Download PDF - Piano Technicians Guild

An International Non-Profit Organization of Registered Piano Technicians 

Taylor Mackinnon, RPT 

President 

520 SE 29th Ave., Hillsboro, OR 97123 

(503) 846-1501 

E-mail: pres@ptg.org 

Richard Bittner, RPT 

Vice President 

519 Melody Ct., Royal Oak, MI 48073 

(248) 398-8721 

E-mail: vp@ptg.org 

Paul Monroe, RPT 

Secretary-Treasurer 

5200 Irvine Blvd., Sp. 310, Irvine, CA 92620 

(714) 730-3469 

E-mail: sec@ptg.org 

David P. Durben, RPT 

Immediate Past President 

2310 E. Romneya Dr., Anaheim, CA 92806 

(714) 491-7392 

E-mail: ipp@ptg.org 

Ruth B. Phillips, RPT 

Northeast Regional Vice President 

3096 Bristol Rd., Warrington, PA 18976 

(215) 491-3045 

E-mail: nervp@ptg.org 

Robert L. Mishkin, RPT 

Southeast Regional Vice President 

1240 NE 153rd St., N. Miami Beach, FL 33162 

(305) 947-9030 

E-mail: servp@ptg.org 

Jack R. Wyatt Sr., RPT 

South Central Regional Vice President 

2027 15th St., Garland, TX 75041 

(972) 276-2243 (H) 

(972) 278-9312 (W) 

E-mail: scrvp@ptg.org 

Robert S. Bussell, RPT 

Central East Regional Vice President 

224 West Banta Rd., Indianapolis, IN 46217 

(317) 782-4320 

E-mail: cervp@ptg.org 

Richard E. West, RPT 

Central West Regional Vice President 

1427 A St., Lincoln, NE 68502 

(913) 631-8227 

E-mail: cwrvp@ptg.org 

Larry Joe Messerly, RPT 

Western Regional Vice President 

2222 W. Montebello Ave., Phoenix, AZ 85015 

(602) 433-9386 

E-mail: wrvp@ptg.org 

Keith Eugene Kopp, RPT 

Pacific NW Regional Vice President 

61283 Killowan Lane, Bend, OR 97702 

(541) 388-3741 

E-mail: pnwrvp@ptg.org 

2 Piano Technicians Journal / November 2000 

Plates in Focus 

Steve Brady, RPT 

Journal Editor 

You may notice a special emphasis on piano plates in this issue; it’s no 

accident. Some time ago the PTG Board of Directors approached me with a 

special request: a “theme” issue dealing with plates. We had hoped to cover 

plates from all angles, from raw materials, to design and manufacture, to 

breakage and repair techniques. Of greatest concern, however, was the matter 

of liability. When a plate breaks, whose fault is it? 

While an article on plate manufacture was not forthcoming, we’ve 

covered most of the other bases. Don Galt’s excellent article gives as much 

detail about the raw materials – not just for plates but for many other iron 

and steel piano parts as well – as most of us will ever need. Although Jim 

Ellis analyzes a specific plate that broke because of poor design, he also 

provides numerous insights into what constitutes a good plate design. From 

Wilford Young comes a welding repair method that has proven effective over 

many, many years and retired engineer Richard Oliver Snelson describes a 

hybrid restoration combining a mechanical repair with welded 

reinforcements. Finally, we’ve addressed liability by relating a tale that 

desperately needs to be told. As Marnie Squire’s story unfolds we see clearly 

that the technical authorities are unanimous: tuning a piano cannot possibly 

break a healthy plate. 

Indeed, a healthy plate can withstand unspeakable acts. In “The Tuner’s 

Life,” Carman Gentile tells of one such experience. I can recall many other 

instances myself. For example, a panicked piano owner called me after she 

discovered that her 11-year-old son had industriously removed all the plate 

rim lags from her small Chickering grand. The piano was horrendously out 

of tune. After replacing the lag screws, I found that the tuning had improved 

somewhat and was able to perform an uneventful tuning. 

In my shop, a concert-grand plate fell some four inches to a concrete 

floor when one of the hoist cables slipped. The plate didn’t break and, 

although this happened nearly 20 years ago, the piano is still humming along 

happily. 

The late Don Galt once related to me that he had seen a six-foot grand 

tumble end-over-end down a flight of 30 steps when a rope the movers were 

using suddenly broke as the piano approached the landing. Broken plate? No. 

“In fact,” Don said, “it was still pretty much in tune when we set it up 

afterward.” And that piano is still in service more than 25 years later.

Iron, Steel & Pianos 

By Don Galt, RPT 

(Reprinted from the Piano Technicians Journal, April, 1970.) 

Editor’s Introduction 

Don Galt served as Technical Editor of this publication from 1969 until 1977. One of the many special 

things that Don brought to his work as a piano technician and his work with the Journal was his extensive 

knowledge of iron and steel, gained over many years in his previous life as a re-bar engineer at 

Bethlehem Steel. I am reprinting this article in its entirety, and in a sidebar I’ve included a brief question 

and answer in which Don replies to a reader’s query on responsibility for plates broken while tuning. 

This item appeared in the November, 1970 issue of PTJ, p.10. – SB 

There is little in the appearance of a piano to 

reveal the massive forces that its members exert 

on one another, without respite, through many 

decades of time. It is only when the instrument 

absorbs the energy of the musician, translates it and throws it 

back as sound energy at small or great dynamic levels that 

the magnitude of these forces is hinted at. 

Even so, the dynamism of the pianist and the stolidity of 

the piano foster the illusion that the former, rather than the 

latter, is actually the source of the sound. 

So perhaps it is natural that piano users are ignorant of, 

and even piano technicians sometimes take for granted, the 

highly stressed metallic members to which the modern 

piano largely owes its dynamic compass. 

To gain a little sympathetic understanding of these 

members, this paper attempts a short description of the 

important iron products used in piano building: their 

manufacture, their physical properties, and their reactions to 

the loads they are asked to carry in the piano. 

For the reader’s convenience the article is divided into 

the following sections: 

General Considerations 

Pig Iron — The Blast Furnace 

Gray Cast Iron — The Cupola 

Steel — Making the Material 

Steel — The Rolling Mill 

Steel — Wire Drawing 

The Iron - Iron Carbide System 

Testing & Properties 

The first five sections of the article, down through 

“Steel — The Rolling Mill,” are fairly general, but with 

occasional references to our special interests. 

The section on “Steel – Wire Drawing” gives a very 

brief description of this procedure, leaving out more than it 

tells (as does every part of the article). 

The last two sections entitled “The Iron — Iron 

Carbide System” and “Testing & Properties,” will probably 

make the greatest demands on the reader’s attention. The 

author hopes that this attention will be rewarded with an 

enlarged understanding of how these materials do their 

work in the piano. If any reader gives up and jumps off 

during “The Iron-Iron Carbide System,” I hope he or she 

will climb back aboard for “Testing & Properties.” 

General Considerations 

Pure elemental metallic iron is a rare thing outside of the 

laboratory and we never encounter it in pianos. The steel 

music wire and the gray cast-iron plates of pianos, as well as 

other familiar iron products such as structural steel, wrought 

iron, white cast iron and so on, are mixtures of iron and 

carbon, iron being by far the predominant ingredient. They 

are not chemical compounds, so their proportions and 

hence their properties can and do vary widely. If the carbon 

content is less than about two percent by weight, the 

material is called steel. If the carbon is more than two 

percent it is called cast iron. This two percent figure is not 

arbitrary, but we will not explore its significance at the 

moment. These terms illustrate the sort of paradox that can 

grow up on a subject when the usage develops gradually. 

Steel is defined as a mixture or alloy of iron and carbon, and 

yet what is called cast iron contains more carbon than steel 

does. 

In practice, most steels have well less than two percent 

carbon, and most cast irons have well more than that 

amount. Gray cast iron, as used in piano plates, contains 

about 3.5 percent carbon, structural steel about 0.25 

percent, tool steels usually one percent or more, piano wire 

about 0.90 percent. The properties of the material depend a 

great deal on the percentage and form of the carbon 

present. 

Continued on Next Page 

November 2000 / Piano Technicians Journal 17


Continued from Previous Page 

All steels and cast irons also contain other elements and 

materials. Some of these, such as sulphur and phosphorus, 

are residual impurities which generally have been reduced 

in manufacture to the economic minimum. Others, such as 

silicon and manganese, may be purposely left in in controlled 

amounts, or be purposely added, to give the material 

special qualities. Examples are gray cast iron, containing 

considerable silicon, and the many alloy steels containing 

chromium, nickel, molybdenum and so on. 

The steel category includes a large spectrum of materials, 

classified by carbon content, as well as by the percentage 

ranges of other alloying elements. Music wire is generally 

made from carbon steel, as distinguished from alloy steel, 

which means that no deliberate alloy additions are used. 

(Except manganese. Almost all steels, either carbon or alloy, 

contain appreciable amounts of manganese.) 

Apart from the chemical distinction between steel and 

cast iron, one of the most important differences is that the 

various steels are generally ductile and malleable in varying 

degrees, while cast iron generally is not. (The amenability of 

a material to plastic deformation under stress without 

fracture is called ductility or malleability, according as the 

stress is tensile or compressive.) 

Wire could not be made from cast iron because the 

manufacturing process and most wire usages demand a 

ductile material. On the other hand, piano plates could be 

made of steel by forging, casting or welding, but among 

other disadvantages they would be costly far beyond any 

strength superiority they would have over plates of gray cast 

iron. 

We will return to some of the other properties and 

reactions of these materials after a short excursion into iron 

and steel making. 

The manufacture of steel divides rather naturally into 

two stages: 

1) making the material and 2) making the product from 

the material (product meaning bars, structural shapes, sheets, 

wire, etc.). With gray cast iron on the other hand, the 

material is generally turned out in product form, as we shall 

see. 

