Professional Documents
Culture Documents
EncShop APP
EncShop APP
EncShop APP
Application Manual
Release 5
Enclosure Designer i
Application Manual
Contents
LEAP EnclosureShop
Application Manual
This document was produced on a Pentium-4 / 2GHz PC with Win2K using Adobe PageMaker 7.0, Adobe
Illustrator 8.0, Adobe PhotoShop 6.0, MathType 4.0 for mathematics typography, and SnagIt 5.2 for screen
captures. Final masters were produced using an Xerox Docutech image setter. Help files were composed and
compiled using Windows Help Designer 3.1.
ii Enclosure Designer
Application Manual
Contents
■ Copyright Ownership
Both the program and the documentation are protected under applicable copyright laws. LinearX is the holder of this copyright. Your right to use the program
and the documentation are limited to the terms and conditions described herein. Use of the software unless pursuant to the terms and conditions of this
license, or as otherwise authorized by law, is an infringement of the copyright.
You may not: (a) distribute copies of the program or the documentation to others, (b) lease, rent, grant sublicenses, or other rights to the program, (c) provide
use of the program in a computer service business, network, time-sharing multiple CPU or multiple users arrangement without the prior written consent of
LinearX, (d) translate or otherwise alter the program or related documentation without the prior written consent of LinearX.
■ Terms
Your license to use the program and the documentation will automatically terminate if you fail to comply with the terms of this agreement. Your license
terminates in the event that you receive a license for an updated version of the product that replaces this product. If a license expiration date is printed on your
documentation, or provided through other means such as a time limited electronic or software key, your license expires on the day as shown in the documen-
tation, or on the day that the electronic or software key expires. If this license is terminated you agree to destroy all copies of the program and documentation.
■ Limited Warranty
LinearX warrants to the original licensee that the disk(s) and or electronic key(s) on which the program is recorded will be free from defects in materials and
workmanship under normal use for a period of ninety (90) days from the date of purchase as evidenced by a copy of your receipt. If failure of the product
components has resulted from accident, abuse, or misapplication of the product, then LinearX or third party licensors shall have no responsibility to replace the
disk(s) or key(s) under this limited warranty.
This limited warranty and right of replacement is in lieu of, and you hereby waive, any and all other warranties, both expressed and implied, including but not
limited to warranties of merchantability and fitness for a particular purpose. The liability of LinearX or third party licensors pursuant to this limited warranty shall
be limited to the replacement of the defective disk(s) or key(s), and in no event shall LinearX or third party licensors be liable for incidental, indirect, punitive,
or consequential damages, including but not limited to loss of use, loss of profits, loss of data or data being rendered inaccurate, or losses sustained by third
parties even if LinearX or third party licensors have been advised of the possibility of such damages. This warranty gives you specific legal rights which may
vary from state to state. Some states do not allow the limitation or exclusion of liability for consequential damages, so the above limitation may not apply to you.
In addition to the foregoing, you should recognize that all complex software systems and their documentation contain errors and omissions. LinearX, its
distributors, and dealers shall not be responsible under any circumstances for providing information on or corrections to errors and omissions discovered at any
time in the product, whether or not they are aware of the errors or omissions. LinearX does not recommend the use of this product in applications in which
errors or omissions could result in loss of life, injury, or other significant loss.
This license agreement shall be governed by the laws of the state of Oregon and shall inure to the benefit of LinearX, its successors, administrators, heirs and
assigns or third party licensors.
Copyright 2002, LinearX Systems Inc. All rights reserved. All other Trademarks are the property of their respective owners.
Technical Support
LinearX provides detailed printed manuals and on-line help within the program as the primary
source for user information and assistance regarding the use of this product. If these sources do
not contain the answers to your questions, contact LinearX via any of the following methods:
Technical support is free and unlimited at this time, however we reserve the right to charge for
this service in the future as conditions, overhead, and support personnel requirements dictate.
When contacting us regarding a technical support issue, PLEASE follow these steps to aid us in
understanding and solving your problem:
(1) The About Box contains a procedure for generating a SYSCONFIG.TXT file. This file can be created by the user
through the About Box and contains all of the information about your computer system and operating system. If you
feel that your question could involve issues relating to your computer/operating system, please produce this file and
attach it along with your fax or Email question.
(2) If your question involves specific details or parameters unique to your project and problem, please include a copy of
your design files with the necessary data so that we can reproduce your problem. This is only possible if you are
communicating via an electronic means such as Email or uploading files directly to our web site.
(3) If the issue regards error messages from the program, please include an exact description of the error message
and/or address information that the program reports.
(4) If there are specific steps involved to reproduce the issue, please note these exact steps required so that we can
reproduce the problem.
Technical support hours are: Monday-Friday 9:00AM to 5:00PM Pacific Standard Time.
iv Enclosure Designer
Application Manual
Contents
Contents
Introduction ix
Enclosure Designer v
Application Manual
Contents
Contents
vi Enclosure Designer
Application Manual
Contents
Introduction
If you have not already read Chapter 2 in the Reference Manual, please do so before
starting the tutorials. It would also be a good idea to read Chapter 3 which will
acquaint you with the many parameter graphs in the program. These will give you
a quick overview of the program. This information will be useful in the tutorials.
Tutorials
There are several tutorials that follow which should help you get started using the
program in the shortest amount of time. Each tutorial introduces new features, and
assumes you are already familiar with the features introduced in previous tutorials.
Note: The first two tutorials introduce the core features of the program.
It is very important that all users read at least Tutorial #1.
It should be noted that there are many ways to operate the program, typically either
from the Main Menu or from the tool buttons. Program operation becomes much
quicker when you learn where a few of the most commonly used tool buttons reside.
You can scan the buttons and view the hints to learn more about their various
functions. The tutorials show the corresponding tool buttons for most of the
commonly used menu items.
There is a high degree of user control over the appearance of graphics in the
program. If you don't like the color schemes used in the examples - simply change
them!
Note: A full color version of the Application Manual is available in the Help file.
Tutorial 1
Sealed Highpass Enclosure
Highlights
Introduction to Transducer Models
Introduction to Enclosure Models
Introduction to System & Guide Curves
Introduction to Parameter Graphs
Differences between Transducer Models
Common Program Features
Objectives
For this design we will begin with the basics, and our enclosure will be a simple
Sealed Highpass box using a 15 Inch (380mm) woofer. We are not interested in
choosing a particular alignment for this example. Rather the box size will be
specified as already known, and the task will then be to setup the necessary
parameters in the program to simulate the design.
Even though this is a very simple enclosure, it will serve well to illustrate the large
volume of data that can be produced through the extensive analysis provided in
EnclosureShop. Comparisons with actual measurements will also be presented.
EnclosureShop 1
Application Manual
Sealed Highpass Enclosure Tutorial 1
Transducer
Before any design can be simulated the transducer model must already reside
in a transducer library file. If this was not the case, we would first need to create
the transducer entry. For this example several transducer models will be used
which have already been prepared in the Tutorial.Ltd library.
First we will create a new design and save it in the Tutorial-1 folder.
2 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
Your screen should now look similar to the lower view. When the graph windows
are maximized, a Graph Select button row appears. You can switch to different
graphs by clicking on one of the buttons below the toolbar. Note the various graph
names given on the horizontal row of selection buttons.
EnclosureShop 3
Application Manual
Sealed Highpass Enclosure Tutorial 1
Transducer Calculate
Parameters Analysis
Parameters
Enclosure 3D Layout
Parameters Parameters
Edit Menu
The Edit menu is central to the operation of the program. This menu contains
five items which can be accessed from either the menu itself or toolbar as shown
above. These five items represent the core functionality needed to define
parameters for any design. You will visit this menu frequently. To a large
extent, the five items must be specified in order from left to right when initially
defining the parameters of a design.
There is also a Graph toggle button located at the bottom which will extend the
dialog and present a preview graph showing curves of various transducer
characteristics for the selected transducer. The various curves can be selected
using the small buttons at the left of the graph.
There are three transducer entries shown in the right list view window that we
will be working with in this example: TL1603/LTD, TL1603/TSL, and TL1603/
STD. These represent three different transducer models of the same 15 Inch
(380mm) woofer.
Note: If you installed the program on a drive other than C:\ you will need to
open the transducer folder to see the transducer files and entries.
4 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
EnclosureShop 5
Application Manual
Sealed Highpass Enclosure Tutorial 1
- DoubleClick on the TL1603, LTD Model entry in the list view window.
This is the transducer editing mode. The dialog is altered to display the
parameters of the transducer. Transducers can be created, imported, or edited
here to define their parameters. Various fields will be enabled/disabled
depending on the model in use.
Since the transducer entries needed for this example are already in the library,
we will not be creating any new transducers here. It should also be noted that
this is not the location where you choose the transducer(s) used in the
enclosure. That is done within the enclosure parameters we will visit shortly.
However, it is important to select the library file here that contains the
transducer(s) you wish to use in the enclosure design.
6 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
Model Menu
The Model menu is used to choose the enclosure model for the design. The menu
and toolbars are shown above and contain eight (8) possible choices. The last
selection is a Custom enclosure model which can be used to construct arbitrary
enclosure structures.
The Model menu performs a 1-of-8 selection function. One of the selections will
always be active, and this selection is shown checked in the menu, or highlighted
in the toolbar. The dialog presented for the enclosure parameters will change
depending on this enclosure model selection.
For this example we will use the Sealed Highpass enclosure model.
EnclosureShop 7
Application Manual
Sealed Highpass Enclosure Tutorial 1
The remaining selections in the Transducer group box do not require changes.
In the Chamber group box, we define any filling material that may be used in
the enclosure. In this example no material will be used.
8 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
You should now see 3.53 Ft³ displayed as the net internal volume.
The last group box Domain allows us to define the environment in which we wish
to simulate the enclosure. We can also specify a finite volume for the domain. For
this example we will use an Infinite Baffle with Infinite volume.
