I changed the PCSTEREO schematic gifs so they now reflect the change in the capacitor value mentioned in V1.50. I also changed the feature list and the notes to reflect the correct information.
I've updated this DOC file and the compatablity listing.
I found a little discrepancy between the V1.00 sound board and V1.50 adapter schematics. The V1.00 specifies that ground on the DB25 connector is pin 20 and the V1.50 adapter specifies pin 25 as ground. Relative to the parallel port, they're both right. Relative to each other, they're wrong. In order for both boards to work correctly, the ground pins on the DB25 connectors have to be the same all the way through. I changed the V1.00 schematic to indicate that both pins 20 and 25 on the DB25 connector should be grounded. In actuality, pins 18 through 25 are all ground. You should connect all of these pins (18 through 25) to ground to insure that the ground is common to all of the boards.
So wouldn't it have made more sense to change the V1.50 schematic instead of the V1.00 schematic? Yes, I should have changed the V1.50 schematic instead of the V1.00 schematic. However, since I was editing the text on the V1.00 schematic, it just seemed easier to add the proper ground pin at that time. Afterall, to get the schematic into GIF format I have to: install Dr. Halo's GRAB utility into memory, load up AutoSketch, load in the schematic, select the portion of the screen I wish to "grab", "grab" the screen, take the resulting CUT file into Dr. Halo and erase AutoSketch's menu and status bars, load up VPIC, load in the CUT file, and then finally save it as a GIF file.
I've also added a picture of what my V1.00 circuit looks like currently. The picture was taken before the V1.50 circuit had been finalized, so it does not appear in the picture. The picture was taken in poor lighting conditions, so the contrast is poor. I was also limited to 400 dpi resolution when I scanned it. Next time, I'll try to include a top view of the circuit. (It is a little difficult to see what's in the picture because of all the wires jumping across the board.)
WHAT'S NEW IN V1.50?
Apparently there are two Stereo-on-1 formats floating around. MODPLAY PRO seems to support the least popular one and my original schematic supports MODPLAY PRO. I have come up with a temporary way to adapt my board to support both of these formats. (I should really clarify that both Stereo-on-1 formats came out of different European countries at about the same time.)
At the end of the summer '92 I actually found the time to build my circuit on a copper clad board. This means I now have a schematic for the power supply circuit. The artwork I drew up for the traces on my version of the board have a few bugs so you won't be seeing that artwork for a while.
Also, I changed a value of one capacitor in the original circuit. The value change isn't critical; the circuit will work the way it was intended if the capacitor isn't or can't be changed. However, I found that by lowering the capacitor value in question you can achieve some interesting effects. Basically, the feature this adds is the almost immediate fallback to mono mode when the stereo signal ends (or pauses).
To this DOC file I have added: sources for parts, more acknowledgements, troubleshooting guide, and a list of what's to come. In a separate DOC file I have included a list of all the files that I found that work with my sound board.
INTRODUCTION:
This circuit schematic is distributed with the hope that someone will find it of some use. This circuit is intended for people who want their computer to NOT sound like an IBM but are too cheap to buy a commercially available sound board or they happen be avid electronics hobbyists and welcome the opportunity to build something more or less for the hell of it. (like me)
Files included in the PCSTGIF2.ZIP file are:
++ PCSTEREO.GIF (V1.00 Sound Board schematic)
++ PCSTTL.GIF (V1.00 Top left corner)
++ PCSTTR.GIF (V1.00 Top right corner)
++ PCSTBR.GIF (V1.00 Bottom right corner)
++ PCSTBL.GIF (V1.00 Bottom left corner)
+ PCSTPWR.GIF (V1.50 Power supply schematic)
+ PCSTPWRL.GIF (V1.50 P.S. left side)
+ PCSTPWRR.GIF (V1.50 P.S. right side)
+ PCSTADD1.GIF (V1.50 Schematic additions #1)
+ PCSTAD1T.GIF (V1.50 S.A. top part)
+ PCSTAD1B.GIF (V1.50 S.A. bottom part)
++ PCSTPIC1.GIF (Picture of the V1.00 board)
++ README.1ST (You are here!)
++ COMPAT.TXT (Compatible files list)
FOCUS1AD.TXT (Advertisement)
FOCUS1AD.TXT is an advertisement for one of the companies I work for. If you are opposed to advertisements, don't read it.
A + in front of the file indicates which files are new or have been altered since version 1.00. A ++ in front of the file indicates which files have been altered since version 1.50.
Note: The partial schematic GIF files contain some overlap. Refer to the whole schematic when using/assembling the pieces.
V1.50 ADDITIONS:
The capacitor that I changed is the one that is connected between the collector and the emitter of the 2N3906 transistor. In V1.00 the capacitor is specified at 0.47 uF. In this version, change that capacitor to 0.02 uF.
Basically, this allows you to do some special effects with the board. Just follow these steps:
1. Obtain a copy of MODPLAY PRO V2.19b and
SPUTTER and set yourself up in the MODPLAY PRO
directory.
2. Type the following at the DOS prompt:
MP /Q=C:\COMMAND.COM FAVORITE.MOD
where:
FAVORITE.MOD is an unzipped MOD file in the
same directory with MODPLAY PRO
3. Get into the directory where you have SPUTTER.
4. Type the following at the DOS prompt:
SPUT FAVORITE.SND
where:
FAVORITE.SND is an unzipped sound file in the
same directory with SPUTTER.
