</strong> produced dross 

amount can diversify significantly through 

<strong>the</strong> increased scale and should <strong>the</strong>refore be 

tested in scale up trials. 

The definition <strong>of</strong> <strong>the</strong> direct reaction prod- 

ucts (mechanism) in accordance with <strong>the</strong> 

structural analysis is difficult, because on <strong>the</strong> 

one hand, a subsequent oxidation <strong>of</strong> possibly 

formed carbides and oxycarbides seems to 

be likely, and on <strong>the</strong> o<strong>the</strong>r hand, due to <strong>the</strong> 

high aluminum content in <strong>the</strong> taken samples 

<strong>the</strong> detection limit for minor components is 

quickly reached. Since carbide or oxycarbides 

in <strong>the</strong> considered region were not detected, it 

is unclear whe<strong>the</strong>r <strong>the</strong> result from FactSage 

issued for equilibrium phases are correct at 

all or if <strong>the</strong> phases are not present because 

<strong>of</strong> <strong>the</strong>ir post-oxidation under normal atmosphere. 

Since <strong>the</strong> <strong>the</strong>rmochemical modelling 

<strong>of</strong> <strong>the</strong> melting experiments refers to <strong>the</strong> reaction 

equilibrium also a kinetic inhibition cannot 

be excluded. 

The remelting losses decrease with decreasing 

oxygen content <strong>of</strong> <strong>the</strong> treatment gases, in 

<strong>the</strong> order from CO 2 , CO, CH 4 . By varying <strong>the</strong> 

feed material it has to be noted that <strong>the</strong> influence 

<strong>of</strong> each gas increases with increasing 

magnesium content <strong>of</strong> <strong>the</strong> melt. This trend is 

confirmed by <strong>the</strong> net metal losses, additionally 

approved by <strong>the</strong> online gas measurement. 


The authors would like to thank <strong>the</strong> Hydro 

Aluminium Rolled Products GmbH for funding 

<strong>the</strong> studies, providing scrap materials for 

trails and also for <strong>the</strong>ir analytical support. 

References 

[1] Jaroni, B., Lucht, A., et al.: Conditions <strong>of</strong> Pyrolythic 

Processes in Multi Chamber Furnaces for 

Aluminium Recycling, Proceedings <strong>of</strong> European 

Metallurgical Conference EMC 2011 

[2] Gnotke, O.: Experimentelle und <strong>the</strong>oretische 

Untersuchungen zur Bestimmung von veränderlichen 

Blasengrößen und Blasengrößenverteilungen 

in turbulenten Gas-Flüssgkeitsströmungen, 

Dissertation, Fachbereich Maschinenbau, TU Darmstadt, 

2004 

[3] von Röpenack, I.: Minimierung von Chlorgasemissionen 

bei der Spülgasraffination von Aluminiumschmelze, 

Dissertation, IME Aachen – Institut für 

Metallhüttenkunde und Elektrometallurgie, 1997 

[4]Antrekowitsch, H. et al.: Spülgastechnik in der 

Kupferindustrie, in Berg- und Hüttenmännische 

Monatshefte, 149. Jahrgang, Heft Nr. 3, S.186-192, 

2004 

� 

Latest News 

www.alu-web.de 



Optimizing <strong>the</strong> energy consumption <strong>of</strong> coil annealing 

furnaces by ma<strong>the</strong>matical modelling <strong>of</strong> <strong>the</strong> annealing process 

Günter Valder and Bernd Deimann, Otto Junker GmbH 

Multi-coil annealing furnaces <strong>of</strong> Otto Junker design 

The coils <strong>of</strong> strip made in <strong>aluminium</strong> mills by 

hot and cold rolling are subjected to a variety 

<strong>of</strong> intermediate and finish annealing treat- 

ments for metallurgical reasons. It is common 

practice that coils <strong>of</strong> different alloy composition, 

geometry and residual heat content 

are placed in shop floor buffer storage and 

allowed to cool down before undergoing <strong>the</strong> 

actual heat treatment in so-called chamber 

furnaces. 

Chamber furnaces may be <strong>of</strong> single or 

multi-coil design. A major benefit <strong>of</strong> single-coil 

units, as <strong>the</strong> name implies, is that <strong>the</strong>y allow 

<strong>the</strong> operator to anneal each coil individually. 

As a result, a change <strong>of</strong> alloy, product geometry 

or start temperature is easily possible between 

batches. 