EnclosureShop 9
Application Manual
Sealed Highpass Enclosure Tutorial 1
This is a large window dialog which fills the entire work area of the screen as
shown on the following page. The large plane represents the infinite baffle, and
the enclosure is embedded in that plane in the center. The radial arrows display
the polar simulation paths, and the primary target is shown at the intersection.
You can pan the view by holding down the Left Mouse button and dragging it
around the screen. As you rotate the scene you will see the other back/rear side
of the infinite plane. You can see the enclosure behind the plane.
The default locations of the entities within the scene should already be correct.
No layout editing will be required for this example.
- Click Ok to close the dialog.
10 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
EnclosureShop 11
Application Manual
Sealed Highpass Enclosure Tutorial 1
We have now defined all of the necessary parameters for the design. To analyze
the design and produce the response curves we use Edit | Calculate.
The progress of the analysis is shown in the lower right fields of the status bar
at the bottom of the screen. There are several phases to the analysis which
depend on the complexity of simulation.
In this example we have chosen an Infinite Baffle domain which places the
baffle board of the enclosure within the infinite plane. Therefore no diffraction
analysis is required. This speeds up the analysis greatly since diffraction
analysis can be very time consuming. The analysis here should be completed
in a matter of seconds.
System curves are the data curves generated by the system analysis. For this
simple design 34 curves are produced. Many of these are polar response
curves. We will enable all of the curves.
- Click the Show All button, and then click Ok to close the dialog.
12 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
■ Select the Scale | Auto or Up/Dn menu items as needed to view the curves.
Your screen should now look something like the view shown below. There are
four SPL curves displayed on the graph. Two are at a level of around 95dB while
the other two are much higher.
The two lower level curves give the response at the primary simulation point and
the overall power response. In this example the primary simulation point is On-
Axis (Polar angle: 0 Horz, 0 Vert) at a distance of 1 Meter. However it could be
located anywhere in space. The power response gives the total power produced by
the design, relative to a half space domain.
Note that at higher frequencies the power response falls due to the directivity of
the transducer. At lower frequencies the radiation is nondirectional with the Power
and On-Axis response identical. A large number of chamber reflections are shown
at the higher frequencies.
SPL 0H,0V
SPL Pwr
EnclosureShop 13
Application Manual
Sealed Highpass Enclosure Tutorial 1
Moving the scale up to see the two higher curves, we find one level reaching
almost 130dB and the other 117dB. These curves represent the internal cham-
ber response and the near-field transducer response respectively. Note that
the sharp primary chamber reflection peak occurs at about 550Hz and shows
up in all the Near field, On-Axis, and Power response curves.
Due to the large number of SPL curves produced from the enclosure simula-
tions, you will often find it necessary to shut off some of the curves to avoid
excessive clutter in the graphs.
There are also a number of other curves produced for every transducer and
port in the enclosure. Examples of these are shown on the following pages
and include: Impedance, Excursion, Velocity, and Acceleration.
14 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
EnclosureShop 15
Application Manual
Sealed Highpass Enclosure Tutorial 1
16 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
The polar response curves are shown in the graph below. Note that there is actu-
ally a double set of curves in the list. These are the Horizontal and Vertical sets.
Since the domain was an Infinite Baffle, the simulation is symmetrical and both
sets of Off-Axis polar response curves are identical.
The effect of the Infinite Baffle domain is clearly visible in the graph. No radia-
tion occurs behind the plane of the domain. All radiation appears on the front side.
It is also possible to display the polar response curves normalized to their values at
zero degrees. That is also a common method for viewing polar plots.
EnclosureShop 17
Application Manual
Sealed Highpass Enclosure Tutorial 1
18 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
The Guide Curve library should appear similar to the view shown below.
- Click Ok to close the dialog.
As shown on the following page, two acceleration curves are displayed. The two
curves show excellent agreement across the entire frequency range, with very tight
correlation below 200Hz. A sharp series of nulls occurs near 500Hz. This is a
standing wave reflection inside the chamber. Remember there is no damping ma-
terial in this example. At very high frequencies the diaphragm no longer behaves
as a piston.
As shown on the following page, two velocity curves are displayed. Again, the
two curves show excellent agreement across the entire frequency range, with very
tight correlation below 200Hz. The chamber resonance appears again in the 500Hz
region. At very high frequencies the measurement bottoms out against the noise
floor of the accelerometer and instrumentation.
EnclosureShop 19
Application Manual
Sealed Highpass Enclosure Tutorial 1
Simulation
Measurement
Measurement
Simulation
20 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
The two simulation and measurement curves for excursion are now shown in the
top graph of the next page. Excellent correlation is again observed below 200Hz,
with the chamber resonance appearing in the 500Hz region.
The two simulation and measurement curves for impedance are now shown in the
bottom graph of the next page. Excellent correlation is now seen across the entire
frequency range. However even the impedance curve shows the effects of the
chamber resonance at 500-1000 Hz.
A large number of curves now appear on the SPL graph. In order to view the
comparisons more easily, we shall only enable the simulation and corresponding
measurement curves for each view.
After adjusting the scale levels, the On-Axis response comparison is shown in the
top graph of the following page. At very low frequencies the measurement bot-
toms out against background noise. This measurement was taken inside a ware-
house thus many small reflections can be seen across the entire frequency range.
However, the average of the measured response shows excellent agreement.
EnclosureShop 21
Application Manual
Sealed Highpass Enclosure Tutorial 1
Measurement
Simulation
Measurement
Simulation
22 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
On-Axis Response
Measurement
Simulation
Near-Field Response
Measurement
Simulation
EnclosureShop 23
Application Manual
Sealed Highpass Enclosure Tutorial 1
After adjusting the scale levels, the Near-Field response comparison is shown
in the bottom graph of the previous page. This measurement was obtained by
locating the nose of the mic approximately 0.1 Inches above the surface of the
cone. At high frequencies the vibration of the cone is no longer uniform, and
again we see the chamber resonance affecting the response between 500-
1000Hz. However, the measured response below 250Hz correlates very well.
After adjusting the scale levels, the Chamber response comparison is shown
below. This measurement was obtained by placing the mic inside the cham-
ber. We now see the enormous reflections which take place within the cham-
ber. The simulation provides a respectable representation of these reflections.
Chamber Response
Measurement
Simulation
24 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
First we will save the current design which employs the LTD model, and then
save it again with a different name which will contain the TSL analysis.
To change the woofer we are using for the enclosure, we return to the Enclo-
sure Parameters dialog.
We can now analyze the design with this new TSL transducer model.
EnclosureShop 25
Application Manual
Sealed Highpass Enclosure Tutorial 1
We are mainly interested in the On-Axis SPL response so we will turn off the
current chamber SPL curves and enable the On-Axis simulation and measure-
ment curves.
After adjusting the scale, the SPL graph below is produced. It may or may not
be obvious, but the response in the vicinity of the knee at 70Hz has slightly
decreased. Note that the measured response now appears somewhat higher.
The remainder of the curve shows similar agreement with the previous LTD
simulation. The next graph will display the differences near resonance in a
more direct and obvious fashion.
26 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
The impedance graph below now shows a clear difference between the measure-
ment and the simulation using the TSL transducer model. The impedance at reso-
nance is considerably lower than the measurement. However, the impedance match
at higher frequencies continues to be largely maintained.
Clearly there is a difference between the LTD and TSL models which affects the
resonance region. The change in the impedance between these two models is a
reflection of their ability to accurately model the suspension loss Rms.
The TSL model assumes Rms to be constant at all frequencies. Rms was originally
characterized at 28Hz, the free air resonance of the speaker. However, the reso-
nance has been pushed up to 55Hz when used in this sealed enclosure. The effec-
tive value of Rms at 28Hz and 55Hz is not the same. This produces an error in the
impedance because the real losses in the transducer suspension are not constant
but are in fact a function of frequency. As the previous impedance graph demon-
strated, the LTD model has the ability to represent Rms much more accurately.
EnclosureShop 27
Application Manual
Sealed Highpass Enclosure Tutorial 1
As before we will save our work with the TSL model, and then save the design
under a different name for use with the STD model.
To change the woofer we are using for the enclosure, we return to the Enclo-
sure Parameters dialog.
We can now analyze the design with this new TSL transducer model.
28 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
The impedance graph below now shows the simulation using the STD model as
compared to the actual measurement. There is significant error at all frequencies.
The standard STD transducer model lacks the ability to represent the motor im-
pedance accurately across frequency. It uses a fixed inductance Levc, and a fixed
resistance Revc, which produce substantial error at mid, and high frequencies. In
fact as the graph below demonstrates, errors in the motor impedance representa-
tion are even apparent at frequencies down to 10Hz.
The accuracy of the impedance response plays a direct role in the accuracy of the
SPL response. The acoustic output of any electrodynamic transducer is dependent
on the voice coil current, and this is a function of the impedance. If the impedance
increases, the current decreases, and along with it the acoustic output.
We can therefore expect to see substantial changes in the SPL response with this
STD transducer model. Since the impedance is lower in the mid band, the SPL
response in the mid band will appear higher.
EnclosureShop 29
Application Manual
Sealed Highpass Enclosure Tutorial 1
The SPL graph below now shows the simulation using the STD transducer
model and the measured response. In the knee region of 70Hz the response
has decreased, while in the mid band region of 200-1000 Hz the simulated
response is several dB higher. This changes the relative balance between the
low and mid acoustic output levels.
The impedance error in the resonance region produces an error in the effec-
tive system Q, and results in a change in the shape of the knee. This error is
similar to the TSL model and is quantitatively about 0.5dB for this example.
However, the STD model produces an error of 2-3 dB between 200-1000 Hz.
The TSL model has no similar error in this region. The most significant dif-
ference between the STD and TSL models is the ability of the TSL model to
represent the motor impedance with much higher accuracy. This would be
especially important for transducers with large motor structures.
30 EnclosureShop
Application Manual
Tutorial 1 Sealed Highpass Enclosure
Summary
We have demonstrated the essential features involved with entering and simulat-
ing a design using EnclosureShop. Some of the basic simulation differences be-
tween the three transducer models has also been examined. Comparisons with
actual measurements show extremely good correlation when the LTD model is
utilized. However, simulation with the TSL and STD models resulted in errors of
approximately 0.5dB and 3dB respectively.
