https://imgur.com/a/5htVl

I think this link should work, I took a bunch of pics of the board (sorry about the extremely poor quality) and uploaded them to imgur. I would embed them but my pics always come out huge. I can take more or maybe better closeups of certain spots if there is interest. I'd really love to have this board back in working condition, but I guess if it's broken it's broken 😢

I have a whole bin of socket 7 processors from a pentium 133 all the way up to a k6-2 350 I can try in it as well. I've verified the graphics card and SD-RAM i had in it before is still ok.

BeginnerGuy wrote:

I think this link should work, I took a bunch of pics of the board (sorry about the extremely poor quality) and uploaded them to imgur. I would embed them but my pics always come out huge.

You need to read over my Imgur tutorial to learn how to resize Imgur embeds: Making clickable embedded images using Imgur image hosting service

Sounds like the motherboard's voltage regulation system comes up fine. BIOS problem maybe?

Stiletto wrote:

BeginnerGuy wrote:

I think this link should work, I took a bunch of pics of the board (sorry about the extremely poor quality) and uploaded them to imgur. I would embed them but my pics always come out huge.

You need to read over my Imgur tutorial to learn how to resize Imgur embeds: Making clickable embedded images using Imgur image hosting service

Thanks for the link and fixing my picture in the other thread Stiletto, I'll get my pictures properly formatted henceforth 😁

gerwin wrote:

Sounds like the motherboard's voltage regulation system comes up fine. BIOS problem maybe?

Gerwin I wish I knew how to tell. My repair abilities are limited to recapping and just plain throwing it in the oven (which I haven't bothered doing). I just posted this on the off chance somebody may be able to spot something.

The BIOS on this board does seem socketed so maybe it can be swapped, the Award chip on the bottom left of the board. Just kind of strange that it would randomly die like that? I thought maybe something shorted against the board while I was tinkering with drives, but I can't see the slightest hint of damage anywhere. I was enjoying this board but I see they go for $100 on ebay, too rich for me to replace 😵

According to http://oldschooldaw.com/asus/1999/p5a-107.pdf, to clear CMOS, you need to short two points on your board - there is no jumper - and power up. Check page 12 for the layout of your board and the clear location and page 14 for the clear instructions.

I imagine it is quite possible that your caps aren't doing their job. It would be awfully frustrating to replace them all and still have a board that doesn't work. It is up to you, though. What is the brand of caps in there, by the way?

I suppose it is also possible that the board could be:
- Held in a reset state
- Having trouble with power states (sleep/off/on)
- Have improper voltages somewhere (caused by bad caps, for instance)
- Voltage monitoring malfunction
- Temperature monitoring malfunction

I suppose with those last two bits of conjecture, you would get some beeps. I don't know that Award BIOS beeps are very extensive in what they provide.

CkRtech wrote:

According to http://oldschooldaw.com/asus/1999/p5a-107.pdf, to clear CMOS, you need to short two points on your board - there is […]
Show full quote

According to http://oldschooldaw.com/asus/1999/p5a-107.pdf, to clear CMOS, you need to short two points on your board - there is no jumper - and power up. Check page 12 for the layout of your board and the clear location and page 14 for the clear instructions.

I imagine it is quite possible that your caps aren't doing their job. It would be awfully frustrating to replace them all and still have a board that doesn't work. It is up to you, though. What is the brand of caps in there, by the way?

I suppose it is also possible that the board could be:
- Held in a reset state
- Having trouble with power states (sleep/off/on)
- Have improper voltages somewhere (caused by bad caps, for instance)
- Voltage monitoring malfunction
- Temperature monitoring malfunction

I suppose with those last two bits of conjecture, you would get some beeps. I don't know that Award BIOS beeps are very extensive in what they provide.

Holy crap! I left the battery out over night with it unplugged and that did nothing, but I shorted the pads pictured in the link you left (it even says in fine print short to reset cmos but I didn't see it). I also swapped the keyboard start jumper on the following page to disabled, and went over the board again randomly pushing on ICs and the bios.. it magically sprang right back to life. SDRAM that was causing it to no post yesterday is working in it too..

