***************************
* LOUDSP.EXE *
* 1988 C. Hoffman *
***************************
LOUDSP.EXE was created using various interpretations of the standard
Thiele-Small equations now used for most modern speaker designs. The
information was compiled from several sources, including:
Speaker Builder Magazine, 1/84, 1/85, 2/85 pub. E.T. Dell Jr.
The Loudspeaker Design Cookbook, by Vance Dickason
How To Design, Build, & Test Complete Speaker Systems, by D.B. Weems
These are all highly recommended as you will be unable to get maximum
benefit from the program without the background information contained in
these references. Now, the program itself:
The first thing the program does on startup is to read the setup file,
LOUDSP.SCR. This sets the screen mode and colors. If the file is not
present, the program outputs a warning message and sets standard CGA mode
with some default colors. The file will be created when the setup option
is chosen from menu #2. If running the program from a floppy, one would
usually set the path for data files to the same floppy or possibly the
other floppy. If running from a hard disk, it is usually best to have a
separate subdirectory for the data files. The path may be set to anything
you wish, but it must be complete and end with a backslash.
Note that if EGA is selected you will have colored graphs. This is the
only real difference (though it looks lots better!), but you will not be
able to print graphs using the DOS graphics program. If you want hardcopy
of the graphs, use CGA, or use a printer driver that works with EGA.
These can be found on many BBSs.
The program was created with Borland's Turbo Basic. It will support a
math coprocessor if present, though it is not really necessary since most
routines are very fast. You may wish to put a new line in your
environment table to insure that the program correctly identifies whether
or not an 8087 is present. Type SET 87=F or SET 87=T at the DOS prompt as
needed. If it won't run, try that first.
Most of the operation of the program is self explanatory. Substantial
error trapping has been incorporated, so it should be difficult to crash.
The usual worst case is that you end up back at the main menu with your
default values still intact.
The last design performed is the design the program remembers. In that
way, the data carries over to whatever routines are selected. If you have
just performed a ported design, the data will carry over to the port
length caluclator on menu #2. In the same manner, ported or sealed
enclosure data will carry over to the maximum power area. To avoid
confusion, just rerun the desired design if you are not sure of what the
last design was.
Misc:
1. The program uses a value of 7 for box loss (also known as Ql). This
has proven quite accurate, and most people will not construct a box,
measure the loss, then modify the box based on the new value. Any box
I have measured has been very close to 7, so as long as you refrain
from making boxes from fiberboard or tool steel, the value should be OK.
2. The crossover design routines are quite limited as this is not really
a crossover design program. Second order APC was chosen, as it is the
most pratical for most builders. First order tends to allow far too
much power into the tweeter, and third order is beyond most peoples
ability to fine tune for best sound. Though the design routines used
here are good, you will probably have to do some fine tuning based on
listening tests. Try to remember that our best mathamatical models
are still a poor representation of what is actually happening in a
complex three dimensional world. Finally, don't forget to use Zobel
networks to flatten the impedance of the drivers in the crossover
area.
3. Many driver files have been included with the program. No guarantee
is made for the accuracy of these files. Manufacturers often change
driver specifications, discontinue drivers, and introduce new ones.
After reviewing these files, you should probably select only a few of
interest and keep the others in a separate area. This will avoid
directory clutter! More data could have been included in the driver
files, but data is not always readily available. The values of Fs,
Vas, and Qts are the minimum required to design an enclosure, and are
all that are stored in the data files.
4. This program is supplied "as is". No guarantee is made for the
accuracy of its predictions. No liability is assumed for any loss or
damage incurred by running the program, inability to run the program,
or consequential damages arising from any use of the program.
5. You may distribute this program in its original form as compressed
with PKPAK. The program may not be sold, and copy fees may not exceed
five dollars. The program may not be distributed if modified.
6. If you have a problem with the program, discover a bug, or have
suggestions for improvement, leave me a message on:
Compuserve 73260,2255
The Serial Port (716) 398-3118
EFX Systems (716) 396-2699
Version 2.10 was released on 11/26/88. The improvements over version
2.06 were the result of suggestions by several users. It is my hope
that you will find this version enjoyable to use, and that it helps to
take some of the mystery out of speaker design.
7. There is no substitute for high quality drivers. I highly recommend:
Madisound Speaker Components
8608 University Green
P.O. Box 4283
Madison, Wisconsin 53711
phone: (608) 831 3433
bbs : (608) 836 9473 N81 300-2400
They are an excellent source for Dynaudio, KEF, Audax, and many others.
8. >>>>>>>> FREEWARE- NO USER FEES REQUESTED, ENJOY! <<<<<<<<
