VOGONS


First post, by SSTV2

User metadata
Rank Oldbie
Rank
Oldbie

Not too long ago, I got a working sample of AWE64 Gold (CT4540), which I've decided to test today more thoroughly. Installed card into the P1 MMX system with Win95C and tried to test all available MIDI synthesizers under the MIDI tab in Multimedia settings, with MIDI file playing in the background. The moment I switched to a different synth, an error message popped up, stating, that a MIDI device cannot be changed while it's in use (or something like that) and at the same time - midi file would stop playing.

After clearing error message, I've set MIDI settings back to defaults, but after that, no midi file I played would output audio, games included. Shortly after, I realised that MIDI music got nothing to do with it, card simply no longer outputs audio.

Reinstalled drivers and all CT related software - still no audio, then changed ISA slot and cleanly reinstalled drivers, still, card would not output any audio.

To completely rule out software related issues, I've installed card into SS7 system with Win98SE. Included Win98 drivers were automatically installed and AWE would still not output any audio, only then I realised that things had gone awry with it...

Started inspecting ICs on it and instantly noticed, that negative voltage regulator appeared unusually hot. Measured its input and output voltages, input was in range, but output was only -4.6V instead of -5V (at least voltage regulator is functional), but such voltage drop on the output indicates that something draws way more current than it should. After some poking, I found another IC that was unusually hot, it was 33079 quad OP amp IC (U16) right next to mixer IC. Every other SOP IC next to the audio out connectors were at normal temps.

It's the first time something like this had happened to me, no sound card ever died on my watch. Prior to that MIDI device switching in control panel, card was working perfectly. I find it difficult to believe, that software related action like that could cause such damage, could anybody explain to me what the hell happened here, was it mere coincidence or something in posibility with these cards (never owned one before)?

Reply 1 of 10, by Jo22

User metadata
Rank l33t++
Rank
l33t++

Hmm. That's strange, indeed. 😐 I know of only two cases so far where software can damage hardware:
* Outputting unsuported frequencies to monitors (esp. TTL monitors)
* Overheating the DMA controller (by using output as input ?)

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 2 of 10, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

Don't forget the time when nVidia released a driver that caaused video cards to overheat and die because the cooling fans never came on.

But in this case, I think it was just a coincidence. I bet you can find that one chip on mouser or similar and replace it if you are good at soldering.

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 3 of 10, by Jo22

User metadata
Rank l33t++
Rank
l33t++

Oh, true. I also forgot: A Windows driver by the maker of a specific USB-Serial adapter chip did intentionally kill clone chips.
The drivers erased the USB ID of the clone chips, thus bricking hundreds or thousends of devices (including medical and scientific ones).

Edit: I don't like to mention any names (Vogons is not the perfect place for that),
but Here's more about that old story: https://tinyurl.com/yad3qvh7

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 4 of 10, by SSTV2

User metadata
Rank Oldbie
Rank
Oldbie

Okay, removed faulty IC and now negative voltage regulator outputs proper -5 volts. Going to test that quad OP amp IC, maybe just one of the amps had failed. Also, I could not find mixers datasheet or pinout, thus I do not know where exactly OP amp connects and what purpose it has in circuit. Could somebody provide CT1745A pinout?

viHaneOl.jpg?1

Reply 6 of 10, by SSTV2

User metadata
Rank Oldbie
Rank
Oldbie
derSammler wrote:

Just try grabbing the audio input from the pins where the IC was soldered. If there's nothing too, the amp is probably not the culprit.

Traced some of the pads down the board and it appears that this quad OP amp IC is in no way related to the R/L line-out. Two OP amps on the right side of the circuit are definitely used for PC speaker header and MIC and two others for CD audio and line-in (not 100% certain).

Tested quad OP amp and it's completely dead. Decided to test card under Windows w/o it and card on OS bootup would output just loud noise. Opened up mixer and muted everything I could, only then the card could output decent enough audio, but it was still a bit distorted. If all of the channels were to be unmuted, card would stop outputting audio completely, audio would return only with a complete PC shutdown, reset does nothing.

Touching pads of missing OP amp reduces audio distortion significantly, so I guess that replacing OP amp should fix the card, also, mixer IC is still responsive. 😊

Reply 7 of 10, by SSTV2

User metadata
Rank Oldbie
Rank
Oldbie

I'd like to know why did that OP amp die, is that mixer chip flawed in some way? Also, how can I be certain that a replacement won't suffer same fate, what else could cause OP amp responsible for MIC/line-in/CD audio/PC-speaker to fail, any ideas?

Reply 8 of 10, by appiah4

User metadata
Rank l33t++
Rank
l33t++
SSTV2 wrote:

I'd like to know why did that OP amp die, is that mixer chip flawed in some way? Also, how can I be certain that a replacement won't suffer same fate, what else could cause OP amp responsible for MIC/line-in/CD audio/PC-speaker to fail, any ideas?

Have you routed something through line-in that output at levels way above normal? Something with pre-amp applied perhaps?

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 10 of 10, by SSTV2

User metadata
Rank Oldbie
Rank
Oldbie

Substituted 33079 OP amp with a far better specs and roughly 10 times more expensive OPA4134UA OP amp. Thought about replacing a current 33079 OP amp responsible for line-out, but decided to leave it as it is... For now... 😀

NG4Hpebl.jpg?1

Card works perfectly now.