Drawbacks <strong>of</strong> this equipment type include 

<strong>the</strong> large space requirement, relatively high 

investment cost and somewhat higher specific 

energy needs, all <strong>of</strong> which must be taken into 

account. In view <strong>of</strong> <strong>the</strong>se factors, multi-coil 

furnaces still have <strong>the</strong>ir justification. Let’s 

briefly review <strong>the</strong> status <strong>of</strong> some recent Otto 

Junker developments here, which are expected 

to implement <strong>the</strong> aforementioned advantages 

<strong>of</strong> <strong>the</strong> single-coil furnace in <strong>the</strong> multi-coil unit 

to a very broad extent. 

One requirement in multi-coil furnace operation 

is that <strong>the</strong> coils placed in <strong>the</strong> furnace 

should have <strong>the</strong> same geometry and start 

temperature so that <strong>the</strong> desired metallurgical 

properties can be obtained in a reproducible 

manner by <strong>the</strong> recipe-based heat treatment. 

Major deviations <strong>of</strong> <strong>the</strong>se two parameters 

can be allowed if <strong>the</strong> coil temperatures are 

measured throughout <strong>the</strong> heat treatment cycle 

and <strong>the</strong>se temperature readings are fed to <strong>the</strong> 

furnace control system. For control purposes 

<strong>the</strong> multi-coil furnace is divided into several 

individually controlled 

heating 

zones with a view 

to heating each coil 

separately by <strong>the</strong> 

jet heating principle 

using specially 

designed nozzle 

fields. 

Coil temperature 

measurements 

can be taken via 

contact <strong>the</strong>rmocouples 

or by 

means <strong>of</strong> shea<strong>the</strong>d 

<strong>the</strong>rmocouples em- 

bedded in <strong>the</strong> coil. 

A disadvantage <strong>of</strong> 

<strong>the</strong>se methods is 

that <strong>the</strong> strip sur- 

face quality may become locally impaired, 

e. g., by imprint marks. Moreover, it is necessary 

to allow set-up time for embedding <strong>the</strong> 

shea<strong>the</strong>d <strong>the</strong>rmocouples into <strong>the</strong> coil, and <strong>the</strong> 

maintenance cost for <strong>the</strong> contact <strong>the</strong>rmocouples 

needs to be considered as well. 

In <strong>the</strong> past few years, Otto Junker has been 

able to accumulate initial experience with 

ma<strong>the</strong>matical modelling <strong>of</strong> heat treatment 

processes on copper strip treatment lines and 

copper billet heaters. It was confirmed that 

integrated ma<strong>the</strong>matical models can enhance 

<strong>the</strong> reliability <strong>of</strong> <strong>the</strong> process while also helping 

to save energy [1]. 

The objective defined for <strong>the</strong> heat treatment 

<strong>of</strong> strip coils in multi-coil furnaces was 

that <strong>the</strong> temperature distribution <strong>of</strong> each coil 

should be calculated in real time, and that <strong>the</strong> 

result <strong>of</strong> that calculation should be used as furnace 

control input. For metallurgical reasons, 

a calculation accuracy <strong>of</strong> better than ± 5 K must 

be repeatably achieved in this application. 

As a general rule, ma<strong>the</strong>matical models are 

intended to describe a system’s response to a 

change in exterior relations. 

One critically important task in <strong>the</strong> present 

context lies in computing <strong>the</strong> energy transfer 

from <strong>the</strong> fluid to <strong>the</strong> coil – a step which calls 

for an accurate understanding <strong>of</strong> <strong>the</strong> underlying 

energy transfer mechanisms. In <strong>the</strong> case 

<strong>of</strong> <strong>aluminium</strong> <strong>the</strong> heating process is clearly 

dominated by <strong>the</strong> forced convective portion, 

although for <strong>the</strong> accuracy required, <strong>the</strong> radi- 

Diagram 1: Temperature curves obtained for a multi-coil heat treatment furnace. 

The diagram shows <strong>the</strong> supply air and return air temperatures (GasInlet, GasOutlet), 

<strong>the</strong> measured coil temperatures (MeasCoil), and <strong>the</strong> difference between 

calculated and measured temperature values (DevCalc Coil3) plotted by way <strong>of</strong> 

example for Coil 3. 

70 ALUMINIUM · EAC Congress 2011 

Images: Otto Junker

ant heat transfer between <strong>the</strong> furnace and <strong>the</strong> 

coil must not be neglected, ei<strong>the</strong>r. To this end, 

basic research was undertaken in cooperation 

with <strong>the</strong> Technical University <strong>of</strong> Aachen 

(RWTH) to gain a physically correct grasp 

<strong>of</strong> key correlations and to ensure that <strong>the</strong>se 

would be duly represented in <strong>the</strong> ma<strong>the</strong>matical 

model. 

The aim in <strong>the</strong>se investigations was to 

minimize <strong>the</strong> number <strong>of</strong> parameters requiring 

adaptation to <strong>the</strong> actual situation during startup 

and in subsequent production operations, 

and to ensure that <strong>the</strong> programmed modules 

would be transferable to similar applications 

elsewhere. 

