BitWrangler wrote on 2024-04-18, 22:15:

devius wrote on 2024-04-18, 20:26:

Ozzuneoj wrote on 2024-04-17, 22:40:

I want someone to figure it out though. The detective work was actually a lot of fun. 😁

Media Vision Thunder and Lightning. A video and sound card in one. I couldn't find the ad that you mentioned, just a schematic of the card on VCFED and it matches your photo.

Heavy Metal VGA ad https://books.google.ca/books?id=gasgHhfj-RAC … ghtning&f=false

Brilliant Deals from Media Vision Ad (tiny pic) https://books.google.ca/books?id=rjsEAAAAMBAJ … ghtning&f=false

Yep! Those are some of the only references I could find, aside from Cloudschatze's post here.

I was about 80% sure it was a Thunder and Lightning (based on the similarities to the Thunderboard and the fact that it looked like a VGA port) until I saw that schematic of the card, that bumped it up to a solid 97% certainty. It's never 100% though... you never know how many variants of cards these crazy people made back in the day. 😁

Cloudschatze wrote on 2024-04-18, 21:12:

Ozzuneoj wrote on 2024-04-17, 22:40:

As it is, even if it's only the back 3 inches of the card and the rest is torn off I believe it will still be the only photos of that much of the card since it was released. Granted, there may be pictures on a non-english speaking site and they just never come up in Google searches.

Re: A gallery of strange sound devices

Awesome! So you did find one! That post doesn't come up in searches because the name isn't actually in text anywhere in the thread.

Anyway, I got the lot today and I am pleased to say that it is in fact a Thunder and Lightning.

Pics are obviously after some cleaning and TLC, but it was overall in good shape. Only major cosmetic issue is that the chrome plating on the back plate is bubbling up in several places, which is not something I can say I've run into with cards in such good condition. I will probably try to clean that up and re-coat it at some point.

And... oh my... there were so many more video cards in the lot than I expected.

I thought there would only be 6 or 8 of them from the pictures I saw, and I definitely did not see the ATI WINTURBO (Mach64 GX 2MB VLB) card. It has a couple badly bent legs on the Mach64 chip, but they appear to be grounds. I still separated the pins very carefully and resoldered them to the pads, but if I beep out the first two legs on each corner of the chip they are almost all grounds. I compared this to an ISA Mach64 GX and it was the same on that one, so even if they had been broken off it'd likely have still worked fine.

Other neat cards in the lot:

Orchid Fahrenheit 1280 Plus VLB (S3 805 I believe), Diamond Speedstar Pro VLB (CL-GD5428), STB Powergraph S3 805 VLB, pile of Trident VLB and ISA, No less than three Acumos AVGA1 ISA cards, Evershine ET4000AX (unusual brand with neat screen-printed logo, PLUS it has someone's writing in sharpie on the back "Jen 11-10-92"... so cool!), Cardinal Technologies ET4000AX, OAK OTI-087 ISA, and several others.

Also got a pile of other misc stuff that wasn't really useful (old Token Ring cards and such), and a possibly-usable Asus P5QL-M (CM5571) micro-atx Socket 775 board with one broken cap.

All in all, I think it was a good purchase (understatement)... but there's a fair amount of work to do here in fixing things, unbending pins (already done a bit) and testing. Also, lots of tantalums to roll the dice on.

I will probably make a thread about the Thunder and Lightning some time. 😀

Holy guacamole that's a thousand bucks worth of vid cards besides.

That is a very nice score !

Yeah, it was definitely a good lot. As mentioned though, there's going to be a lot of work here. They were wrapped a bit but were boxed with some pretty hefty stuff (junk) so pretty much everything is scuffed and has bent things. I imagine there will be a LOT of pin reflowing in my near future.

The absurd thing is that I don't even have a VLB test setup going right now. I need to set one up before I can really start testing these.

... there really was a LOT more here than I had anticipated. 😅

Oh, now that I think of it, I do have a question for the more circuit-savvy individuals here.

I was curious and did a quick test for shorted tantalums on the Thunder and Lightning, just using the basic continuity\diode tester mode on my DMM. I can do repairs, but I am not good with understanding circuit designs, so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted? It is near the top edge of the card, the center cap in that group of 5. It is labeled CT940. None of the others seem to be shorted.

I also have a basic LCR meter which has helped to diagnose some things in the past. When I put it across the legs of that cap it shows around 44uf and I think like .5 ohm ESR (could be wrong on the resistance). I know this is mostly useless info since it is in-circuit, I'd just like to do whatever I can do avoid a possible cap blow out. I've only experienced a couple in all the years I've been doing this and the many hundreds of old cards I've messed with. The danger of tantalums exploding seems greatly exaggerated to me... but I'd rather not experience one on this particular card if at all possible given the rarity.