Pig Iron — The Blast Furnace 

The first step for either steel or cast iron is to recover iron in 

usable form from iron ore, which is iron oxide (rust) in 

varying mixtures with earth, sand and rock. This recovery is 

mostly a process of getting rid of the oxygen by heating the 

ore in the presence of carbon and limestone. This takes place 

in a blast furnace, which is a shaft, typically 25 feet or more 

in diameter by 75 feet or more in height, charged with 


layers of coke, ore and limestone, which form a descending 

column. Air is forced in at the bottom, and the coke burns 

partially to carbon monoxide, which in turn reduces the 

iron oxide ore. The limestone forms a molten, fluid slag, 

which, floating on the molten iron, accumulates and carries 

off much of the waste matter. The blast furnace operates 

continuously, with materials charged at the top and the slag 

and molten pig iron drawn off at the bottom. This is a hot 

process, with a temperature gradient in the furnace from a 

few hundred degrees at the top to about 2,750 degrees F. at 

the bottom. 

Pig iron, the blast furnace product, contains fairly large 

percentages (totaling seven percent or more) of impurities 

such as silicon, manganese, sulphur, phosphorus and an 

excess of carbon. “Impurities” is a relative term, as some of 

these inclusions are impurities only as they are in excess for 

the purpose at hand. 

Gray Cast Iron — The Cupola 

If the end product is to be gray cast iron, the pig iron from 

the blast furnace is refined in an oxidizing furnace known as 

a cupola. In foundry practice the cupola charge usually 

includes cast iron and steel scrap and ferro-silicon, as well as 

the pig iron. Coke for fuel and limestone for flux are also 

included. Because little or no chemical correction is possible 

in the cupola after melting, the charge must be carefully 

planned as to proportions of entering materials, based on 

the constitution of these materials and the desired constitution 

of the product. A typical melt for piano plates might 

contain 3.5 percent carbon and 2.4 percent silicon, about 

which more will be said later. 

The molten “cast iron” is drawn off and poured into 

molds, usually of sand, in which it takes the shapes of the 

patterns used in preparing the molds — piano plates, for 

example. 

Steel — Making the Material 

If the end material is to be steel, the pig iron from the blast 

furnace is refined in one of various types of oxidizing 

furnace permitting closer control than the cupola of the 

iron foundry. The reader will have heard of the Bessemer 

converter and the open-hearth furnace, both long used in 

steel making. The electric furnace, once limited to special 

steel manufacture, is now used extensively in the production 

of more common grades. Steel music wire may be made of 

either open hearth or electric furnace steel, never Bessemer. 

(The very fast Bessemer process is not deliberate enough to 

allow the analysis and chemical corrections necessary to the 

careful manufacture of high carbon steel.)

After various tests show that the desired constitution has 

been achieved in the furnace, the molten steel is poured 

into molds, where it cools and solidifies into ingots, oblong 

in form and varying in size from a ton or less to many tons. 

When the steel freezes in the ingot mold it has been in the 

molten state for the last time and the steel making might be 

said to be complete. The solid state processing which follows 

does not change the constitution of the material, that is, the 

proportions of its elemental ingredients. 

However, the various forming procedures and heat 

treatments do influence greatly the grain structure of the 

steel, and hence its physical properties such as strength and 

hardness. The special character of music wire depends as 

surely on the forming process as on the high temperature 

chemistry of the steel furnace. So now let us examine some 

of these forming processes in general and wire making in 

particular. 

Steel — The Rolling Mill 

The greatest tonnage of steel ingots goes next to the rolling 

mill (See Figure 1) in which a white-hot ingot is passed 

between rolls having appropriately shaped circumferencial 

grooves. The ingot is thus elongated and reduced in crosssection. 

Figure 1 — Rolling Mill 

Generally ingots are “broken down” into blooms, slabs 

or billets in the “blooming mill,” etc. (These industry terms 

identify the shape and size ranges of the products of the 

initial rolling operations.) The blooms, slabs and billets go 

on to smaller mills for rolling into various finished shapes. 

Larger shapes may be rolled directly from ingots. 

A sequence of many different roll passes is required to 

reduce an 18” square by 6’ long ingot, for instance, to more 

than 3/8 of a mile of 2” by 2” by 1/4” angles. 

The “hot working” of the steel by rolling, squeezes and 

elongates the grains or crystals, making the steel somewhat 

fibrous in structure, with considerably increased longitudinal 

strength and toughness. (Steel should not be thought of, 

however, as actually having fibers.) Flat rolled products, that 

is, sheets and wide plates, generally are cross-rolled early in 

the rolling sequence, not only to gain width but also so that 

the improvement in properties will not be limited to the 

longitudinal direction. 

Steel destined for piano wire making is rolled on a rod 

mill to a diameter of slightly over 1/4". It comes off of the 

mill in a continuous coil instead of being straightened and 

cut to length on the hot bed. 

Up to this point in steel music-wire making, all of the 

processing has been at high temperatures, beginning in the 

molten state in the blast furnace and the steel furnace where 

the temperatures approach 3,000 degrees F. The forming 

work on the rolling mill in the solid state requires temperatures 

of the order of 2,000 degrees or more. 

Steel — Wire Drawing 

If the hot work of rolling added strength and toughness by 

elongating the grain structure, the cold work of the drawing 

process which follows in wire manufacture has an even 

greater effect on the properties of the steel. 

After cleaning and coating, the rolled steel is drawn cold 

through a fixed hole, or die (See Figure 2), which results in 

elongation together with reduction of cross-section, and a 

great increase in hardness, tensile strength and yield strength. 

(This will be discussed in “Testing & Properties.”) As used 

here, “cold” does not necessarily mean cold to the touch. 

Work at any temperature below the recrystallization tem- 

perature (several hundred degrees) is “cold work.” 


Figure 2 — Wire Drawing 




A corollary result of the increase in 

hardness is a reduction in ductility; that is, 

some of the ductility is “used up.” The 

fairly high carbon content of music wire 

steel limits its ductility to a reduction in 

area of about 30 percent or less per draft. 

Obviously many drafts are required to 

reduce a 1/4” rod to the diameter of music 

wire and the ductility must be restored 

between drafts. This is done by patent 

annealing, which consists of reheating the 

wire to a selected temperature, perhaps 

1,500 degrees F., followed by slow cooling 

in air or in a medium maintained at a 

rather high temperature. The result is 

recrystallization into a ductile grain 

structure. The cycle of drawing and 

patenting must be repeated several times. 

The wire is finished on the final draft, with 

no further heat treatment, so its strength 

and hardness depends a great deal on the 

amount of reduction after the last anneal. 

The traction for pulling the wire 

through the die is supplied through the reel, or block, on 

which it is wound. Since the tension is quite high, the wire, 

which comes through the die straight, is strained or cold 

worked to the curvature of the block’s circumference. A 

technician struggling to untangle a small coil of recoiled 

piano wire that “got away” can readily estimate the diameter 

of the draw block, unless, of course, the wire was straightened 

before recoiling. 

At this point let us leave the solid-state processing of 

steel, and look at the constitution and micro-structure of the 

various iron-carbon mixtures. 

The Iron-Iron Carbide System 

The molten metal which comes from the cupola or the steel 

furnace is largely a solution of iron carbide in iron, or more 

properly, a mutual solution of the two in each other. As 

cooling proceeds, the metal undergoes a series of changes 

(reversible on reheating), of which freezing is merely the 

first. The nature of these changes, and the temperatures at 

which they occur, depend largely on the percentage of 

carbon in the metal. 

Figure 3, which is a simple version of the Iron-Iron 

Carbide Equilibrium Diagram, will help to visualize the 

state of different iron-carbon mixtures at various temperatures, 

and some of the changes that occur during solidification 

and cooling. 


Figure 3 – Iron-Iron Carbide Equilibrium diagram 

The horizontal scale represents the percentage of 

carbon by weight, and covers the range from pure iron (0 

percent carbon) to pure iron carbide (6.67 percent carbon). 

The vertical scale represents the temperature in degrees 

Fahrenheit. We will not concern ourselves with the intricacies 

in the extremely low carbon area at the left side of the 

diagram. 

The lightly cross-hatched area of Figure 3 represents a 

“mushy,” partly frozen state, bounded at the top by the line 

marked “liquidus” and at the bottom by the line marked 

“solidus.” Above the liquidus the material is completely 

molten. Below the solidus it is completely frozen or solidified. 

The mushy, partly frozen state between has an analogy 

in salt-water solutions, which are mushy through a considerable 

temperature range between the freezing point of pure 

water and the rather lower total freezing point of the 

solution. The reader will note that, for both pure iron and 

the eutectic cast iron (4.3 percent carbon) the liquidus and 

the solidus coincide — there is no mushy state, whereas at 2 

percent carbon the mushy state persists through a range of 

about 450 Farenheit degrees. 

Below the solidus, solid state changes occur during 

cooling, which, to simplify, consist of the separation of iron, 

carbon, and iron carbide in various forms as their mutual 

solubility changes with falling temperature. These separations 

are, of course, microscopic and result in changes in 

crystalline structure. Volumes could be written analyzing

these changes and their practical significance. We will limit 

ourselves to a few observations relevant to this article. 

To better understand Figure 3, one must realize that in 

the solid state iron exists (at different temperatures) in at 

least two allotropic forms. Below 1,333 degrees F. it is stable 

as alpha iron. Above 1,333 degrees it is stable as gamma iron. 

These forms have different crystalline structures and their 

interest to us is in the fact that the solubility of carbon in gamma 

iron is fairly high, but in alpha iron it is extremely low. 

With these facts before us we are ready for definitions 

of some of the terms appearing in Figure 3. 

Austenite: A solid solution of carbon in gamma iron. 

The maximum carbon in austenite varies 

between 2 percent at 2,065 degrees F. and 

0.8 percent at 1,333 degrees F. (See A cm 

line.) Austenite does not exist below 1,333 

degrees F. because gamma iron changes to 

alpha below that temperature. 

Ferrite: Almost carbon-less alpha iron. (Limited to 

about 0.02 percent carbon, as carbon is 

virtually insoluble in alpha iron.) 

Cementite: Iron carbide (Fe 3 C). An extremely hard and 

brittle compound, hard enough to scratch 

glass. Contains 6.67 percent carbon by 

weight. 

Pearlite: A low-temperature (below 1,333 degrees 

F.), rather homogenous laminar mixture of 

ferrite and cementite, containing about 0.8 

percent carbon. 

Now, referring to Figure 3, let us follow a few ironcarbon 

materials from the molten to the cold state. 