The result of all this is that MODPLAY PRO will start to play your favorite song continuously in stereo while allowing you to freely roam around DOS. When you execute step #4, the MOD player will halt allowing SPUTTER to play the sound file in mono. When SPUTTER is done, the MOD file resumes playing in stereo. The problem with V1.00 is that if you try to do this, the fallback to mono is too slow and you lose the first second of the sound file. All you hear is silence for that second. By the way to get rid of the MOD file playing in the background, simply type EXIT at the DOS prompt.
A note about roaming: I have a 386-20 and roaming for me includes looking at text files with a text viewer and playing MAHJONGG by Nels Anderson. Everything else seems to crash the system. I have tried my sound board on my father's 486-50 and have found that you can get almost anything to run as long as the file isn't real big and isn't real processor intensive. (processor intensive on a 486? yeh, right.)
The second modification to this circuit involves making the sound board compatible with both forms of Stereo-on-1 formats. When I first ran across MODPLAY PRO, I was only aware of that particular version of the Stereo-on-1 format. In fact, back then, I really hadn't stumbled across very many programs that utilized a parallel port sound board in either stereo or mono. Recently though, I have found a couple of MOD players and 3 to 4 demos that support this other format. While, MODPLAY PRO is still my favorite program, I have found that these other MOD players actually deliver better sound quality. (they're less sophisticated)
You'll find the entire schematic for this modification in the PCSTADD1.GIF file. There are a few things that should be noted about this modification:
1. This circuitry is only temporary. In V2.00 I
plan to have this circuitry switched in and out
of circuit with some auto-detect circuitry when
needed.
2. I've designed this circuit so that all you have
to do is stick it in-line with the input of the
V1.00 sound board. No modification to the
actual sound board is required, so if by chance
you've already built it, you're in luck.
3. This circuit is impedance dependent, so both
the inverter chip on the V1.00 sound board and
the inverter required by this circuit should be
74HCT04 in order for the circuit to work as I
intended it to.
4. I have tested this circuitry only on my machine
for every mixing frequency allowed by my
machine and it seems to work fine. However,
I cannot guarantee that it will work perfectly
on all systems, as of yet. It is simple enough
though that building it and testing it on your
system shouldn't require much effort. If you
have problems with it: I would like to hear
from you.
5. When I crank up the volume on my amplifier, the
left channel seems to have a little noise on
it. (Of course, this could just be my
amplifier.) I'm almost sure that this is the
result of this "band-aid" approach to making
the original V1.00 board compatible with V1.50
stereo sound, and I'm fairly sure that when I
do a complete overhaul that this problem will
disappear. As it stands, the noise isn't
really noticeable until you turn the amplifier
up to about 20 watts or so. If you can
actually hear it, try fudging with the
capacitor value associated with the left
channel of the V1.50 circuit. If you select a
value that is too low, a clicking noise shows
up. If you pick a value that is too large then
the sound from the left channel becomes less
and less stereo. Chances are, you won't even
notice unless you are looking for the noise.
FEATURES:
1. Stereo compatibility with MODPLAY PRO,
2. Mono compatibility with SPUTTER, MODPLAY PRO, WOW200
3. Auto-detection and processing of stereo signals
4. Auto-fallback to mono when stereo signal ends
5. Compatible with most software supporting Covox
Speech Thing
6. Indicators for stereo and power on
7. Limited compatibility with Disney Sound Source
LIMITATIONS AND COMPATIBILITY:
Now, for a discussion of the limitations of this and all similar designs. This and all other sound boards that reside on the parallel port are no substitute for the SoundBlaster or Adlib cards. Both the SoundBlaster and Adlib cards contain a dedicated sound processor which allows for the playing of relatively high quality sound during graphics intensive programs such as games. Because this design does not contain its own processor and most types of sound files are highly processor intensive, most software programmers would not even consider making sound files for their programs compatible with this or any similar design. So what does that leave us compatible with? The answer to that question is all programs written for parallel port
sound boards that don't try to do a lot of heavy graphics manipulation at the same time.
Basically this comes down to two file types: .MOD and .VOC or .SND. .MOD files are supported by MODPLAY PRO (my favorite) written by Mark Cox whose program also contains the specifications which the core part of the V1.00 circuit is loosely based on. .MOD files are sound files that may contain up to 31 digitized instruments that are played four at a time (2 through the left channel and 2 through the right channel). Since each instrument is only digitized once and the computer calculates how fast to play the samples back to achieve different pitches, each file may only be a few hundred K bytes long yet playback for several minutes. For a more detailed description of .MOD files refer to the documentation files included with MODPLAY PRO by Mark Cox. I have also found some other .MOD file players that support parallel port sound boards. There is a sister program to MODPLAY PRO called MODEDIT by Norman Lin which allows you to edit or create your own .MOD files.
The other type of sound files are .VOC, .SND, and a whole host of other file extensions, including .WAV sound files which are associated with Windows 3.x. These files are digitized so that when t=0 the instantaneous voltage is digitized and recorded. This is repeated for t=1,2,3,... for t equal to some preset constant time interval. Hence, you can have files that playback for only a couple minutes that may take up several megabytes of space. Almost every computer (IBM, Amiga, Mac, etc.) has their own format of this type of sound file and with the right program you can playback all of these file types on whatever computer you have. SPUTTER written by Adrienne Cousins is one such program. It will play just about anything and also includes a text to speech synthesizer.