The temperature distribution over space 

and time within <strong>the</strong> coil can be described with 

sufficient accuracy by a partial differential 

equation, assuming consistent metal characteristics. 

Although in <strong>the</strong> case <strong>of</strong> coils it is necessary 

to note significant differences between 

<strong>the</strong> radial and axial <strong>the</strong>rmal conductance. 

The differential equation is solved by <strong>the</strong> 

finite differences (FD) method, dividing <strong>the</strong> 

coil into <strong>layer</strong>s and <strong>the</strong> time into intervals. For 

<strong>the</strong> FD process to become stable, <strong>the</strong> time 

resolution and spatial resolution must at all 

times be mutually matched on <strong>the</strong> basis <strong>of</strong> <strong>the</strong> 

coil‘s <strong>the</strong>rmal conductivity. 

Once <strong>the</strong> ma<strong>the</strong>matical model is calibrated 

and stabilized, <strong>the</strong> following benefits are to 

be anticipated: 

1. Given <strong>the</strong> ability to support greater geometrical 

and temperature differences between 

coils, <strong>the</strong> furnace can be filled to 100% capacity 

more frequently, i. e. furnace capacity 

utilization will be improved while <strong>the</strong> energy 

consumption is reduced. 

2. Since greater temperature differences between 

coils are allowed, it will be possible, 

ideally, to charge coils in <strong>the</strong>ir ‘as rolled’ temperature 

state. As a result, <strong>the</strong> mean temperature 

at <strong>the</strong> start <strong>of</strong> <strong>the</strong> heat treatment cycle 

will be clearly increased. The energy requirement 

will diminish accordingly. 

3. Unlike <strong>the</strong> physical temperature measuring 

methods mentioned earlier, a ma<strong>the</strong>matical 

model will need no set-up time or maintenance 

in this respect. 

The first results obtained with a ma<strong>the</strong>- 

Spouts and Stoppers Ceramic Foam Filters 

www.drache-gmbh.de · mail@drache-gmbh.de 


matical model developed along <strong>the</strong> principles 

described above are shown in diagram 1. The 

illustration shows <strong>the</strong> temperature pr<strong>of</strong>iles <strong>of</strong> 

three coils as measured during <strong>the</strong> heat-up 

phase. The coils were placed in <strong>the</strong> furnace 

with starting temperatures <strong>of</strong> 25 to 75°C, and 

were heated to <strong>the</strong> same target temperature 

<strong>of</strong> 320°C ± 3 K from <strong>the</strong>ir respective state. 

For testing purposes, <strong>the</strong> furnace was controlled 

using data from a ma<strong>the</strong>matical model 

which has been in use since mid-2010. In <strong>the</strong> 

case <strong>of</strong> Coil 3, <strong>the</strong> difference between <strong>the</strong> 

measured temperatures (MeasCoil 3) and <strong>the</strong> 

calculated ones are not greater than -1 K and 

+ 8 K, respectively, at any time (DevCalcCoil 

3). During <strong>the</strong> last third <strong>of</strong> <strong>the</strong> heat-up cycle, 

this difference was even better than -1 K and 

+ 2 K throughout. 

The rise in mean coil temperature prevailing 

at <strong>the</strong> start <strong>of</strong> <strong>the</strong> heat treatment will save 

energy while also reducing CO 2 emissions. 

Diagram 2 illustrates <strong>the</strong> CO 2 savings achievable 

at different annual annealing capacities 

by increasing <strong>the</strong> mean coil temperature at <strong>the</strong> 

start <strong>of</strong> <strong>the</strong> heat treatment cycle. The compar- 

For Aluminium DC Casting 

ALUMINIUM · EAC Congress 2011 71


Diagram 2: CO 2 savings as a function <strong>of</strong> <strong>the</strong> mean coil temperature at <strong>the</strong> start <strong>of</strong> heat treatment 

ison was made between two state-<strong>of</strong>-<strong>the</strong>-art 

annealing furnaces <strong>of</strong> <strong>the</strong> multi-coil type. The 

reduction in CO 2 output has been calculated 

on <strong>the</strong> basis <strong>of</strong> specific CO 2 emission figures 

for <strong>the</strong> current mix <strong>of</strong> electricity sources in 

Germany. Around two-thirds <strong>of</strong> this reduction 

are attributable to fuel savings (assuming 

that <strong>the</strong> furnace is running on <strong>natural</strong> gas), <strong>the</strong> 

remaining one-third reflects electrical power 

savings (assuming that <strong>the</strong> increased coil temperature 

at <strong>the</strong> start <strong>of</strong> <strong>the</strong> heat treatment will 

shorten <strong>the</strong> heat-up cycle and hence, reduce 

fan operating times). 