Suppose we have molten steel of 0.8 percent carbon, 

which is permitted to cool slowly. Freezing commences at 

about 2,690 degrees F. and is complete at about 2,520 

degrees. The solid austenite then cools without change right 

down to the 1,333-degree line. At this point the gamma 

iron changes to alpha iron and the austenite transforms to 

pearlite as the carbon is cast out of solution in iron carbide 

laminae. This all-pearlite 0.8 percent carbon steel is said to 

have eutectoid composition. 

Pearlite actually retains its identity in non-eutectoid 

steels, but since its carbon content is uniform at 0.8 percent, 

these non-eutectoid steels are necessarily not pure pearlite. 

If the steel is hypo-eutectoid (less than 0.8 percent carbon) 

it consists of pearlite and ferrite grains mixed. If the steel is 

hypereutectoid (more than 0.8 percent carbon) it consists of 

pearlite and cementite grains mixed. 

The precipitated cementite in hyper-eutectoid steel 

forms at the pearlite grain boundaries, and this network of 

cementite increases the hardness and strength of the steel, 

compared with that of steel which is lower in carbon. 

It might be of interest to trace the cooling cycle of a 

hyper-eutectoid steel of, say 1.25 percent carbon. This steel 

starts to freeze at about 2,650 degrees F. and freezing is 

complete at about 2,350 degrees. It now remains solid 

austenite containing 1.25 percent carbon until it cools to 

the A cm line at about 1,675 degrees. At this point the solution 

is saturated with carbon and the excess carbon starts to 

precipitate in cementite. This precipitation of cementite 

continues on down to 1,333 degrees, at which point the 

remaining austenite, which has now reached eutectoid 

composition, transforms to pearlite, and we have a mixture 

of pearlite and Cementite grains as mentioned in the 

preceding two paragraphs, and a rather hard steel. 

Hypo-eutectoid steel (less than 0.8 percent carbon), 

after freezing, continues to cool as austenite down to the A 1 

line, at which point it begins to cast out ferrite crystals. This 

continues on down to 1,333 degrees F. at which point again 

the remaining austenite has reached eutectoid composition, 

and transforms to pearlite, resulting in pearlite grains in a 

ferrite matrix. Ferrite is similar in properties to pure iron. 

Hence steels low in carbon are relatively soft and of lower 

tensile strength than the high carbon steels. 

The temperature-related changes we have been discussing 

are also time-related. The equilibrium states represented 

by the various areas of Figure 3 assume that the cooling 

metal has the necessary “leisure” for the changes to take 

place. Sudden cooling may force non-equilibrium changes 

that alter the crystalline structure and the properties of the 

material. 

These facts are the basis of the heat treatment of steel, 

which consists of slowing or speeding the cooling rate 

through critical temperature ranges to facilitate or inhibit 

change. Thus, in the wire-drawing process, ductility was 

restored between drafts by reheating the wire above the A 3 - 

A cm lines to the all-austenite condition, then cooling it 

slowly to below the A 1 line. A rapid quench cooling on the 

other hand would have increased the hardness and reduced 

the ductility by forcing a non-equilibrium change into one 

of the harder forms, as, for instance, Martensite (not defined 

in this article). 

Steel for piano wire, having about 0.85 to 0.90 percent 

carbon, is near eutectoid composition and hence mostly 

pearlite. It affords an excellent compromise between high 

tensile strength and high ductility. 

Now let us look at the cast iron area of Figure 3. We see 

at once that a cast iron of 4.3 percent carbon has the 

minimum freezing point of all the iron-carbon mixtures 



(2,065 degrees F.). We see also that it has no mushy stage, 

since the liquidus and solidus lines coincide at that carbon 

ratio. In foundry work this material would have the obvious 

advantage of remaining fully fluid at a relatively low temperature, 

flowing freely in intricate molds such as those 

required for piano plates. Also, with almost no temperature 

change during freezing, shrinkage during freezing would be 

minimal, resulting in sound castings free of shrinkage 

defects. While the carbon in gray cast iron rarely exceeds 3.5 

percent, the advantages named above are still partly retained 

at this percentage. 

In view of what we have said about the hardness of 

mixtures high in carbon (cementite), a 3.5 percent carbon 

cast iron might seem much too hard and brittle to be useful. 

Indeed it would be if all the carbon remained in combination 

as cementite. But the silicon present in the metal comes 

to the rescue. As freezing and subsequent cooling progresses, 

most of the carbon in excess of the eutectoid ratio (0.8 

A plea for advice about a distressing occurrence 

comes from New York State: 

“As a former member, retired, I am seeking some 

information. I pulled up an old piano 1/2 note to 440 

pitch, informing the man it was risky to raise the pitch. In 

doing so, after the pull-up, the plate [cracked] in the treble 

section. 

What I want to know is, who is responsible? 

This is the first time this has happened to me in 58 

years tuning.” 

Some [technicians] tune pianos for a lifetime 

without ever experiencing a broken plate. But it does 

occur once in a while. It is always sudden and unpredictable, 

the result of some unseen internal strain in 

the plate. It may be the result of metal fatigue in a 

plate that was never quite perfectly fitted to the heavy 

wooden frame of the piano, but was forced to conform 

to it by the many bolts and screws. It may be the 

result of slight and gradual changes in the wooden 

frame itself, to which the cast-iron plate is not able to 

accommodate itself. It may be the result of a hidden 

weakness in the casting, which finally “lets go.” It is, as 

I say, a rare thing, but it does occur. It has been known 

to occur when the piano was sitting idle by itself, 


percent) is precipitated as graphite flakes, interspersed with 

the ferrite and cementite crystals, so that little cement-ite is 

present except in pearlite form. This formation of graphite 

is essential to the character of gray cast iron, and the silicon 

is the chief agent in its formation. Hence the great importance 

of silicon in gray cast iron. (Figure 3 does not try to 

show this effect.) 

Testing & Properties 

Since the properties of a material determine its suitability 

for a particular purpose, the testing of these properties is 

very important. 

Some vital properties of engineering materials are 

toughness, hardness, and fatigue strength. Toughness is 

measured by the energy absorbed before fracture in a 

standardized impact test. Hardness is measured by the 

penetration of a standardized indenting die. Fatigue strength 

is measured as the maximum stress tolerable under indefi- 

Broken Plate! Who Is Responsible? 

with no tuner within miles! 

It is safe to assert that a “healthy” plate, properly 

designed, properly fitted and secured in the piano, 

will not break under the ordinary stresses of the 

tuning process, even when the string tension is being 

raised back to standard after long neglect. It is designed 

to stand much higher stresses than those set up 

by strings tuned to standard pitch. 

There is even some question whether such a 

plate could be broken by deliberately over pulling the 

strings or whether the strings would not break first. 

But since the tuner does not do this, the question is 

academic. 

So there is no basis for considering the tuner to 

be responsible for plate breakage that occurs during 

or after tuning. This fact is easier for customers to 

accept if they are forewarned that there is risk involved 

in raising the pitch of a piano. That this risk is 

extremely small is shown by the fact that our inquirer 

tuned for 58 years, including, I am sure, hundreds of 

pitch raises as drastic as this one, without ever experiencing 

a broken plate. 

— Don Galt

nitely repeated cyclic loads. 

The properties of steel most important to us in connection 

with piano wire are those called tensile properties. 

Because there are some striking similarities between wire 

drawing and the tensile testing of steel, we will examine the 

latter in some detail. 

The tensile test consists of stretching a specimen of 

known cross sectional area to failure in a testing machine. 

The maximum tensile stress endured by the specimen 

before failure, divided by the original cross-sectional area, 

gives the ultimate tensile strength per unit of area, usually 

called simply the tensile strength. In countries using the 

English system, the tensile strength and the other tensile 

properties, such as elastic limit and yield strength, are given 

in pounds per square inch. It is usual to use a specimen 

accurately machined to a diameter of 0.505". This has a 

cross-section of 0.2 square inch, which is a convenient 

divisor for converting the measured load on the specimen 

into the stress per square inch. 

While the tensile strength is easily determined with 

proper equipment, it is not the most important tensile 

property of steel. More important is the amount of stress it can 

stand without permanent deformation. The specimen elongates 

as it undergoes constantly increasing loads in the testing 

machine. At first this elongation is elastic; that is, if the 

tension is removed the original length will be recovered. As 

the load is further increased an elastic limit is reached, and 

plastic, or permanent deformation begins. The precise elastic 

limit is seldom determined in practice, as the process is 

cumbersome. It requires the alternate application and release 

of increasing loads, with measurement of the gauge length 

each time to determine if the behavior is still elastic. 

The tensile property usually obtained instead of the 

elastic limit is the yield point, which, in steel, is only slightly 

above the elastic limit. This is the unit tensile stress at which 

the specimen continues to elongate for a period without 

any increase in load. The term “yield” is quite descriptive of 

what takes place. At and above the yield point permanent 

slips occur along planes in the crystals, until they are arrested 

by crystal boundaries and broken grains. At first, yielding is 

distributed throughout the length of the specimen, but as 

the load rises a “neck” starts to develop in the specimen and 

further elongation is concentrated at this neck. After 

necking starts, elongation continues under decreasing loads, 

because the effective cross-section is decreasing, and this 

continues until fracture occurs at the neck. 

The cold work of stretching actually hardens and 

strengthens the stretched steel, so that the true unit stress on the 

material at fracture (the load divided by the instantaneous 

cross-section of the neck) is considerably greater than the tensile 

strength (the maximum load achieved, divided by the 

original cross-section). If one thinks of wire drawing as a 

sort of controlled tensile test, in which a “continuous neck” 

of uniform cross-section is formed, it is not hard to see how 

the drawn wire develops strength and hardness superior to 

those of the hot rolled rod from which it is drawn. The 

drawn wire exhibits a new elastic limit and yield point, 

higher than those of the steel in as-rolled condition, and is 

capable of elastic behavior within these limits. 

It is imperative that the cold work of drawing not be 

overdone, lest the wire become too brittle. 

(The tensile test and the drawing process are not perfectly 

analogous. In drawing, the lateral compression of the metal by the 

die is an effect similar to cold forging—an effect absent from the 

tensile test.) 

What we have discussed is the testing of rolled steel. 

Wire can be tested similarly, though it is customary to quote 

the breaking strength of wire rather than the tensile 

strength. Tensile strength is a property of the material 

independent of size, whereas breaking strength is a property 

of the specimen, depending on both tensile strength and 

sectional area. 