There is one important note to make about the comparison of a SoundBlaster Pro to my sound board. SoundBlaster Pro can only achieve a maximum mixing frequency of 23 KHz while playing sound files or MODs in its stereo mode. So far, I haven't found the limit on this sound board. I have found programs that will support my board with a mixing frequency of up to 63 KHz. What does this mean? It means that with a fairly decent computer and my sound board, my sound board can sound about 3 times better than a SoundBlaster Pro in the stereo mode. Even my computer (a 386-20) can achieve about 2 times better stereo sound output. (For comparison, a low end CD player samples at 44.1 KHz)
You'll note that feature 5 says something about compatibility with Covox Speech Thing. Some programs (including a few decent games) support these parallel port sound boards. So far I haven't run into any compatibility problems with Covox Speech Thing except those that specifically require Smooth Talker (a copyrighted program by Covox).
In the last version of this DOC file, I said that the sound board can support some Disney Sound Source programs as long as they don't try to auto-detect the DSS card. Well, since then, I have purchased a DSS and some of its supporting software, and found out that this version of the sound board really doesn't support the DSS format at all. (Sorry, about that.) It turns out that the only DSS program, that I have found, that supports my sound board is MODPLAY PRO. But since the DSS is a mono board and my board is in stereo, there isn't any logical reason to set MODPLAY PRO up in the DSS mode.
So far I am only aware of software for the IBM that supports this hardware design. Although, any computer that will handle a parallel printer should be able to accommodate this sound board (if you know someone who wants to write software for it).
The software listed above will work with any of the IBM models. However, MODPLAY PRO requires at least a 286 at 10 MHz. In theory, .SND files aren't as processor intensive as .MOD files and should be able to accommodate slower machines. Case in point, a friend of mine owns a Tandy Color Computer 3 which can play .SND sound files through roughly equivalent internal circuitry yet the processor runs slower than the slowest IBM PC-XT. (He recently upgraded to an Amiga 500. Unfortunately, he isn't as interested in collecting MOD files as I am.)
FUNCTIONAL DESCRIPTION V1.00:
When a stereo signal is present, pins 1 and 14 on the parallel port are active. (pins 1 and 14 are normally high) Thus, initially when pin 1 or 14 goes low, the 2N3906 is properly biased and the .47 uF capacitor connected to the threshold pin of the 555 timer is discharged to ground. This causes the output of the 555 timer to go high which in turn lights the stereo indicator and also allows the G input on each 74HCT373 latch to follow an inverted signal from their respective strobes (pin 1 for the right channel, pin 14 for the left channel).
When either latch receives a high on their G input the output lines on that latch then change to reflect the same states found on their corresponding input lines. When the signal on the G input falls low again, the output lines remain at the states they were most recently set to. The capacitor connected to the G input serves a double purpose. First, it isolates the signal found at the G pin from feeding back to the inverter driving the 2N3906. Second, it cuts the pulse width of the original strobe signal for the hell of it.
Each DAC0808 then takes the output from their respective latch and converts it into a corresponding voltage below ground.
The LM324 Op Amps take the voltage at the output of their respective DAC0808 and amplify the voltage as well as bring the voltage positive with respect to ground.
As long as pins 1 and 14 stay active the .47 uF capacitor never gets a chance to charge up, thus the circuit remains in the stereo mode. Once the activity stops on pins 1 and 14 the .47 uF capacitor charges up causing the output of the 555 timer to swing low. This causes the stereo indicator to go out and also forces the G inputs on both latches to go and stay high. This results in the output lines on the latches to continually follow the their
corresponding input lines, which allows for mono compatibility.
FUNCTIONAL DESCRIPTION V1.50:
Basically the functional description of V1.50 is the same as V1.00 and in fact it is the same if you are using the sound board in the V1.00 mode. The main difference between the two Stereo-on-1 modes is that the V1.00 mode uses two strobe lines to decode the stereo signal and V1.50 uses only one strobe line. The way that V1.50 decodes the stereo signal is that it latches the right channel on the negative-going edge of the strobe and latches the left channel on the positive-going edge. Thus the data on the bus changes when the strobe line is either a stable high or a stable low. This method is actually more efficient than the V1.00 method, not only with conserving port lines, but also from a programming standpoint. On the other hand, if I had stumbled across this method first, it would be considerably more difficult to adapt the V1.00 method to it.
HOW TO PRINT GIF FILES:
I don't know of any nice, easy ways to print GIF files. I have found a couple of ways to do it, though. Both of which require you to have a program called VPIC by Bob Montgomery which can be obtained from almost any BBS.
The first way requires you to also have a copy of CSHOW which also can be obtained from almost any BBS. The first step is to load the GIF file into VPIC and invert the colors (so black is white, and white is black, etc.) and save the file in GIF format. Second, load the inverted GIF into CSHOW and print it out. Note: Some colors don't seem to print, so you will have to draw in some of the lines manually.
The second way requires you to purchase Dr. Halo. The first step is to load the GIF file into VPIC and convert it to a CUT file. Then load the CUT file into Dr. Halo and print it. (I haven't actually tried this, so I am not sure how well this method works.)