Therefore, it is important to understand the role which the transducer model plays
in simulation accuracy and its limitations. The LTD model offers advanced capa-
bilities which permit highly accurate simulations to be obtained. The next tutorial
will explore this model further with analysis performed at multiple power levels.
However, the derivation of LTD model parameters for a given transducer is far
from trivial. The data measurement process is more complex, tedious, and must
be performed carefully if accurate parameters are to be obtained. The TSL model
offers the best compromise between simulation accuracy and ease of model deri-
vation. For this reason its use is highly recommended as the primary model for
general design work.
You have now scratched the surface of EnclosureShop and been exposed to some
of the common heavily used features in the program. There is still much more in
the program to be discovered, and the following tutorials will serve to demonstrate
additional aspects of enclosure design and analysis.
Note: The completed tutorial design is given in the files Tutorial-1.led, Tutorial-
1_TSL.led, and Tutorial-1_STD.led.
EnclosureShop 31
Application Manual
Sealed Highpass Enclosure Tutorial 1
32 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Tutorial 2
Ported Highpass Enclosure
Highlights
Ported Enclosure Design
Quick Design Alignments
High Power Analysis
Port Linearity Analysis
Interpreting System Curves
Enclosure Layout Editing
Objectives
A ported Highpass enclosure using a 15 Inch (380mm) woofer with air vent. The
Quick Design utility will be used to select a general alignment for the enclosure and
provide some starting parameter values. The design will then be simulated at
various power levels to demonstrate the large signal behavior. The results will be
compared to actual measurements.
Transducer
Before any design can be simulated the transducer model must already reside in a
transducer library file. If this was not the case, we would first need to create the
transducer entry. For this example a transducer using an LTD model will be
required, and is already prepared in the Tutorial.Ltd library.
EnclosureShop 33
Application Manual
Ported Highpass Enclosure Tutorial 2
First we will create a new design and save it in the Tutorial-2 folder.
34 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Your screen should now look similar to the lower view. When the graph windows
are maximized, a Graph Select button row appears. You can switch to different
graphs by clicking on one of the buttons below the toolbar. Note the various graph
names given on the horizontal row of selection buttons.
EnclosureShop 35
Application Manual
Ported Highpass Enclosure Tutorial 2
36 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The Quick Design dialog provides the ability to quickly obtain enclosure design
parameters based on the parameters of the woofer and the type of enclosure we wish
to construct. Transducer data may be entered here manually or taken from the
currently selected LTD library. For this reason it is important to select the
transducer library before performing any other design tasks.
A general assumption of acoustic loss must be specified for the enclosure. Since we
are not using filling material in this example, a nominal Qb value of 10.0 will be
used. If there had been filling material involved, a Qb value of 5 to 7 would have
been more appropriate.
Three different alignment group boxes are shown at the bottom of the dialog. These
contain Vab chamber volume and Fp port frequency values for three possible
alignments based on your choices and values. As each of the selected options and
values are performed, the values in these alignment groups are calculated.
EnclosureShop 37
Application Manual
Ported Highpass Enclosure Tutorial 2
Note the Fp and Vab values of the three alignments. They range from 25 - 32Hz
and from 4.8 - 9.3 cu Ft³ respectively. We can see what the response curves look
like for each of these choices by clicking on the Graph tab.
Given the three possible curves, it is noted that the Quasi 3rd Butterworth has
the flattest looking response. There are many possible criteria that could be
used for selecting a curve. However, as we shall see shortly the actual response
that results from the detailed simulation will rarely match this highly simplified
prediction. We will select this alignment to begin the design.
The two key parameters which we need to produce this alignment for Vab and
Fp are: 4.8Ft³ and 32Hz Note these values as shown here.
38 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
We must now select the type of enclosure model for the design.
A simple rectangular box will be used. We will need to select dimensions which
produce the needed net volume of about 4.8 Ft³. We wish to work with units of
Inches and cubic Feet so we will also change the units. We will assume a wall
thickness of 3/4 Inch and an occupied volume of about 0.35 Ft³.
EnclosureShop 39
Application Manual
Ported Highpass Enclosure Tutorial 2
The Domain selections should already be correct. Since we are not using any
fill material, the Vfill % value is zero in the chamber.
The port itself can now be described by use of the object button in the Fp editing
field of the Port group box.
The first task is to define the area of the port, so we again use an object button
but this time in the Sp field.
40 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Note that the physical length of the port Lp has been calculated as 4.1 Inches. This
yields our tuning frequency of 32 Hz. We could also enter the port length and the
port frequency would be calculated.
The Vfill and Dfill port values allow you to model the effects of filling material
inside a port. In general one would not wish to use filling material within a
conventional port designed for a typical bass-reflex application. This would only
decrease the effectiveness of the port. However, there are special cases such as
transmission line and adiabatic enclosure modeling where these are useful.
The enclosure layout is displayed similar to what appears in the views on the
following page. Note that two sources are present: the transducer and the port. We
will leave the transducer in the center of the baffle board, but we will need to move
the port to a location above the transducer.
EnclosureShop 41
Application Manual
Ported Highpass Enclosure Tutorial 2
42 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The general desired location is shown on the preceding page. Note: If you have
trouble operating the 3D editing controls, see the reference manual for help.
You can try rotating around the scene to visualize the entire layout. If you click on
the enclosure it will become transparent, and you can then see the internal length
of the port and rear of the transducer.
We will now save our design to disk to keep/update our changes to this point.
EnclosureShop 43
Application Manual
Ported Highpass Enclosure Tutorial 2
The progress of the analysis is shown in the lower right fields of the status bar
at the bottom of the screen. There are several phases to the analysis which
depend on the complexity of simulation.
In this example we have chosen an Infinite Baffle domain which places the
baffle board of the enclosure within the infinite plane. Therefore no diffraction
analysis is required. This speeds up the analysis greatly since diffraction
analysis can be very time consuming. The analysis here should be completed
in a matter of seconds.
System curves are the data curves generated by the system analysis. For this
design we will enable all of the curves.
44 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
■ Select the Scale | Auto or Up/Dn menu items to view the lower curves.
Your screen should now look something like the view shown below. There are five
SPL curves displayed on the graph. Two are at a level of around 95dB while the
other three are much higher. We shall begin by discussing the lower curves.
The two lower level curves give the response at the primary simulation point and
the overall power response. In this example the primary simulation point is On-
Axis (Polar angle: 0 Horz, 0 Vert) at a distance of 1 Meter. However it could be
located anywhere in space.
The power response gives the total power produced by the design, relative to a half
space domain. Note that at higher frequencies the power response falls due to the
directivity of the transducer. At lower frequencies the radiation is nondirectional
with the Power and On-Axis response identical. At very low frequencies some
difference is seen between the power response and On-Axis response. This is due
to the strong cancellation between the port/spkr and their non-axial locations.
SPL 0H,0V
SPL Pwr
EnclosureShop 45
Application Manual
Ported Highpass Enclosure Tutorial 2
■ Select the Scale | Up/Dn menu items to view the higher curves.
Your screen should now look something like the view shown below. The
three higher curves consist of the two near field pressure curves for the trans-
ducer and port along with the chamber pressure response.
Near field SPL is the pressure found directly at the surface and center of the
diaphragm, or in the case of the air vent, the pressure at the mouth of the port.
At low frequencies the chamber behaves in a constant pressure manner for all
locations within the chamber. At higher frequencies there are reflections.
For a ported Highpass enclosure we see that the chamber response has a simi-
lar shape as the port near field response, although somewhat higher in overall
level. The pressure response of the transducer shows a sharp null at the port
resonance (32Hz) where the other two curves are maximum.
The comb filtering nulls at higher frequencies of the near field transducer
response are due to cancellations across the diaphragm. However, these
cancellations are based on the assumption of a rigid piston. Actual transduc-
ers are not rigid at these frequencies and can be expected to show variations.
SPL Chamber
46 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The impedance graph is shown below. There are actually two curves on this graph,
both identical. One curve is the impedance for the transducer and the other is the
total impedance for the enclosure. Since there is only a single transducer, the
curves are the same. However, in more complex enclosures there may be many
different impedance curves.
The excursion for both the transducer and port is shown on the following page.
The excursion of air in the port is many times higher than that of the transducer.
This follows directly from the ratio of the area of the transducer Sd and the area of
the port Sp. In this case Sd is 136 In² while Sp is 12.5 In². This is a ratio of 11:1.
At very low frequencies the same quantity of air displaced by the transducer must
also pass through the port. Since the area of the port is smaller than the transducer,
the motion of air within the port is higher. This is a motion magnification factor
equal to the Sd/Sp ratio. The air in the port must move 11 times further than the air
at the surface of the transducer.
Transducer
Enclosure
EnclosureShop 47
Application Manual
Ported Highpass Enclosure Tutorial 2
Transducer Port
Port
Transducer
48 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The velocity graph is shown on the preceding page. Here we clearly see again that
the linear velocity of air in the port is much higher than that of the transducer. If
the same quantity of air is to move through the smaller area it must move faster.
The volume current graph is shown below. This is the volume of air per second
moved by the transducer or port. As expected this graph shows that the volume
current of both sources at very low frequencies is the same. This is why the air
must move faster through the smaller port.
For this example the fluid in the port must move at 11 times the velocity of the
transducer. This creates a potential problem for large signal performance, where
the volume of fluid in the port must move at extremely high velocity. Remember
that the analysis completed thus far was performed at a drive level of 1 Watt. It is
therefore of interest to investigate the behavior of the port at different drive levels.
Comparing these results will give us a general indication port linearity.
Transducer
Port
EnclosureShop 49
Application Manual
Ported Highpass Enclosure Tutorial 2
If you were really paying attention to the previous On-Axis SPL response curve,
then you may have noticed that the response did not turn out as expected, based on
the previous alignment we chose in Quick Design. We saved that alignment curve
in the Guide Curve library, and we shall now use it for direct comparison with the
full simulated results.
