SPEAKERS - Goldmund

Goldmund White Paper
SPEAKERS
The materials, manufacturing techniques, and design concepts behind Goldmund
speakers
When designing speakers, Goldmund engineers use
computer-based modeling, which takes into account
the characteristics of the speaker drivers and
enclosure and allows them to perfect the speaker’s
acoustical performance. While we do listen to various
prototypes during the design process to judge the
efficacy of engineering changes, we design and build
our speakers primarily to meet strict technical
standards, not to suit the taste of a particular
listener’s ears.
Goldmund employs numerous advanced engineering
and manufacturing techniques in the development
and production of its speakers.
Heavy aluminum enclosures minimize extraneous
resonances and vibrations. Mechanical grounding
techniques route what little vibration is left into the
floor, where it is immediately dissipated. Modular
construction further decreases vibration and
enhances mechanical grounding, while making
future upgrades practical and simple. And active
amplification options improve performance and
system versatility.
“Perfection in an imperfect world”

Aluminum enclosures
Extraneous vibration is one of the most challenging
problems in loudspeaker design. The intent of the
speaker designer is that only the cones of a speaker’s
woofers and the domes of its tweeters produce sound
waves. If a loudspeaker cabinet vibrates along with the
woofers and tweeters, it contributes unwanted sounds of
its own. The unwanted sound waves may partially cancel
certain audio frequencies and reinforce others, creating
an uneven frequency response and an unnatural sound.
A vibrating speaker cabinet also acts as an acoustical
“spring,” absorbing sound from the back of the woofers
then re-radiating it into the cabinet and out through the
ports to the listener.
The result is that sounds “ring”. In other words, the
speaker continues to emit a musical note long after the
note was supposed to end. This emission has two
deleterious effects: It smears the musical note, spoken
dialogue, or special effect being played; and it obscures
the attack of the sound that follows.
Most speaker enclosures are built from medium-density fiberboard (MDF), a wood-based material
that, while much less resonant than natural wood panels or plywood, still vibrates relatively
easily.
Goldmund has chosen to build its speaker cabinets from a superior, although much more costly
material: dense slabs of aluminum measuring at least one-half of an inch thick. Aluminum bracing
inside the speaker cabinet further reduces what little vibration exists.
The use of aluminum results in a speaker that is heavier and denser than most speakers of
comparable size. The result of this robust construction is a more natural, clean, and uncolored
sound.
“Perfection in an imperfect world”

Modular construction
All Goldmund speakers are built in modular fashion, which
both improves their sound and allows easy upgrades later
(Signature versions). Each speaker module is sized ideally
for its purpose, and using multiple smaller cabinets
reduces resonance and vibration when compared with a
single larger enclosure.
If more modules are desired—to add, perhaps, a centerchannel
speaker, or to augment bass response—multiple
modules can be combined in metal rack systems. These
racks hold each module in the proper position for ideal
acoustical performance and firm mechanical grounding.
The Epilogue system offers great expansion capabilities.
The system’s core component is the relatively small
Epilogue 1 speaker or Epilogue 1 Signature.
For more bass response, Epilogue 2 or Epilogue 2 Signature bass modules may be added. The Full
Epilogue system extends bass response even deeper with the addition of the Epilogue 3 Signature
subwoofer modules and an additional Epilogue 1 in each rack to increase the acoustical output for
stereo listening, and to reproduce center-channel sound in home theater systems.
Internal amplification
Passive speakers are the best choice for those who prefer to select the power and quality of their
amplifiers. These speakers also allow easy upgrading of a system to more powerful amplifiers
when the need for greater output and sonic precision arises.
The Goldmund Metis 10 Acoustic Processor and Metis Active Speakers
“Perfection in an imperfect world”
Active speakers are best for
those who prefer to leave all
of the specification to
Goldmund. The amplifiers in
these speakers are designed
expressly to meet the demand
presented by the speaker
drivers. Goldmund speakers’
enclosures make comfortable
homes for these amplifiers;
the extraordinary thermal
mass of the thick aluminum
enclosures eliminates the
need for amplifier ventilation
and cooling fins.