If GIf files are so hard to print, why would I pick them as the format to distribute my schematics in? As far as I know GIF files are the easiest to view. I was originally distributing the V1.00 schematic in AutoSketch and AutoCad formats but I figured that the people most likely to be interested in my circuit probably wouldn't have either of these programs. I could distribute the files in TIF format, but I haven't found any way to display this format.
NOTES:
I have tested the sound board by replacing the 74HCT04 on the V1.50 adapter with a 74LS04 and verified that my suspicions are correct. The sound board wasn't able to successfully separate the stereo signal. Therefore, from here on in, only construct this circuit using HCT logic as is specified by all of my schematics. Although the basic function of a 74HCT04 and a 74LS04 are the same, they don't appear to be directly substitutable for this application. I am going to assume that trying to substitute all other logic types will result in similar problems. The bottom line is: build the circuit with the parts that I have specified in the schematics. If you've built the circuit using a different logic type and you are having problems getting the circuit to work properly, be aware that this might be the problem.
You should always turn this sound board on before or at the same time as the computer. MODPLAY PRO has trouble detecting the sound board if you don't. If you have a printer also connected to the same parallel port, you will need to turn it on as well. If you don't, sound through the board will sound noisy and distorted.
Trying to operate the sound board on parallel port where there is either a hardware or software printer buffer won't work very well, if at all.
MODPLAY PRO is freeware. SPUTTER and WOW200 are shareware. All of these programs can be found on your favorite BBS's.
Although the schematic says that the voltage input should be + and - 12 volts, you can actually get by with anything between + and - 9 to 15 volts, just make sure that the power supply is balanced. I have run this circuit off of two nine volt batteries for over an hour.
SPUTTER by Adrienne Cousins also has a TSR that allows you to play a sound file in place of the system beep, whenever you open or close a file, a certain time of the day, whenever you press a certain key, or just about
anything else on your computer. As an example of what this program does, I have my autoexec.bat set up so that whenever my computer boots up the computer plays a sound file of Data from Star Trek - T.N.G. saying "Sir, I have completed a level one diagnostic. There are no malfunctions." I have the TSR set up to play "Excellent" whenever I open a file and "Bogus" whenever I delete a file. Both of those sound clips are from the movie, Bill & Ted's Excellent Adventure.
Pin 7 on the 74HCT04 should be connected to ground. Pin 14 on the 74HCT04 should be connected to +5 volts. (the output of the 7805) It is an assumption on most schematics that if using gate chips (inverters, AND's, OR's, etc.) that the power and ground lines have to be hooked in order to ensure proper operation even though you may never see anything on the schematic indicating that these chips have power and ground lines. Usually, but not always heres how power and ground pins are connected on a digital I.C.:
Remember, the purpose of decoupling capacitors is to eliminate noise on the power supply lines as close to each chip as possible to ensure reliable operation of those parts. For that reason, I do not show decoupling capacitors on the schematics. Inexperienced people have a tendency when they see them on a schematic to group all of them in the same area on the board or replace them with one large capacitor instead. This completely defeats their purpose. So remember to wire in a decoupling capacitor of 0.1 uF across the power and ground line to each chip as close to the chip as possible. The only requirement on the type of capacitor that should be used is that to stay away from electrolytics, usually disc capacitors work best.
There is one note to make about the V1.50 circuit. If the sound coming from the sound board seems to be distorted, noisy, or missing altogether in one channel, then you probably have the switch set to the wrong position. Try the other switch position and see what happens. Remember that for most toggle switches that when the switch is in the "top" position the "bottom" pins on the back are connected. However, for a slide switch, the pins that are directly under the sliding part of the switch are in contact.
I make it a point on all my circuits to use I.C. chip sockets with all of the I.C.'s. If you've ever had to try and remove a 14 pin chip from a soldered board, you know why. Socketing chips makes it easier to debug the circuit as well.
If you obtain a copy of SPUTTER by Adrienne Cousins, make sure you add the following line to the AUTOEXEC.BAT:
SET SPUTARG=/P378
The above line specifies that the sound board is on LPT1. If your sound board is on a different parallel port, you will have to look in the accompanying DOC files to see what number you should replace 378 with.
On the back of the computer there are two types of ports that look almost identical, but aren't. The parallel port is characterized by a female (one with holes) DB25 connector. Serial ports can have a male (one with pins) DB25 connector. Don't connect the sound board up to a serial port, it won't work. And no, using a gender changer on a serial port doesn't convert it to a parallel port.
It is always a good idea to enclose all circuits in metal boxes. However, when circuits are enclosed in a metal box, the box should be grounded in some way. There are a couple of main reasons for this. If the hot wire of the high voltage side should ever come loose (and make connection with the box), the fuse or the house's circuit breaker will blow (if the box is earth grounded) protecting the user from a potentially fatal shock. High voltage wiring and power transformers make good antennas for 60 Hz hum. An earth grounded box will effectively eliminate almost all of this problem. On low voltage circuits, the box should be connected to the circuit ground. This prevents almost all noise (EMI, RFI, AFI, etc.) from this circuit from being transmitted to other circuits. It also prevents almost all noise (EMI, RFI, AFI, etc.) transmitted by other circuits from reaching this one. If possible, it is a good idea to connect the earth ground to the circuit ground. Before making this connection permanent, it is a good idea to test it first. Occasionally, you will run into a situation where the earth ground will feed more noise in than it will take out. (This is usually a result of the earth ground having to much resistance in the line. Be aware that if this is the problem, then that outlet should be considered very unsafe.)
