Did you replace caps? A fried chip?

I have considered replacing caps first and going from there. I was just curious if the symptoms pointed to bad caps or a fried chip.

I cannot tell directly from this picture but blue electrolytic to the right of C13 is suspect. It is leaning a good bit which may mean it has built enough pressure inside to start pushing the rubber plug out of the bottom.

It looks like it's in pristine condition.

So you hear something, like just noise or faint music sound even?

Maybe the volume pot is crusty and needs cleaning? Have you tried just adjusting the volume up and down several times to clean the pot a bit?

Or maybe the ISA connector needs cleaning? It does look very good to me, but some contacts do have some oxidation on them.

Other than that, usual procedure of fixing things is to debug what is wrong with it and then repairing it.
What I mean is don't blindly replace the caps, the odds are it won't fix it.
It would help if you have access to an oscilloscope, so you could systematically probe what works and what doesn't.

Since you only hear some noise, it appears the amplifier chip is OK, so maybe the CMS chips don't output audio, maybe it does not see writes from ISA bus.
Or maybe just the pot is broken so audio is cut between CMS chip and amp.

Jepael wrote:

It looks like it's in pristine condition. […]
Show full quote

It looks like it's in pristine condition.

So you hear something, like just noise or faint music sound even?

Maybe the volume pot is crusty and needs cleaning? Have you tried just adjusting the volume up and down several times to clean the pot a bit?

Or maybe the ISA connector needs cleaning? It does look very good to me, but some contacts do have some oxidation on them.

Other than that, usual procedure of fixing things is to debug what is wrong with it and then repairing it.
What I mean is don't blindly replace the caps, the odds are it won't fix it.
It would help if you have access to an oscilloscope, so you could systematically probe what works and what doesn't.

Since you only hear some noise, it appears the amplifier chip is OK, so maybe the CMS chips don't output audio, maybe it does not see writes from ISA bus.
Or maybe just the pot is broken so audio is cut between CMS chip and amp.

I think the pot is ok, the noise floor moves smoothly with the dial. What is interesting is how the system locks up when audio playback is attempted.

Does it lockup when you try games?

You don't have an Adlib driver like ADLIB.COM or SBFMDRV.COM loaded as well by any chance? Any driver that assigns itself to INT 80H might cause a crash.

Great Hierophant wrote:

Does it lockup when you try games?

You don't have an Adlib driver like ADLIB.COM or SBFMDRV.COM loaded as well by any chance? Any driver that assigns itself to INT 80H might cause a crash.

Yes, it locks up with games, I first tried with Revenge of DOH. Some of these systems had a CT1350B in there, but I pulled it and removed the text from AUTOEXEC. I am curious as to what causes the driver to assign itself to INT 80H.

Formulator wrote:

I think the pot is ok, the noise floor moves smoothly with the dial. What is interesting is how the system locks up when audio playback is attempted.

That's good info! There is a rumour that says when you try to write to C/MS chips that are not installed on old Sound Blasters 1.x cards the PC will hang. This is because writes to C/MS chips are slow there is circuitry to trigger on the IO write and it will trigger a signal towards the ISA bus to add wait states to an IO cycle, and this signal is removed when the C/MS chip indicates it has gotten the write.

So when the C/MS chips are not installed on SB1.x (or not responding), the wait states are added up to infinity and PC hangs. Some people say it should not hang because there is supposed to be a timeout so that a device cannot hold the bus cycle for too long.
(I'd try this, but my SB 1.5 has the C/MS chip soldered on, not on sockets).

My next question is, can the games and demo programs detect the presence of the sound card and its address automatically?
If so, then at least half of the card is working, but the sound chips may have some issue.

Also, if possible, remove the CT1350B when doing these experiments. I think the PAL chip on CT1350B missing prevents the ISA bus hold circuitry from working, so trying to access C/MS chips on CT1350B does not hang the PC. And obviously the sound cards need to be in different base addresses to work. The point being, the games look for a Creative card from all base addresses and it's usually from address 210h onwards. There may be differences, but they might first detect if it is a CMS card or SB card, and they will usually settle for the first card they find.

Jepael wrote:

So when the C/MS chips are not installed on SB1.x (or not responding), the wait states are added up to infinity and PC hangs. Some people say it should not hang because there is supposed to be a timeout so that a device cannot hold the bus cycle for too long.
(I'd try this, but my SB 1.5 has the C/MS chip soldered on, not on sockets).

Interesting, I do have a CT1350B with the socketed C/MS chips. I wonder if this experiment would work with the C/MS jumper open and the sockets removed. Probably not, but worth a try.

Measure the voltage on pin A10 (I/O CH RDY) of the ISA bus when the computer is stuck. If it's logic low the system was hung by the wait state cirquit.

The electrolytic capacitors cause no/bad sound when defective but will not freeze the system.