Because of the additional demands placed on subwoofer amplifiers, Goldmund’s Epilogue 3
subwoofer module is internally powered to ensure maximum performance and reliability.
Goldmund Media Room applications
Although most of Goldmund’s speakers are designed for freestanding use, a new line of modular
speakers has been created for in-wall and home theater applications. The available modules are
the U-Logos 1 main speaker, the U-Logos 2 woofer module, and the U-Logos 3 subwoofer.
All of the modules are designed to be mounted in a wall, with an acoustically transparent
decorative fabric of the homeowner’s choice covering the speaker. The shapes and dimensions of
the U-Logos speakers may also be customized to fit into any space with sufficient volume.
The U-Logos Media Room speakers are unusual in that they are specifically designed for use with
an external power amplifier/crossover. Goldmund has created a Media Room amplifier that
combines amplification circuitry of the same quality found in Goldmund’s freestanding amplifiers
with a digital signal processor that performs crossover functions. Each driver in each Media Room
speaker is assigned its own amplifier channel, with a power rating of 580 W if the amplifier is run
in mono mode, and 290 W per channel if the amplifier is run in stereo mode. The DSP contours
the sound to suit that driver’s characteristics—i.e., tweeters receive only high-frequency sound,
while woofers receive only low frequencies.
Related web pages:
Goldmund Epilogue Speakers:
http://www.goldmund.com/products/epilogue_1
http://www.goldmund.com/products/epilogue_1_2
http://www.goldmund.com/products/epilogue_full_system
Metis Speakers and Subwoofer:
http://www.goldmund.com/products/metis_speaker
http://www.goldmund.com/products/metis_sub
“Perfection in an imperfect world”

Goldmund White Paper
SPEAKERS
MODELING PROCEDURE
With the aim of ensuring the perfect
functioning of our loudspeaker systems, we
have implemented a modeling procedure
enabling any component, as well as the whole
systems themselves, to be under control.
This procedure allows us to be very efficient in
case of driver and filter modification or
replacement. On top of that, it allows any
possible problems encounter by our dealers to
be analysed with the aim of correcting them.
Finally, it will constitute a sound theoretical
basis for any of our future loudspeaker system
development.
“Perfection in an imperfect world”

The procedure comprises 5 steps, which are explained below with examples:
1. Thiele and Small parameter measurements (added-mass method)
2. Preliminary calculations:
- Electrical, mechanical and acoustical lumped-constant parameters
- Input impedance, volume velocity and membrane displacement (infinite baffle,
closed-box, vented-box...)
- First estimation of drivers and port radiating sound pressure
3. Crossover modeling and transfer function calculation
4. Modeling comprising crossover, driver and enclosure lumped-constant parameters:
- Input impedance with and without cross-over
- Volume velocities with and without cross-over
5. Final loudspeaker system calculations:
- Introduction of the modelled filtered volume velocities into calculation code
- Calculation of the loudspeaker system sound pressure in amplitude and phase,
taking into account driver finite radiating surface and diffraction at the enclosure
edges
To sum up, this procedure offers various possible loudspeaker response calculations, allowing our
systems to be studied and analysed in different ways depending upon the application under
consideration:
- Loudspeaker system input impedance (amplitude and phase)
- Electrical impedance at any location, like for example at the driver terminals
- Volume velocities in amplitude and phase (active and passive drivers, port…)
- X-over transfer functions
- Sound pressure level valid in the near field as well as in the far field, and taking into
account the diffraction at the enclosure edges (GTD and UTD methods)
- Sound pressure phase and group delay (active and passive drivers, port,
loudspeaker system)
“Perfection in an imperfect world”

STEPS DETAILED EXPLICATIONS WITH EXAMPLES
1. Thiele and Small parameter measurements (added-mass method)
This well-known method enables the TS parameters to be measured without test enclosure. The
measurement is carried out manually and/or automatically to ensure a high level of accuracy.
The figure below shows an example of automatic measurement (Logos Sub active driver).
10
8
6
4
2
0
10 1
10 2
LOGOS SUB
Coming from the average of several driver measurements, the TS parameters (Re, Le, fs, Qms, Qes,
Qts, Vas) are finally compared with the constructor ones before being introduced in calculation
code.
“Perfection in an imperfect world”
10 3
10 4
Hz

2. Preliminary calculations
First we need to determine:
- The force factor Bl
- The mechanical parameters linked to the mobile system: mass ms, resistance Rms and
compliance Cms
- The acoustic equivalent parameters of enclosure, port and losses due to leakage: Cab,
map, Rleak
- The acoustical radiation impedance mar and Rar
These lumped-constant parameters enable input impedance, membrane displacement, as well as
drivers and port volume velocities to be calculated. According to the hypotheses of membrane
coincidence and monopole radiation, they lead to a first estimation of drivers and port radiating
sound pressure.
The figure below shows the example of the Minilogos calculation results:
ohm
m3/s
35
30
25
20
15
10
5
0
6
5
4
3
2
1
0
- Drivers input impedances
- Drivers membrane displacements
- Port and drivers volume velocities
- Port and drivers radiating sound pressure estimated at 1W/1m
DRIVERS INPUT IMPEDANCE
10 2
x 10 -3 PORT & DRIVERS VOLUME VELOCITY
10 2
10 3
Hz
10 3
Hz
10 4
10 4
mm
dB SPL
2.5
“Perfection in an imperfect world”
2
1.5
1
0.5
0
100
95
90
85
80
75
70
65
60
55
50
DRIVERS MEMBRANE DISPLACEMENT
10 2
PORT & DRIVERS SOUND PRESSURE at 1W/1m
10 2
10 3
Hz
10 3
Hz
10 4
10 4