The response curve saved during our Quick Design process was given the units of
Ratio since it is a dimensionless transfer function. The response is shown below,
and displays the quasi-3rd order alignment transfer function.
If we wish to display this curve on the SPL graph for comparison with the detailed
design simulation, we will need to change the units to SPL and scale the curve up
to the appropriate level.
50 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The alignment curve now appears on the SPL graph at a level of 94dBspl as shown
in the top graph of the following page. This is due to that ratio value of 1.0 V/V
becoming a value of 1.0 Pascals which equals 94dBspl. We will want to move it up
about 1dB to better match our detailed simulation.
Now the alignment response should appear as shown in the lower graph of the
following page. We see that the On-Axis response peaks at the knee near 80Hz and
generally has a valley between 100Hz and 500Hz. The response did not turn out
flat as predicted by Quick Design, and this is a fairly typical result.
EnclosureShop 51
Application Manual
Ported Highpass Enclosure Tutorial 2
52 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
The alignment predictions given by Quick Design must be viewed in the context of
rough approximations. The value of these approximations can sometimes be useful
and other times all but useless. In many cases it may be easier and faster to make
a simple guess for the initial volume and port tuning frequency. The design can then
be improved through trial & error analysis until a suitable response is achieved.
If we wish to reduce the peaking behavior at the corner of the knee, we can lower
the port resonance frequency. This can be done by increasing the length of the port.
Since an 8 Inch long tube was already available during the prototype construction,
we shall merely change the length to that of our available precut tube.
EnclosureShop 53
Application Manual
Ported Highpass Enclosure Tutorial 2
The resulting SPL response with the new longer port is shown below. While
the peaking at the knee has been reduced, the shape of curve still differs
significantly from that of the Quick Design alignment response. The knee now
appears too soft and the response at very low frequencies is higher than that of
the alignment response.
Since the response is stronger at the very low frequencies, we should probably
try a reduction in the chamber volume. This will decrease the response at very
low frequencies and also tighten up the region near the knee.
To reduce the chamber volume we will change the depth of the box from 18
Inches to 15 Inches.
Simulation
Quick Design
54 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Since we have now changed the dimensions of the enclosure, we must reopen the
Layout Parameters dialog to pick up the changes. We do not need to make any
adjustments there, but it is mandatory to re-save the layout in order for the
dimensional changes of the enclosure to be picked up.
The radiation and diffraction models are constructed when the Layout Parameter
dialog is closed. Anytime we make external changes to the enclosure or transducer
the Layout Parameters must be revisited. If we do not, EnclosureShop will issue
a warning message and refuse to perform the analysis.
EnclosureShop 55
Application Manual
Ported Highpass Enclosure Tutorial 2
The resulting SPL response with the new smaller chamber is shown below. The
response at very low frequencies is fairly close to that of the Quick Design
alignment. However, once again the knee remains significantly different. At
40Hz the simulation is below the alignment but at 80Hz it is above. Based on
our brief efforts so far we can even now make a reasonable observation:
No combination of Vab or Fp values exist which will ever allow the response
to match the predicted alignment to any high degree of precision.
We will now save our design to disk to keep/update our changes to this point.
Quick Design
Simulation
56 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
In preparation for the analysis at different power levels, we will turn off the
polar analysis. This will reduce the number of curves and facilitate easier
copying of the curves as a single block.
We now have a set of System Curves run at the 0.1 Watt level. We will copy
these curves and paste them into the Guide Curve library. We will then edit
the name on the first curve to indicate the 0.1W analysis.
EnclosureShop 57
Application Manual
Ported Highpass Enclosure Tutorial 2
58 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
EnclosureShop 59
Application Manual
Ported Highpass Enclosure Tutorial 2
60 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
We will now save our design to disk to keep/update our changes to this point.
■ Select the File | Save menu item (or use CTRL-S)
EnclosureShop 61
Application Manual
Ported Highpass Enclosure Tutorial 2
In order to see just the SPL On-Axis curves at the different power levels, we
will initially turn off all the Guide Curves and then just enable the curves we
desire. We will also shut off the Phase display to reduce the clutter.
100W
10W
1W
0.1W
62 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
While the SPL curves seem to indicate that something is happening at low fre-
quencies, it is difficult to visualize clearly due to the overall change in level. We
can see the changes more clearly if we remove the overall 10dB change in the
curves, and normalize them all to the same level, in this case 0.1W. We will copy
the three 1W, 10W, and 100W curves and then scale those down appropriately.
EnclosureShop 63
Application Manual
Ported Highpass Enclosure Tutorial 2
Your SPL graph should now look something like the view shown below. The
three normalized 1W, 10W, and 100W curves are just below the 0.1W curve.
At 100 Watts we see that the overall midband level drops about 1dB. This is
due to voice coil heating and a rise in Revc. However, there is a larger drop at
low frequencies of several dB. There is also a change in the shape of the
curves at these lower frequencies.
Due to the varying slopes and large dynamic range of acoustic response curves,
it is often difficult to visualize changes across different power levels. A supe-
rior way to observe these change is via the impedance response. All of the
electrical and mechanical changes are reflected here as well. But the need to
perform scaling is eliminated and the overall scale is more visible. Further-
more, the causes of the nonlinear changes in the response can be more clearly
determined as due to the port or transducer.
100W
10W
1W
[Normalized]
64 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Your screen should now look something like the view shown below. There are
four Impedance curves representing the various power levels. Here we can clearly
see a substantial change occurring at the lower frequency hump. At the 100W
level we see a general impedance increase everywhere due to the increase in Revc.
The low frequency hump is where the majority of nonlinearity takes place. The
portion of the enclosure which is most responsible for this region is the port. As
the drive level increases the acoustic resistance inside the port also increases, and
this is responsible for most of the nonlinearity observed here.
These results are quite remarkable. They show a noticeable nonlinearity even
between drive levels of 0.1W and 1W. Bear in mind that this is a large enclosure
with a 15 Inch woofer, yet nonlinearity begins even at 0.1 Watts.
At 100 Watts the port shows substantial resistance, and the lower hump is nearly
destroyed. The valley at the port resonance has also increased. Very high distor-
tion is present in this region of operation.
0.1W
1W
10W
100W
EnclosureShop 65
Application Manual
Ported Highpass Enclosure Tutorial 2
This example design was built and measured. We can compare these simula-
tions with the actual measurements of the impedance response at the four
power levels, using the data files provided in the Tutorial-2 folder. These
measurements were obtained using LMS and a VI-Box. First we shall turn off
the simulation impedance curves to prevent clutter.
66 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Your screen should now look something like the view shown below. Compare
these measured curves with the simulation curves on the previous graph. The
simulations show very good agreement to the actual nonlinear behavior. Here
again we see nonlinear changes occur at low levels between 0.1W and 1W. At
100W the low frequency impedance hump is severely compressed. The shape is
slightly different than that of the previous simulation, but very similar strong non-
linearity is measured.
It should be noted that the simulations rely on equivalent RMS sinewave analysis.
When there is high nonlinearity the signal waveform becomes severely distorted.
Observations taken with an oscilloscope on the SPL, excursion, and current wave-
forms show massive waveform deterioration. Sinewaves become squashed asym-
metrically, exhibit peaks, and often show multiple zero crossings. These distorted
waveforms change shape throughout the low frequency region, and differ in shape
between the various quantities at the same frequency.
Actual Measurement
0.1W
1W
10W
100W
EnclosureShop 67
Application Manual
Ported Highpass Enclosure Tutorial 2
- Check the boxes in the [G] column of the grid for #11, #29, #47, #65.
- Click Ok to close the dialog.
68 EnclosureShop
Application Manual
Tutorial 2 Ported Highpass Enclosure
Your excursion graph should now look similar to the view shown below. There are
four pairs of simulation and measured excursion curves. Each pair of curves are in
good agreement across the various power levels and frequencies.
There is a noticeable difference in the 100W curve pair below 20Hz. The mea-
sured excursion is lower than the simulation. This is due to the highly distorted
waveforms in the actual measurement, and the assumption of sinusoidal wave-
forms in the simulation. The large harmonic content cannot be represented accu-
rately by equivalent RMS sine wave analysis at this very high level of nonlinearity.
Near field acoustic measurements were not taken on this ported design, as they are
generally less informative for comparison with simulations. The simulations are
pure mathematical calculations of the near field pressure at the transducer and port
both in isolation. Equivalent measurements cannot easily be obtained since they
contain leakage between both sources. For example, when a mic is placed near the
cone of the transducer it experiences a deep null at the port resonance frequency.
However the port is producing maximum pressure at this frequency which leaks
into the mic and contaminates the transducer measurement.
EnclosureShop 69
Application Manual
Ported Highpass Enclosure Tutorial 2
Summary
We have demonstrated some of the essential features involved with a ported
enclosure design using EnclosureShop. The design was also analyzed at vari-
ous power levels to demonstrate port nonlinearity. Comparisons with actual
measurements show very good correlation across a wide range of power lev-
els and frequencies.
For this design a transducer with full LTD parameters was utilized. If mean-
ingful and accurate large signal analysis is desired, LTD models are really a
necessity. While most of the nonlinearity in this design was due to the port,
significant changes in the transducer can and do occur under large signal
analysis.
To improve the linearity of this design one would enlarge the port area Sp,
either as a single port or with multiple ports. To maintain the same tuning
frequency the port length would of course also need to be increased. How-
ever increasing the port area is crucial to reduction of the velocity magnifica-
tion factor. In general it is best to keep the port area between 25-50% of the
transducer piston area whenever possible to minimize nonlinearity. In this
example the port area was only 9% of the transducer area.
70 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
Tutorial 3
Transducer Location & Diffraction
Highlights
Sealed Highpass Enclosure
Woofer & Tweeter Transducers
Baffle Board Location Effects
Domain Differences
Diffraction Analysis
Enclosure Layout Editing
Objectives
For this example two different transducers will
be used: an 8 Inch (200mm) woofer and a 1 Inch
(25mm) dome tweeter. A square baffle board
will be used to provide identical axial modes
which emphasize the effects.