In mid-2011, Otto Junker supplied five 

multi-coil furnaces – along with <strong>the</strong> charging 

machine and a complete automation 

system – to Aluminium Norf GmbH. At <strong>the</strong> 

automation level, a computer cluster linking 

Numerical analysis <strong>of</strong> <strong>the</strong> seam welds <strong>of</strong> industrial extrusion pr<strong>of</strong>iles 

T. Kloppenborg 1 , M. Schwane 1 , M. Fiderer 2 , A. Reeb 3 , 

N. Ben Khalifa 1 , K. A. Weidenmann 3 , A. Brosius 1 , A. Erman Tekkaya 1 

1 Institute <strong>of</strong> Forming Technology and Lightweight Construction, Technische Universität Dortmund, 

Germany; 2 Kistler-IGeL GmbH, Schönaich, Germany; 3 Institute for Applied Materials – Materials 

Science and Engineering (IAM–WK), Karlsruher Institute <strong>of</strong> Technology (KIT), Karlsruhe, Germany 

The increasing number <strong>of</strong> regulatory requirements 

as well as <strong>the</strong> rising expectations in comfort 

result in a continuous weight increase <strong>of</strong> 

automobiles. The weight <strong>of</strong> <strong>the</strong> vehicles influences 

<strong>the</strong> fuel consumption and increases air 

<strong>the</strong> PLC, HMI and modelling computers is 

created and tied in with <strong>the</strong> manufacturing 

requirements planning (MRP) system. An <strong>of</strong>fline 

version can be used to load data <strong>of</strong> coils 

currently in shop floor buffer storage; <strong>the</strong>se 

coils can <strong>the</strong>n be arranged into energy-optimized 

or time-optimized furnace batches 

comprising four coils in each case. The batches 

determined by means <strong>of</strong> <strong>the</strong> optimizing calculations 

are <strong>the</strong>n returned to <strong>the</strong> MRP system 

and cleared for heat treatment. The heat 

treatment process is <strong>the</strong>n controlled with <strong>the</strong> 

online version, which relies on <strong>the</strong> same computing 

core. 

This innovation project is subsidized by <strong>the</strong> 

Federal Ministry <strong>of</strong> Environment because <strong>the</strong> 

integration <strong>of</strong> a ma<strong>the</strong>matical model is expected 

to yield a significant reduction in energy 

pollution. The automobile industry has pursued 

<strong>the</strong> strategy to reduce <strong>the</strong> weight <strong>of</strong> <strong>the</strong> 

vehicles by an intelligent car body for a long 

time. Therefore, an increasing number <strong>of</strong> extrusion 

pr<strong>of</strong>iles made <strong>of</strong> aluminum have been 

consumption and hence, CO 2 emissions. Since 

<strong>the</strong> comparison was carried out against ‘earlier 

generation’ multi-coil furnaces, <strong>the</strong> anticipated 

savings <strong>of</strong> approx. 8,300 tonnes <strong>of</strong> CO 2 at 

an annual production output <strong>of</strong> 180,000 

tonnes <strong>of</strong> <strong>aluminium</strong> still exceed <strong>the</strong> results illustrated 

in diagram 2 [3]. The project is slated 

to be completed by <strong>the</strong> end <strong>of</strong> 2011, so that 

<strong>the</strong> first reports on operating experience are to 

be anticipated for 2012. 

References 

[1] Cüppers, J.: Betriebserfahrungen mit verschiedenen 

prozeßrechnerunterstützten Erwärmungsstrategien 

an Hubbalkenöfen in der Buntmetallindustrie 

[Operating experience gained with diverse process 

computer-assisted heating strategies on walking 

beam furnaces in <strong>the</strong> non-ferrous metals industry], 

Blech Rohre Pr<strong>of</strong>ile 28, 1991, pp. 882-890 

[2] Bölling, R.: Allgemeine Systemtechnik – Einführung 

in die Finite-Volumen-Methode, Lösung strömungs- 

und wärmetechnischer Probleme des Industrie<strong>of</strong>enbaus 

mittels numerischer Methoden [General 

systems engineering – Introduction to <strong>the</strong> finite 

volume method, solution <strong>of</strong> fluid and <strong>the</strong>rmal engineering 

problems in industrial furnace construction 

using numerical methods], Lecture Notes, Technical 

University <strong>of</strong> Aachen (RWTH), Winter Term 2010 

[3] Federal Ministry <strong>of</strong> Environment, www.bmu.de. 