Black magic!?

I'm so glad this board came back on. I hope it stays that way now 🤣

edit: To answer your questions about the caps, the black ones are all rubycon. The two larger green ones at the top of the board don't seem to have a brand label on them. I suppose they weren't the problem though. I find this very odd. No power source and no CR2032 would have done the same thing as shorting the cmos reset pads right?

BeginnerGuy wrote:

it magically sprang right back to life.

That is good news!

I just prepared this image so I will post it anyways. My experence with an asus P2B (not a P5A) board is that when the VRM does not give the Powergood signal, then the CPU fan will not come up and there will be no beep signal. Hence my guess that VRM was OK but BIOS may have been the problem.
Also some boards like to play dead for a while, I don't know what mechanism is at play in that case.

Thanks for helping out gerwin. I'm not sure how to use the image you provided above, am I able to check the output on the VRM with my multimeter? Unfortunately I'm a programmer and my understanding of electronics is minimal.

The issue has come back, unfortunately. I added drives and my mouse back to the system and it went right back to the same issue. Fiddling around with it had it posting or not posting what seems to be at random, most times i could short "reset" and hear a distinct clicking sound from the speaker, and it would post. Once it posts it stays on until I shut it down with no issue booting into an OS, all drives and peripherals connected and working as usual.

Now it seems to be back where I left off last night, no beeps at all. any fans connected to the boards 3 pin headers turn on, including CPU. The only things to note are that the CD-ROM will not light up or open/close IF it is plugged into an IDE port while this is happening, if I remove the IDE cable it powers on fine and opens and closes. On the random times that it would start, the drive works fine. Issue happens with or without the drive.

shorting power on/off for 5 seconds seems to turn it off as expected now too.

So we know the board works, it just doesn't seem to like starting up.. Does this update shine any light on the issue? It seemed bullet proof yesterday when I first started using it. I can't figure out what I've done to the sucker 😒

I'm going to take a break on this for now, but I guess maybe the only thing I can think to do is to desolder the caps and test them out, which would have to wait until next weekend.

BeginnerGuy wrote:

I'm not sure how to use the image you provided above, am I able to check the output on the VRM with my multimeter? Unfortunately I'm a programmer and my understanding of electronics is minimal.

Yes, you can monitor that pin with a DMM. It'll go to 0 volts (plus or minus a little bit) if the output voltage doesn't regulate to the set point as defined by the VID# pins.

Can't find my multimeter anywhere to try. I've ordered a new one and full replacement caps, so I guess I'll update with the final results in about a week 😢 . All of the original caps look pristine but I can't test them to be sure, I probably should have just recapped the board before I ever even tried it. I guess this is growing pains trying to go back to 90s computers after a 20 year vacation.

I tried swapping out to a p5 100 I know for sure is good, and now the CPU doesn't even seem to heat up (Same with the K6-2 when I put it back in). Jumpers were all per the manual, I guess this board was on it's last legs or maybe if I'm lucky a recap will bring it back from it's grave.

I have a P5A that I got maybe a year ago. First time I powered it up it worked.. then after that, nothing.

I've tried it a few times since then with different known good CPUs and still nothing.

Looking forward to seeing if a recap fixes it for you.

Hmm... I have a GA-5AX that seems dead now (GPU fan spins). Wonder what gives.

Here's an interesting twist. In a little last ditch (fit of rage), I decided to press down on the CPU while the machine was on, I noticed the speaker started making the clicking sound again, the CPU instantly became hot. I hit reset, boom it posted and booted like nothing happened. The moment i let go of the CPU, it froze. I remounted the heatsink thinking the added pressure would be enough but no such luck.

So now I'm to believe the issue is something with the socket itself, what would you guys recommend I do to clean the contacts on a socket 7 connector? I wonder if it's dirty contacts or maybe something with the socket itself?

Still waiting on my new multimeter and caps in the mail, but this seems to indicate the socket itself.