EnclosureShop 71
Application Manual
Transducer Location & Diffraction Tutorial 3
Transducer
Before any design can be simulated the transducer models must already reside
in a transducer library file. If this was not the case, we would first need to create
the transducer entry. For this example a transducer using an LTD model will
be required, and is already prepared in the Tutorial.Ltd library.
First we will create a new design and save it in the Tutorial-3 folder.
72 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
Your screen should now look similar to the lower view. When the graph windows
are maximized, a Graph Select button row appears. You can switch to different
graphs by clicking on one of the buttons below the toolbar. Note the various graph
names given on the horizontal row of selection buttons.
EnclosureShop 73
Application Manual
Transducer Location & Diffraction Tutorial 3
74 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
We must now select the type of enclosure model for the design.
EnclosureShop 75
Application Manual
Transducer Location & Diffraction Tutorial 3
The Domain selections should already be correct. We will start with an Infinite
Baffle domain but will later change this to Full Space.
76 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
The enclosure layout is displayed similar to what appears in the views on the
following page. The woofer is already in the center of the baffle board, and the
baffle board (face) of the enclosure is imbedded into the plane of the infinite baffle
plane. No editing is required at this time. You can pan around the scene, or select
the enclosure to see the woofer inside the enclosure.
This dialog controls the analysis parameters for the design. Most of the parameter
values here should already have acceptable default values. Verify that the power
is 1W per transducer or total, and that all of the optional analysis is enabled. For this
example we will enable the Horizontal polar analysis, but leave the Vertical
disabled.
For this enclosure we will set the Diffraction resolution and order to 3kHz and 4th
order respectively. This provides adequate convergence for this size enclosure and
shape. For more details on these parameters see the Reference Manual.
EnclosureShop 77
Application Manual
Transducer Location & Diffraction Tutorial 3
78 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
The progress of the analysis is shown in the lower right fields of the status bar at the
bottom of the screen. There are several phases to the analysis which depend on the
complexity of simulation. In this first run we have chosen an Infinite Baffle domain
which places the baffle board of the enclosure within the infinite plane. Therefore
no diffraction analysis is required. This speeds up the analysis greatly. The analysis
here should be completed in a matter of seconds.
For this example we are only interested in the On-Axis SPL response curve, so we
will disable the other two SPL curves. All other curves we will leave enabled.
EnclosureShop 79
Application Manual
Transducer Location & Diffraction Tutorial 3
■ Select the Scale | Auto or Up/Dn menu items to view the curve.
Your screen should now appear similar to the graph shown below. There is a
single curve shown which is the On-Axis response of the woofer on an infi-
nite baffle domain.
Note that the response is very smooth showing only a slight rise above 2kHz.
We will use this Infinite Baffle analysis as a reference against the Full Space
response curves which will follow.
Infinite Baffle
80 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 81
Application Manual
Transducer Location & Diffraction Tutorial 3
This analysis run will take much longer, probably several minutes depending
on the speed of your computer. If you receive an error message that you do not
have sufficient memory, you will not be able to continue with this example.
82 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
The progress of the analysis is shown in the lower right fields of the status bar at the
bottom of the screen. There are several phases to the diffraction analysis which
consume the majority of the analysis time.
When the analysis has finished you should see two curves on the SPL graph as
shown here below. The previous Infinite Baffle response is shown in Black and the
new Full Space response is shown in Red.
At low frequencies the level has dropped about 6dB. This is a result of the spatial
volume doubling which cuts the pressure in half. At very high frequencies there is
no change in the response since the transducer is highly directional.
However, at mid frequencies we now see a peak at 450Hz and a dip at 1kHz. This
is a result of the diffraction around the box given the center location of the woofer.
Since the woofer is directly at the center of the baffle board, the diffraction effects
due to each of the four face edges are identical and quadruple the amplitude. This
produces a very strong diffraction resulting in the severe peak/hole combination.
As before we will want to save this response before changing the location of the
woofer for the next run.
EnclosureShop 83
Application Manual
Transducer Location & Diffraction Tutorial 3
We will need to perform two editing operations on the woofer and enclosure.
First we select the woofer and move it down towards the bottom of the
enclosure. Next we select the enclosure and move it up to place the woofer back
On-Axis with the primary target again.
We must keep the woofer On-Axis with the target in order to maintain the same
high frequency response. The results of the two operations are shown on the
following page.
- Use the Position Cube to move the Woofer DOWN towards bottom.
- Use the Position Cube move enclosure UP, align woofer axis with target.
84 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 85
Application Manual
Transducer Location & Diffraction Tutorial 3
When the analysis has finished you should see three curves on the SPL graph as
shown here below. The previous Infinite Baffle response is shown in Black, the
woofer center response is shown in Green, and the new woofer at the bottom
response is shown in Red.
Moving the woofer off the dead center of the baffle board and towards the bottom
has reduced the mid band ripple to less than half the previous amount. Even though
the woofer is still centered on the width of the face, it is no longer centered vertically
on the face. This is a major improvement.
The diffraction paths from the woofer to the four edges are no longer identical. The
distance from the woofer to the top and bottom are very different. This causes the
frequencies of the diffraction ripples to be staggered, rather than piled on top of each
other. Staggering this distances spreads out the diffraction across a wider frequency
region and reduces the amplitude of the peaks and dips.
Woofer Center
Infinite Baffle
Woofer Bottom
86 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
Again we will need to perform two editing operations on the woofer and enclosure.
First we select the woofer and move it left towards the corner of the enclosure. Next
we select the enclosure and move it right to place the woofer back On-Axis with the
primary target again.
We must keep the woofer On-Axis with the target in order to maintain the same high
frequency response. The results of the two operations are shown on the following
page.
- Use the Position Cube to move the Woofer LEFT towards corner.
- Use the Position Cube move enclosure RIGHT, align woofer axis with target.
EnclosureShop 87
Application Manual
Transducer Location & Diffraction Tutorial 3
88 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
When the analysis has finished you should see four curves on the SPL graph as
shown here below. Once again the new curve is in Red showing the response with
the woofer now in the corner of the baffle board.
Moving the woofer to the corner has continued to reduce the peak at 450Hz, but has
now actually changed the dip at 1kHz to a peak as well. This is yet another different
combination of staggered diffraction paths from the woofer to the edges around the
enclosure.
For our final location we will move the woofer slightly higher again, thus making
the distances from the woofer to all four edges unique. As before we save this curve.
Woofer Corner
EnclosureShop 89
Application Manual
Transducer Location & Diffraction Tutorial 3
Again we will need to perform two editing operations on the woofer and
enclosure. First we select the woofer and move it up slightly higher about half
way more towards the center of the enclosure. Next we select the enclosure and
move it down to place the woofer back On-Axis with the primary target again.
We must keep the woofer On-Axis with the target in order to maintain the same
high frequency response. The results of the two operations are shown on the
following page.
- Use the Position Cube to move the Woofer UP towards but not to center.
- Use the Position Cube move enclosure DOWN, align woofer axis with
target.
90 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 91
Application Manual
Transducer Location & Diffraction Tutorial 3
When the analysis has finished you should see five curves on the SPL graph as
shown here below. Again the new curve is in Red showing the response with
the woofer now fully offset from all baffle board edges by different amounts.
This response now shows a new small dip at approximately 1600Hz, but overall
is probably the best compromise. Diffraction produces ripples, and moving a
driver around on the baffle board can only push these ripples to different
locations. The diffraction has not been eliminated but merely staggered.
Woofer Offset
Woofer Corner
92 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
Tweeter Analysis
We will now change the transducer in the enclosure to use a 1 Inch (25mm) dome
tweeter. The same sequence of previous baffle board locations will be applied.
We can use the tool button features to reset the position of the transducer and
enclosure without using the Position Cube.
When you have completed the changes, close the dialog. The scene should appear
shown on the following page.
EnclosureShop 93
Application Manual
Transducer Location & Diffraction Tutorial 3
Infinite Baffle
94 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 95
Application Manual
Transducer Location & Diffraction Tutorial 3
Tweeter Center
Infinite Baffle
96 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
When the analysis has finished you should see two curves on the SPL graph as
shown on the preceding page. The previous Infinite Baffle response is shown in
Black and the new Full Space response with the tweeter in center shown in Red.
We now see diffraction ripples just as we did in the case of the woofer, but for the
tweeter they are now somewhat different. The peak at 450Hz is still present as is
the dip at 1kHz. However there are many more ripples extending through 10kHz.
The reason behind this is the Off-Axis radiation differences. The woofer becomes
directional at the higher frequencies and produces little Off-Axis pressure. Remem-
ber that all of the diffraction energy comes from the 90 degree Off-Axis response
of a transducer, since it is mounted on the enclosure surface. The tweeter on the
other hand is a much smaller source with substantial Off-Axis radiation to much
higher frequencies. As such the diffraction interference patterns continue to much
higher frequencies.
As before we will want to save this response before changing the location of the
tweeter for the next run.
EnclosureShop 97
Application Manual
Transducer Location & Diffraction Tutorial 3
We will need to perform two editing operations on the tweeter and enclosure.
First we select the tweeter and move it down towards the bottom of the
enclosure. Next we select the enclosure and move it up to place the tweeter
back On-Axis with the primary target again.
We must keep the tweeter On-Axis with the target in order to maintain the same
high frequency response. The results of the two operations are shown on the
following page.
- Use the Position Cube to move the Tweeter DOWN towards bottom.
- Use the Position Cube move enclosure UP, align tweeter axis with target.
98 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 99
Application Manual
Transducer Location & Diffraction Tutorial 3
When the analysis has finished you should see three curves on the SPL graph as
shown here below. The previous Infinite Baffle response is shown in Black, the
tweeter center response is shown in Green, and the new tweeter at the bottom
response is shown in Red.
