What drivers are you talking about? How did you configure the card?

You can always just run UNISOUND and than the game setup to select sound card.

If this doesn't work, post hires photos of the front and back of your card.

I think I tried drivers from this link: http://www.vogonsdrivers.com/getfile.php?file … menustate=45,38

There are several archives inside, not sure which one I've chosen. Run the install program in DOS and let it configure everything. The card gets detected everywhere but produces only buzz. Tried original Win98SE drivers which are included on the CD. The same result.. I can try UNISOUND but I think there is some HW problem on the card.

Will post some picutures tomorrow.

Here's pictures. Please note that I've removed one electrolytic cap (C101, middle down) that looked damaged to verify it's ok. Measured good so the problem is elsewhere.

Front:

Back:

I have a Vibra16S which was exhibiting a horrible static/screeching sound. The actual audio was there, but it was buried by the noise.
Turned out it was the amp chip, which should be this one I've highlighted on your card.

Replaced the chip (easily found on eBay) and the noise was gone, and the audio was as it should be.
Can't say it's the same problem you have, but worth looking in to, if you don't have any better ideas.

Paar wrote on 2022-04-05, 08:28:

Here's pictures. Please note that I've removed one electrolytic cap (C101, middle down) that looked damaged to verify it's ok. Measured good so the problem is elsewhere.

The caps that can kill your sound are the caps in the top right quarter of your card. A quick check you can do to test whether the problem is in the analog part (so most likely the caps) or the digital part of your card: Set the mixer to record "What-U-hear", i.e. the audio output of your card. This enables loopback from the output of the 1745 mixer chip to the input without needing the caps and amplifiers in the top right part. You can then play a MIDI file in Windows and record the output using the Sound Recorder. Play back the recording with a working sound card. If the recording sounds OK, some caps in the top right part are most likely the source of your problem.

RockstarRunner wrote on 2022-04-05, 12:13:

I have a Vibra16S which was exhibiting a horrible static/screeching sound. The actual audio was there, but it was buried by the noise.
Turned out it was the amp chip, which should be this one I've highlighted on your card.

I have the problem even when listening to the audio through Line Out, bypassing the amplifier. So it probably is not the source of the problem. Probably.

mkarcher wrote on 2022-04-05, 12:21:

The caps that can kill your sound are the caps in the top right quarter of your card. A quick check you can do to test whether the problem is in the analog part (so most likely the caps) or the digital part of your card: Set the mixer to record "What-U-hear", i.e. the audio output of your card. This enables loopback from the output of the 1745 mixer chip to the input without needing the caps and amplifiers in the top right part. You can then play a MIDI file in Windows and record the output using the Sound Recorder. Play back the recording with a working sound card. If the recording sounds OK, some caps in the top right part are most likely the source of your problem.

That's a good idea. Do I get it right that I should record some music with some software in Windows? Any tips for suitable SW compatible with Win9X? EDIT: You mentiones Sound Recorder... I'll try that.

Paar wrote on 2022-04-05, 13:07:

RockstarRunner wrote on 2022-04-05, 12:13:

I have a Vibra16S which was exhibiting a horrible static/screeching sound. The actual audio was there, but it was buried by the noise.
Turned out it was the amp chip, which should be this one I've highlighted on your card.

I have the problem even when listening to the audio through Line Out, bypassing the amplifier. So it probably is not the source of the problem. Probably.

Typically, there are multiple amplifier chips on sound cards. Your test of the line-out jack most likely rules out the power amplifier chip as root for your problem, so it definitely helps narrowing down the issue, but often sound cards also have low-power amplifier chips like the TL074 or the NE5532 between the mixer chip and even line-out. Electrolytic caps are considerably more prone to failure, so amplifier failure is unlikely, but I won't put all my money on the amplifiers being fine.

I've done some testing today in Windows 98. Unfortunately, I haven't managed to get mixer output on the Line In so I was unable to record anything. Everytime I tried to play something, the player looked fine (device manager didn't show any problems either), but when checking Mixer utility from Creative, there was none activity shown on the visual equalizer. Tried SB32 and equalizer was nicely active. I'm suspecting that the mixer chip could be a culprit.
But not all was a failure. I've connected Dreamblaster X1 to the AWE32 and connected headphone directly to the DB throu headphone jack and heard MIDI music just fine. Tried Line Out and heard only buzzing. That means data goes to the wavetable header properly.

Paar wrote on 2022-04-05, 19:00:

But not all was a failure. I've connected Dreamblaster X1 to the AWE32 and connected headphone directly to the DB throu headphone jack and heard MIDI music just fine. Tried Line Out and heard only buzzing. That means data goes to the wavetable header properly.

OK, so the main problem is clearly on the analog side of the card. The most likely culprits are electrolytic caps, amplifier chips voltage regulators and the mixer chipfor the analog part (ordered by how likely I expect the thing to be the culprit).

Thanks for the tip, I'll try to recap the analog part of the card first then. Hopefully it'll work.

Finally got to this card again after having it in a storage for a long time. Turned out the main culprit was bad voltage regulator generating 5V. After replacing it almost everything started to work.

Almost. For some reason there are no sound effects, even though FM and AWE32 synth works fine (tried with line out). I don't know if it's a driver issue or HW one. More investigation needed.