Press release dated February 8, 2011: Innovative 

Glühöfen für die Aluminium-Industrie [Innovative 

annealing furnaces for <strong>the</strong> <strong>aluminium</strong> industry] 

Authors 

Dr.Ing. Günter Valder, Technology Business Unit 

Manager, Thermoprocess Equipment Division, Otto 

Junker GmbH, Simmerath, Germany 

Dipl.Ing. Bernd Deimann, Senior Sales Manager, 

Thermoprocess Equipment Division, Otto Junker 

GmbH, Simmerath, Germany 

used as semi-finished products. The complexity 

<strong>of</strong> <strong>the</strong> pr<strong>of</strong>ile cross sections varies depending 

on <strong>the</strong> automobile manufacturer and on 

<strong>the</strong> number <strong>of</strong> produced vehicles. 

Currently, <strong>the</strong> estimation <strong>of</strong> <strong>the</strong> produc- 

72 ALUMINIUM · EAC Congress 2011

Fig. 1: Tool concept for <strong>the</strong> visioplastic analyses in a porthole die 

ibility <strong>of</strong> complex cross sections is based on 

expert’s knowledge and cost-intensive prototyping. 

A more efficient method to predict <strong>the</strong> 

producibility and to reduce <strong>the</strong> risk <strong>of</strong> production 

failure in advance is presently not available. 

Hence, during <strong>the</strong> conceptual designing 

<strong>of</strong> new automobiles not all possibilities can be 

considered. One factor is <strong>the</strong> position and <strong>the</strong> 

quality <strong>of</strong> <strong>the</strong> longitudinal seam welds. Insufficient 

seam welds can influence <strong>the</strong> mechanical 

properties <strong>of</strong> <strong>the</strong> pr<strong>of</strong>iles. 

In this paper, <strong>the</strong> recent results <strong>of</strong> a transfer 

project within <strong>the</strong> scope <strong>of</strong> <strong>the</strong> Transregional 

Collaborative Research Center / TR10, which 

is kindly supported by <strong>the</strong> German Research 

Foundation (DFG), are presented. In this 

project, extrusion companies, die makers, 

automobile companies, extrusion s<strong>of</strong>tware 

companies, and <strong>the</strong> Gesamtverband der Aluminiumindustrie 

e.V. work toge<strong>the</strong>r to evaluate 

how present simulation s<strong>of</strong>tware based 

on finite element methods is able to simulate 

complex industrial extrusion processes. 

For <strong>the</strong> validation <strong>of</strong> <strong>the</strong> numerical results 

in porthole die extrusion by visioplastic plastic 

analysis a new modular die concept <strong>of</strong> a porthole 

die was developed. This tooling concept 

allows preparing <strong>the</strong> material out <strong>of</strong> <strong>the</strong> die 

without post plastifications on <strong>the</strong> material 

surface. Especially <strong>the</strong> analysis <strong>of</strong> <strong>the</strong> tribological 

effects near <strong>the</strong> die walls can be done 

[1,2]. The die is used for <strong>the</strong> production <strong>of</strong> 

Fig. 2: Modeling <strong>of</strong> <strong>the</strong> process in Deform and HyperXrude 

a rectangular pr<strong>of</strong>ile with a cross section <strong>of</strong> 

40 mm x 10 mm. It consists <strong>of</strong> twelve single 

parts. The fundamental part is a divided conical 

part, which is necessary to extract <strong>the</strong> material 

out <strong>of</strong> <strong>the</strong> die. Additionally, <strong>the</strong> material 

which is prepared for <strong>the</strong> visioplastic analysis 

can be inserted into <strong>the</strong> die again (Fig. 1). 

Only a billet preparation would result in 

high deformations <strong>of</strong> <strong>the</strong> contrast material inside 

<strong>the</strong> feeders. In this case a determination 

<strong>of</strong> <strong>the</strong> material flow and <strong>the</strong> tribological effects 

on <strong>the</strong> die wall would be inaccurate. Due to 

this, to analyze <strong>the</strong> material flow inside <strong>the</strong> 

billet, <strong>the</strong> feeders and <strong>the</strong> welding chamber <strong>of</strong> 

<strong>the</strong> die were filled by extruding an EN AW- 

6063 alloy. The material was <strong>the</strong>n extracted 

and prepared with <strong>the</strong> grid pattern technique 

[1, 3]. In <strong>the</strong> vertical symmetry plane cylindrical 

pins with a diameter <strong>of</strong> 5 mm were used as 

contrast material. To visualize <strong>the</strong> horizontal 

and <strong>the</strong> vertical material flow in one experiment, 

<strong>the</strong> pins were inserted vertically and 

horizontally in <strong>the</strong> symmetrical upper and 

lower part <strong>of</strong> <strong>the</strong> die. An EN AW-4043A alloy 

was used as contrast material. The distance 

between <strong>the</strong> pins was 15 mm. But in <strong>the</strong> area <strong>of</strong> 

transition from <strong>the</strong> billet to <strong>the</strong> feeder as well 

as in <strong>the</strong> area <strong>of</strong> <strong>the</strong> welding chamber, different 

distances were necessary. After preparation, 

<strong>the</strong> material was reinserted in <strong>the</strong> modular die 

for <strong>the</strong> experiment. After <strong>the</strong> preparation <strong>of</strong> 