Moving the tweeter off the dead center of the baffle board and towards the
bottom has reduced the peak at 450Hz and the dip at 1kHz significantly. Even
though the tweeter is still centered on the width of the face, it is no longer
centered vertically on the face. This is a major improvement.
The diffraction paths from the tweeter to the four edges are no longer identical.
The distance from the tweeter to the top and bottom are very different. However
the small ripples still remain at the higher frequencies. While some have been
reduced, others remain unchanged or pushed to different frequencies.
Tweeter Bottom
Tweeter Center
Infinite Baffle
100 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
Again we will need to perform two editing operations on the tweeter and enclosure.
- Use the Position Cube to move the Tweeter LEFT towards corner.
- Use the Position Cube move enclosure RIGHT, align tweeter axis with target.
EnclosureShop 101
Application Manual
Transducer Location & Diffraction Tutorial 3
102 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
When the analysis has finished you should see four curves on the SPL graph as
shown here below. Once again the new curve is in Red showing the response with
the tweeter now in the corner of the baffle board.
Moving the tweeter to the corner has continued to reduce the peak at 450Hz, but has
now actually changed the dip at 1kHz to a peak as well. This is yet another different
combination of staggered diffraction paths from the tweeter to the edges around the
enclosure.
For our final location we will move the tweeter slightly higher again, thus making
the distances from the tweeter to all four edges unique.
Tweeter Corner
Tweeter Bottom
Tweeter Center
Infinite Baffle
EnclosureShop 103
Application Manual
Transducer Location & Diffraction Tutorial 3
Again we will need to perform two editing operations on the tweeter and
enclosure. First we select the tweeter and move it up slightly higher about half
way more towards the center of the enclosure. Next we select the enclosure and
move it down to place the tweeter back On-Axis with the primary target again.
We must keep the tweeter On-Axis with the target in order to maintain the same
high frequency response. The results of the two operations are shown on the
following page.
- Use the Position Cube to move the Tweeter UP towards but not to center.
- Use the Position Cube move enclosure DOWN, align tweeter axis with
target.
104 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 105
Application Manual
Transducer Location & Diffraction Tutorial 3
When the analysis has finished you should see five curves on the SPL graph as
shown here below. Again the new curve is in Red showing the response with
the tweeter now fully offset from all baffle board edges by different amounts.
Tweeter Center
Infinite Baffle
106 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
We now have a ratio curve which is the transfer function representing the diffrac-
tion of the woofer at the center position relative to the infinite baffle response.
Your screen should now appear similar to the graph shown on the following page.
Here we can clearly see the effects of the diffraction in isolation. At low frequen-
cies there is a reduction of 6dB, a peak of nearly 5dB at 450Hz, and a dip of -3dB
at 1kHz. We observed these effects in the original woofer/center acoustic response
but here we see them shown in isolation as a pure transfer function.
EnclosureShop 107
Application Manual
Transducer Location & Diffraction Tutorial 3
108 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
We will need to shut off the tweeter curves in the System and Guide curve librarys
and turn back on the equivalent woofer curves in the Guide Curve library.
EnclosureShop 109
Application Manual
Transducer Location & Diffraction Tutorial 3
Your screen should now appear similar to the graph shown above. Four curves
are shown: the pair of simulations, and the pair of measurements. While they
are similar the response at high frequencies is entirely different. The woofer
model was not tuned to replicate the actual transducer used in the test.
We can now divide the two measurement curves to produce the equivalent
diffraction transfer function as was done for the simulation data.
110 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
We now have a transfer function ratio curve from the measurements. We will
setup a color for this curve and enable the previous ratio curve as well.
- Check the boxes in the [G] column of the grid for #15, #20.
- Change the color on curve #20 to Red.
- Click Ok to close the dialog.
EnclosureShop 111
Application Manual
Transducer Location & Diffraction Tutorial 3
Your screen should now appear similar to the graph shown below. The dif-
fraction simulation is in very good agreement with the measurement. These
results are typical of the diffraction simulations provided by EnclosureShop.
The diffraction engine calculates the entire radiation field 360 degrees around
the enclosure as shown by the two polar graphs on the following page. Gen-
eration of a complete field is by far one of the toughest tests for diffraction
analysis. Excellent convergence is demonstrated by the lack of step/offsets in
the low frequency polar response curves all around the enclosure. This indi-
cates that nearly all of the radiated and diffracted energy has been included.
The rear field behind and to the sides of the enclosure is produced entirely
from diffracted radiation.
112 EnclosureShop
Application Manual
Tutorial 3 Transducer Location & Diffraction
EnclosureShop 113
Application Manual
Transducer Location & Diffraction Tutorial 3
Summary
We have demonstrated some of the essential capabilities involved with dif-
fraction analysis using EnclosureShop. Many different types of diffraction
can be explored for a wide variety of enclosure shapes and transducer/port
locations. However, diffraction analysis is extremely computer intensive and
places enormous demands on the hardware capabilities.
This example showed the effects of changing the transducer location, while
the enclosure shape was maintained. We could just as easily changed the
enclosure shape as well, say by using beveled corners, or a non rectangular
baffle board. Countless possibilities can be explored.
From the analysis provided here it is clear that locating transducers at equal
distances to the enclosure edges should be avoided. If we wish to minimize
the diffraction ripple amplitudes the transducer(s) should be placed at unique
distances to all edges whenever possible. However, there may be occasions
where the diffraction ripples can be pushed around in such a way as to com-
pliment and/or cancel some of the natural response defects of the transducer.
114 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
n Highlights
Ported Highpass Model
Port Area & Length Effects
Port Standing Waves
Port Loss & Damping
Media Characteristics
Model Comparisons
n Design Objectives
Design a transmission line enclosure. Triple Double
Woofer to be used, 8 Inch (200 mm). Folded Folded
Maximize the low frequency response.
Minimize the port reflection ripples.
Compare response to other model types.
This class of enclosure can probably best be defined by two fundamental charac-
teristics: a small chamber volume and a very long port. Due to the longer port the
standing waves within the port begin at much lower frequencies. These standing
waves (reflections or pipe modes) are very strong and produce aberrations in the
response of the enclosure. To minimize this the port is typically filled with a
batting media such as fiberglass, polyester, etc. which dampens the reflections.
However, the losses produced from the batting also reduce the effectiveness of the
port at very low frequencies. Therefore TL designs often come down to a tricky
balancing act between adding loss to reduce the reflections, while still achieving a
suitable low frequency response. The different types of media and how they are
applied all play a critical role in the final response.
EnclosureShop 115
Application Manual
Transmission Line Enclosures Application Note 1
The port itself is the primary focus of the design. The port may be configured
with constant area or tapered. In the later case it becomes a horn/waveguide.
TL designs have been made with either expanding or contracting area from
the transducer to the mouth. However, in this example we will only consider
the constant area type.
Therefore the two fundamental parameters for the port are its area Sp and its
length Lp. Due to the vital role played by the filling material; the type of
media, fill volume, and media density also become important parameters.
In some cases TL designs are constructed with batting material in only certain
locations. These highly specialized nonuniform configurations are difficult
to model in any convenient fashion. For this example we will only consider
the specification of batting by media type, density, and fill percentage.
116 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
n Transducer Parameters
A simple 8 Inch woofer model was created for this example. A TSL model was
use for simplicity. The driver is located in the Tutorial.LTD file of the Transducers
folder. The driver TSL parameters are shown below.
EnclosureShop 117
Application Manual
Transmission Line Enclosures Application Note 1
0o On-Axis
Infinite Baffle
- 3D Layout Views
The 3D layout for this setup is shown on the following page. The scene was
rotated to show both the frontal and rear views of the infinite plane. The
primary simulation position is shown by the target object, and the polar paths
are shown by the arrow arcs.
- Polar Response
The graphs on the next following page show the polar pattern. Two versions
are given: absolute and normalized (to zero degree on-axis values).
118 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Infinite Baffle
Front View
Infinite Baffle
Rear View
EnclosureShop 119
Application Manual
Transmission Line Enclosures Application Note 1
Infinite Baffle
Spkr @ 1M OnAxis
120 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Infinite Baffle
EnclosureShop 121
Application Manual
Transmission Line Enclosures Application Note 1
Infinite Baffle
122 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Infinite Baffle
EnclosureShop 123
Application Manual
Transmission Line Enclosures Application Note 1
0o On-Axis
Infinite Baffle
Vented Highpass
Enclosure Model
As a quick & dirty approach the parameters for this vented enclosure were
chosen as follows: Vab equal to Vas, and Fp equal to Fs. A 50% fill of 1 lb/Ft³
density fiberglass was assumed for the chamber. A rectangular shell/chamber
and port were used with area Sp of about 1/3 the Sd value of the transducer.
- 3D Layout Views
The 3D layout for this model setup is shown on the following page. The
scene was rotated to show both the frontal and rear views of the box. The
primary simulation position is shown by the target object, and the polar paths
are shown by the arrow arcs.
- SPL Response
The graphs on the next following page show the acoustic response for the 1
Meter field location, and the localized regions of the enclosure. This includes
the internal chamber, near field speaker, and near field port response. The
response has quite good bass response and is 3dB down at 35Hz. The pri-
mary port reflection occurs at about 500Hz due to the 10.4 In duct length.
124 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Vented Highpass
Enclosure on
Infinite Baffle
Front View
Vented Highpass
Enclosure on
Infinite Baffle
Rear View
EnclosureShop 125
Application Manual
Transmission Line Enclosures Application Note 1
Vented Box
Spkr @ 1M OnAxis
Internal Chamber
126 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
The port reflection just shown at 500Hz is a transmission line effect, and under-
scores the previous point that all ports are transmission lines. In this case the port
had a length of only 10.4 In. Longer ports will cause the reflection frequencies to
move down accordingly.
- Impedance Response
The graph below shows the impedance response, and displays the classic double
hump behavior as one would expect for a vented enclosure.
Vented Box
EnclosureShop 127
Application Manual
Transmission Line Enclosures Application Note 1
Vented Box
Port
Spkr
Port
Spkr
128 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Vented Box
Port
Spkr
Spkr
Port
EnclosureShop 129
Application Manual
Transmission Line Enclosures Application Note 1
0o On-Axis n TL Design #1
Much has been written about the design
criteria for transmission line enclosures.
