VOGONS


Gigabyte GA-486VS missing caps

Topic actions

First post, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Hi all, I've got one of these boards that appears to have had all its caps ripped off (there's just a couple left around the ISA slots). I realize this might be fruitless, but if anyone has one of these boards (rev 8B) that could tell me the value of all the caps I'd be super appreciative.

https://imgur.com/a/EKfQEtw

Reply 1 of 25, by Horun

User metadata
Rank l33t++
Rank
l33t++

Wow ! looks like the caps were cut in half or all were Tantalums and blew up,. Also missing the cache chips. Sorry do not have the board but just had to look.

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 25, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

I have a Rev.8a board but there shouldn't be many differences regarding the caps. In my board these are the values:

  • TC1,TC2,TC6,TC7,TC8,TC11,TC13,TC15,TC16,TC20,TC21,TC23,TC27,TC29: Tantalum, 10uF 16V
  • TC4,TC5: Tantalum, 10uF 25V
  • TC3,TC26,TC28: Electrolytic, 330uF 10V

Hope this helps. By the way, if you remove diode D7 to prevent recharging the battery you can solder a CR2032 holder to the BAT2 area stenciled below the power connector.

Reply 3 of 25, by jakethompson1

User metadata
Rank Oldbie
Rank
Oldbie

Out of curiosity is there any reason why someone would do this? Do you think they were practicing desoldering? Or wanted to re-cap the board but then quit before the new ones came in?

Reply 4 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie
TheMobRules wrote on 2021-07-18, 23:40:
I have a Rev.8a board but there shouldn't be many differences regarding the caps. In my board these are the values: […]
Show full quote

I have a Rev.8a board but there shouldn't be many differences regarding the caps. In my board these are the values:

  • TC1,TC2,TC6,TC7,TC8,TC11,TC13,TC15,TC16,TC20,TC21,TC23,TC27,TC29: Tantalum, 10uF 16V
  • TC4,TC5: Tantalum, 10uF 25V
  • TC3,TC26,TC28: Electrolytic, 330uF 10V

Hope this helps. By the way, if you remove diode D7 to prevent recharging the battery you can solder a CR2032 holder to the BAT2 area stenciled below the power connector.

That's super helpful, thank you. Thanks for the battery tip too, that'll be easier than trying to replace it with a rechargable jobby. I'll have to order these and give it a go.

jakethompson1 wrote on 2021-07-19, 00:09:

Out of curiosity is there any reason why someone would do this? Do you think they were practicing desoldering? Or wanted to re-cap the board but then quit before the new ones came in?

I was wondering that too. Considering the battery was also removed I was wondering if they ripped them all off to prevent leaking in storage as there's no visible damage suggesting anything blew up. At the same time, however, ripping them off and then storing the board doesn't seem very smart, unless they catalogued everything.

Reply 5 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Update: Got all the caps replaced, flashed a new BIOS chip and attempted to power the board on. No magic smoke, which was good, but also nothing at all happens. The PSU fan doesn't come on, it makes a very faint almost buzzing sound, and that's it. The PSU is good (verified it was still good after this test), before I call this board dead, does anyone have any ideas what might cause this?

Reply 6 of 25, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

PSU refusing to turn on means there's a short in the board (especially considering that you have verified the PSU to be in good working condition). I would guess one of the tantalums is shorted, most likely in the 12V or -12V line. Did you replace all the tantalums or are there some of the old ones left?

In any case, I would start by grabbing a multimeter and testing the resistance between the power connector pins and ground to find which line is shorted.

Reply 7 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie
TheMobRules wrote on 2021-08-04, 19:28:

PSU refusing to turn on means there's a short in the board (especially considering that you have verified the PSU to be in good working condition). I would guess one of the tantalums is shorted, most likely in the 12V or -12V line. Did you replace all the tantalums or are there some of the old ones left?

In any case, I would start by grabbing a multimeter and testing the resistance between the power connector pins and ground to find which line is shorted.

I replaced them all. After I'd finished I probed around with my multimeter in continuity mode to check what I could and what I could trace seemed okay, but of course there are plenty that go into some invisible layer. I never thought to check the power connectors. I just quickly checked in continuity mode and all my +5v connectors are grounded, which I'm assuming is my issue. I guess at this point I should start removing caps until all the shorts go away?

Reply 8 of 25, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

Yeah, you can start by checking the caps that are connected to 5V. Tantalums are usually the culprits, but other components may have shorted. Also, make sure the 5V line is actually shorted, sometimes it measures very low resistance and the multimeter beeps but it's OK, an actual short should appear as 0.00 or something close to that. Another good place to check for shorts are the legs of the SiS chips.

But the buzzing sound from the PSU and fan refusing to start is 100% caused by a short, the PSU does that to protect itself and the rest of the system. So I'd say if you manage to discover the cause there is good chance to revive the board.

Reply 9 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie
TheMobRules wrote on 2021-08-04, 20:44:

Yeah, you can start by checking the caps that are connected to 5V. Tantalums are usually the culprits, but other components may have shorted. Also, make sure the 5V line is actually shorted, sometimes it measures very low resistance and the multimeter beeps but it's OK, an actual short should appear as 0.00 or something close to that. Another good place to check for shorts are the legs of the SiS chips.