<strong>the</strong> billet <strong>the</strong> extrusion process was performed 


on a 10 MN extrusion press at <strong>the</strong> IUL to visualize 

<strong>the</strong> material flow inside <strong>the</strong> die. 

For <strong>the</strong> numerical analysis <strong>of</strong> <strong>the</strong> material 

flow, <strong>the</strong> finite element method was used. The 

simulation <strong>of</strong> <strong>the</strong> process was done with <strong>the</strong> 

commercial s<strong>of</strong>tware HyperXtrude and Deform3D. 

As initial data for <strong>the</strong> threedimensional 

finite element models <strong>the</strong> geometrical 

data <strong>of</strong> <strong>the</strong> experimental investigations was 

used. Due to <strong>the</strong> fact that each s<strong>of</strong>tware uses 

varying formulations a different modeling <strong>of</strong> 

<strong>the</strong> process becomes necessary. 

The implicit finite element code from Altair 

Engineering uses <strong>the</strong> Euler formulation 

where <strong>the</strong> material is meshed with a fixed 

mesh. Hence, <strong>the</strong> overall material flow in <strong>the</strong> 

process has to be modeled. It includes <strong>the</strong> 

70 mm long billet (diameter <strong>of</strong> 105 mm), <strong>the</strong> 

material inside <strong>the</strong> die, and <strong>the</strong> exiting pr<strong>of</strong>ile. 

Additionally, <strong>the</strong> container and <strong>the</strong> die were 

considered as rigid bodies to calculate <strong>the</strong> 

heat transfer. Only a symmetrical part was 

modeled to reduce <strong>the</strong> calculation time for <strong>the</strong> 

quasi stationary simulation. The inflow velocity 

was set to 10 mm/s. 

In Deform3D, <strong>the</strong> Lagrange formulation is 

used. Here, <strong>the</strong> finite element mesh is linked 

to <strong>the</strong> material flow. Particularly in <strong>the</strong> simulation 

<strong>of</strong> extrusion processes, where high deformation 

occurs, frequent remeshing is required 

in order to maintain a good mesh quality. In 

this model components were discretized with 

linear tetrahedral elements. The block with a 

length <strong>of</strong> 80 mm and <strong>the</strong> material in <strong>the</strong> filled 

die were modeled as one part, which has been 

assigned a homogeneous initial temperature <strong>of</strong> 

430°C. Both <strong>the</strong> container and <strong>the</strong> die were 

assumed to be rigid. They were also discretized 

in order to incorporate <strong>the</strong> heat exchange between 

material and tools during <strong>the</strong> process. 

The diameter <strong>of</strong> <strong>the</strong> block was chosen to be 

equal to <strong>the</strong> inner diameter <strong>of</strong> <strong>the</strong> container 

to increase <strong>the</strong> comparability to <strong>the</strong> HyperXtrude 

model. Due to this, <strong>the</strong> upsetting <strong>of</strong> <strong>the</strong> 

ALUMINIUM · EAC Congress 2011 73


Fig. 3: Visualization <strong>of</strong> flow lines from <strong>the</strong> experiment and <strong>the</strong> simulation 

block was also not considered in <strong>the</strong> Deform 

model. Fur<strong>the</strong>rmore, only a quarter model was 

setup and a constant punch speed <strong>of</strong> 10 mm/s 

was chosen. 

Flow lines were calculated based on <strong>the</strong> 

velocity field out <strong>of</strong> <strong>the</strong> HyperXtrude simulation. 

The integration time was set to <strong>the</strong> press 

time to make a qualitative validation with <strong>the</strong> 

experimental directly visible. Due to <strong>the</strong> results, 

a high strain distribution occurs in <strong>the</strong> 

area <strong>of</strong> <strong>the</strong> welding chamber which is shown 

by <strong>the</strong> plastification <strong>of</strong> <strong>the</strong> contrast material 

(Fig. 3). The calculated vertical flow is in accurate 

agreement with <strong>the</strong> visioplastic result. 

Additionally, it can be seen that <strong>the</strong> material 

sticks on to <strong>the</strong> die wall. This effect is also 

shown by <strong>the</strong> finite element model. It can be 

seen that <strong>the</strong> material shears near <strong>the</strong> die wall. 