Infinite Baffle While it would be highly desirable to uti-
lize a simple set of formulas or tables to
TL
Enclosure
arrive at values for port length and area,
Model this is often not practical or even possible.
A trial & error methodology is probably more effective. The response can be
simulated with much higher detail and accuracy to better optimize a particular
set of design objectives. In this respect EnclosureShop provides an ideal means
of simulating many different TL designs quickly and accurately.
130 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
We really have no idea what to choose for the port length. However, one general
rule does apply here. The low frequency cutoff will be somewhat related to the
length of the port. The longer the port, the lower the cutoff frequency.
To begin we choose a port length of 6 Ft (72 In). In some cases the port length is
simply the longest length possible to fit in the desired enclosure size.
EnclosureShop 131
Application Manual
Transmission Line Enclosures Application Note 1
TL Design
132 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
The resulting acoustic response for this first TL design is shown below. There are
an abundance of reflections throughout the mid band region. These are the stand-
ing wave reflections within the port, and they are substantial. Note that the level
drops over 20dB in the first primary hole. Such is the need for damping.
The acoustic response at both ends of the port and near the speaker is shown on the
following page. Likewise the impedance graph is also shown. These are the clas-
sic impedance reflections which occur from finite length tubes.
The excursion and velocity graphs are shown on the next following page. The port
reflections are extremely pervasive throughout the system. They affect every re-
sponse parameter.
Note also that unlike the previous vented design, the excursion and velocity am-
plitudes for both the transducer and port are now equal. For the TL case the port
area was chosen to be equal to the transducer area. It is probably fair to say that
very few people would desire a listening experience with this design.
On-Axis Response
EnclosureShop 133
Application Manual
Transmission Line Enclosures Application Note 1
Spkr Near
134 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Port
Spkr
Port
Spkr
EnclosureShop 135
Application Manual
Transmission Line Enclosures Application Note 1
n TL Design #2
The obvious improvement
needed from the first design is
to add acoustic losses through
the port to dampen the nasty
reflections.
The resulting acoustic response for this PE damped TL design is shown on the
following page. This is a tremendous improvement over the previous case.
The large 20dB hole is now 6dB, and the numerous other ripples have also
been greatly reduced.
The impedance graph now shows a very damped response, with the reso-
nance and reflection humps largely removed. Only a broad hump rise re-
mains.
However, the excursion curves and their new values at 10Hz should be noted.
The amplitudes have now dropped from their previous values of about 2.6
mm to less than 2.0 mm. The Polyester within the port adds considerable
resistance, and this restriction results in less overall output at all frequencies.
In order to reduce the port reflections, port resistance must be added. At the
same time this also reduces the low frequency response. Such is the compro-
mise associated with a TL design.
Of course we may take the position take we desire a very soft knee for the
response, and are not bothered by the loss of low frequency output. In that
case we should increase the port losses further.
136 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
On-Axis Response
Spkr Near
EnclosureShop 137
Application Manual
Transmission Line Enclosures Application Note 1
Spkr
Port
138 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
n TL Design #3
We could use several different tech-
niques to increase the port losses.
For this attempt we will merely
switch to a different media.
The resulting acoustic response for this FG damped TL design is shown in the top
graph on the following page. The large ripples are now nearly gone. However
some small sharp ripples remain.
We must remember that a small chamber volume was included in the TL design.
We have not assigned any damping media to this chamber, and it too has reflec-
tions - small ones.
It may be of interest to compute the amount of batting material used to achieve this
level of damping. The internal volume of the port is about 1.4 Ft³. Using the 1/2
lb Fiberglass media, about 3/4 of a lb would be utilized per enclosure. If Polyester
was to be used, a much higher density and quantity would be required.
EnclosureShop 139
Application Manual
Transmission Line Enclosures Application Note 1
On-Axis Response
On-Axis Response
140 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Spkr
Port
EnclosureShop 141
Application Manual
Transmission Line Enclosures Application Note 1
n Model Comparisons
We can now compare the various response characteristics of the #3 TL design
to the different models simulated previously. The acoustic and impedance
response graphs are shown on the following page, and the diaphragm excur-
sion on the next following page.
Without question, the most efficient bass design is the standard vented highpass
enclosure. The response extends very low with substantial output to 35Hz.
There is the issue of a 500Hz reflection from the port, but this generally falls
outside the frequency range of use in a three-way crossover system. A small
amount of damping inside the port would easily reduce this at such a high
frequency of 500Hz, where batting materials are highly absorbent.
The infinite baffle acoustic response is without question the smoothest, since
there is no chamber or port where reflections can occur. The low frequency
knee is soft, and the response rolls off at the 12dB/Octave slope.
The TL design is essentially a heavily damped 4th order vented design, and
approximates a 3rd order response, between that of the infinite baffle and
conventional vented. The knee is softer than the vented design. Ripples still
remain in the passband, but are much smoother with this level of damping.
The impedance curves show the expected characteristics. The TL design has
the lowest Q resonance behavior. However, either of the other two curves
could be easily squashed down as well by merely paralleling a resistor, with
no changes in the acoustic response.
The excursion response for the TL design is heavily restricted at low frequen-
cies as compared to the other two. This is of course due to the acoustic resis-
tance damping within the port.
Comparing the excursions between the conventional vented design and the
TL design near 32Hz highlights the efficiency of the vented design. The ex-
cursion for both is essentially the same. Yet the acoustic output from the
vented design is 8dB higher. The vented design effectively utilizes the radia-
tion from both sides of the diaphragm.
142 EnclosureShop
Application Manual
Application Note 1 Transmission Line Enclosures
Model Comparisons
Vented
Infinite
Baffle
TL #3
Infinite
Baffle
Vented
TL #3
EnclosureShop 143
Application Manual
Transmission Line Enclosures Application Note 1
Infinite Baffle
Vented
TL #3 Speaker
Excursion
n Summary
Transmission Line (TL) enclosures can be simulated with relative ease and
high accuracy using the capabilities of EnclosureShop. Many different de-
signs are possible and can be explored very quickly to determine the best
possible combination of parameters.
Their construction is typically far more difficult than most conventional en-
closures, and may greatly affect the results. Correlation with simulations are
more problematic than conventional designs, since so many non ideal factors
all contribute to the behavior of the port. Bracing, filling consistency, and
area/corner changes can significantly affect the response.
144 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
n Highlights
Sealed Highpass 2x15 Array
Custom Enclosure Model
Import 3D Enclosure Shell
Multi-Enclosure Analysis
Diffraction Analysis
Array Comparisons
n Objectives
Design & simulate an array.
Evaluate Polar response.
Five 2x15 cabinets.
Compare array curvature.
In most cases a custom 3D object will need to be created and imported in order to
define the shell of the array. This is a panelization process required to decompose
the shape of the array into a proper 3D polygon object. This process will be
demonstrated here for three different array shells.
EnclosureShop 145
Application Manual
Loudspeaker Arrays Application Note 2
n Transducer Parameters
A 15 Inch (380 mm) woofer will be used for this example. The TL1603A
model was chosen. The driver is located in the Tutorial.LTD file of the Trans-
ducers folder. The driver LTD parameters are shown below.
146 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
40 In 24 In
20 In
16 In
EnclosureShop 147
Application Manual
Loudspeaker Arrays Application Note 2
148 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
Spkr @ 1M OnAxis
Internal Chamber
EnclosureShop 149
Application Manual
Loudspeaker Arrays Application Note 2
150 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
Horizontal
Vertical
EnclosureShop 151
Application Manual
Loudspeaker Arrays Application Note 2
152 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The acoustic response of the single enclosure in the anechoic domain is shown
below. The response is quite different from that of the Infinite Baffle domain.
Note that the response rises above 300 Hz. This is due to the directivity of the
enclosure relative to those frequencies. At lower frequencies the enclosure has no
directivity and radiates into full 4p space, thus producing a lower On-Axis level.
The polar plots on the following page give the same information. At low frequen-
cies the enclosure behaves omnidirectional as indicated by the circular response
curves. At higher frequencies the enclosure, and then the transducers, become
directional.
The horizontal curves show the typical lobing patterns that arise from multiple
sources. The vertical curves show only the behavior of a single transducer, since
the two transducers are located in-line horizontally.
We can expect that an array of these enclosures will increase the directivity dra-
matically at low frequencies due to the larger baffling effect of their combination.
Spkr @ 1M OnAxis
EnclosureShop 153
Application Manual
Loudspeaker Arrays Application Note 2
Horizontal
Vertical
154 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The array is now viewed as a single multi-enclosure, with five isolated chambers
and ten transducers, two in each chamber. To construct this enclosure the Custom
Multipass model dialog is used as shown below. Virtually any arbitrary enclosure
structure with any combination of chambers, transducers, and ports can be defined
using this dialog. Here we configure five chambers with two transducers each.
For this straight array a standard shell could be chosen from the dialog. However,
since we will need to use an imported 3D shell for the next arrays, we shall intro-
duce the process now. A polygon version of the array must now be defined with
the proper vertex and face information. This information can then be used to
create an OBJ text file defining the shell for import into EnclosureShop. The
panelization of the array shell is shown on the following page.
EnclosureShop 155
Application Manual
Loudspeaker Arrays Application Note 2
6 5
SPEAKER ARRAY
MULTI ENCLOSURE SHELL
VERTEX & FACE LAYOUT FACE-4
No Curvature
2 1
5 2 2 1 1 5
6 6
Y
FACE-1 Y FACE-2
FACE-6 FACE-3
X Z
7 8 8 4
4 3 3 7
4 3
Vertex List (In)
Face List
1 +20.000 +50.000 -00.000 (counter clockwise) FACE-5
2 -20.000 +50.000 -00.000
3 +20.000 -50.000 -00.000 1 1243
4 -20.000 -50.000 -00.000 2 1375
5 +20.000 +48.000 -24.000 3 6842 8 7
6 -20.000 +48.000 -24.000 4 6215
7 +20.000 -48.000 -24.000 5 4873
8 -20.000 -48.000 -24.000 6 5786
156 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The array panelization contains a simplified version of the rear surfaces. The back
of the array is represented as a flat surface, while in reality there might be small
angular voids between the individual enclosures for a trapezoidal shape.