Generally, smaller sizes of piano wire have higher tensile 

properties than larger sizes because of the additional drafting 

or cold work. Also, wire used for bass string cores generally 

is made with lower properties, in order to preserve enough 

malleability so that the wire can be swedged, or flattened at 

the start and finish of the winding. Moreover, some manufacturers 

make more than one grade of wire, for example, 

two grades of bass and two grades of treble wire. For the 

most part these different grades and sizes come from the 

same steel and owe the difference in their properties to the 

amount of drafting and the spacing of annealings in the 

manufacturing sequence. 

The yield strength of the wire, which cannot be exceeded 

in piano stringing and tuning without serious 

damage, is determined by a rather arbitrary process known 

as the 0.2 percent offset method. It is generally found to be 

about 70 percent of the breaking strength. I do not mean to 

pass lightly over the elastic limit and yield strength. The 

non-determination of elastic limit, and the use of an 

arbitrary method for yield strength, are dictated by practical 

considerations, not by any lack of importance of these 

properties. Long and careful practice has shown that this 

testing method for yield strength gives a valid measure of 

the two properties in piano wire. 

In common with many other materials, steel under high 





tensile stress continues to elongate slightly for an indefinite 

time. This elongation is called creep. In most steel uses creep 

is considered negligible if the temperature is less than 40 

percent of the melting point on the absolute scale, i.e., less 

than about 700 degrees F. With piano strings, whose pitch is 

so sensitive to a very small change in tension, creep at 

ordinary temperatures is probably a factor in both the quick 

loss of pitch in newly strung pianos (so-called primary 

creep), and in the long-term loss of 

pitch (secondary creep). 

Fatigue in metal has been mentioned 

briefly, fatigue strength being the 

tolerance of indefinitely repeated cyclic 

loads. The high frequency reverse 

bending that occurs constantly in a 

highly tensioned vibrating string, 

particularly at the ends where the 

transverse waves are reflected, is surely 

high stress cyclic loading, and many 

string breaks in playing must be regarded 

as fatigue failures. 

Whenever a string is placed in a 

piano, it is necessarily cold worked at 

several points, such as the bridge pins, 

the agraffe, etc. Every non-elastic bend 

that is put into the wire tends to harden 

the wire by effectively cold working the 

steel at that point. The fibers on the 

convex side of any plastic bend have 

probably been stretched beyond the 

yield strength. This point will then be 

slightly more brittle than other parts of 

the string, and a likely candidate for 

ultimate fracture. One such point is the 

agraffe. Another is the point of tangency 

where the string starts to wind around 

the tuning pin. Repeated small tuning 

changes subject a short section of the 

string to alternate bending and straightening, 

which, even though slight, tend 

to work-harden the steel. This is 

probably why so many “old age” string 

breaks occur at the tuning pin. 

It should be obvious that piano 

strings are hard-working elements of 

the musical structure of the piano and 

we should be careful not to do anything 

to make their lot harder. Specifically, we 

should avoid subjecting them to 

unnecessary plastic bends by kinking or 


any other means, and to excessively high tension. Remember 

that any tension approaching 70 percent of the breaking strength is 

dangerously close to the yield strength. 

A few more words are in order about the properties of 

gray cast iron, which is a more prosaic cousin of piano wire. 

Its eminent suitability from the standpoint of manufacturing 

convenience in piano plate work has been mentioned. 

Its low freezing point (compared with other iron-carbon

material), and the proximity to each other of its liquidus and 

solidus temperatures combine to make for sound, uniform 

castings. 

Gray cast iron also has about the same coefficient of 

thermal expansion as steel music wire, which is a factor 

favoring tuning stability. Lighter metals sometimes used for 

piano plates have somewhat higher coefficients of thermal 

expansion. 

Gray cast iron has good machinability, as anyone who has 

drilled a piano plate has observed. This characteristic is 

enhanced by the presence of the carbon in graphite flake 

form. 

Gray cast iron has low tensile and compressive strength 

compared with steel, its ultimate tensile strength being about 

25,000/30,000 pounds per square inch. Structural steel has a 

tensile strength about three times as high, and cold drawn 

piano wire about ten times as high. In piano plates this relative 

low strength is not a particular disadvantage. The plate needs 

to be heavy enough to afford a solid platform for the strings 

to stand on while they are shaking the soundboard. Furthermore, 

the parts of the plate must be large enough so that there 

will be little deflection under load. These two requirements 

WIRE DRAWING A.D. 1540 

Pure Sound 

Top quality Stainless 

Steel Piano Wire 

from 1600 N/mm 2 - mid 19th century 

to 2200 N/mm 2 - modern pianos (short scaling) 

Most versatile high tech piano wire based on latest 

developments in stainless steel processing. 

Tuning stability better than average. 

Lower inharmonicity. Lovely sound! 

Prices in Euros: approx. 95¢ to the Euro at June 2000 

500 g - 15.00 250 g - 8.50 - 125 g - 5.50 (0.700 mm - 1.50 mm) 

500 g - 21.00 250 g - 11.00 - 125 g - 7.00 (0.500 mm - 0.675 mm) 

Complete list of breaking strengths, yield points 

and other data available. 

Pure Sound 

Juan & Mary Más Cabré 

Eline Verestraat 46 • 1183 KZ AMSTELVEEN, Netherland 

Phone: +31.20.6418099 Fax: +31.20.6407621 

E-mail: info@puresound-wire.com 

permit use of a material of relatively low unit strength. 

One of the properties of gray cast iron, which makes it 

particularly suitable for machinery bases, is also to its advantage 

in piano plates. This is its tendency toward internal selfdamping 

of vibrations. This property, largely due to the 

honeycombing with graphite carbon, means that plates of 

gray cast iron are not apt to show objectionable resonance. 

They do not “ring.” Happily, the words “bell metal,” which we 

all have seen cast into some piano plates, are simply not true. 

The graphite carbon of gray cast iron makes it very 

difficult to weld. Many piano technicians have had successful 

experiences in repairing broken plates by welding. It must be 

said, however, that the weldability of gray cast iron is poor, and 

that these successful repairs are in spite of, not because of, the 

properties of the metal. After welding, the homogeneity of the 

casting is gone, and there is sure to be a zone of weakness 

somewhere. Fortunately the plate is usually over-designed as 

far as structural strength is concerned and the weakness of the 

weld repair is not necessarily fatal. Sometimes there is no 

alternative to attempting repair by welding and it is not the 

author’s intent to write against it. But due to the microstructure 

of the material, welding cannot be thought of as a 

reliable means of making broken gray 

cast iron “as good as new.” 

Of course, there are other iron 

materials to be found in pianos in 

smaller quantities. Tuning pins are made 

of cold-drawn steel wire, not so hard, 

considerably larger in diameter than 

piano wire. The many steel screws are 

made of steel that is made brittle by the 

presence of sulphur, so that it will thread 

easily and cleanly. The leg plates of 

grand pianos are made of a tougher cast 

iron than the gray cast iron of the string 

plates. They are lower in carbon, much 

lower in silicon and higher in manganese, 

and their carbon is not in graphite 

flake form. They exhibit a fair degree of 

malleability and less likelihood of 

fracture under shock than gray cast iron. 

But these two cousins in the iron 

family, steel music wire and gray cast 

iron plates, are the real backbone of the 

modern piano. Without them it would 

be, as it once was, a very different 

instrument. 


Piano Plate Breakage: A Case Stud 

By Steve Brady, RPT 

Journal Editor 

The story I relate here is about an event, or series 

of events, in the life of Marnie Squire, an 

Associate member of the Cincinnati, OH, 

chapter of PTG. She has been kind enough to 

provide the documentation from her case for use in this 

issue of the Journal. “If it can help save someone else from 

the kind of nightmare I went through,” she says, “I’m happy 

to share my story.” 

At about 10:30 a.m. on Friday, July 2, 1993, Marnie 

Squire arrived to tune a small Fischer grand piano at a home 

in Middletown, OH. The piano, an Aeolian product, had not 

been tuned in 13 years. Squire played the piano briefly to 

evaluate its condition and found several keys not playing as 

well as some damper problems. The piano was 37 cents flat. 

After bringing the piano back to a condition of rough 

playability, Squire began the process of raising pitch. Using a 

Sanderson Accu-Tuner, she completed a “normal” first pass 

and had nearly completed a second pass. Then, “I was about to 

tune the second or third string from the bottom of the bass 

section. I played the keys and a huge ‘bang!’ happened. I had 

no idea what had happened and was very shaken.” Looking 

over the piano she saw a crack in the second plate strut from 

the top and another crack in the tuning pin area at the bass/ 

tenor break. Mortified, she called the owner of the piano, who 

was at work and explained what had happened. 

The piano owner filed a lawsuit over the broken plate. 

The owner enlisted the aid of another piano technician in the 

area as an “expert” witness and this technician (whose name is 

omitted here) told the piano owner that Squire had brought 

the pitch up too fast, that she didn’t know what she was doing, 

and that she had actually broken the plate! 

In September of 1993, Marnie Squire retained an attorney 

to defend herself in the lawsuit and the long process of 

gathering evidence began. Several PTG members sprang to 

Squire’s aid by examining the piano and writing opinions, 

some even performing sophisticated analyses based on the 

physical evidence. From over two dozen written opinions 

placed at my disposal by Marnie Squire, I have excerpted a 

number of relevant quotations. 

Willard Sims, piano service manager at Baldwin from 1946 to 

1984, wrote on September 17, 1993: 

“The tuning of a piano by an experienced technician 

will not cause the string plate to fail.” In another letter dated 

April 10, 1994, Sims reiterated this stance: “I repeat my statement 

that the tuning of a piano will not cause plate failure. If 

a rebuilder refurbished and perhaps rescaled the piano, re- 


moving and resetting the plate, then that action may result in 

plate failure.” 

Sandy West, a later piano service manager for Baldwin, elaborated: 

“It is my considered opinion that a broken plate cannot 

be blamed or attributed to a typical tuning/service call. My 

experience is that strings will break before the plate will. A 

broken plate is usually the result of some major trauma, such 

as the piano being dropped or the result of a defect in the 

manufacture of the plate. In both such instances the actual 

crack or break may not show up for quite some time. It will 

develop over time as an eventual result of the continued pressure 

on the fault by all the strings. Simply tuning the piano 

would not cause such damage.” 