But the buzzing sound from the PSU and fan refusing to start is 100% caused by a short, the PSU does that to protect itself and the rest of the system. So I'd say if you manage to discover the cause there is good chance to revive the board.

I'm not a wizard with my multimeter, basically only used it for continuity but I'm quite certain I checked resistance correctly. Every other connector except the +5v lines were measuring resistance in kiliohms or miliohms (I believe). The +5v lines were all below 1 ohm.

Correct me of I'm wrong but without a schematic trying to find which caps are connected to +5v would be a shot in the dark. Starting with no caps installed and going one by one may be simpler.

Thanks for all your help by the way.

Reply 10 of 25, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie
Hydrohs wrote on 2021-08-04, 22:57:

Correct me of I'm wrong but without a schematic trying to find which caps are connected to +5v would be a shot in the dark. Starting with no caps installed and going one by one may be simpler.

You can check for continuity between any of the +5V pins of the power connector and the positive leg of the cap.

Once you identify those you can remove them one by one until the 5V line no longer shows continuity to ground.

Reply 11 of 25, by JimWest

User metadata
Rank Newbie
Rank
Newbie

TheMobRules is right about the short circuit. I have often noticed this behavior of the PSU in the event of a short circuit.

I think I would check everything again. All soldering points, polarity of the caps, short circuit on the caps, etc. If you can't find anything here, then I would also check all jumpers.

I had the board once too. I still have a few photos of the sale. If you find them useful, I can post them.

Reply 12 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie
JimWest wrote on 2021-08-05, 09:17:

TheMobRules is right about the short circuit. I have often noticed this behavior of the PSU in the event of a short circuit.

I think I would check everything again. All soldering points, polarity of the caps, short circuit on the caps, etc. If you can't find anything here, then I would also check all jumpers.

I had the board once too. I still have a few photos of the sale. If you find them useful, I can post them.

Yeah I took another quick poke around with the multimeter before work today and most of the caps are shorted together, so I'm probably just gonna remove them and start over.

I followed along with the manual and I'm pretty sure I got all the jumpers right, but perhaps I should check again with them all removed before I start desoldering.

EDIT: Feel like a total idiot, still had a short after removing the jumpers, so started desoldering everything until I'd run out of caps and still had a short. Then noticed I missed a couple jumpers. Removed them, no more short. Gotta reinstall all those caps now.

Reply 13 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Alright caps all reinstalled, jumpers reset, no shorts. PSU comes on normally but I get no video, CPU doesn't feel hot at all.

I've double checked jumpers and I'm confident they're set correctly. I may have missed some, there are a lot on this board but the only ones I skipped had to do with setting for Cyrix CPUs, DX4s (I'm using a 486SX for testing) or Overdrives.

Reply 14 of 25, by JimWest

User metadata
Rank Newbie
Rank
Newbie

That’s a step forward. 😃
If you post us a current photo of the board, we can take another look on it. Maybe someone notices something. Do you have such a POST card analyzer?

Reply 15 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Unfortunately I don't have a POST card, should probably pick one up for situations like this.

Here's the board currently: http://imgur.com/a/meDGoOk

I have an 8MB, 60ns EDO SIMM installed.

Reply 16 of 25, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

EDO RAM won't work with that chipset. For 486 and earlier you want FPM memory modules, only some of the later 486 boards support EDO.

Until you get the proper RAM and maybe also POST card you can try the following: remove all memory modules from the board, plug in a PC speaker to the appropriate header and turn on the board. If the board is alive, it should beep several times to indicate it hasn't found any memory.

Reply 17 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Ah, got them mixed up. Thought EDO predated FPM. Tried with some FPM RAM but made no difference. Removed the RAM and plugged in a (known working) speaker, and nothing. Speaker gets power but makes no sound. I also tried the CPU in another system I have just to make sure it didn't die in the process and it is still working. Got a POST card on the way which hopefully will help. The only other thing of note I can think of is that I have no battery installed whatsoever, not sure if that would make a difference on a system like this or not.

Reply 18 of 25, by Hydrohs

User metadata
Rank Newbie
Rank
Newbie

Alright, delayed update but the first POST card I bought didn't actually work in ISA slots. Got a working one now; all my voltages are good, clock LED is solid, frame LED is solid and the reset LED lights when the board is turned on or off, but I get no post codes whatsoever. Since the BIOS chip was removed along with all the caps when I got the board, I had to flash a new one. I went according to this post: https://www.vcfed.org/forum/forum/genres/late … 486vs-diagnosis and got a couple Atmel AT27C512R which according to the datasheet had the same pinout as the chip that was in that guy's board and flashed it with the BIOS from here: http://th2chips.freeservers.com/ga486vs/index.html

I read the chip back after the flash and it matched the image file, so I'm not sure if the BIOS is bad, this chip doesn't actually work, or this board is unfortunately just dead.

Reply 19 of 25, by Robin4

User metadata
Rank l33t
Rank
l33t
Horun wrote on 2021-07-18, 23:21:

Wow ! looks like the caps were cut in half or all were Tantalums and blew up,. Also missing the cache chips. Sorry do not have the board but just had to look.

I think the previous owner tried to re-cap the motherboard. But gave up.

~ At least it can do black and white~