As you can see the one Bridge with the wire in red insulation wasn't a big deal. How ever I found the second smaller bridge with the wire in black to be quite difficult. There was a bit of trace on the board that was connected to the pin on the card edge. Not much to work with and it shortened it self while I was attempting to tin the trace with solder. Not much luck until it was just the last bit of trace on the board until it widened up and became the pin on the card edge. A blob of solder attached and I started attaching the small bridge. I turned my Iron down to 300 and touched the small wire close to the blob of solder and that was that. I don't like how it looks, but it seems solid enough with the probes. The connections check where they should and I'm not finding shorts where I shouldn't.

What are my chances with this card working as it is when the Memory gets here? If there are chips on the PCB that went bad is there a way to source them?

It looks like if there was a component that would be bad it would be C39 on the front of the card closest to the notch in the card edge. It looks fine. Is there a way I can test whether that capacitor is bad or not with out desoldering it?

JimJim wrote:

#1 is there a chance the chips were untouched by the extra voltage?

#2 Once the Traces are bridged, is there a chance I will damage a motherboard that I will test this card in?

1. There was no extra voltage;
2. +5V rail traces would burn in different locations on the sound card.

Something is dead-shorting on that card, that's why +5V rail traces burned out like fuses.

For start, remove red wire you've added, also that solder blob on the edge. Once done, solder a short jumper wire as thin as possible in the blobs place, that would act like a fuse. Connect low voltage power source, a 3.3v rail from ATX PSU would do, inspect element(s) on the card that emit unusual ammounts of heat, remove or disconnect them from VCC, continue testing till there are no shorts left.

That is the brute method of ruling out faulty components, but is pretty effective.

You could also insulate D16 pin on edge connector (ISA pinout) and insert card in any bad motherboard in order to perform same test.

SSTV2 is right - what you are doing here is like bridging a burned fuse. The traces burned for a reason. Bridging them with a wire that can take even more current will destroy additional components. Find the faulty component(s) first, then repair the traces.

Perhaps a silly observation, but there seems to be a jumper shunt lodged on the ISA bus pin header, just at the beginning of the AT portion of the card edge connector. Could that not be shorting something. I was under the impression that those headers were to be left alone, except when installing modules like the 16-bit recording option.

That's the jumper for address selection.

derSammler wrote:

That's the jumper for address selection.

Of course, it's right on the silk screen, too, and I ignored it; silly me. 😊

Thank you for your replies.

"For start, remove red wire you've added, also that solder blob on the edge. Once done, solder a short jumper wire as thin as possible in the blobs place, that would act like a fuse. Connect low voltage power source, a 3.3v rail from ATX PSU would do, inspect element(s) on the card that emit unusual ammounts of heat, remove or disconnect them from VCC, continue testing till there are no shorts left."

Okay. The Red wire I can remove with out much fuss. However the Solder blob is what's connecting the small wire with a bit of black insulation on it. The solder blob is not the bridge, and it's certainly not a very large wire. Infact it's probably already the smallest wire I have. I should get some of the delicate thread wire, but I don't have any on hand. And if I remove it I'm not sure if I'll ever get another to attach. Am I correct in reading the ISA pin out that the side I soldered is the Data and Address side of the ISA slot?

As for testing. I'm not sure how to proceed with the 3.3v rail from the PSU. Is that in a motherboard with just the 3.3v rail connected? I might need the procedure explained to me further on how to diagnose this short.

Nope. It's the B side. I'm constantly confusing my self with the pinouts of sockets and card edge. Okay... so both B3 and B29 which I bridged are the +5V

So with the picture. I was probing around with my multimeter and found the resistor which I circled. Which is a Red Red Red Gold Resistor which should be a 2,200 ohm resistor. My meter picked up one ohm of resistance between it's leads. If there's a short in the board could that be it?

Measure resistance between +5V rail and GND, maybe shorting component was already removed, eg. RAM. If that's not the case and resistance between +5V and GND is very low, then strip everything possible from a board and remeasure resistance.

If it became way higher, then most likely you've removed shorting component, which again, can be tested separately for short by measuring resistance between its VCC and GND pins and if resistance is still low - short persists.

In that case, you'll have to make some sort of testing stand, which should have protective device - a fuse made from hair-thin copper string or some protective load - mini light bulb of up to 12V or low resistance, high power resistor (1 to 10 ohm) and a low voltage DC power source - battery or some other PSU.

Resistors do not short, they just burn open just like fuses.

SSTV2 wrote:

Measure resistance between +5V rail and GND, maybe shorting component was already removed, eg. RAM. If that's not the case and r […]
Show full quote

Measure resistance between +5V rail and GND, maybe shorting component was already removed, eg. RAM. If that's not the case and resistance between +5V and GND is very low, then strip everything possible from a board and remeasure resistance.