Dr. Albert Sanderson pointed out that Marnie Squire’s pitchraising 

method had been entirely appropriate, then added: 

“It has been my observation that plates that break under 

normal tuning stress have a flaw in the casting that can be 

seen when the break is examined. A flaw could be a bubble 

in the casting or a crack that has been growing gradually over 

the years owing to metal fatigue.” 

Noted piano rebuilder Tony Geers reiterated the now-familiar 

theme: 

“Based on the information we have at hand, most notably 

the fact that the piano plate broke in both the tuning pin 

area and treble bar, it would be our conclusion that faulty 

installation of the plate during the manufacturing process is 

the most likely cause for the breakage. It is impossible for a 

tuner to break a plate by tuning alone. There must be other 

circumstances present, i.e., faulty manufacturing, flaw in the 

cast iron, piano dropped, etc. Tuning works against the strength 

of the cast-iron plate. Over-tuning would cause string breakage 

long before any possible damage to the plate could occur. 

“If the plate was improperly installed at the factory, plate 

breakage is a very real possibility. Improper installation could 

be the bending of the plate over the pinblock or securing 

bolts or screws. When tension is added by tuning, extreme 

stress is focused on the bent portion of the plate; such as in 

the area of the tuning pins and treble bars.” 

A letter from prominent piano technician and educator Jim 

Geiger stated: 

“The conclusion is that a normal piano plate, designed 

to withstand 40 tons of pressure would not be broken by the 

tension from the piano strings regardless of the applied tension, 

how fast the tension is applied and at what point of the

ase Study 

scale the tension is applied. In the piano factories the tension 

is applied as fast as the tuner can bring the strings up to pitch. 

Indeed, it should not be possible to break a normal piano 

plate with string tension alone, because the strings would 

break first. There is just not enough margin between the actual 

string tension and the tension at which failure will occur 

for the piano wire to be able to produce the force necessary 

to cause a good piano plate to fracture.” 

University technician Rolf von Walthausen added some background 

on the material, then pointed out that some piano models 

frequently suffer cracked plates. In this particular case, it turned out 

that many Aeolian grands had suffered the same fate. von Walthausen 

wrote: 

“Piano plates are made of cast iron, which is a material 

that is extremely hard, but also brittle. Properly cast and installed, 

it is capable of withstanding tremendous pressure from 

the strings, which are fastened to and held in tension by the 

plate. Improperly cast or installed, a cast-iron plate could easily 

break or crack. Even if piano wire, which is a steel alloy 

with great tensile strength, is stretched quickly beyond a certain 

stiffness, the wire will break far, far before exerting a 

force on the piano plate that would cause breakage or any 

type of damage to the plate. 

“Some brands of pianos have frequent occurrence of plate 

breakage or cracking. It is rare to find an old Bechstein grand 

piano, for example, without a crack in the plate. Opinions 

from experts differ as to why this is so (poor casting, design or 

installation), but one thing is never disputed: tuning or pitch 

raising was never the cause.” 

Nevin Essex, another highly regarded technician from the 

Cincinnati area, wrote: 

“I have been teaching piano tuning and technology 

through the Guild, at universities and on my own since 1982. 

I have researched teaching methods and developed my own. 

Nowhere have I ever seen or heard any scientific evidence 

that suggests that a piano tuner can break a plate. My understanding 

is that plates are designed to withstand much more 

tension than exists in any piano. I have never heard any credible 

account of a piano tuner breaking a plate. I was never 

taught nor do I teach any technique or method designed to 

prevent a plate from breaking while tuning. Tuning methods 

that emphasize raising pitch evenly do so for the purpose of 

achieving a good tuning, not for preventing the plate from 

breaking. 

The author of The Piano Book, Larry Fine, weighed in with 

an opinion that even contributed a touch of humor to the situation: 

“There are only two ways I know of that a tuner can 

break a piano plate that is not defective while servicing a 

piano in the home. One way is to excessively tighten the 

nose bolt that supports the plate in the center area of the 

piano. This is an adjustment not normally made outside of a 

piano rebuilding shop. The other way is to take a sledge hammer 

to it. In other words, it is virtually impossible for a piano 

tuner to break a plate during the normal tuning and pitchraising 

of a piano unless the plate is already defective and 

ready to break, in which case any tuner, regardless of skill or 

method of tuning, will be the unwitting agent of such breakage 

by fate alone. Even in a worst-case scenario, in which a 

tuner sought to sabotage a piano by stretching all the strings 

far above standard pitch, chances are that the strings would 

break long before the plate would. Tuners are sometimes 

blamed for plate breakage by understandably distraught piano 

owners, but in every such case the blame is misguided 

and completely unjustified.” 

In expectation that the lawsuit would come to trial, 

Marnie Squire asked Jim Ellis to look at the piano and to 

render an opinion from his background as an engineer. After 

examining the piano, Ellis wrote a formal analysis (included in 

this issue of PTJ) proving that the plate was poorly designed. 

“What I couldn’t understand,” he said, “is why the plate hadn’t 

broken when the piano was first strung in the factory.” 

Enter Paul Monachino, who had worked for the now-defunct 

Aeolian Corporation during the years when the subject piano was 

manufactured. Delivering the death-blow to the plaintiff’’s case, 

Monachino wrote: 

“This problem is nothing new in this style piano. I have 

seen this particular plate cracked in the same place many, many 

times. The fault lies in the construction of the plate and not in 

the tuning of the piano.” (Monachino’s emphases —SB) 

The night before the case was scheduled for trial, the 

piano owner’s “expert” witness backed out, leaving the 

prosecution with no case at all. The shame of the whole story 

is that Marnie Squire had been placed in such a position to 

begin with. Besides having to spend hundreds of dollars in 

attorney’s fees, she was “a nervous wreck” for the year that 

passed before resolution. The silver lining to this cloud is that, 

because of what she went through, and the unanimous 

opinions provided by the real experts, this kind of nightmare 

— a lawsuit obviously without basis in fact — should not 

have to be suffered by any piano technician again. 


An Analysis of a Broken Plate 

By Jim Ellis, RPT 

Knoxville, TN Chapter 

Background 

When Steve Brady called and asked me if I would 

contribute an article about broken piano plates 

(to replace an article promised by someone else, 

but which had never actually materialized), his deadline was 

just five days away and I was leaving on a trip in three days. 

When I returned it would be too late. Because of the time 

constraint Steve and I decided that I should just use an 

analysis that I did back in May, 1994 for the Journal article. 

The piano was a 1973 J&C Fischer that had been 

neglected for several years and allowed to go 37 cents flat. 

The scale design had only three major divisions, no agraffes, 

and the forward termination for the strings was a long 

curved capo bar that ran all the way from #1 in the bass to 

#88 in the treble. The middle section spanned 32 triplestring 

unisons without any additional support and the 

bearing angle of the strings against the bar was excessive. 

The dimensional cross section of the bar was minimal, and 

there were no shoulder (nose) bolts to secure the plate struts 

to any beams underneath. Immediately after the tunertechnician 

brought the piano up to standard pitch and 

began to check the tuning, the middle section of the capo 

bar broke. 

Another technician claimed that the plate broke 

because the tuner brought the piano back up to standard 

pitch in one tuning rather than in several small increments 

spread out over a period of days, weeks or months. The 

owner filed a lawsuit against the tuner for an amount that, in 

my opinion, was far in excess of the actual worth of the 

piano. I was asked to be an “expert witness” for the tuner. 

Although the piano was located in the Cincinnati area some 

260 miles from where I live, I agreed to do it for net 

expenses only. The intentions of the piano owner may have 

been perfectly honest, but his/her decision to sue was based 

upon an erroneous conclusion by another technician and 

the result could have set a precedent that would have been 

absolutely wrong! 

After taking a good look at this particular piano I was 


surprised that that section of the capo bar had not broken 

when the piano was first chipped at the factory. I was later 

told that some of them did. 

The following is the analysis that I presented to the 

tuner-technician’s attorney and I was well aware that I 

might be called upon to present it in court later on. Fortunately 

for everyone the suit was withdrawn the day before 

the hearing. This was such a no-winner! The design of the 

piano was, in my opinion, just asking for trouble. 

The original analysis included four figures, which are 

included here, and eight photographs — primarily for the 

education of the attorney — that do not appear here 

because they are no longer available. 

Tuning Procedure Used by Mrs. Squire 

Mrs. Squire and I discussed the procedure she had used to 

tune the piano just before the plate broke. The piano had 

been neglected and not tuned for more than a decade. Mrs. 

Squire measured its pitch and found it to be about 37 cents 

flat. In tonal nomenclature, a “cent” is 1/100 part of a 

semitone; a “semitone” is 1/12 part of an octave; and an 

octave represents a ratio of 2:1 in frequency, or pitch. In 

going up the musical scale, each of the 12 semitones in an 

octave increases by the 12th root of 2, or 1.059463094 

above the one below it. Therefore, being “37 cents flat” 

means that the frequency (pitch) of the notes on the piano 

was about 98 percent of what it should have been at standard 

pitch (A=440Hz). 

When a piano is flat (low in pitch) by this much, 

current procedure calls for raising the pitch of each string 

very, very slightly above its normal frequency so that when 

it settles after tuning it will be at, or very near, the desired 

pitch. It is well within the limits of good tuning practice to 

raise the pitch of a piano by 37 cents at one time. Obviously, 

the tuner then repeats the tuning in order to obtain a finer 

tuning, since the piano will always settle back some. This 

procedure is accepted and recommended throughout the

industry. It is my understanding that Mrs. Squire had 

actually finished the tuning, and was playing chords to 

evaluate the job, when the plate finally broke. 

Mrs. Squire was following a tuning procedure that was, 

and is, appropriate for the occasion. I can find no fault at all 

with what she did. 

The Actual Cause of the Plate’s Failure 

In order to describe clearly what happened, and what 

caused this plate to break in this piano, I must first outline 

the most basic principles of cast-iron plates in grand pianos. 

Basic Construction of a Piano Plate 

The plate of a piano is the structure that provides a very 

strong, rigid and stable framework inside which all the 

strings are strung. It is what makes the modern piano 

capable of staying in tune for weeks and months at a time. 

