Not very scientific but I fixed a sound card once with similar issue by simply giving a good wash - i.e. warm soapy water, let it soak for a couple mins, light scrub with a toothbrush or equivalent, rinse with fresh water, and let it dry for a week +.

Is it a temporary or permanent fix?

Edit: Is soaking the card in isopropyl alcohol a good idea?

bjwil1991 wrote:

Is it a temporary or permanent fix?

Edit: Is soaking the card in isopropyl alcohol a good idea?

isopropyl alcohol can leave residue if you're not careful.

I(d use DeOxit to clean it.

Jade Falcon wrote:

bjwil1991 wrote:

Is it a temporary or permanent fix?

Edit: Is soaking the card in isopropyl alcohol a good idea?

isopropyl alcohol can leave residue if you're not careful.

I(d use DeOxit to clean it.

Arrgh. Apparently, there's no store closer to me that has it, and they charge $20 + Shipping and Handling for a 5oz can of deoxit. I tried putting on a new jack since the one was falling apart on the inside of the jack, and it still won't play nice, and it only outputs music to the right speaker only with the new jack. Maybe I didn't add enough solder to the new jack.

Edit: replaced the remaining jacks, the problem still persists.

First, see if it's the same problem on line out and speaker out. Then, start mapping the signal path for the line out to try to find any caps or semiconductors that need to be bypassed, tested, or replaced.

gdjacobs wrote:

First, see if it's the same problem on line out and speaker out. Then, start mapping the signal path for the line out to try to find any caps or semiconductors that need to be bypassed, tested, or replaced.

Looked like all the caps are intact, not bulging or anything. Maybe it'll be a good idea to replace those caps since it's possibly damage from humidity (super hot in the US right now) or bad packaging when I received the card.

Have you washed that bad boy yet?? Super hot you say? It'll dry in a couple of days. Easier than replacing caps.

badmojo wrote:

Have you washed that bad boy yet?? Super hot you say? It'll dry in a couple of days. Easier than replacing caps.

Just looked at the card and one cap is loose, and about to come off, while the other ones are stable. Is the Amp IC the three-legged chip held in with a screw and nut?

bjwil1991 wrote:

badmojo wrote:

Have you washed that bad boy yet?? Super hot you say? It'll dry in a couple of days. Easier than replacing caps.

Just looked at the card and one cap is loose, and about to come off, while the other ones are stable. Is the Amp IC the three-legged chip held in with a screw and nut?

That's a voltage regulator.

I dropped some words in the retro activity thread regarding this project. Good luck with your card.

On this board shot, the amp is a TDA1517 from Phillips. In my experience, the majority of boards use a TDA (or TEA) unit in the same form factor, although the pinouts can differ. Some board designs use two smaller amps (each is mono), often two JRC units.
http://www.amoretro.de/wp-content/uploads/azt … r_pro_32-3d.jpg

As far as troubleshooting, any possibility of borrowing or using an oscilloscope?

gdjacobs wrote:

On this board shot, the amp is a TDA1517 from Phillips. In my experience, the majority of boards use a TDA unit in the same form factor, although some board designs use two smaller amps (each is mono), often two JRC units.
http://www.amoretro.de/wp-content/uploads/azt … r_pro_32-3d.jpg

As far as troubleshooting, any possibility of borrowing or using an oscilloscope?

The amp IC has a bit of rust on some of the pins, not all of them. Just one cap is loose, and the other one is not too far behind by the jacks. Also, I don't know anyone who has an oscilloscope, and the ones used today are USB only and all of my laptops and desktops are in storage for the time being. Should I order new caps and amp IC, as well as a desoldering iron to remove the bad chip, and caps?

Also, tell me if this is a good or bad idea: putting an 18-pin socket on the sound card to easily install or remove the amp IC.

You don't need anything special to remove the chip. A desoldering iron just makes it easier. Order one if you see yourself doing lots of this sort of thing, otherwise I'd just make do with braid and a pump.

I'm not a big fan of USB scopes, but they would work I guess. Probably the cheapest option is to instead buy an old style cathode ray scope (CRO), although they do take twice the space of a newer LCD scope. Bandwidth on the order of 20 mhz would be plenty for working in audio. Again, only do this if you want to be equipped to troubleshoot audio gear or otherwise have a use for a scope.

As far as replacing parts, I'd probably hold off. I feel there's a strong possibility of the issue being with the caps. Several are installed in series for bias removal and would negatively affect the output volume if their ESR were to increase greatly due to failure. This wouldn't always be visually apparent, even amongst the electrolytics. I can't, however, rule out the possibility that it's a more serious problem, such as with the codec chip. If that were the case, the card would probably be beyond economic repair.

The best option for confirming any diagnosis is to use a scope, but I suppose it's possible to make a probe that could be fed to an audio output* so you can trace the signal path. Any caps you suspect of being problematic could be removed and, if they're in series, jumpered.

* Don't blame me if something gets blown up, though.

The amplifier chip would certainly have an effect on the speaker output as well as, potentially, on the line out if it's become very leaky. The line out and TDA signal inputs appear to be connected with series caps, so you can fully isolate the line out as an early troubleshooting step.

I looked at my other sound card, a Creative Labs Sound Blaster 16 WavEffects CT4170 that has the same chip, as well as traces for the Amp IC to the capacitors as well. I'm going to look at the Aztech card to see if there're traces from the Amp to the capacitors as well. If there is, would it be a good idea to swap chips around, and replace the caps, or just but a new IC chip and appropriate caps for that chip? Either that, or I can wait a while to replace those caps and the chip. When I have the volume for the left channel maxed out and the right channel 6 below the left channel, both speakers play audio without issues, but balanced volume or sliders in Windows 98SE or MS-DOS doesn't work out too well.

Edit 1:

Here's a snapshot of the traces between the chip and caps:

Edit 2:

Would it be a good idea to add solder to the pins on the chip that have rust or is that a bad idea?

Bypass the chip entirely. Either remove it or remove the caps required to disconnect it.

I mentioned a cheap way of tracing previously. This is what I was thinking of.
http://billmaudio.com/wp/?page_id=1254

gdjacobs wrote:

Bypass the chip entirely. Either remove it or remove the caps required to disconnect it.

I mentioned a cheap way of tracing previously. This is what I was thinking of.
http://billmaudio.com/wp/?page_id=1254

I removed the caps, and unfortunately,when I set the volume levels for the left and right channels in the mixer program, it reverts back to the default settings when running Doom.

Before running Doom:
Master: 100%
SB Voice:
L: 100%
R: 63%
FM/WT:
L: 100%
R: 63%
Line in:
L: 0%
R: 0%
Mic: 0%
Master balance: equal.
Stereo: ON.

After running Doom:
Master: 100%
SB Voice:
L: 94%
R: 94%
FM/WT:
L: 94%
R: 94%
Line in:
L: 0%
R: 0%
Mic: 0%
Master balance: equal.
Stereo: OFF.

So apparently, the caps are used to save the mixer settings, and without it, the sound card reverts back to the settings listed above, and refuses to save the volume settings as well.

Looking at the board shot I linked to above and consulting the TDA1517 pinout, the bias filtering caps are labeled C16 and C31. I'm fairly certain those are okay to remove and maintain line out functionality (unless the board does something really weird). Additional caps on the signal path (the conductors that carry the audio output signal between the line out jack and the codec) can either be removed (for shunt caps, usually connected in parallel with ground) or jumpered (for series caps).

Depending on their function, filter caps grouped with logic chips should not be removed without replacement as the result could be spurious pulses. I think there's a good chance of a chip being reset by noise that is causing your problem.