Outside this area, <strong>the</strong> velocity is mainly constant 

in <strong>the</strong> feeders. Errors occur due to <strong>the</strong> fact 

that HyperXtrude is not able to calculate <strong>the</strong> 

billet upsetting <strong>of</strong> <strong>the</strong> block at <strong>the</strong> beginning <strong>of</strong> 

<strong>the</strong> process. The error can be established by <strong>the</strong> 

decreasing error in press direction in <strong>the</strong> billet 

as well as in <strong>the</strong> increasing error radial from 

<strong>the</strong> billet center to <strong>the</strong> container. 

Additionally <strong>the</strong> friction modeling in Deform 

was analyzed. The shear friction model 

according to Tresca with m = 1 was used in 

<strong>the</strong> calculations. Physically, this corresponds 

Fig. 4: Qualitative comparison <strong>of</strong> HyperXtrude and Deform3D results 

to sticking between material and tool surface, 

since <strong>the</strong> friction stress equals <strong>the</strong> shear yield 

stress <strong>of</strong> <strong>the</strong> material. In order to evaluate <strong>the</strong> 

impact <strong>of</strong> <strong>the</strong> friction boundary conditions on 

<strong>the</strong> material flow and <strong>the</strong> flow grid pattern, 

respectively, friction was also modeled using 

<strong>the</strong> sticking condition solely as well as with <strong>the</strong> 

sticking condition in conjunction with m = 1. 

The movement <strong>of</strong> nodes on a contact surface 

is prevented when <strong>the</strong> sticking condition is applied. 

The results revealed that <strong>the</strong> flow lines 

partially debond in case <strong>of</strong> <strong>the</strong> shear friction 

model. Fur<strong>the</strong>rmore, <strong>the</strong> material slides along 

<strong>the</strong> die wall in <strong>the</strong> feeder, which is also clearly 

shown by <strong>the</strong> depicted velocity pr<strong>of</strong>ile. These 

effects do not match <strong>the</strong> experimental results. 

Here, full sticking between material and die 

can be observed, so that shearing in <strong>the</strong> subsurface 

<strong>layer</strong>s occurs. This flow characteristic 

is qualitatively achieved by an activation <strong>of</strong> 

<strong>the</strong> sticking option. The difference between 

sticking and sticking in conjunction with shear 

friction turned out to be insignificant. Due to 

<strong>the</strong> presented results, <strong>the</strong> model incorporating 

volume compensation as well as sticking condition 

was determined to give best results. 

The Deform results were validated using 

<strong>the</strong> HyperXtrude results, which already had 

turned out to be reliable. Hence, a flow net 

with equally spaced lines was computed with 

both Deform3D and 

HyperXtrude. It can be 

shown that <strong>the</strong> results 

computed by HyperXtrude 

and Deform3D 

are in good accordance 

in terms <strong>of</strong> flow 

line pattern as well as 

flow velocity prediction 

(Fig. 4). 

The hints regarding an 

appropriate modeling <strong>of</strong> 

<strong>the</strong> material flow in port- 

hole dies, which where gained by <strong>the</strong> visioplastic 

analysis will also be incorporated into <strong>the</strong> 

DEFORM models <strong>of</strong> <strong>the</strong> industrial extrusion 

tests, which were conducted by <strong>the</strong> collaborating 

companies <strong>of</strong> this project. First results <strong>of</strong> 

<strong>the</strong> numerical process simulation are already 

in good correlation with <strong>the</strong> experimental results. 

In particular <strong>the</strong> simulation <strong>of</strong> <strong>the</strong> pr<strong>of</strong>ile 

exhibiting different wall thicknesses shows 

some significant characteristics. The different 

wall thicknesses <strong>of</strong> <strong>the</strong> pr<strong>of</strong>ile generate an inhomogeneous 

velocity distribution at <strong>the</strong> die 

orifice which, in turn, causes a curved shape 

<strong>of</strong> <strong>the</strong> exiting pr<strong>of</strong>ile. A higher flow velocity 

is predicted at <strong>the</strong> thicker walls. Fur<strong>the</strong>rmore, 

<strong>the</strong> extruded pr<strong>of</strong>iles showed slight underfillings 

at <strong>the</strong> outer face. The process simulation 

is also capable <strong>of</strong> representing <strong>the</strong>se failures. 

It should be mentioned here that <strong>the</strong> numerical 

analysis <strong>of</strong> this process still requires some 

effort since <strong>the</strong> material flow is not symmetric. 

Therefore, <strong>the</strong> material and <strong>the</strong> mesh, respectively, 

has to be merged manually at <strong>the</strong> welding 

lines when Deform s<strong>of</strong>tware is used. 