Piano plates are made of gray cast iron, and not “bell metal,” 

as some people believe. 

Cast iron is chosen because it is economical, mechanically 

stable and has a low coefficient of thermal expansion. 

It is very strong under compression, but weak under tension, 

Figure 1 

and brittle. Its properties depend upon the amount of 

carbon and other impurities that it contains, and these can 

vary widely. For these reasons, a quality grand piano is 

designed so that the areas of high stress concentration in the 

plate are those that are under compression, not tension. For 

maximum stability and strength, piano plates are almost 

always cast in one piece. They are usually finished with a 

bronze-lacquer-base paint. 

For purposes of illustration, Photo 1 is a photograph of 

the tuning-pin area of a high-quality grand piano, not the 

piano with the broken plate. [EDITOR’S NOTE: Remember 

none of the photos referenced in this article are available for 

publication. — SB] The photo shows the reinforcing bars 

that are a part of the plate structure, tuning pins, agraffes, 

capo bar and strings. The bars that run parallel to the strings 

and carry the load of the tension of all the strings lie mostly 

above the strings. The total tension can be anything from 

30,000 to 50,000 pounds, depending upon the size and 

scaling of the piano. Because the strings are all pulling 

inward on these bars in a plane that lies below their 

centerlines, the bars have a tendency to arch upward in the 

middle (See Figure 1). 





When a bar is bent, the material in the inner part of the 

bend is compressed, but that of the outer part of the curve is 

under tension and is elongated. Cast iron will not withstand 

great bending forces because of its low tensile strength. 

The tendency of a piano plate to bow upward in the 

middle is normally restrained by anchoring it to massive 

wooden beams under the soundboard using shoulder bolts 

that extend through the soundboard. There are other 

shoulder bolts in this piano, but they are not visible in the 

photo. The outer perimeter of the plate is bolted to a very 

rigid rim made of laminated hardwood to hold the plate 

perfectly flat, and not allow it to bend under the tension of 

all the strings. By anchoring the plate in this way, the only 

major forces acting on it are compressive, not tensile. 

The plate flange at the front of the piano, the bar that 

extends across the width of the piano, and the bars that 

connect A-B and C-D together, all form a very rigid 

structure. Another part of the casting extends downward 

from point X and attaches to a massive “cross-beam” that 

traverses across the width of the piano below the 

soundboard. The point where several of the plate bars 

converge, is one of the regions of highest stress concentration 

in a grand piano. The part of the casting that ties this 

part of the plate to the cross-beam (sometimes referred to as 

the “horn” because of its shape), greatly improves the 

strength and stability of the plate by securing it (out in the 

span across the piano) to the massive structure under the 

piano. 

Photo 2 is a close-up of the tuning-pin area of the 

piano shown in Photo 1. It clearly shows how the strings in 

a grand piano extend through the agraffes, which form the 

forward termination of the speaking lengths of the strings. 

Agraffes are made of machined brass, with threaded studs at 

the bottom, which are screwed into threaded holes in the 

plate. In this division of the piano, they have three eyelets, 

one for each of the three strings in each unison (note). 

Making and installing agraffes is time-consuming and costly 

because it is detailed work. Nevertheless, this is the preferred 

way to terminate the strings of a grand piano, except for the 

high treble divisions, where a capo bar is usually used. 

Construction of the Plate That Failed 

Photo 3 is a picture of the inside of the Fischer piano with 

the broken plate. The bar (H-J) in the foreground is broken 

at point J, and protrudes upward and slightly to the rear of 

the piano. 

Photo 4 is a closer view of the broken bar (H—J). A 

total of 30 three-string unisons can be seen in this division 

of the piano. 


Photos 5 and 6, taken at slightly different angles, are 

close-ups of the break at the right-hand end of the bar. 

Photo 7 shows the cracked plate at the left-hand end of 

the bar. 

Photos 5-6-7 clearly show that no agraffes are present. 

Instead, the treble capo bar has been extended all the way 

through the piano to act as a common forward termination 

Pinblock 

Figure 2 

for all the strings. This appears to have been a cost-cutting 

measure by the manufacturer. Photos 5 and 6 show that the 

bar jumped upward and toward the rear of the piano when 

it broke. Judging from my first-hand observation of the 

piano on May 2, 1994, and from the photographs I made, 

the angle of rise of the strings as they came forward from 

under the bar must have been at least 30 to 35 degrees 

before the bar broke. I consider this much bearing angle to 

be excessive. An angle of 16 to 18 degrees would have been 

more appropriate (See Figure 2). 

Photo 8 is a view looking inside the action compartment 

of the piano, with the keys and action removed. The 

sostenuto rod is in the lower foreground, and the damper 

flanges and wires are behind it. The bottoms of the dampers 

can be seen above the wires. The “belly rail” (to which the 

front edge of the soundboard is glued) lies behind the 

damper wires. The cross-beam is just under the belly rail. 

The open space between the damper wires (middle of the 

photo) under the cross-over between the bass and tenor 

strings is where the “horn” (described earlier) would be, if 

there were one, but there is not. Neither are there any 

shoulder bolts to secure the plate bars or string plate to a 

massive under-structure. 

The pinblock is the laminated hardwood plank that 

securely holds the tuning pins. It is normally fitted to, and 

rests against, a flange on the under-side of the plate, behind 

the tuning pins, so that the pinblock cannot move under the

tension of the strings. In this Fischer piano, a wide gap exists 

between the pinblock and the plate flange, a condition that 

is considered poor construction by most piano builders and 

rebuilders. 

Scenario of the Break 

There are 30 unisons (notes) in this middle division of the 

piano. Each unison has three strings. We may assume that 

the total tension of all the strings in this piano, when tuned 

to standard pitch, would be somewhere between 35,000 and 

38,000 pounds, and that the average tension of each string 

in this division would be about 155 pounds. As I indicated 

earlier in this report, I believe the bearing angle of the 

strings at the capo bar was at least 30 degrees, and perhaps as 

much as 35 degrees before the bar broke. It is impossible 

now to measure the angle, because the bar is broken, but if I 

imagine that it is intact in the right place, that is the approximate 

number that I get. 

Therefore: 

Tension of each string (pounds) = 155 

Number of unisons in this division = 30 

Number of strings per unison = 3 

Estimated string bearing angle under capo = 30 

degrees minimum, 35 degrees maximum. 

Figure 3 is a vector analysis of the combined force of all 

the strings against the capo bar that broke. The force vectors 

are taken from the origin (O), which represents the point of 

contact of the strings against the capo bar in this division of 

the piano. The horizontal axis represents the plane of the 

Figure 3 

strings in the piano. The length of line (O-G) represents the 

total tension in the speaking length of the strings. The 

length of lines (O—A) and (O-B) represent the total 

tension in the short lengths of the strings between the capo 

bar and the plate bearing surface for the estimated bearing 

angles of 30 and 35 degrees respectively. 

The original scale design of this piano is not available to 

me, therefore I am constrained to estimate what the total 

tension of the piano, and of the strings in this division 

would be. I believe that it is reasonable to assume an average 

tension of 155 lbs. for the strings in this division, as stated 

above. The total string tension in this division would then 

be 3 X 30 X 155 = 13950 pounds. I am also constrained to 

assume that the tensions of the strings were all equalized 

across the capo bar. The graphic analysis assumes this. 

However, because of the excessive bearing angle at the capo 

bar and therefore excessive friction, this may not be the case, 

which would make the situation actually worse than what I 

have shown. 

If the string bearing-angle across the capo bar were only 

30 degrees, line (O-A) would represent that vector. Completing 

the parallelogram (lines A-F and F-G), gives the 

resultant force represented by line (O-F), and indicates that 

the force was 7300 pounds upward at an angle of 15 degrees 

toward the rear of the piano. This breaks down into component 

forces of 7000 pounds straight up and 2000 lbs. toward 

the rear. If, on the other hand, the bearing angle had been 

35 degrees, then the resultant would have been 8450 





pounds upward at an angle 

of 17.5 degrees to the rear. 

The component forces on 

the capo bar would then 

have been 8000 pounds 

straight up and 2650 

pounds toward the rear. 

Figure 4, attached, 

shows the general pattern 

of stress in this division of 

the capo bar. Forces 

pressing upward and 

outward against the curve of the bar would create “stress 

risers” at the bottom of each end. Tensile strains (for which 

cast iron is weak) would then focus at these points and tend 

to tear the bar apart here. 

When the tensile strain exceeded the low elastic limit of 

the casting at the right end of the bar, it broke. The fracture 

appears to have begun at the bottom, near the front, and 

moved upward in a matter of a few microseconds until the 


Figure 4 

break was complete; the 

right end of the bar flew 

up an inch or two, and 

then fell back down to 

where it is now, as the 

plate cracked at the other 

end of the bar due to the 

sudden prying action of 

the bar as its right end flew 

upward. 

In my opinion, the full 

88-note capo bar in this 

piano was an inexpensive and inferior substitute for much 

better construction. The long curved span of the capo bar 

was unrestrained in the middle. The string bearing-angle 

was excessive. In my opinion, this bar was strained to its 

limit as soon as the piano was tuned at the factory. It held 

for the first few years, and then for the final decade of the 

piano’s life while the string tension slowly decreased as the 

piano was neglected and allowed to go flat. Then, when the 

piano was finally tuned back up to 

standard pitch, the weakened bar 

broke. 

Conclusion 

In my opinion the plate in this piano 

broke as a result of inferior and faulty 

design. It would have broken, no 

matter who tuned the piano. Fortunately, 

piano plates rarely break, but 

when one does, it is not uncommon 

for it to do so some years after the 

piano was built, and the cause is almost 

always a fault of some sort in the plate. 

Stress fatigue in metal does occur with 

the passage of time, and microscopic 

cracks do grow until the part finally 

fails. In my opinion Marnie Squire was 

not at fault in any way, and to conclude 

that the plate in this piano broke 

because of any negligence or improper 

technique on her part would be 

utterly absurd.

Welding Cracked Plates: 

A Proven Method 

By Wilford Young, RPT 

Salt Lake City, UT Chapter 

A 

broken plate can be welded if one wants to go 

to all the work. First, one must determine how 

much the plate has warped or changed shape. It 

makes no sense to weld the plate in its broken 

configuration. That’s like tuning a low-pitched piano 

“where it’s at.” The problem with welding a plate “where 

it’s at” is that forevermore the piano will be in that same 

shape — the shape of “where it’s at.” 

