home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Shareware Overload
/
ShartewareOverload.cdr
/
busi
/
perf200.zip
/
PERFECT1.DOC
< prev
next >
Wrap
Text File
|
1988-06-16
|
10KB
|
160 lines
PERFECT BOX Version 3.00 7/88
Thank you for your support of PERFECT BOX. I have spent
hundreds of hours massaging PERFECT BOX into its current version.
Initially, the program was only intended for my own purposes, but
after time I realized that a very large hole existed in the
software domain for the serious loudspeaker designer.
I have seen a few very simple routines that do the "classic"
ideal box calculations, and concluded that there must be a way to
make those hand calculations mean more that dB's here and there!
That is when I though a graphic plot would make the design process
go faster. Thus, the old standby program, "BOXRESPONSE," became
the foundation on which PERFECT BOX was built.
Now, the BOXDATA database file holds all the information in
one nice, concise form, readable in seconds, for program use!
Gone are the days when you can't finish (or start) a new design
because the T/S parameters are nowhere to be found!
Speaking of the BOXDATA FILE, I have added a few new
routines, external from the main program, that will APPEND new
records from another BOXDATA "like" file. I say "like" because
the NEW file is of the same format as BOXDATA, but is named
"DATAUP.DAT." You will find a few new BOXDATA records on the
enclosed disk in this updated file. To load them into your
current file, use the addata.exe routine. It will either add to
or replace the records in your BOXDATA file (in the current DOS
default directory) with the records in the "DATAUP.DAT" file, by
your choice.
The other new routine, BOXSORT.EXE, alphabetically sorts
the BOXDATA by the manufacterer (1st Key) and model number (2nd
Key.) To start the sort, execute the BSORT.BAT file from the DOS
prompt (you can use the new DOS SHELL option from the opening menu
of PERFECT BOX.) This routine will also delete any BOXDATA
records with the word "DELETE" as the model or manufacturer, or
any records with three spaces first in both of those fields. You
must use EDIT to enter those spaces or the word DELETE (in caps)
for the BOXSORT program to recognise them for deletion. The sort
program itself only creates a new temporary file called
BOXDATA.SRT. Feel free to modify the batch file to your needs.
You will notice a few changes in the PERFECT BOX program
itself including...
1. Color is now supported! You may change the colors (a
temporary change) by typing "COLOR" as the "New path" for the
BOXDATA file.
You may use either CGA, EGA, or Hercules(Tm) graphics
modes. The mode can be set by changing the "PERFECT.CTL" file.
Use "CGA", "EGA", "HERCULES", or "NO", for no graphics. (You must
use the quotes.) If you are going to use Hercules graphics, you
must first run the QBHERC.COM program before starting PERFECT BOX.
If you choose EGA mode, you will have to use the Toggle
Graphics Mode" function to do hard-copies of the graphics plots.
I do all of the main work in EGA mode, then switch to CGA for the
screen print function. For you Apple fans, I am currently working
on converting PERFECT BOX to run on 512K Macintosh machines (full
window operation!.)
2. A SHELL to DOS is now supported.
3. When you are ADDing records to the database, the Xmax
prompt will accept either inches or millimeters, and the Vas
prompt will accept either cubic feet or cubic liters. Just press
RETURN to change back at forth at those prompts.
4. A "Ideal box" calculation for all the drivers in the
BOXDATA file can be printed easily from the "Automatic Parameter
Entry area.
5. The Port Table screen has been fixed up to line up on
the CRT, and is color coded for each port size; RED-Negative port
length, (not usable), YELLOW-Ok, but a bit long, and GREEN-Good
choices.
6. The Ideal box values are rounded to two places past the
decimal for CRT output, but not in calculations.
7. Coprocessor support is automatic. If you don't have
one, you will notice a slight loss of speed from previous
versions. I have adjusted the calculation of data points to skip
over the odd frequencies above 100 Hz. If you don't use an 8087,
just go get another cup of Java!
8. The BOXDATA file is stored in a different form than in
previous versions. Your old file is NOT compatible(!) UNTIL you
run the conversion program included with PB. The new data file
name is BOXDATA4. You can Append your old parameters to the
BOXDATA4 file after they are converted. Just be careful not to
overwrite the new BOXDATA4 file included with this release. The
2BOX3.EXE conversion program takes your old BOXDATA file as input,
and produces the new BOXDATA4 file. I suggest you do the
conversion then rename the converted file to DATAUP.DAT to be used
as input to ADDATA.EXE program. Making (keeping) a backup copy of
all of your data files is strongly recommended before attempting
the changeover.
If you are doing all of this conversion stuff in a different
directory (or drive) than where the runtime module, BRUN40.EXE is
located, you may get an error that says .."BRUN40 cannot be found
, enter new path.." Just give the path to the directory where
BRUN40 is located, (like C:\PERFECT.)
I want to comment on the BOXDATA records that I have been
receiving on your disks. First, I think that many users are
confusing the definition of the X-max parameter. The typical
ranges for X-max range from 0.1 inches to 0.25 inches with 0.5 on
the JBL 18" 2245H being the largest I've seen (Have you ever seen
one, they are huge.) If you don't know the X-max value,
approximate it with 0.1 to 0.15, 0.2 for large suspensions.
Note that X-max is NOT the total possible cone movement
allowed by the suspension, which on large drivers could be at
least an inch. I have seen specification sheets from a few
manufacturers (Credence is one) that give an "X-max" and a "X-lin"
(for X-linear?) value. As far as I can tell from these, the X-lin
is really the value of X-max as normally defined. And the "X-max"
value is the maximum cone travel possible. Contact the
manufacturer if you are confused about their specifications. I
hope that a standard for "X" measurements will surface soon.
The X-max measurement defines the peak (not to be confused
with peak-to-peak) distance that the voice coil can travel without
extending beyond the magnetic field as determined by the pole
piece length. The voice coil may be longer in length than the
pole piece. This is defined as "voice coil overhang." The
program uses the X-max measurement in calculating the Maximum
Sound Pressure Level (SPLmax), and thus is a key factor in the
Excursion Limit Plot. Theoretically, the driver is "linear" only
if the cone travel is within the X-max limits! Think about this
the next time you watch your low-end drivers move 3/4 " peak, from
the rest position!
In general, (not always,) low Q drivers are intended for 4th
order systems, and typically have what is coined as "long-throw
suspension." Also, remember that PERFECT BOX model is valid up to
Qts of 0.7, and the Marquis 12G38BPP1H is slightly over that value
at 0.72. Of course, each individual driver will vary from the
manufacturers published specs, so you might want to measure them
yourself to be sure.
Good luck in your new designs, thanks again for all of your
generous support. If you have any specific questions or would
like chat about loudspeaker topics, drop me a letter, or call my
new BBS, "UCF Health Topics" at (407)-281-5522, 2400 N-8-1.
You will find the latest release of Perfect Box posted here.
My BBS is an independent node in FidoNet, (1:18/507), and thus
will accept "Electronic Mail" 24 Hours a day. If you are a
FidoNet SysOp or know a friendly one with a "BinkleyTerm
compatible Mailer", you can file request with the magic filename
"PERFECT", for the current MS-DOS version, or MACPERF for
Macintosh (when available.) Of course, you could always File
Request by the current archive name "PERF300.ARC." Also, I host a
local echo-mail roundtable called, "Audio Topics". You can find
many questions regarding Loudspeakers, etc. answered there.
Warren A. Merkel
*** End of Document ***