The ZNRs and zener diodes in the power supply are provided for surge protection. They are not "needed" for the circuit to operate. The ZNR on the high voltage side will prolong the life of the power switch. The two ZNRs on the low voltage side of the transformer serve to help keep high voltage spikes from getting across the transformer. (the one on the primary side does most of the work) The two zener diodes clamp reverse voltage and transient spikes that could be potentially damaging to the sound board circuitry or create loud snaps through the audio outputs during power up, power down, and when large appliances "kick" in or out.
The transformer in the power supply schematic is a 25.2 V AC centertapped stepdown transformer. As long as the transformer is centertapped and stepdown (from 120 V AC) the secondary voltage can be anywhere from 24 volts to 30 volts AC.
TROUBLESHOOTING GUIDE:
Problem:
No sound whatsoever.
Possible causes and solutions:
This could be caused by a lot of things. Is the power on? If you are using batteries, are you sure that they are not dead? Is the sound board input hooked up to the computer? Is the sound board output connected to the line inputs of a suitable amplifier? Is it on? Are you sure that there aren't any hardware or software printer buffers in the system? Are you sure that the program you are using supports the sound board and that the program is setup to use it? Are you sure that the board is wired properly? If your amplifier has a source selector switch, make sure that it is set to the position corresponding to the sound board. If there is a fuse in the power supply, is it good?
If all else fails, try another parallel port. If it is necessary to purchase another parallel port, remember the cheap ones seem to work best. (Cheap multi-I/O cards are typically about $11 mailorder.)
Problem:
There is sound, but it sounds like shit.
Possible causes and solutions:
If you have a printer sharing the parallel port, make sure that it is on, or just unplug its data cable. Make sure you use the line inputs on the amplifier, if you use the microphone inputs the signal may sound distorted. Make sure that you have the audio ground hooked up between the sound board and the amplifier. You may need to use shielded patch cords between the sound board and the amplifier. (I always do.) If you are using batteries, they may be almost dead. It could be a loose connection, try shaking the board. (If this temporarily fixes it, then the problem is definitely a loose connection.) It could be that you mixed up the data bus wires going to the 74HCT373s or to the DAC0808s. It could be that you forgot to hook up the power and ground pins to one or more of the chips. Make sure that if the V1.50 circuit is hooked up, that the switch is in the right stereo mode. Make sure that at least pins 20 and 25 on all of the DB25 connectors are connected to ground on the V1.50 adapter and the V1.00 sound board. (Remember, pin 25 on the input of the V1.50 adapter is not only ground to that board, but it also runs through to the output DB25 connector.)
Problem:
The stereo light comes on and stays on or flashes, when there isn't a stereo signal present.
Possible causes and solutions:
Something in the auto-detect circuitry isn't wired properly. (The auto-detect circuitry region consists mainly of the KS555, the 2N3906, and the 74HCT04, and pins 1 and 14 on the DB25.) There also may be a loose connection.
Problem:
The fuse in the power supply keeps blowing every time I turn it on.
Possible causes and solutions:
Somewhere in the circuit a (+5, +12, etc.) power line has been accidentally connected to ground. If the fuse seems to be good up until it starts trying to play sound, then it is likely that the output of a chip has been accidentally connected directly to ground (or power). (The chip in question, may also be damaged.) This problem can also occur if a chip has been placed in backwards.
Problem:
After several hours of the sound board playing perfectly, the sound will start to cut out (and the power LED may dim).
Possible causes and solutions:
If you are using batteries, they're probably getting low. Most likely, though, the power regulators are shutting down because they are overheating. The 7805 and the other voltage regulators have built in thermal protection; when they get too hot, they shut down. This can be fixed by mounting a bigger heatsink on the voltage regulators. The immediate fix is just to shut off the power to the sound board and to let it cool off naturally. (The above problem only happened to me when I was using a logic probe that I was powering off of the sound board's power.)
Problem:
The program that I am trying to run, either says that there isn't enough memory or that I have some memory managers in the memory that need to be removed.
Possible causes and solutions:
Dust off the DOS manual and look up how to make a bootable disk. Once you have the bootable disk, insert it into the floppy drive and reset the computer either by shutting the power off for a few seconds or pressing the RESET button on the computer. (Ctrl-Alt-Delete won't work correctly.) You should be able to run the program now. By booting off a floppy, the AUTOEXEC.BAT and CONFIG.SYS files are bypassed, and thus all the driver files that normally reside in memory aren't loaded into your computer when it boots. If you need the mouse to work, you will have to find your MOUSE.COM (or similar) file and run it.
If, however, you are one of those unlucky persons that has STACKER (Yuck!) or a similar program running on your hard drive, it is not recommended you boot off of floppy. You'll have to refer to the STACKER manual to determine how to free up memory. If you're feeling lucky and decide to boot off of floppy and your drive is STACKed then accessing the drive (modifying, adding, or deleting files) without the STACKER drivers in memory may cause fatal errors to occur, in other words you would be simulating your own software induced hard drive crash. If you don't know if your hard drive contains STACKER, just ask whoever setup your computer if they installed it. For the most part, only people that have hard drives smaller than 130 Mb actually go out and buy STACKER.