If it became way higher, then most likely you've removed shorting component, which again, can be tested separately for short by measuring resistance between its VCC and GND pins and if resistance is still low - short persists.

In that case, you'll have to make some sort of testing stand, which should have protective device - a fuse made from hair-thin copper string or some protective load - mini light bulb of up to 12V or low resistance, high power resistor (1 to 10 ohm) and a low voltage DC power source - battery or some other PSU.

Resistors do not short, they just burn open just like fuses.

Alight. Thank you for your reply. Those are some excellent diagnostic suggestions. I will give those a try tonight and post a reply with my findings. Again, thank you for your suggestions.

The Resistance between the +5 rail and GND is about 360 to 370 Ohms. I can probe all the VCC and GND on the chips in the board and find this resistance too. The only trouble is I don't know what resistance I should be expecting? I can also tell you that the resistor I circled before is shorted to the +5V rail on both sides of it. Not sure if that's something to pay attention to or not either. What resistance should I expect between +5V and Gnd?

I imagine someone put the DRAM chips backwards in the card and that's why the traces burned out.
360...370ohm is ok between GND and 5V and nothing that can produce any burnout. That particular resistor looks like it is part of signal path of some sort and shouldn't be anything to worry about.

It wouldnt surprise me if one of the 74 logic chips is dead. I see this on arcade boards _all_ the time.

360 ohm resistance is pretty low, but not low enough to cause traces to blow up. I'd say that normal resistance between VCC and GND should be at least in kilo-ohms range, eg. 2k+ would be totally fine.

As long as there is no dead-short on board, you could try testing that card in a PC, just strip insulation off of that red wire and bridge +5V rail just with a single string of it (use at least 1cm string), insulate remaining two +5V pads on the edge connector.

BloodyCactus wrote:

It wouldnt surprise me if one of the 74 logic chips is dead. I see this on arcade boards _all_ the time.

Thank you for your reply.

Are the chips I circled in my picture what you're talking about?

I found these on OctoPart : https://octopart.com/t74ls05b1-stmicroelectronics-40887023 / https://octopart.com/t74ls244b1-stmicroelectronics-2278500 / https://octopart.com/t74ls245b1-stmicroelectronics-2278516 / https://octopart.com/t74ls266b1-stmicroelectronics-2278535

Are those what I'm looking for if one or some of them are bad? What about the second set of numbers on these chips. The 99227 and 99231? Are those significant?

Which of these chips would you suggest I check first?

Tiido wrote:

I imagine someone put the DRAM chips backwards in the card and that's why the traces burned out.
360...370ohm is ok between GND and 5V and nothing that can produce any burnout. That particular resistor looks like it is part of signal path of some sort and shouldn't be anything to worry about.

Thank you for your reply.

My Memory hasn't arrived yet. From the few pictures I've found of the GUS 2.4 these are the ones that were fully populated. It looks like all the Pin 1's of these memory chips point at the Large GF1 Chip.

I assume I should probably wait and test this card out in a motherboard before I populate it with the Memory I'm waiting on? At what point would it be safe to populate it with the memory? And at what point should I try and install the GUS software to see if I can get a response from the Card in DOS?

SSTV2 wrote:

360 ohm resistance is pretty low, but not low enough to cause traces to blow up. I'd say that normal resistance between VCC and GND should be at least in kilo-ohms range, eg. 2k+ would be totally fine.

As long as there is no dead-short on board, you could try testing that card in a PC, just strip insulation off of that red wire and bridge +5V rail just with a single string of it (use at least 1cm string), insulate remaining two +5V pads on the edge connector.

Thank you for your replies.

"just strip insulation off of that red wire and bridge +5V rail just with a single string of it (use at least 1cm string), insulate remaining two +5V pads on the edge connector." I'm afraid I don't quite follow. It sounds like you're at the next level of diagnostics when it comes to PC cards, and this is the first card I've done any diagnostics like this on. The majority of my previous fixes for any PC board has just been replacing capacitors.

#1 Bridge +5V rails with 1cm thick string?

#2 Insulating pads on the ISA Card Edge. What do you recommend using to accomplish that.

#3 And what is the purpose for stripping the insulation off of the wire? Heat Related?

BloodyCactus wrote:

It wouldnt surprise me if one of the 74 logic chips is dead. I see this on arcade boards _all_ the time.

I also found these from the site I ordered my memory from. http://www.arcadecomponents.com/catalog/item/ … 648/1705685.htm / http://www.arcadecomponents.com/catalog/item/ … 648/1705284.htm / http://www.arcadecomponents.com/catalog/item/ … 648/1705192.htm / http://www.arcadecomponents.com/catalog/item/ … 648/1705277.htm

Those listings lack the B1 that I see on my chips and the OctoPart site. Are they the same chips or does that matter?