3. Crossover modeling and transfer function calculation
The above calculation of port and drivers sound pressure radiated at 1W/1m, enables the
crossover cut-off frequencies and slopes to be estimated.
The figures below show the example of the MiniLogos crossover modeling and transfer function
calculation.
Vm
Vtw
Rg
1 2
0
Rg
1 2
0
V+
V+
1
Cm
2
1
Rm
2
2
1
Cm1
1
2
Rm2
1
Lm
Rem
2
V+
2
V- V-
1
Rtw
2
1
Ctw
1 2
1
Ltw
2
Rtw2
2
V- V-
1
1
V+
Retw
2
10
dB
0
Hz
10 100 1000 10000 100000
-10
-20
-30
-40
-50
-60
-70
-80
4. Modeling comprising crossover, driver and enclosure lumped-constant
parameters
The points 2 and 3 lead to the final system modeling. The latter enable the filtered input
impedance and volume velocities to be calculated.
The figures below show the MiniLogos final modeling and the resulting calculated filtered input
impedance and volume velocities:
Vwr
Rg0r
Rtwr
0
Cmr
Rmr
Ctwr
Rem0r
Cm1r
Rm2r
R3r
Ltwr
R2r
Lmr
Lemr
+ -
-Bl
Bl
+
-
msmr
Rsmr
Csmr
+
-
+
-
Sd
Rtw2r Re_twr Letwr
mstwr Rstwr Cstwr
martwr
“Perfection in an imperfect world”
+ -
0
Bl
+
-
-Bl
0
-Sd
marmr
+
-
+
-
Sd
Cabwr
0
-Sd
Rapwr
0
mapwr

ohm
deg
deg
35
30
25
20
15
10
5
0
-50
m3/s
50
10 1
0
10 1
6
5
4
3
2
1
10 1
0
150
100
50
0
-50
-100
-150
10 2
10 2
10 2
x 10 -3 VOLUME VELOCITIES (medium & port)
10 1
10 2
MINILOGOS INPUT IMPEDANCE
10 3
10 3
Hz
Hz
deg
m3/s
10 3
0
10 3
10 3
“Perfection in an imperfect world”
6
5
4
3
2
1
150
100
50
0
-50
-100
-150
10 4
10 4
x 10 -5 VOLUME VELOCITIES (tweeter)
10 3
10 4
10 4

5. Final loudspeaker system calculations
The modeled volume velocities of drivers and port are introduced in the calculation with the aim
of calculating the sound pressure radiated by the loudspeaker system.
The sound pressure is then calculated in amplitude and phase taking into account the driver finite
radiating surface, the drivers and port locations in the enclosure and the diffraction at the
enclosure edges (GTD / UTD methods). The figure below shows the MiniLogos sound pressure
level at 1W/1m with and without box edges diffraction calculation:
dB SPL
dB SPL
deg
100
80
60
40
20
0
100
80
60
40
20
0
3
2
1
0
-1
-2
-3
Without edges diffraction
10 2
10 2
10 2
Hz
Hz
Hz
10 3
10 3
10 3
10 4
10 4
10 4
dB SPL
dB SPL
deg
100
80
60
40
20
“Perfection in an imperfect world”
0
100
80
60
40
20
0
3
2
1
0
-1
-2
-3
10 2
10 2
10 2
With edges diffraction
Hz
Hz
Hz
10 3
10 3
10 3
10 4
10 4
10 4

The irregularities are calculated in order to compensate those of the driver. Thus, in case of a new
loudspeaker system development, it is necessary to measure the driver frequency responses
(baffle assembly). It is also necessary of course to measure the final product in an anechoic
chamber in order to validate the calculation procedure.
Finally, it should be noted that the sound pressure phase derivative enables the group delay to be
calculated with a great accuracy. This last loudspeaker response is very important for the
development of in-phase drivers and analysis of fast response systems.
ms
18
16
14
12
10
8
6
4
2
0
10 2
MINILOGOS GROUP DELAY CALCULATED at 1m
Hz
10 3
“Perfection in an imperfect world”
10 4