Great Haul. What chipset does that Orchid VLB card have I wonder?

I have no Trident VLB specimens and am pretty envious to see, what, like 5 of them in there 🤣!

Enjoy man.

Ozzuneoj wrote on 2024-04-19, 06:34:

so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted?

Not that I know of, I certainly cannot think of any situations. Only way resistance that low would read is if there's another parallel component that would exhibit such a resistance, or maybe a shorted piece of silicon but in this case I'm pretty sure that tantalum needs replacement. You should certainly test it out the circuit.

Ozzuneoj wrote on 2024-04-19, 06:34:

Yeah, it was definitely a good lot. As mentioned though, there's going to be a lot of work here. They were wrapped a bit but wer […]
Show full quote

Yeah, it was definitely a good lot. As mentioned though, there's going to be a lot of work here. They were wrapped a bit but were boxed with some pretty hefty stuff (junk) so pretty much everything is scuffed and has bent things. I imagine there will be a LOT of pin reflowing in my near future.

The absurd thing is that I don't even have a VLB test setup going right now. I need to set one up before I can really start testing these.

... there really was a LOT more here than I had anticipated. 😅

Oh, now that I think of it, I do have a question for the more circuit-savvy individuals here.

I was curious and did a quick test for shorted tantalums on the Thunder and Lightning, just using the basic continuity\diode tester mode on my DMM. I can do repairs, but I am not good with understanding circuit designs, so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted? It is near the top edge of the card, the center cap in that group of 5. It is labeled CT940. None of the others seem to be shorted.

I also have a basic LCR meter which has helped to diagnose some things in the past. When I put it across the legs of that cap it shows around 44uf and I think like .5 ohm ESR (could be wrong on the resistance). I know this is mostly useless info since it is in-circuit, I'd just like to do whatever I can do avoid a possible cap blow out. I've only experienced a couple in all the years I've been doing this and the many hundreds of old cards I've messed with. The danger of tantalums exploding seems greatly exaggerated to me... but I'd rather not experience one on this particular card if at all possible given the rarity.

Just me but I would be replacing every one of them evil little bastards regardless of how they test, tantrum caps are pretty cheap to buy in bulk and certainly worth the cost for a card as rare as this one. As for the time .. I mean its a rare card likely worth a bit in working condition so .. worth the cost in time to just replace every tantrum cap.

Ozzuneoj wrote on 2024-04-19, 06:34:

Yeah, it was definitely a good lot. As mentioned though, there's going to be a lot of work here. They were wrapped a bit but wer […]
Show full quote

Yeah, it was definitely a good lot. As mentioned though, there's going to be a lot of work here. They were wrapped a bit but were boxed with some pretty hefty stuff (junk) so pretty much everything is scuffed and has bent things. I imagine there will be a LOT of pin reflowing in my near future.

The absurd thing is that I don't even have a VLB test setup going right now. I need to set one up before I can really start testing these.

... there really was a LOT more here than I had anticipated. 😅

Oh, now that I think of it, I do have a question for the more circuit-savvy individuals here.

I was curious and did a quick test for shorted tantalums on the Thunder and Lightning, just using the basic continuity\diode tester mode on my DMM. I can do repairs, but I am not good with understanding circuit designs, so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted? It is near the top edge of the card, the center cap in that group of 5. It is labeled CT940. None of the others seem to be shorted.

I also have a basic LCR meter which has helped to diagnose some things in the past. When I put it across the legs of that cap it shows around 44uf and I think like .5 ohm ESR (could be wrong on the resistance). I know this is mostly useless info since it is in-circuit, I'd just like to do whatever I can do avoid a possible cap blow out. I've only experienced a couple in all the years I've been doing this and the many hundreds of old cards I've messed with. The danger of tantalums exploding seems greatly exaggerated to me... but I'd rather not experience one on this particular card if at all possible given the rarity.

Learn from my mistake Smoke coming from PAS16

Still waiting for my 16x tantalum’s to arrive and audio chip.

Tiido wrote on 2024-04-19, 07:53:

Ozzuneoj wrote on 2024-04-19, 06:34:

so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted?

Not that I know of, I certainly cannot think of any situations. Only way resistance that low would read is if there's another parallel component that would exhibit such a resistance, or maybe a shorted piece of silicon but in this case I'm pretty sure that tantalum needs replacement. You should certainly test it out the circuit.

Would a capacitor in parallel with a diode read short on both legs?

appiah4 wrote on 2024-04-19, 11:03:

Tiido wrote on 2024-04-19, 07:53:

Ozzuneoj wrote on 2024-04-19, 06:34:

so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted?

Not that I know of, I certainly cannot think of any situations. Only way resistance that low would read is if there's another parallel component that would exhibit such a resistance, or maybe a shorted piece of silicon but in this case I'm pretty sure that tantalum needs replacement. You should certainly test it out the circuit.