As an aside.. I also just bought a slot 1 board (advertised as working from the seller....) with the same symptom of no post, no beeps, fans connected to board headers work. Once again trying RAM, PSU, GPU, HDD from a currently working slot 1 rig. I thought maybe the psu was no good on 5v since this is now two boards with the same problem, so I also pulled the power supply right out of a slot 1 rig running a p3 800, same symptom in both boards persists.

My luck with getting a win98 rig running for myself is becoming a laughable joke. I'm the laughing stock of vogons 🙁

Aside from support capacitors for Vcore, etc., it is possible you have a cracked solder joint underneath the power socket that you temporarily fix by applying pressure to the board - your CPU socket is right next to the power plug. I couldn't really tell much from the photos - they are rather blurry.

I don't think you have to worry about a dirty ZIF socket. I don't think your CPU should move at all once you have that guy in the socket and locked down. Not to mention that the ZIF socket operation slides laterally to make the connections.

You can pull the board and do some visual inspections to see if you notice anything physically, but it would be best if you monitored your power with a multimeter and see how they look.

I don't know what types of electrolytic caps you have in there, but I know one thing for sure - they are old. I'd at least replace your Vcore capacitors.

I suppose you could try pushing your board (do you have something insulated to use?) near the mobo's edge right next to the power socket. That would give a physical push in the same area as your CPU socket and possibly also let it boot. Not nearly as much surface area for a physical push, but a lazy way to potentially see if a different path (than that of needing to clean the CPU socket) is worth pursuing.

Thanks CkR for the reply, I went and opened another thread about cleaning the socket thinking this one vanished 😀. Sorry about the picture quality, my only camera is on a busted up old smart phone that has trouble focusing.

I did inspect the socket a little while ago carefully and noticed nothing unusual on any traces near the socket and I can clearly see the gold contacts running laterally to the top for each pin as you said, they all seem to be clean and shining gold.

I did order new caps for the board along with a replacement multimeter, just waiting on them to come in the mail. I guess I really shouldn't keep screwing with the board until I can properly test and replace the caps. I don't think that pressing around the socket/atx connector would temporarily solve a cap issue though, but I could be wrong. I'll try to get some clear shots of the ATX connector tomorrow, now that you mention it that sounds very plausible.

I have the board pulled and out on a box now, that's where I managed to get it to post.

A good way to test for lose pin socket contact is to use a pin about the same diameter as the cpu pins. Rise the socket lock down handle, insert pin into hole, close handle, then gently pull pin out. You should feel some resistance when pulling on the pin. Repeat this procedure for all the other pin holes.

CkRtech wrote:

I don't know what types of electrolytic caps you have in there, but I know one thing for sure - they are old. I'd at least replace your Vcore capacitors.

The P5A has mostly Rubycon YXG and some Sanyo WG.

I have the same symptoms as the author, and I usually just clear the CMOS and it works ok. I don't think it's the caps though. YXG is an excellent series, and they often are within spec even after 15 years.

Hello BeginnerGuy. I have a P5A that I bought a month ago at RE-PC. Store that recycles electronics and pc's. Well by the time I got around to assembling and starting up, in power management under the Voltage monitor there's an error on -5 volt. It's reading 0.0 (err). Had a 435 watt atx psu then swapped to a 485 watt psu same thing. Even swapped out cpu. I already had a K6 II 500 Mhz. Same thing. Don't know if the board needs to be recapped and possibly replace NEC K3296 transistors and or driver chips. OF course by the time I assembled and found problem it was past the 30 day return so was stuck and I paid $120 for board. They are as much as $230 on E-bay opening bid. If anybody else has had this issue and if someone knows where to get the replacement caps/transistor would be nice to know. Would be nice to know if you ever recapped your board and what the result was. Doesn't look like you've been here in a long while though. I'll check back.

In your case it's unlikely to be a board fault - most ATX PSUs made after 2003 do away with the -5v rail, since it was only needed for the ISA bus, which no PCs had by that point.

Check the PSU label on your two power supplies and see if they mention "-5V", they probably don't have it. The lack of a -5v rail is only really a problem if your sound card needs it, most of them don't.