The idea, if I'm understanding his intent, is to make sure you have only one source of 5V, and to connect it with only one very short (1 cm or less) piece of very thin wire, so that if something goes wrong, it will (hopefully) break the connection, somewhat like a fuse. The wire you're using appears to be stranded; that is, made up of many very thin wires in a bundle. He is suggesting that you take only one thin strand out of the insulation and use that to make the temporary connection for testing purposes.

I wouldn't try adding memory to the card until you confirm that it is usable. Find the source of the problem first, then test that it actually works to some extent. Consider adding memory only once you know that it can actually produce sound and is safe to keep in your system.

640K!enough wrote:

The idea, if I'm understanding his intent, is to make sure you have only one source of 5V, and to connect it with only one very short (1 cm or less) piece of very thin wire, so that if something goes wrong, it will (hopefully) break the connection, somewhat like a fuse. The wire you're using appears to be stranded; that is, made up of many very thin wires in a bundle. He is suggesting that you take only one thin strand out of the insulation and use that to make the temporary connection for testing purposes.

I wouldn't try adding memory to the card until you confirm that it is usable. Find the source of the problem first, then test that it actually works to some extent. Consider adding memory only once you know that it can actually produce sound and is safe to keep in your system.

Exactly.This is how it should look like:

For pads insulation, use any scotch tape.

640K!enough wrote:

The idea, if I'm understanding his intent, is to make sure you have only one source of 5V, and to connect it with only one very short (1 cm or less) piece of very thin wire, so that if something goes wrong, it will (hopefully) break the connection, somewhat like a fuse. The wire you're using appears to be stranded; that is, made up of many very thin wires in a bundle. He is suggesting that you take only one thin strand out of the insulation and use that to make the temporary connection for testing purposes.

I wouldn't try adding memory to the card until you confirm that it is usable. Find the source of the problem first, then test that it actually works to some extent. Consider adding memory only once you know that it can actually produce sound and is safe to keep in your system.

Thank you for your replies.

Oh yea... the Red Wire is a stranded copper wire, alright. So of the two traces that I patched. I should only have one of those patched? And then have every other voltage source insulated on the card edge? And how do I go about insulating those?

Also does anybody know what you can do with these cards with no memory? This isn't a sound blaster. This is a very early General Midi type Synth card for Music. It's designed to be patched with another Sound Blaster Card in your system and to be configured to run along side it. Right now this one is set to use Address 250 I believe. The Sound Blaster produces Sound FX and the GUS produces the Music from a "Sound Font" that is loaded into its memory upon Boot up. I know there's something you can do with it with out the Memory. I'm just not sure what.

*I see SSTV2's post. Okay.... I see. And Scotch tape on the pins. Any other pins I should insulate on either side. Or just those two?

You haven't powered that card yet, yet you worry about it so much 😀. Maybe it's damaged to the point where you won't be able to fix it yourself.

Anyway, insulate just marked pads and give it a go in a dummy motherboard, who knows, maybe card will show signs of life right away.

While you're doing this sort of testing, make absolutely certain that the bare strand of wire doesn't touch anything else on the card, or the result may be worse than what you started with. You could possibly try a bit of electrical tape over the other contacts on the board that are nearby (meaning the pins of the header and nearby components, no additional contacts on the card edge, other than those you were already instructed to insulate) to reduce the chances of an unfortunate accident.

That generation of card won't do much without memory, in terms of sound reproduction, but the software should be able to communicate with the main ICs and initialise them, to the point where it will complain that there is no memory on-board. That works on newer Gravis cards, at least. Once you get to that stage, it will tell you that at least the main, mostly irreplaceable chips are functional. If nothing else is causing traces to burn up or parts to over-heat, then you will be in a position to consider a more permanent patch for those traces, removing the tape and doing some more testing, before finally taking the chance and installing the memory.

Today got the same card (ver 2.4) in similar condition: +5V trace is burnt out. It seems like it is a common problem to this PCB version.
Just patched the card with a thin isolated wire. The burn to the right of patched one is alive indeed. I left it as is because card works for now. Maybe it will affect on RAM chips functioning (at now only two of eight is installed = 256K), but they will arrive in about a month. Time willl show.

Could anyone in this thread who owns one of these take a look at their card to see what is written on the crystal at Y1 near the GF1 socket? I'd also like to know what is written on the trimmer in the center of the board at R88 (to see if different ones were used). If anyone knows what exactly the trimmer controls and how to set it properly once replaced, it would be much appreciated! I received one from a scrap lot with both those components missing but in otherwise decent condition.

If you want, you can post here:

Need help repairing a Gravis Ultrasound Classic 2.4!

Thank you guys!