Would a capacitor in parallel with a diode read short on both legs?

No.

Looks like CT940 goes between -12V and +5V analog supplies (based on the copper areas on the back), so seems unlikely there should be a short between those two. Might want to try just removing it and making sure there isn't a short between the two holes. Only thing I can think immediately think where there might be a reason to have a short across a capacitor would be maybe an output smoothing capacitor on a linear regulator that can have a input-output bypass jumper, so you could e.g. switch from 5V to 3.3V.

appiah4 wrote on 2024-04-19, 07:12:

Great Haul. What chipset does that Orchid VLB card have I wonder?

I have no Trident VLB specimens and am pretty envious to see, what, like 5 of them in there 🤣!

Enjoy man.

I believe the Orchid Fahrenheit 1280 Plus is an S3 805. Some day I would love to find one with the sound card components populated (that's what the upper right corner is for). I have an ISA model and now a VLB, and I have seen others and they always have the locations clearly marked but unpopulated. There are some other variants I believe, but this one is meant to have an (ancient) ESS ES488 onboard. If I ever find an ES488 with a damaged PCB I would be tempted to move the components over to one of these. That'd be so cool. 😁

Trashbytes wrote on 2024-04-19, 08:41:

Just me but I would be replacing every one of them evil little bastards regardless of how they test, tantrum caps are pretty cheap to buy in bulk and certainly worth the cost for a card as rare as this one. As for the time .. I mean its a rare card likely worth a bit in working condition so .. worth the cost in time to just replace every tantrum cap.

Honestly, unless you do board repairs on devices like this for a living and do it constantly, I would put the risk of accidental PCB damage during repairs much higher than the risk of tantalums popping when they aren't already shorted. As mentioned, I've handled and used tons of cards and old PCs and have only ever had three failed tantalums, and only two of those were "eventful".

One was on a Trident 8900C ISA card, which popped violently the first time the card was powered. I replaced it and the card was fine... though I was much more aware of the problem after that. The second was in an Everex EV-659 which I bought brand new in box to put in my PC 5150 (to provide EGA and Parallel on one card). The PC wouldn't power on with the card installed. I located a shorted tantalum, replaced it with whatever old electrolytic cap I had laying around of similar value and it has been working fine for 8 years now. Third failed tantalum was when I plugged in a Gridcase 386 laptop that was given to me and hadn't been used in probably 30 years. There was a fizzle in the PSU and it died. I replaced it and it was good to go... at least until the RIFA X cap blew and probably took a couple years off my life with fumes... those are caps you do not leave in devices. Ever!

In each of these cases the tantalums failed after not being used for 25-30 years. I've personally never had one go on a card that I had powered on previously, and I have likely put power through tens of thousands of tantalums (dipped or SMD).

zuldan wrote on 2024-04-19, 08:59:

Learn from my mistake Smoke coming from PAS16

Still waiting for my 16x tantalum’s to arrive and audio chip.

Thankfully, I learned from that on a the Trident card mentioned above many years ago. I check them for shorts now. 😁

snufkin wrote on 2024-04-19, 15:21:

Looks like CT940 goes between -12V and +5V analog supplies (based on the copper areas on the back), so seems unlikely there should be a short between those two. Might want to try just removing it and making sure there isn't a short between the two holes. Only thing I can think immediately think where there might be a reason to have a short across a capacitor would be maybe an output smoothing capacitor on a linear regulator that can have a input-output bypass jumper, so you could e.g. switch from 5V to 3.3V.

I could be wrong, but I don't think those solder pads actually connect to the -12v and +5v planes. I don't know what the proper terminology is, but the solder pads are on "islands" with no bridge to the surrounding plain. Their traces are likely on the other side of the PCB.

Tiido wrote on 2024-04-19, 07:53:

Ozzuneoj wrote on 2024-04-19, 06:34:

so is there ever a normal (not failure) situation where both legs of a tantalum, in circuit, would appear shorted?

Not that I know of, I certainly cannot think of any situations. Only way resistance that low would read is if there's another parallel component that would exhibit such a resistance, or maybe a shorted piece of silicon but in this case I'm pretty sure that tantalum needs replacement. You should certainly test it out the circuit.

Well, there must be something going on. I just removed the tantalum and it is fine... not shorted. Those pads are still shorted somewhere, and I'm beginning to think it is intentional. The lower (+) leg goes to the plane around it. The other leg goes directly above to the closest (bottom left) pin of the U15 IC. The "short" seems to be to ground. Maybe this is just a filter for that IC?

That is strange, I would think in a case like this, the tantalum wouldn't be doing anything... it would help to know exactly what the circuit is (little bit of schematic). Is it a dead short or just low resistance (that doesn't change with time) ? And does it measure same in both directions ?