In grands, the most common break results in the plate 

and pinblock dropping a few millimeters. Sometimes the 

pinblock will make contact with the action. Removing the 

action when this has happened can be a real challenge, but 

with persistence it can be done. Loosening the strings is a 

must. 

Preliminaries 

To begin the repair, unhook the strings from their hitch pins 

in the area to be worked on (about two octaves). Tie them 

to one side. This is required before grinding, drilling and 

welding begins. Once the action is removed, a small hydraulic 

jack can be used to raise the plate back to its original 

position. 

In vertical pianos, the most common break occurs midsection 

just behind the keybed. The plate will likely have 

moved forward to where it has lodged against the action 

A typical break in a vertical piano. Straps of steel were used for reinforcement. 


and keybed. It will have pulled a nearby nosebolt out of its 

supporting post — stripping the threads. 

I use a jig to pull vertical plates back into their original 

shape. (See Figure 1) After releasing string tension and 

removing the strings in the break area I drill a 3/4" hole 

through the soundboard. Then I insert a long 3/4" continuous-threaded 

bolt 

through the hole and 

through blocks placed in 

the front and back of the 

piano, and begin pulling 

the plate back to where 

it belongs. The correct 

position is determined 

by checking the 

downbearing of a string 

laid across the bridge and 

fastened to its hitch pin. 

In some cases it may not 

be possible to bring the 

plate back that last one 

or two millimeters 

without danger of 

breaking the plate in a 

different place. But don’t 

throw the piano away, 

View showing the completed weld and how the straps 

were placed across the break. 

even if you are unable to achieve positive downbearing. 

After all, Vladimir Horowitz’s Steinway had negative 

downbearing during his later years of concertizing. 

Just try closing the gap in the crack as much as possible. 

One will find that a lot of force is required to move the 

plate even a small amount. As the plate moves closer to 

position, the nut on the jig will get so tight that it nearly 

strips the threads of the 3/4" bolt, so don’t try using a 

smaller bolt. 

Preparation for the Weld 

Proper procedure requires that a “V” groove be made where 

the weld is to take place. Cast iron is easy to grind and drill. 

A small hand-held, high-speed grinder works quickly. If 

limited space prevents use of a grinder, then try a 3/8" drill

Shows how a groove is cut into the broken strut to allow the weld to 

bond the two broken parts together. 

with various size bits (sharpened). 

The ideal situation is to have the broken plate removed 

from the piano and delivered to one’s shop where it is free 

from encumbrances. Another advantage of this is that it 

allows the plate to spring back into its original shape, 

eliminating the need for jigs, etc. Most of my welding, 

however, takes place in the customer’s living room. 

To protect the piano from sparks, get some old towels or 

cloths at a thrift shop. Dampen them (but not dripping wet). 

Spread them out to protect the keybed from beads of hot 

metal. Also, stuff damp cloth between the plate and the 

wood behind it (bridge and soundboard). One can safely 

weld a half-inch away from wood if there is a wet cloth in 

between. Protect the customer’s floor with sheets of plywood. 

Open some doors and windows for circulation of air. 

Welding creates a certain amount of smoke, but ordinarily it 

is not enough to leave a smell in the house when you are 

gone. 

For a welding machine, I use a 180 amp, 220 volt stick 

welder. I connect to the clothes-dryer outlet in the house. 

Most of the time I need an extension cord (No. 8 wire). 

Mine is 50 feet long. A word of caution: the clothes-dryer 

outlet is not suitable for continuous heavy-duty welding. 

But since we are welding in short intervals of just a few 

seconds at a time, this should not be a problem. 

It would be a mistake for a person to attempt welding a 

plate without any previous welding experience. If you are 

inexperienced, it is best if you go over the welding part with 

someone who has done cast welding before. 

The Actual Weld 

Take your time. Weld only 3/8 inch to _ inch at a time, then 

quickly peen the weld metal with a ball-peen hammer. This 

causes the metal to “flow” as it shrinks during cooling, at the 

same time adding strength to the weld and reducing the 

chance of small cracks showing up. Before proceeding, let 

the weld cool until you can place your hand on it. If you 

proceed faster, cracks will occur somewhere else in the plate, 

maybe as far away as the opposite end. Do not try to speed 

up the cooling process with a fan, or cold cloths or water. It 

will mean failure. 

What happens if you get in a rush? A friend asked me to 

weld his initials on the cast-iron lid of his Dutch oven. I 

asked him if he wanted it done in the manner I do piano 

plates (very slowly and costly) or a quick blacksmith job 

with cheap welding rod. He chose the fast method, where I 

would weld continuously nonstop till finished. So I did it. 

When I finished, I observed that the lid was laced with 

spider-web cracks. I was surprised the lid did not shatter, but 


Figure 1 — Jig for pulling vertical plates to original shape. 


Welding Cracked Plates 


the outer rim was holding it together – good enough for 

making stew, we decided. 

A special welding rod is used for doing cast iron. It 

contains mostly nickel and is not cheap. I can circle my 

fingers around a $200 bundle. The reason nickel is good is 

because of its flowing characteristics. It also shrinks less and 

is more ductile than iron, and bonds well. 

The complete repair. (Steinway grand) 

I use 1/8 th -inch welding rod. I set the current at between 

100 and 120 amps. 

Reinforcing the Repair 

To ensure that the repair will not fail, I reinforce it with a 

strap (or more than one) of cold-rolled steel, usually one 

inch wide, 1/8" thick and about four inches long. If the 

crack is a long one (more than five inches long) I will use 

more than one strap. I do this to ensure that the repair will 

be stronger than the original plate. Simply to weld the break 

without reinforcement leaves the chance that it will break 

again close to where it broke in the first place. One never 

knows for sure just what caused the plate to break. On 

occasion I have seen small bubbles in the break area – a 

result of faulty casting at the foundry. Plate failure is not a 

respecter of brand names; it happens even with the best of 

them — Steinways, Bechsteins — the spectrum runs from 

large to small, expensive to cheap. 


I smooth off the burrs and bumps with a grinder. In a 

grand, where the break is exposed to view, I finish the job 

with Bondo in the fashion of doing an auto body repair. 

Feathering off the outer edges, sanding, then spraying with 

gold paint, completes the job to where one would not be 

aware of any repair. In a vertical piano, I may omit the 

Bondo application and simply end the repair by smoothing 

the weld with a grinder then spray painting. 

Success Rate 

This technique has worked for me for more than 40 years. 

In all this time, I have never had a single call telling me that 

the repair failed. I have seen welds by other people fail and I 

have learned from their mistakes. I allow three working days 

to do a welding job and charge according to the amount of 

time taken away from my regular tuning appointments. To 

do the job right requires moving a pickup-load of equipment 

to the job site. 

How About Brazing? 

Brazing is a process in which the bond is made by using 

melted brass. This requires taking the plate out of the piano, 

putting it in a furnace and heating it to red hot, then 

through an access hole in the furnace door performing a 

brass braze using an acetylene torch. It is then allowed to 

cool slowly to room temperature. Since brass is not as strong 

as cast iron, the repair will likely fail. I have seen it happen 

several times and have been asked on two occasions to redo 

the repair correctly. 

To try to arc weld over a brass repair is futile. It can’t be 

done unless all the brass is first removed. Brass will vaporize 

(almost explode) when subjected to the heat of an arc 

welder. 

Other Types of Plate Repair 

One plate that I was asked to weld had previously been 

bolted together in an attempt to keep the parts from 

shifting. But the procedure had not worked because the 

string tension was simply too great for the plate to sustain a 

tuning. 

I have also heard of another repair method called “lockstitch 

welding.” It is unusual in that no heat is used. Not 

having seen it in operation, though, I can’t comment on its 

success. All I can say is that I would be afraid to trust it. To 

me, the work involved in a plate repair is too great an 

expenditure of time and effort to risk having it end up in 

failure and getting a call telling me that it did not work.

A Double-Safe 

Engineered Plate Repair 

By Richard Oliver Snelson 

Central Illinois Chapter 

Those of us that have heard a piano plate crack know the 

sick, depressing feeling that comes at the end of the great 

chord that sounds as the relaxing strings and cracking plate 

shout in fortississimo relief. It would be even more depressing 

if you had written a rather large check for the piano the 

day before. I’m sure we would feel just as bad if the piano 

belonged to a customer. In this business we pick ourselves 

up and go to work. Right? 

Searching through back issues of the Piano Technicians 

Journal, I found a number of articles on plate repair – most 

of them using welding or bracing with bolted on steel 

gussets. A lot of these articles had notes from the technical 

editor that piano plate repair isn’t always successful. That 

wasn’t too encouraging, but probably true. I had met 

Wilford Young at the Pacific Northwest Conference in 

Provo, UT, and had discussed with him a previous repair 

that I had completed on a Cable Conover plate. Wilford has 

a lot of expertise in plate repair and gave me some good 

suggestions on plate welding. From this information and my 

own experience as an engineer I started to design a new 

approach to plate repair. 


Photo 2 

Photo 1 Photo 3 

My Design Criteria 

1. Repair the plate in the piano, with only a few strings 

and dampers removed. This is important because of the 

large cost involved in removing, reinstalling and stringing 

the plate. 

2. The repair should not rely solely on welding or castiron 

stitching but should use both methods. The 

welding should remain simple and able to be done by 

any good local welder using a high nickel rod.

Photo 4 

3. The repair should blend with the plate and not stick out 

like a sore thumb. 

The Repair 

The piano was a Baldwin 6’3” and had a cracked web at the 

juncture of the third strut from the bass end where it joins 

the primary bass strut which is held down by the nose bolt 

and cap nut. Advice from Delwin D. Fandrich was to find 

what may have caused the break in the first place and fix 

that first. After a careful search I found the bass strut nose 

bolt was totally floating in the woodblock that provides 

threads and support for it. I had checked the nut for tightness 

before I started, but had not checked the bolt’s tightness 

in the wooden block. Loss of the bolt’s hold-down power 

let the strut rise as the piano was brought to pitch. This 

small rise was enough to crack the joining strut. The block 

was replaced with a new hard maple block with the bolt 

screwed into position. 