QUICKIE CIRCUIT TROUBLESHOOTING GUIDE:
This section will attempt to point you in the right direction if you come to suspect that something is incorrectly wired on the circuit board. It is impossible for me to predict all of errors that one could make when wiring this board, so here are the most common or most likely.
If you are going to troubleshoot this board you are going to need a DC volt meter (preferably a DMM) and a logic probe. The logic probe should be set to TTL and the sound on it should be set to pulse mode so that if the signal is pulsing, the probe makes noise, and if the signal is stable, the probe is quiet. The logic probe's power leads should either be connected across the +5 and ground on the sound board or the probe should be connected to an independent 5 volt DC power supply with the ground on the independent source also tied to the ground on the sound board.
Unless otherwise specified, all of the troubleshooting techniques require that the power to the sound board be on and some sort of program should be outputing sound to the sound board. In order to isolate problems that I may not have thought of, it is important to follow the troubleshooting directions in order.
Often times in this guide I will specify that you should be reading a specific voltage AC or DC, this means that you have to set the DMM or volt meter to either the AC or DC range before you try to read it. For example, if I specify that with the meter hooked up in a certain way that you should be reading + 12 volts DC, then you have to set the DMM or voltmeter to read DC volts before trying to read it.
Problem:
The board sounds fine but the power light never comes on.
What to check:
In all likelihood, the LED was wired in backwards. Try removing it, turning it around, and putting it back into the circuit. (You may experience a similar problem with the stereo indicator if it is wired in backwards.)
Problem:
The board doesn't output any sound and the power and stereo lights don't come on or are very dim.
What to check:
Trouble Shooting the Power Supply:
This usually indicates that something is not right with the power supply. If you have a continuity meter handy, verify that the fuse in the power supply circuit hasn't opened up. Next, place the leads of the DMM across the primary side of the transformer, the meter should read about 110 to 120 volts AC. If it doesn't then it is likely that either the fuse, the fuse holder, the power switch, or the cord is to blame.
From here on in when I mention the transformer, I am referring to the secondary side, the low voltage side. Make sure that don't confuse the primary and secondary sides of the transformer. Hooking anything to the primary that belongs on the secondary, whether circuitry, test equipment, or people can cause serious damage to the transformer, the circuitry, the test equipment, and you or any other living creature nearby.
Connect the black lead (negative) of the DMM to the sound board's ground (I suggest the ground pin of a chip). First, check the transformer's ground, by placing the red probe on the secondary centertap of the transformer. The appropriate reading is zero volts AC and DC. If there is a voltage here then it is possible that you forgot to tie the secondary centertap of the transformer to ground or that there is a loose connection at this point. Next, use the red probe to check each of the poles of the secondary side of the transformer. If you are using a 25.2 volt AC centertapped transformer, then the meter should read about 12.6 volts AC on each pole. If it doesn't, then the transformer may be bad.
Next, place the red probe on the positive output side of the bridge rectifier. The voltage should read greater than + 14 volts DC. If it does not, then one or both of the two diodes connected at this point may be in backwards or may just not work. If it is because they don't work, then something else down the line may have caused them not to work. Check the negative output side of the bridge rectifier in the same way, it should read less than - 14 volts DC, numberwise.
Next, test the inputs to the voltage regulators. The inputs of the voltage regulators should test out as follows:
7812 same as positive bridge output
7912 same as negative bridge output
7805 about + 12 volts DC
If you don't then it is likely that there is no connection. Be advised that you should go through all of the testing instructions pertaining to the 7812 and 7912 and then go back and test the 7805.
Next, test the ground pin on all of the regulators. All three should test out to be 0 volts DC. If not, then they probably have a bad connection.
Next test the output of all of the voltage regulators. The output pins should test out as follows:
7812 about + 12 volts DC
7912 about - 12 volts DC
7805 about + 5 volts DC
If not, then the regulator is bad or it has been wired incorrectly.
Check to make sure that the power lines running from the 7812, 7912, and ground actually make it to the sound board.
This just about raps it up for the power supply. You should test for voltage across each of the capacitors and the other components in the power supply. If you aren't reading any DC voltage, then it is likely that there is a bad connection. If you are reading a stable but wrong voltage, then you may have a short between the line you are testing and somewhere that contains that voltage.
Special Case:
There is a special case if the fuse keeps blowing every time the power is turned on. The first thing to do is disconnect the sound board from the power supply, in other words cut the + 12 volt and the - 12 volt line. (You can leave the ground connected.) Turn the power on with a new fuse installed. If the fuse doesn't blow then you have a short circuit in the sound board somewhere. If the fuse blows then you have a short circuit in the power supply.
In either case, make sure you check all of the chips to make sure that one or more of them aren't in backwards. Check all of the electrolytic capacitors to make sure that they have been installed correctly, reversing an electrolytic is like shorting the power to ground. Make sure that all of the diodes are installed correctly. Now, check all of your wires (or traces) to make sure that they all run to the correct places.
Problem:
The power light comes on, the power supply has been checked out and it works, but there is no sound.
What to Check:
The first thing to check is whether or not there is a data signal on the parallel port. You will need to have a program running that is continually sending a signal. (preferably mono) Use the logic probe to check pins 2 through 9 on the DB25 connector. All of the pins should have a pulsing signal on them, most of the time. If they don't, doublecheck to make sure that the program is sending the data to the parallel port. If the lines are still dead, then try flipping switch on the parallel port card that disables its interrupt. Reset the computer and try this test again. Also, make sure that you have connected the sound board to a parallel port and that you don't have any printer specific programs hogging the parallel port. If it fails the test again, then you probably need another parallel port.