That particular pin on U15 (an opamp) is the power supply to that chip. It shouldn't be measuring against pin4 (GND).

Tiido wrote on 2024-04-19, 22:16:

That is strange, I would think in a case like this, the tantalum wouldn't be doing anything... it would help to know exactly what the circuit is (little bit of schematic). Is it a dead short or just low resistance (that doesn't change with time) ? And does it measure same in both directions ?

That particular pin on U15 (an opamp) is the power supply to that chip. It shouldn't be measuring against pin4 (GND).

Doh... so... the continuity test gives a solid beep. However, when I measure resistance across the two points it is 9.9 ohms in both directions and doesn't change. It's just relatively low resistance I guess... not a short to ground.

... so this was probably all a waste of time, right? Other than being able to test that one tantalum out of circuit to verify that it wasn't shorted.

You'd think after all these years of dabbling in this stuff I'd know how to diagnose circuits better than this.

EDIT: Also, the datasheet for that opamp shows that pin is Vee, which is apparently "negative supply", where Vcc is "positive supply".

In this case, it looks like this part of the circuit is: Pin4 of Opamp (negative supply) --> - Tant + --> GND

There's one school of thought that goes; connect 12V directly across tantalum with some improvised shielding around it, and if it goes bang it's bad.

Okay, I think this was a wild goose chase. I have a Thunderboard here, and it has the same LM351 and pin4 is connected to ground with effectively zero resistance (0.1 ohm). Maybe in the case of the Thunder and Lighting they didn't want the video and sound portions of the card to interfere with one another so they put that tantalum between pin 4 and ground as a filter... even though there is very little resistance between those points anyway? I don't know. 🤣

Ozzuneoj wrote on 2024-04-19, 23:12:

Doh... so... the continuity test gives a solid beep. However, when I measure resistance across the two points it is 9.9 ohms in […]
Show full quote

Doh... so... the continuity test gives a solid beep. However, when I measure resistance across the two points it is 9.9 ohms in both directions and doesn't change. It's just relatively low resistance I guess... not a short to ground.

... so this was probably all a waste of time, right? Other than being able to test that one tantalum out of circuit to verify that it wasn't shorted.

You'd think after all these years of dabbling in this stuff I'd know how to diagnose circuits better than this.

EDIT: Also, the datasheet for that opamp shows that pin is Vee, which is apparently "negative supply", where Vcc is "positive supply".

In this case, it looks like this part of the circuit is: Pin4 of Opamp (negative supply) --> - Tant + --> GND

Whoops, I swapped the pin purposes...

But yeah, beeps can be deceptive. It seems like this is normal and this capacitor is probably not going to throw a tantrum 🤣.
10 ohm is fair bit of load though (depending on voltages involved of course), but if it is on the same rail as the speaker amp it is probably quite normal.
EDIT: Ok, the mystery is solved then I suppose.

Tiido wrote on 2024-04-19, 23:42:

Whoops, I swapped the pin purposes... […]
Show full quote

Ozzuneoj wrote on 2024-04-19, 23:12:

Doh... so... the continuity test gives a solid beep. However, when I measure resistance across the two points it is 9.9 ohms in […]
Show full quote

Doh... so... the continuity test gives a solid beep. However, when I measure resistance across the two points it is 9.9 ohms in both directions and doesn't change. It's just relatively low resistance I guess... not a short to ground.

... so this was probably all a waste of time, right? Other than being able to test that one tantalum out of circuit to verify that it wasn't shorted.

You'd think after all these years of dabbling in this stuff I'd know how to diagnose circuits better than this.

EDIT: Also, the datasheet for that opamp shows that pin is Vee, which is apparently "negative supply", where Vcc is "positive supply".

In this case, it looks like this part of the circuit is: Pin4 of Opamp (negative supply) --> - Tant + --> GND

Whoops, I swapped the pin purposes...

But yeah, beeps can be deceptive. It seems like this is normal and this capacitor is probably not going to throw a tantrum 🤣.
10 ohm is fair bit of load though (depending on voltages involved of course), but if it is on the same rail as the speaker amp it is probably quite normal.
EDIT: Ok, the mystery is solved then I suppose.

Okay, I put the original cap back in place and the card works fine.

BIOS came right up the first time on my 440BX system...

I ran a quick DOS sound setup program for a game and digital and FM sound are both working fine, though there is a pretty pronounced crackle in the background. It could be just about anything (I have -5v, but maybe the system is too new\fast for it), but I'll tinker with it a bit to see if I can get it to go away. Aside from that, it is working exactly as it should. 😀

No explosions, no smoke, no fire. Just... thunder and lightning. 😆

I also tested every tantalum on every other card in this lot for shorts and didn't find a single one, which is a nice, but pretty much what I have experienced in the past.

Excellent ~