EDIT: Tried it in another machine with Windows 95 and the result is the same. It looks like card is trying to play the sound, but whenever it does I can only hear faint ringing sound. So it's probably a HW issue. Maybe bad opamp or something like that.

I have similar problems with two ct2760 cards.
Both of them worked when I got them, then suddenly spazzed out while in use and were never the same again.

One of them exhibits pretty much the same behaviour as yours, Paar - OPL and synth seem to play fine, but sound effects are barely audible.
The interesting thing I found was that running unisound actually makes the problem worse - everything on the card gets turned down to a barely audible level. Ironically, the card is actually fairly functional like that, as long as I turn my speakers all the way up. This is not a great solution, however, because I get heaps of noise, and the sound is still not very high.
And no, using a mixer program to turn up the levels doesn't help.

The other dead card makes intermittent screeching sounds, but doesn't seem to output anything else but white noise.

I have replaced nearly every through-hole component I can think of on the second card - totally recapped it, changed the amp chip, and replaced both of the voltage regulators. I think I might have messed something up, because I seem to recall that it used to play everything quiet and scratchy through only one channel. Now it just screeches. I checked over all my work with a fine tooth comb though, and can't find any mistakes, so I don't know what's going on there. (My soldering can best be described as a hatchet job, but there's no short circuits or anything, and all the components are in the right spot and properly aligned.)

I haven't done much to the first card because my soldering skills aren't that great and I don't want to mangle another card unless I have reasonable confidence it will work.

<b>Just wondering, if anybody knows about these things, if the DAC chip (CT170x) or the mixer chip (ct1745) are at all prone to failure?</b>
Either/both of those failing would, to my mind, explain the issues with these cards.

I happen to have a badly damaged CT2290 sb16 (Pulled it out of a dumpster. It has extensive and weird/hard to identify damage but many of the SMT components are still fine) and I was thinking of using it as a donor card. It also has the supposedly noise-free CT1703-A!

My issue is that I have two cards I want to repair, I have only one donor card, and I've never done SMT work before. I'm more willing to screw around with the one I've already pulled apart, because it's already pretty ugly. But I'm also worried that I've messed something up on that card already and my efforts will be wasted anyway.

Anyway, I'll probably take the plunge and make some decision in the near future. If I don't totally screw up I'll post my findings here and hopefully it might be useful to other people having similar issues.

AFAIK bad op-amps are pretty common thing on Sound Blasters. I would check them one by one, that's what I'm planning to do when I have the time. I have to indentify which one is used for wave and check whether it works properly or not. And possible replace it.

Screeching sounds may be caused by bad voltage regulators. Have you checked them?

Yep, I've replaced both voltage regulators on both cards with brand new ones and it didn't make a lick of difference.
I've just spent the afternoon teaching myself to desolder surface mount components. Not as scary as I thought! I just have to be careful of the surrounding tiny resistors and capacitors when waving my heat gun around.
Anyway, I'll replace the op-amps as you suggested. Again, luckily there's perfect replacements on the smashed sb16.
If that doesn't work I might move up to the DAC and the mixer chip.

I would definitely recommend buying yourself an oscillator (they're rather cheap on Aliexpress these days) and try to pinpoint the issue before doing any soldering work. You'll avoid many headaches and will save a lot of time.

Paar wrote on 2025-04-11, 06:50:

I would definitely recommend buying yourself an oscillator

Do you really mean "oscillator", to inject "beeps" into the analog circuit? Or did you intend to write "oscilloscope"?

Silly me, I meant oscilloscope of course.

I have watched the latest Necroware's video about faulty SB16s and noticed that one of them had similar problems to my AWE32 - no PCM sound. In his case absent/wrongly configured CPS chip was the culprit so I wondered if it is somehow applicable to my card. So I have checked all the jumpers again and noticed that I had CPS chip disabled. I did that purposely as figured it isn't used that much in any software anyway. I have enabled the chip and PCM started working. Hurray!

Never thought the CPS on CT2760 AWE32 is mandatory for PCM sound. Now I know. So I consider this card fixed.

Paar wrote on 2025-04-12, 19:00:

Never thought the CPS on CT2760 AWE32 is mandatory for PCM sound. Now I know. So I consider this card fixed.

It's not mandatory to have a CSP for PCM sound, but it is mandatory that the IFSD/APSD jumpers properly represent whether a CSP chip is physically present on the card. For cards with a CSP installed, the serialized digitial audio is sent from the bus interface chip into one pin of the CSP, and the CSP outputs the (possibly modified) digital audio on another pin, which is connected to the DAC. This allows the CSP to intercept and modify the digital data, e.g. add some kind of reverb or equalizing to it. Obviously, if no CSP were installed, the data from the bus interface would be sent just to an empty solder pad or connection in an unpopulated CSP socket, and no audio is heard. This is what happened with NecroWare's SB16 that had its CSP removed. To have the card work without a CSP, there are two jumpers that can connect the CSP "input" and "output" pins, so the audio data is passed directly from the bus interface chip to the DAC.

If you set the "no CSP installed" jumpers and a CSP is installed, both the CSP and the bus interface chip drive the same line to the DAC, which causes a bus conflict and (in case the CSP wins), blocks digital audio. As the CSP is initialized in "pass-through" mode, having the audio data routed through the CSP does not create any disadvantage compared to removing the CSP and setting the bypass jumpers.