Second, once you've established that there is a signal on the parallel port, the next thing to check is whether or not sending a stereo or mono signal out to the port has any effect on its condition. To do this run a program that outputs mono, then run a program that outputs stereo. If one mode works and the other doesn't then the problem most likely lies in the auto-detect circuitry. If both modes don't work then the problem is most likely to be one of the following:
1. power to the LM324 is not connected.
2. 5 V power to the entire board is not
connected.
3. the data bus is not connected to the
sound board.
4. one or more chips are not connected
correctly or are in backwards
5. the G pins on the 74HCT373s are being
held low indefinitely
If you confirm that the problem is #5 then something in the auto-detect circuitry isn't working properly.
Problem:
There is sound, but its noisy and distorted.
What to check:
Check to make sure the data bus lines aren't crossed at some point. If you are playing a stereo signal make sure the stereo mode select switch is in the correct position. Make sure the negative voltage supplied to the DAC0808 is at - 12 volts DC (if you built the power supply).
Well this covers the main problems. I promised someone I would have this file done by January 10, so I'm gonna wrap this section up kind of early. In the next version I'll improve this section and possibly add a schematic diagram with all the voltages you should test for.
WHAT'S NEXT? (V2.00):
By my next major release I plan to have support for both Stereo-on-1 formats finalized and an auto-detect feature that will automatically determine which type of signal is present and process it accordingly. It'll be a while before I get around to this, so don't expect to see it before the summer '93. At that time I'll release two schematics, one that will be like this modification and require very little or no modification to the V1.00 sound board and the other which will be a complete overhaul of the V1.00 sound board.
Also by my next release, I plan to make my sound board compatible with Stereo-on-2 formats. This is actually a simple (and cheap) modification to the circuit itself, but if you already have a V1.00 built, it might be easier to start from scratch. I don't know if this feature will have associated auto-detect circuitry. Auto-detecting this type of signal is a bit tricky. I know of few programs that use this type of stereo, and it is understandable why. This method is actually easier and faster from a programming stand-point.
I'm not sure if I'll get to my next feature by my next release and I'm not sure how messy it is going to be to integrate it into sound board. I plan to add more Disney Sound Source compatibility. Disney (and Mozer, the company that actually made their sound board,) have several patents and copyrights on their design, so I am NOT going to attempt 100% compatibility. Basically, the principle behind their sound board is that they send out an auto-detect signal on one of the lines and then wait for answer from the DSS on another line. This is how they know whether or not a DSS exists in your system. Then, if they find one, they send out the sound code out on the 8-bit bus with a latching signal. While their sound board isn't looking at the bus they send out a garbage signal to keep people like us from having compatibility with their programs. To keep the people at Disney and Mozer from hunting me down and shooting me, I am not going to fake their auto-detect signal, but only offer an alternative to listening to DSS sound through their board. That means in order to take advantage of this feature you would still have to buy a DSS.
What advantage would this have? Well, I bought a DSS to figure out how they work. So here's what they offer: a maximum of a half watt (1/4 nominal) of power output through their internal amplifier, output through a two inch speaker, not easily adapted to external amplifier circuitry for better sound, a headphone jack that doesn't sound much better than the speaker (I tested it with Sony digital-ready headphones), and a nine volt battery power source that dies every few hours. On the bright side, the DSS is completely portable and easy to use.
Also planned for the next release is a bill of materials listing specific suppliers and prices and artwork for etching a double-sided copper clad board. I have artwork drawn up right now for the V1.00, however, when I went to construct a V1.00 for myself using the artwork, I found a few minor bugs in it. Included in the version that I built for myself over the summer, I added a volume control, a switch that turns the sound off when I want to print something, and the ability to daisychain a printer through my sound board to the computer. I also substituted a blue point source LED for the red power indicator. (yep, it was worth it.) I'll try to include all of these features next time. (except the blue LED, it costs $2.80 through Digi-Key)
Sometime in the future, I would like to get around to making a sound digitizing board. Right now, I imagine that the board would consist of 5 or 6 chips and various other discrete components. Parallel ports are particularly easy to use when outputing data, however, they don't lend themselves very well for inputing data. Therefore, I would most likely use a serial port. (Actually, after doing some preliminary calculations, the serial port may prove to be too slow. No, I will NOT consider using the game port instead. Contrary to popular belief, the game port doesn't contain any ADCs, otherwise it would be a perfect solution.)
Possible specifications:
- 8 or 10 bit input (depends on the price of 10 bit)
- Stereo input (a must)
- Software selectable sampling frequency
(Possibly better than SoundBlaster Pro's)
- Possible hardware filtering
Unfortunately, there seem to be a few blocks in my way. Currently, I know about 4 different assembly languages, but none of them are utilized by the IBM. I don't think that it will be anytime soon that I will be diving into 80386 assembler either. Therefore, if anyone knows of any shareware or freeware programmers (Assembler or C preferred) out there that would like to write the software for my hardware, have them get in contact with me. Perhaps I can have the hardware schematic release done for it by next year.
