majestyk wrote on 2023-06-27, 05:38:You should use a PLCC extractor tool for that. These chips are sometimes stuck in their sockets and you risk breaking the socket […]
Show full quote
You should use a PLCC extractor tool for that. These chips are sometimes stuck in their sockets and you risk breaking the socket if you are unexperienced.
I would not suspect the Rubycon caps (and the Nichicons) at the moment.
But you can try with some low end socket 370 CPU like a ~500-800 MHz Celeron to see if the lower power demand makes a difference.
And you should clean the dusty RAM slots carefully with IPA (if you haven´t done this already). I´d recommend using a small paintbrush.
Have you tried with a single RAM stick in all 3 sockets?
Thanks for the suggestions, I actually tried two different Celerons and Pentiums. Cleaned AGP, RAM slots and tried different RAM sticks individually on each socket. Also tried different GPUs and PSUs, CMOS batteries, all to no avail, board lights on, but won't power on. Also shorted the two solder pins to reset the CLR clock, and tried to start the board with the case power on button, and also by simply shorting the pins. Re-checked the caps solder joints that I reflowed, all looks fine.
aaronkatrini wrote on 2023-06-28, 14:46:
Yes, that is the one, be very careful. It is very easy to break that socket if you do it to fast or apply too much force.
Thank you as well, I'll keep that in mind.
I still can't help but feel that the cause is somewhat related to something being cracked somewhere along the layers of the board or chipsets as it was suggested. From the board arriving three years ago with a compression hit on one of its outer tips during shipping, me unknowingly installing it in a PC case with three different varieties and heights of standoffs, the PC starting to not POST every single time it was moved or shuffled around which happened more and more, the PC hanging from me just hitting it slightly with my hand, going from that to not POSTing at all after I removed and disassembled it to install the new PSU, and finally to not even power on after multiple tests, tries and reassemblies... It just feels that every time the PC was moved or touched it started causing more and more problems, and the more I manipulated, removed and reinstalled the board in the troubleshooting process its supposed crack somewhere may have just finally gave way. I just feel it was too consistent of a symptom to be a coincidence.
As it has become obvious I don't have any formation in IT or computers, I don't have a large or even medium sized hardware collection, what I have is all at use barring some extras that I ended up getting when I got the cases to assemble these builds. I use those extras to help diagnose eventual problems like these, and over time I got a cheap solder kit for some small fixes, which proved very useful. But in this case I really think it's above my skills (and tools). It's becoming apparent that a multimeter is the next thing to get, but to go from that to diagnostic cards and beyond it it's just not worth it for someone with as little hardware and knowledge as I have, might as well just channel those funds directly to a replacement part. Even if it indeed was a cracked point somewhere on the board I wouldn't have the skills to find it.
I'll keep all the suggestions and when I can I'll try and see if someone more prepared than I am can help me with this. And who knows, if my suspicions are wrong and it ends up being just the caps then great, I'll even recap it myself when I have the time in the future
