Maybe check the cpu core voltages? I have the same problem, but with an INTEL D865PERL mobo.

Measured voltages at the CPU socket - they are as follows:

VCC2 -- Takes about 30 seconds to come up to 1.48v.
VCC3 -- Takes about 5 seconds to come up to 1.19v.
VSS -- No voltage detected.

Seems like the time to get to a stable voltage is a problem. PSU voltages are stable from the connector as well.

With voltages taking this long to stabilize, I should probably replace electrolytics. They check out visually but things might not be as they seem. Am I off base here?

Additionally, this isn't the only problem I have with boards held in reset. I have another immediately obvious that has this problem though it's a 486. Perhaps they're all just in need of new electrolytics.

clintonoddfellow wrote on 2024-03-25, 16:29:

.....

So far, according to my POST card, the board doesn't make it to POST. Looks like it's being held in reset for some reason.
......

There is a Led on POST cards, which shows the status of the reset line.
Held in Reset you would easily see: The Reset LED is on all the time.
Normal behaviour is reset is on, and then is turned off quickly (less than half a second).

clintonoddfellow wrote on 2024-03-25, 17:06:

Measured voltages at the CPU socket - they are as follows: […]
Show full quote

Measured voltages at the CPU socket - they are as follows:

VCC2 -- Takes about 30 seconds to come up to 1.48v.
VCC3 -- Takes about 5 seconds to come up to 1.19v.
VSS -- No voltage detected.

Seems like the time to get to a stable voltage is a problem. PSU voltages are stable from the connector as well.

With voltages taking this long to stabilize, I should probably replace electrolytics. They check out visually but things might not be as they seem. Am I off base here?

Additionally, this isn't the only problem I have with boards held in reset. I have another immediately obvious that has this problem though it's a 486. Perhaps they're all just in need of new electrolytics.

https://theretroweb.com/motherboards/s/asus-p5a-b

Vcore should be 2.2V
Vi/o 3.3V

I'd check the mosfets.
Post a pic of the area to identify what's in there.

On TRW pics:
https://pdf1.alldatasheet.com/datasheet-pdf/v … /PBYL1025B.html

one is a diode, basically you can try to replace it with another diode? mine is a guess.

https://pdf1.alldatasheet.com/datasheet-pdf/v … IL/HIP6008.html
This creates the voltages.

CoffeeOne wrote on 2024-03-25, 18:37:

There is a Led on POST cards, which shows the status of the reset line. Held in Reset you would easily see: The Reset LED is on […]
Show full quote

clintonoddfellow wrote on 2024-03-25, 16:29:

.....

So far, according to my POST card, the board doesn't make it to POST. Looks like it's being held in reset for some reason.
......

There is a Led on POST cards, which shows the status of the reset line.
Held in Reset you would easily see: The Reset LED is on all the time.
Normal behaviour is reset is on, and then is turned off quickly (less than half a second).

Yep, that's what I'm seeing. Documentation on this thing is sketchy, but I assumed the reset light staying lit meant it was being held in reset. Trying to troubleshoot that, and came up with weird voltages at the CPU.

I'd check the mosfets.

I'm a bit sketchy with testing MOSFETS in circuit. I'm getting some weird results when I run resistance tests to ground, but that could just be the in circuit part of things, or expected behavior and me being a goof.. At this point if I were going to remove them to test out of circuit, I'd just replace them. Pardon my ignorance, but is there a good guide to checking these things in circuit?

Thanks!

I think replacing the capacitors in the VRM circuit is a good place to start. They can be dead, even if they are not bulged. I had few boards that did not POST initially and ran perfectly fine after capacitor replacement.

havli wrote on 2024-03-25, 19:53:

I think replacing the capacitors in the VRM circuit is a good place to start. They can be dead, even if they are not bulged. I had few boards that did not POST initially and ran perfectly fine after capacitor replacement.

Given the fact that the voltages take a good long while to stabilize, I figured this was a good place to start. Electrolytics have a shelf life and it's probably good practice to replace them.

Interestingly on closer inspection, a couple of the capacitors appear to have been replaced in the past. All of the electrolytics on the board are identical values, but two of them are rubycons. Makes me think that if those two died in the past, the rest are going as well, giving me my weird voltages.

I've ordered replacements, and we'll see if that knocks it out.

-C

Got it figured out!

Nichicon caps came in, and I replaced them. Voltage improved but were still weird, telling me the caps were in need of replacement anyway, but still no boot

Was investigating the MOSFETs and was getting ready to desolder legs to test, and then the MOSFET just popped right off the board without so much as a by your leave.

Both of the MOSFETs on the board were in the same condition. Reflowed, and back in business. Board now boots.

Problem solved!

Moral of the story, check or just try reflowing your MOSFETs if you've got wonky issues. It's very little effort and might just work!

-C