Next, I removed strings and dampers on each side of 

the strut and protected the remaining strings with cardboard 

around the repair area, using old blankets to cover everything 

else on the piano except 10 inches of the broken strut. 

A special heat-resistant fiberglass welder’s material was 

pushed under the strut and over the protected strings. A thin 

steel plate was then slipped under the strut and on top of 

the fiberglass material. The piano had been moved to a 

concrete floor away from all carpet and flammable material. 

This article is not intended to be a tutorial on the use of 

metal stitching pins. However I will cover the basic process, 

which will give you a good idea as to what’s involved. For 

detailed information on Lock-N-Stitch, contact the Lock- 

N-Stitch International Company in Turlock, CA. Their 

phone number is 800-736-8261. They have a very good 

video that explains and demonstrates the process. The 

special taps, drills, spot facer and pins to complete a repair 

run around $150. I purchase backups of those items because 

of problems with breaking both the drills and taps. 

The next step in the repair was to install metal stitching 

pins down the length of the crack. Each pin is installed so 

that it overlaps the previous pin by a small amount. Pin 

length is selected so that it goes nearly through the plate 

strut. 

The Steps for Installing the Pins 

1. Drill the required hole using the bit furnished by Lock- 

N-Stitch. (See Photo 1) It’s important to maintain 

the drill at right angles to the plate web. A good drilling 

lube is used for each of the following steps. 

2. Determine the required depth for the pin’s shoulder 

and adjust the stop on the spot facer. Again with the 

drill at right angles spot face the hole. 

3. Clean out the hole using compressed air and then using 

the supplied Lock-N-Stitch tap cut the threads in the 

hole. Not holding the tap at right angles as it starts will 

quickly ruin the hole! 

4. Clean out the hole again and now the pin is screwed 

into place. (See Photo 2) Coat it with a liquid locking 

solution like Loc-Tight before driving it. As the shoulder 

seats in the spot face, the pin is tightened until it 

shears off at the thin neck placed there for just that 

purpose. This leaves about 1/8” of the head of the pin 

sticking out of the hole. Use a grinder on this and 

simply grind in flush with the plate web. To solidly seat 

the shoulder in the spot face I use a prick punch and 

tap it several times around the outside edge. 

5. Continue installing the metal stitching pins down the 

crack; each pin installed so that it overlaps the previous 

pin. (See Photo 3) 

Photo 5 


A Double-Safe Engineered Plate Repair 


Photo 6 

A few words of caution about metal stitching – I had 

learned the hard way that the cast-metal pieces to be joined 

using metal stitching (Lock-N-Stitch®) must be held tightly 

together or the seam will grow wider as the pins are installed. 

C-clamps are not enough. Also, attempting to metal stitch a 

plate area that had been previously welded is very difficult 

because of the extreme hardness of the cast around the weld. 

Drilling for the pins is all but impossible. 

From reading the “pianotech” discussions on plate repair, 

I understand that individuals are using only Lock-N-Stitch 

for plate repair and have been successful with it. I wanted to 

go beyond metal stitching and build the plate up to where it 

would be as strong or stronger than when it came from the 

factory. This would involve a gusset and some welding. 

Steps for Welding 

1. A 1/8”-thick metal gusset was cut from mild steel to 

extend several inches on each side of the plate crack. It 

was then ground to provide 1/4" welding slots in a 

horizontal direction. Each slot would allow a 1” long 

horizontal weld. Slots were placed at the top and bottom 

of the gusset on both sides of the crack. 

2. The gusset was then fit to the plate strut by heating it 

and pounding with a ball peen hammer. Very little 

shaping was needed for this repair. 

3. The gusset was then clamped to the plate strut and holes 

were drilled and spot faced for Lock-N-Stitch pins. 

These pins would hold the gusset in place for welding, 

plus add a lot of additional strength to the repair. After 

tapping the holes, each pin was installed through the 

gusset. After grinding them off, nothing showed on the 

sides. No bolts sticking out! (See Photo 4) 


4. Next came the welding using a high nickel rod and 

electric arc. Each one inch weld was completed and 

immediately peened across its length with a pointed 

welding hammer. I think this tends to cut down on high 

stress areas in the weld and surrounding cast. Take your 

time at this point and let each weld cool before doing 

the next one. Since each weld is at right angles to the 

strut crack and over 1” long they have a lot of strength 

and will solidly hold the gusset in place. (See Photo 5) 

With the welding complete and a then a little grinding 

to clean up both the welds and gusset, it was time to test the 

repair. It took about two hours to remove the protective 

covers, replace the strings and bring the piano to pitch. (See 

Photo 6) Thanks to the good help from Del Fandrich and all 

the individuals that had written Journal articles on plate 

repair, it held. 

To complete the repair took several more hours to 

contour the ends of the brace and flatten the tops on each 

side of the welds. Auto-body putty and lot of elbow grease 

Photo 7 

did the job. (See Photo 7) I sprayed a coat of red-oxide 

primer on the steel and then finished with a coat of gold 

paint. This didn’t quite match, but will be changed later 

when I put a new pin block in the piano. Several piano techs 

have looked at the piano and each has asked, “Just where did 

you repair it?” If they can’t find it, that’s good enough for me. 

I don’t mean to hide the fact that a plate repair has been 

done. I place an engraved plaque on the plate stating that this 

piano and plate has been repaired by Oliver Piano Services.

Photo 1 

Unbelievable! 

By Carman Gentile, RPT 

Redwood, CA Chapter 

Sometimes an experience will confirm what 

we already know: that a properly designed 

and installed piano plate is a lot stronger than 

it needs to be. 

I received a call from the owner of a 1906 Starr 5’4" 

grand (#64759). She said that in 1995 she removed all 

the plate screws and nosebolt nuts in the sincere but 

mistaken belief that she could lift out the “harp” to 

clean the soundboard. She gave up trying to lift the 

plate and eventually all the fasteners were lost. Now, 

five years later, she called me to ask if I could replace 

the fasteners and tune the piano. She said that another 

piano technician “almost fainted” when he saw the 

unsecured plate and, after lecturing her, refused to 

undertake the task of replacing the fasteners. I also 

wasn’t sure if I wanted to undertake this project, but 


decided I had to at least see this piano myself. 

When I arrived at the client’s home, the piano was 

indeed just as described (See Photo 1). My examination 

revealed no cracks or fractures, the plate resting 

securely on the spacers, (See Photos 2 & 3) and the 

pitch 50 cents flat. Imagine my astonishment when my 

new client added that the piano had been moved three 

times since the fasteners were lost in 1995. 

Somehow this piano’s plate had survived having all 

14 plate screws and seven nose bolt nuts removed while 

under tension and then tipped and moved three times 

over a five-year period. This piano’s history was as 

unbelievable as it was unprecedented. By now the 

owner was aware she had “made a terrible mistake” 

and readily agreed to sign a statement I drew up which 

states that “...the plate may crack or break at any time.” 

Then I measured the diameter and depth of each plate 

screw hole and the diameters and thread counts of the 

nose bolts and also took a few photographs. 

My next step was to consult my colleagues for 

advice and locate replacement fasteners. I described 

this piano on the PTG Archives message board and 

received an informative and encouraging e-mail from 

Del Fandrich. The members of my PTG Chapter were 

amazed and amused at the unprecedented nature of 

this project and encouraged me to pursue it. One of 

our senior chapter members wisely saves old plate 

screws and I was able to procure the correct sizes from 

his collection. The nosebolt nuts were another matter. 

The nut size needed was 5/16" by 20 threads per 

inch. My search for that particular thread count was 

fruitless; hardware stores have sizes that are 5/16" by 18 

or 24, but not 20. I could not even find that size in the 

catalogs. Fortunately, the owner of the piano works 

with an engineer who had the correct size tap and he 

created the nosebolt nuts for her. 

Thus I re-approached this piano armed with

Photo 2 (TOP) and Photo 3 (ABOVE) — Mirrors show plate resting securely on 

spacers. 

correct fasteners, a signed release from my client and 

encouragement from my colleagues. I lowered the 

string tension, installed all 21 fasteners, performed 

three pitch raises, tuned it to A=440 and my client 

proceeded to play her beloved Starr at concert pitch. 

(See Photo 4) Three weeks later my follow-up visit 

revealed the piano needed a touch-up tuning, but was 

still up to pitch. 

Herewith is the e-mail message I received from Del 

Fandrich after I described this unbelievable project: 

Actually, Carman, this is not quite as unbelievable 

as it seems. Starr mostly built relatively small pianos, so 

I’m assuming this one is something under 5' 7" (170 

cm). Regardless of its size, few pianos of this type were 

really ‘designed’ or engineered as is commonly thought. 

That is, many of them were essentially copies of something 

else that the builders saw and liked. Or saw and 

thought they could make some money on. This has 

long been a problem in the piano industry. Wolfenden 

recognized it in 1916, “Some pianoforte makers are 

unwilling to incur the trouble and cost of properly 

calculated and drawn scale designs. The more usual 

way is to copy instruments of suitable dimensions, 

which seem to possess pleasing qualities. Naturally this 

method of procedure not only reproduces such errors 

of design as there may be in the instruments copied, 

but probably introduces others.” (Wolfenden, Samuel, 

A Treatise on the Art of Pianoforte Construction, 1916, 

Unwin Brothers, Ltd., Surrey, England, pp. 4 & 5.) A 

plate would be measured and, if there was any doubt 

about its strength, a bit more iron would be thrown 

in. Sometimes – a lot more. Not much has changed 

over the decades — many pianos are still “designed” 

this way. Piano plates, especially those in very small 

pianos, are really intended to support the entire stress 

of the string load unless they are equipped with “horns” 

such as those found in the Steinway, etc., plate designs. 

These devices actually do couple some of the string 

load to the belly bracing mechanism. Otherwise, all of 

the string tension stress is held by the plate. Its fastening 

to the rim only serves to stabilize it. In some cases 

it would even be possible to remove the pinblock 

screws without causing catastrophic failure. (No, I am 

not recommending doing this — just suggesting that 

in some cases it probably would be a reversible procedure.) 

Anyway, good luck to you and your fastidious client. 

I sure hope that soundboard got clean. 

Del 

Photo 4

Download PDF - Piano Technicians Guild

Create successful ePaper yourself

Delete template?

Save as template?