Simplification of real physical shells is often needed and desirable for diffraction
analysis. The inclusion of small details, especially on the back of the enclosure,
does not greatly benefit the results. Yet these small details can dramatically in-
crease the computational requirements and produce a model which requires un-
reasonable resources.
All of the vertex (or nodes) of the enclosure faces are first calculated and listed.
The faces are then labeled and listed using their appropriate vertex numbers. The
faces must be have their vertex numbers listed in counter clockwise direction when
the face is viewed from the outside of the enclosure. This is the winding order.
This is important so that EnclosureShop can understand which side of each face
belongs on the inside and outside of the enclosure. The vertex list for any face can
begin at any vertex around the face.
The OBJ file can then be created using any text editor as shown below. The vertex
are listed first beginning with the letter v, followed with the x,y,z coordinates on
each line. The faces follow beginning with the letter f.
f 1 2 4 3
f 1 3 7 5
f 6 8 4 2
f 6 2 1 5
f 4 8 7 3
f 5 7 8 6
EnclosureShop 157
Application Manual
Loudspeaker Arrays Application Note 2
158 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The 3D shell definition OBJ file is then imported. The transducers are arranged on
the multi-enclosure as shown in the view of the 3D scene on the preceding page.
The single enclosure was simulated at a distance of 1 Meter. However, a larger
distance is appropriate when simulating the response of the array. A distance of 3
Meters was chosen for the analysis of these arrays.
The acoustic On-Axis response is now shown in the graph below for this straight
array (no curvature). As expected there is a remarkable increase in the low fre-
quency amplitude. Indeed, the level at 100Hz now exceeds that of 1kHz.
The response also shows a null which occurs at 1300Hz. This is due to the cancel-
lation between the multiple drivers spread across the vertical distance. The array
is over 8 Feet in height.
The horizontal and vertical polar response curves are now shown on the following
page. Note that the Directivity Index (DI) for the vertical profile is now over 4dB
even at 80Hz. At 160Hz the level behind the array is only about 10dB below that
in front of the array.
Loudspeaker Array, No Curvature
Spkr @ 3M OnAxis
EnclosureShop 159
Application Manual
Loudspeaker Arrays Application Note 2
Horizontal
Vertical
160 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The array panelization for this configuration is shown on the following page. As
before the back of the array is simplified and merely represented as a flat surface.
The vertex and face lists are given in the pictorial on the following page, and in the
associated OBJ file shown below.
The transducers are arranged on the multi-enclosure shown in the view of the 3D
scene on the next following page. Again a simulation distance of 3 Meters is used.
The front of the array now contains five distinct panels, representing the baffle
boards for each of the single enclosures.
f 5 6 8 7
f 3 4 6 5
f 7 8 10 9
f 1 2 4 3
f 9 10 12 11
f 13 14 2 1
f 11 12 16 15
f 14 16 12 10 8 6 4 2
f 1 3 5 7 9 11 15 13
f 13 15 16 14
EnclosureShop 161
Application Manual
Loudspeaker Arrays Application Note 2
14 13
SPEAKER ARRAY
MULTI ENCLOSURE
FACE-6
VERTEX & FACE LAYOUT
Small Curvature
2 1
2 1
2 1
13 14 14 13
FACE-4
4 4 3 3
FACE-2
6 6 5 5
Y
Y
FACE-10 X Z FACE-9
FACE-8 FACE-1
8 8 7 7
FACE-3
10 10 9 9
FACE-5
16 16 15
15
Vertex List (In) 12 11
12 11
01 +20.000 +49.640 -04.930 12 11
02 -20.000 +49.640 -04.930
03 +20.000 +29.920 -01.640
04 -20.000 +29.920 -01.640 Face List FACE-7
05 +20.000 +09.990 +00.000 (counter clockwise)
06 -20.000 +09.990 +00.000
16 15
07 +20.000 -09.990 +00.000 01 5687
08 -20.000 -09.990 +00.000 02 3465
09 +20.000 -29.920 -01.640 03 7 8 10 9
10 -20.000 -29.920 -01.640 04 1243
11 +20.000 -49.640 -04.930 05 9 10 12 11
12 -20.000 -49.640 -04.930 06 13 14 2 1
13 +20.000 +43.730 -28.270 07 11 12 16 15
14 -20.000 +43.730 -28.270 08 14 16 12 10 8 6 4 2
15 +20.000 -43.730 -28.270 09 1 3 5 7 9 11 15 13
16 -20.000 -43.730 -28.270 10 13 15 16 14
162 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
EnclosureShop 163
Application Manual
Loudspeaker Arrays Application Note 2
The acoustic On-Axis response is shown in the graph below for this slightly
curved array (small curvature). Again we see the low frequency increase, but
the null has moved from 1300Hz down to 900Hz. Arcing the array pushes the
outer transducers further away from center and more off-axis, thus causing
more path delay and lower frequency cancellation.
The horizontal and vertical polar response curves are now shown on the fol-
lowing page. Compare these polar curves to those of the previous straight
array. The DI values are slightly reduced. Clearly the result of arcing the
array.
The amplitude of the higher frequency lobes in the vertical direction has slightly
increased. Arcing the array points the outer transducers more off-axis and
increases the higher frequency coverage to more distant vertical angles.
However, the curvature of this array was small. We will now arc the array to
the maximum allowed by the enclosures and observe those changes.
Spkr @ 3M OnAxis
164 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
Horizontal
Vertical
EnclosureShop 165
Application Manual
Loudspeaker Arrays Application Note 2
The array panelization for this configuration is shown on the following page.
As before the back of the array is simplified and merely represented as a flat
surface. The vertex and face lists are given in the pictorial on the following
page, and in the associated OBJ file shown below.
The transducers are arranged on the multi-enclosure shown in the view of the
3D scene on the next following page. A simulation distance of 3 Meters is
once again used.
f 5 6 8 7
f 3 4 6 5
f 7 8 10 9
f 1 2 4 3
f 9 10 12 11
f 13 14 2 1
f 11 12 16 15
f 14 16 12 10 8 6 4 2
f 1 3 5 7 9 11 15 13
f 13 15 16 14
166 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
14 13
SPEAKER ARRAY
MULTI ENCLOSURE
VERTEX & FACE LAYOUT FACE-6
Large Curvature
2 1
2 2 1 1
13 14 13
14 FACE-4
4 4 3 3
FACE-2
6 6 5 5
Y
Y
FACE-10 X Z FACE-9
FACE-8 FACE-1
8 8 7 7
FACE-3
10 10 9 9
16 FACE-5 15
15 16
12 12 11 11
Vertex List (In)
12 11
01 +20.000 +48.600 -09.790
02 -20.000 +48.600 -09.790
03 +20.000 +29.720 -03.300 Face List
FACE-7
04 -20.000 +29.720 -03.300 (counter clockwise)
05 +20.000 +10.000 +00.000
06 -20.000 +10.000 +00.000 01 5687 16 15
07 +20.000 -10.000 +00.000 02 3465
08 -20.000 -10.000 +00.000 03 7 8 10 9
09 +20.000 -29.720 -03.300 04 1243
10 -20.000 -29.720 -03.300 05 9 10 12 11
11 +20.000 -48.600 -09.790 06 13 14 2 1
12 -20.000 -48.600 -09.790 07 11 12 16 15
13 +20.000 +38.920 -31.850 08 14 16 12 10 8 6 4 2
14 -20.000 +38.920 -31.850 09 1 3 5 7 9 11 15 13
15 +20.000 -38.920 -31.850 10 13 15 16 14
16 -20.000 -38.920 -31.850
EnclosureShop 167
Application Manual
Loudspeaker Arrays Application Note 2
168 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
The acoustic On-Axis response is shown in the graph below for this maximum
curved array (large curvature). Again we see the low frequency increase, but the
null has moved further from 900Hz down to 600Hz. Arcing the array further has
continued to push the outer transducers further away from center and more off-
axis, thus causing more path delay and lower frequency cancellation.
The horizontal and vertical polar response curves are shown on the following page.
Compare these polar curves to those of the previous straight array. The DI values
are much more reduced. At 80Hz the DI has decreased from 4.3dB to 3.9dB.
Note also that the DI values between 160Hz and 640Hz have generally fallen to
one-half those of the straight array (5 vs. 10). This is a considerable broadening of
the coverage pattern in the vertical direction.
It should also be mentioned that the On-Axis null will decrease somewhat at larger
simulation distances, and the response will generally become more uniform. The
closer to the array, the larger the path length differences between the drivers.
Spkr @ 3M OnAxis
EnclosureShop 169
Application Manual
Loudspeaker Arrays Application Note 2
Horizontal
Vertical
170 EnclosureShop
Application Manual
Application Note 2 Loudspeaker Arrays
n Summary
Many different types of array characteristics can be explored. The examples given
here are merely an introduction to array modeling. Different types and shapes of
arrays can be investigated both for low and high frequency transducers.
The simulations provided in this example utilized 3kHz diffraction resolution, and
4th order diffraction analysis. Simulation of the single 2 x 15 enclosure only
required 5 minutes of computation time and 75MB of memory. However, simula-
tion of each array required 3 hours of CPU time and consumed over 500MB of
memory. Large arrays can exhaust the entire 2GB address space of the Win32®
operating system.
It is therefore wise to plan the design carefully and eliminate any unnecessary
small surfaces from the shell. Their contribution to the overall result is typically
minimal, but can greatly increase the computational burden. Proper choice of the
diffraction resolution and diffraction order also play a critical role in determining
the computation requirements.
EnclosureShop 171
Application Manual
Loudspeaker Arrays Application Note 2
172 EnclosureShop
Application Manual