I've been kicking around the idea to sponsor a programming contest. In the next release, I'll tell you if it's a go or no-go. Right now, I'm not sure that there is enough interest in my sound board or not for it to be an effective way get more software out there to support it. Besides which, I'm a college student majoring in electrical engineering at a private school (M.S.O.E.) and don't have any money to put out for really nice prizes. If I do go ahead with the contest, the prizes would go something like this:
1 - First Prize - A V2.00, fully assembled and tested
2. The program must be a game (recreational or educational)
3. The program must support VGA (256 colors)
4. It must support my sound board in one of it's stereo
modes, although only programs utilizing Stereo-on-1
will be considered for First Prize.
5. The program cannot exist before the contest begins.
6. You must be prepared to release the program as either
freeware or shareware after the contest is complete.
7. The program doesn't have to be 8088 compatible. It
should be compatible with at least a 386. (a 486 is
highly recommended, a 286 is optional)
8. Don't use extended or expanded memory, unless the
program will function without either.
9. Keyboard control is a must, but feel free to use
joystick and/or mouse.
10. Yes, multiple entries from the same person/group would
be accepted. (1 year usually isn't enough time to come
up with more than one program, though)
11. The program should be written in either Assembler, C, or
possibly even Pascal. (NO BASIC PROGRAMS ALLOWED!!!)
and whatever else I can think up. At the time of the contest, all of the applicants would receive a copy of the official rules, detailed info on my sound board so that they can design the program whether or not they have a sound board, and a copy of the judging sheet so they will know exactly what I am looking for. Like I say though, right now I'm not sure if I'll try this contest or not.
If you or anyone you know is interested participating, please drop me a line. I'm sure the determining factor of whether or not I do this is going to be determined by how many letters (from serious programmers) I get in 1993 saying "Go for it."
SOURCE LIST:
There are two main companies where I buy parts. They are:
JameCo
1355 Shoreway Road
Belmont, CA 94002
1-800-831-4242
and
Digi-Key Corp.
701 Brooks Ave. South
P.O. Box 677
Thief River Falls, MN 56701-0677
1-800-344-4539
Each of these companies has their own idiosyncrasies. Both of them will be happy to send you a free catalog, though.
JameCo:
- has all of the components needed to build this circuit
- $30 minimum order
- component prices are usually better than Digi-Key
- shipping charges: UPS ground $7.25
- offers individual component data sheets for $0.50 ea.
- offers no alternatives to backordering
Digi-Key:
- has most (if not all) of the parts required
- no minimum order: $5 fee assessed to orders under $25
- shipping charges: UPS ground $0.00 (FREE!)
- offers a few alternative parts to backordering
If you find it necessary to purchase an additional parallel port, the sound board has been tested on a part # AT-IO. It is a multi-I/O card compatible with 286, 386, and 486 computers. It was purchased from:
National Computer Accessories
769 N. 16th St., Ste. 200
Sacramento, CA 95814
1-916-441-1568
The card costs $7.99. Shipping costs = $2.95 + UPS ground shipping costs. The multi-I/O card also has a serial port and a game port.
ACKNOWLEDGEMENTS:
I'd like to thank Mark Cox for writing a really good program that makes the IBM sound a lot better than you can ever imagine. I'd also like to express thanks for developing the specifications that I based my V1.00 circuit on.
I'd like to thank Adrienne Cousins for writing a program that can play almost any sound file format.
I'd like to thank Disney and Covox for setting standards that a few programs actually chose to support.
I'd like to thank Apogee, Id, and Sierra in advance for their DSS support. I hope you guys (and gals) will consider expanding your software support to include my and any other parallel port sound boards.
I'd like to thank the SpacePigs, Skyhawks, Twilight Zone, Cascada, and whoever else for making all of those kick-ass demos that support my sound board.
I'd like to thank DongleWare Publishing for making a decent game with lots of sounds. Next time, how about easing up on the ram requirements? (You know, most of us do have hard drives.)
I'd like to thank the makers of Adlib and SoundBlaster cards for making their products so damn expensive that it is well worth it to build this circuit.
TESTING:
The V1.00 sound board has been tested and found to work on the following systems:
a 386DX-20
a 386DX-33
a 486DX-50 (had to use cheap multi-I/O card, instead of
the one integrated on to the floppy/hard
controller)
The V1.50 circuit has been tested and found to work on the following systems:
a 386DX-20
I haven't had the chance to test it on anyone else's computer, yet.
If you don't have one of the systems listed above, don't worry. This is only a list of what the sound board has been tested on. The sound board should work fine on all computers that have a parallel port, since the parallel port is the most important factor in determining whether or not the sound board is compatible, and there are several guidelines that manufacturers of parallel ports must follow to keep it a standard format.
DISCLAIMER:
Any damage you do to yourself, your computer, your stereo, or anything or anyone else is your own fault, not mine.
To the best of my knowledge there are no errors in the circuit schematic.
This circuit schematic has been given to you in hopes that you will copy it, distribute it, and use it to build this circuit.
Altering this file (like removing my name) or any of the other files in this ZIP file is strictly prohibited and illegal. Selling this file for a profit is strictly prohibited and illegal.
I don't expect any fee for using this file. However, if you find this schematic to be of great use, a donation of $5 would be greatly appreciated. This schematic took me more time to draw up and document than the circuit took to construct and debug.
Send any questions, comments, and donations to:
Dave Boyd
W2740 Rock River Paradise
Watertown, WI 53094
You can also Email me via EXEC-PC BBS at (414) 789-4210.