This paper is based on investigations <strong>of</strong> <strong>the</strong> 

subprojects B1 – ‘Integral design, simulation 

and optimization <strong>of</strong> extrusion dies’ and T6 – 

‘Efficient industrial extrusion simulation’ <strong>of</strong> 

<strong>the</strong> Transregional Collaborative Research 

Center / Transregio 10, which is kindly supported 

by <strong>the</strong> German Research Foundation 

(DFG). 

Additionally, <strong>the</strong> support <strong>of</strong> our industrial 

partners Audi AG, Altair Engineering GmbH, 

F.W. Brökelmann Aluminiumwerk GmbH & 

Co. KG, Daimler AG, Gesamtverband der 

Aluminiumindustrie e.V., Honsel AG, Kistler- 

IGel GmbH, and S+C ETS GmbH is greatly 

acknowledged. Particularly, we want to thank 

Wilke Werkzeugbau GmbH for <strong>the</strong> die manufacture. 

We also want to thank <strong>the</strong> Institute for Innovative 

Mechanics and Management <strong>of</strong> <strong>the</strong> 

University <strong>of</strong> Padova in Italy for <strong>the</strong> experimental 

tests for <strong>the</strong> flow curve evaluation. 

Literature 

[1] L. Donati, L. Tomesani, M. Schikkora, N. Ben 

Khalifa and A.E. Tekkaya: Friction model selection 

in FEM simulations <strong>of</strong> aluminum extrusion, Int. j. 

Surface Science and Engineering, Vol. 4, No. 1 (2010) 

[2] H. S. Valberg: Applied metal forming: including 

FEM analysis, Cambridge University Press (2010) 

[3] H. S. Valberg: Experimental techniques to characterize 

large plastic deformations in unlubricated 

hot aluminum extrusion, Key Engineering Materials 

Vol. 367, pp. 17-24 (2008) 

74 ALUMINIUM · EAC Congress 2011

EFFICIENT 

Yes, I/we would 

like to subscribe to 

ALUMINIUM+AluminiumePaper 

straight away for 

<strong>the</strong> period specified below. 

ALUMINIUM appears 10 x a year; 

I/we will be notified by e-mail as soon as 

<strong>the</strong> current ePaper edition is available online 

for downloading. 

� 1-year subscription: 

in Europe EUR 289.00/ overseas USD 382.00 

� 2-year subscription: 


� 3-year subscription 


Hans-Böckler-Allee 9 

30173 Hannover · Germany 

Tel. +49 511 8550-2638 

Fax +49 511 7304-233 

www.giesel.de 

vertrieb@giesel.de 

Company 

First name, last name 

Department 

Address 

Postcode, city 

Country 

Telephone Fax 

E-mail (for <strong>the</strong> notification) 

VAT ID number 

The subscription fee is to be paid by credit card: 

Number 

VISA American Express Euro-/Mastercard 

Cardholder 

Valid until 

The fee is to be debited from our account (only possible within 

Germany). 

Bank 

Bank code 

Account no. 

www. 

alu-web.de 

Knowledge makes 

an impact! 

ALUMINIUM now 

provides you with 

effective information 

in no less than 

three ways! 

What’s more, if you take out your subscription straight away 

for 2 or 3 years, you’ll also be saving up to 25 %! 

The fee will be remitted on receipt <strong>of</strong> <strong>the</strong> invoice. 

Date Signed 

Anz.Efficient.BWH

800 mm 

150 MN 

Smoothly supplying, erecting and commissioning 

complex plants is our business. 

But as a fl exible partner we can also 

satisfy unusual wishes. For our customer 

Shandong Yankuang Light Alloy Company 

in China, we erected one <strong>of</strong> <strong>the</strong> world’s 

most powerful <strong>aluminium</strong> extrusion presses 

in our works because <strong>the</strong>y wanted to 

see for <strong>the</strong>mselves <strong>the</strong> capabilities <strong>of</strong> <strong>the</strong> 

gigantic machine in advance. That is what 

we mean by customer orientation. 

1 

BILLET 

DIAMETER 

PRESS 

FORCE 

Competence and experience, reliability and 

fl exibility – <strong>the</strong> hallmarks <strong>of</strong> SMS Meer. 

We give our customers worldwide a 

lead over <strong>the</strong> competition in <strong>the</strong>ir markets. 

SMS Meer <strong>of</strong>fers high-performance machines 

and plants for steelmaking and 

continuous casting, tubes, long products, 

forging, NF metals and heat treatment. 

Our solutions – a good reason to be proud! 

SATISFIED 

CUSTOMER 

www.sms-meer.com

Influence of the natural aluminium oxide layer on ... - ALU-WEB.DE

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?