Mysterious non-powering mobos

[PCBeal23]

I just started working as a hardware tech for a small data conversion firm. One of my first projects was to diagnose and attempt to repair six dead computers. out of the six, five PSUs were blown, but when plugging the mobos into a working PSU, the power-to-board LED would light up, but nothing would turn on. I traced the main issue to the VRM, and replaced the capacitors. One brought one back to life that way, which leaves five dead boards. (Intel D845EPI/845GVSR boards, to be exact)

There are no signs of wear (charring, scorching, heat bubbles) on the board or any of its components, which lead me to believe that this would be a Northbridge issue. I removed the heatsink on three of them, and all three chipsets have discoloration around the die. I don't know if this is normal (as a good D845 that was already functioning had the same discoloration on it's NB. Would this inhibit the mobos from powering up?


Any help?

 

[CryHavoc]

Originally posted by: PCBeal23
I just started working as a hardware tech for a small data conversion firm. One of my first projects was to diagnose and attempt to repair six dead computers. out of the six, five PSUs were blown, but when plugging the mobos into a working PSU, the power-to-board LED would light up, but nothing would turn on. I traced the main issue to the VRM, and replaced the capacitors. One brought one back to life that way, which leaves five dead boards. (Intel D845EPI/845GVSR boards, to be exact)

There are no signs of wear (charring, scorching, heat bubbles) on the board or any of its components, which lead me to believe that this would be a Northbridge issue. I removed the heatsink on three of them, and all three chipsets have discoloration around the die. I don't know if this is normal (as a good D845 that was already functioning had the same discoloration on it's NB. Would this inhibit the mobos from powering up?


Any help?

I will save you some trouble and time.

I have alot of experience with the D845 platform. GVSR and GLVA boards.
Sometimes they quit with no external reasons such as a popped chip, meltdown etc. Never took the time to diagnose it down to a component level as there is a point where time spent diagnosing chip repair outweighs the cost of a new board..


I still have several GVSR boards and GLVA boards in perfect working condition.
Let me know if you would be in the market for them and I'd sell them off fairly cheap.

Just a thought.

 

[PCBeal23]

How many boards do you have that are working? I'll let my boss know, but we may not go with the same model line (as the boards I just fixed may quit on us for no apparent reason.).

Thanks!

 

[gammaray]

I know it's stupid but, check if the wire from the power up button is plugged at the right place on the mobo...

 

[PCBeal23]

lol, yes, I'm sure the power switch is plugged in to the right place. In fact, when I test each mobo, I make sure to change the orientation of the cable to check if I made a mistake (which I always take into account, lol). Perfect people make perfect mistakes

 

[PCBeal23]

If anyone else has any more insight, I could really use it