Can a case short/fry a CPU? Nope. It is the motherboard.

[txlonghorn]

This sounds preposterous, but I lost 2 CPUs and I am not going to take a chance and risk frying a 3rd CPU. Here is what happened.

1. Built a WHS with spare parts on the workbench. Everything is working fine and it ran for a week on the workbench.
2. Yesterday, I put all the parts in a Thermaltake Slement S case. Fired it up. Nothing. No POST.
3. Took everything out of the case. Tried everything I can think of. Still no POST.
4. Today, I swapped in a new CPU while the setup is still on the bench. It worked just fine.

Conclusion: must be fried CPU. I did not use anti-static strap while putting the parts in the case. My bad.

5. Let's try again to put all the parts in the Thermaltake case. This time with anti-static strap and extra care.

6. After all is done, fired it up. No POST. Blank screen, just like yesterday.

Conclusion: Fried CPU again.

I don't have another CPU to verify this, but the symptom is exactly the same as before. I am totally stumped. I have no idea what the problem is. Has anyone heard of this problem?

 

[ahenkel]

I'd take things back out and retest on your work bench, just test with minimal components.

 

[txlonghorn]

I'd take things back out and retest on your work bench, just test with minimal components.

Done that. No POST. Trying get get a third CPU to bring the system back to life on the bench. However, I am afraid to put the parts back in the case until I know what is going on. I can't afford to continually burn CPUs trying to figure this out.

 

[SpeedTester]

If the standoffs are all in the right spots and there are no flaws on the motherboard tray
I can't see how the case could be killing your CPU. What processor keeps dieing on you?

 

[Fallen Kell]

I know it might be fugly, but you can try coating the motherboard tray area with paper tape. It might not be the case either, but the motherboard. There might be a short over one of the standoffs. Most motherboards have a reinforcing metal right which might also be grounded. If they screwed up on one of them and then ground isn't ground, it could cause a short over through a metal standoff to the case, since many cases are grounded themselves. The only way to test this is to use plastic standoffs, but I am not sure if your case will support that.

 

[txlonghorn]

The 1st CPU is Athlon 1640. The 2nd is Sempron 140. The mobo is ECS A740GM-M. I checked the standoffs, they are standard height and tall enough so that no part of the back of the mobo comes close to touching the mobo tray. The mobo is slightly curved, but that is not unusual.

I am going to drop an old Athlon XP board in the case and see what happens. Like I said, this sounds preposterous, but I have 2 non-working CPUs and 0 explanation.

 

[txlonghorn]

I know it might be fugly, but you can try coating the motherboard tray area with paper tape. It might not be the case either, but the motherboard. There might be a short over one of the standoffs. Most motherboards have a reinforcing metal right which might also be grounded. If they screwed up on one of them and then ground isn't ground, it could cause a short over through a metal standoff to the case, since many cases are grounded themselves. The only way to test this is to use plastic standoffs, but I am not sure if your case will support that.

Interesting theory. I hope it is correct because I really like the case. I don't have any plastic standoffs. Even if I did, the case does not support it. Too bad I don't have any more AM2 CPU left to try it on another mobo. I am going to use an Athlon XP board to give it a try.

 

[SpeedTester]

The 1st CPU is Athlon 1640. The 2nd is Sempron 140. The mobo is ECS A740GM-M. I checked the standoffs, they are standard height and tall enough so that no part of the back of the mobo comes close to touching the mobo tray. The mobo is slightly curved, but that is not unusual.

I am going to drop an old Athlon XP board in the case and see what happens. Like I said, this sounds preposterous, but I have 2 non-working CPUs and 0 explanation.

Have you installed one of the first two cpu's that appreared to be dead? You have it running with the third cpu on the bench now right? It's hard to think why a case could be killing your processors if anything is mounted correctly with no shorts.

You can also try mounting the small nylon washers that came with cases a while ago under the motherboard but on top of the standoffs if you can verify its a grounding short.

Keep us updated.

 

[aigomorla]

so the board killed 2 cpu's, and u want to try a third?
Ever think of a bad board after losing 2 cpus?

 

[txlonghorn]

Ever think of a bad board after losing 2 cpus?

Didn't occur to me the board may be bad until Fallen Kell mentioned it. It has worked, after all, flawlessly on the bench for many months. Now that I know it may be the board that is killing my CPU, I am not going to try the 3rd one. Just going to wait until the next Fry's combo deal and start fresh.

 

[aigomorla]

Didn't occur to me the board may be bad until Fallen Kell mentioned it. It has worked, after all, flawlessly on the bench for many months. Now that I know it may be the board that is killing my CPU, I am not going to try the 3rd one. Just going to wait until the next Fry's combo deal and start fresh.

the only things i can think of is you touched a mosfet pin in the rear, and shorted something.

However most of time that is just a bad ground.

The board and cpu should of booted up right away after u fixed it.

Not kill the cpu twice.

 

[SpeedTester]

I believe that board has a 3 Year Parts / 2 Year Labor Warranty on it. I only bought a handfull of ECS boards but never had to deal with any RMA's with the company.

 

[txlonghorn]

OK. I just dropped an old socket A board in the case and it worked flawlessly. Conclusion: the case is fine. It is got to be motherboard. That's too bad. I really like the feature set on that board for a server build. Now I got to find a replacement combo..... (don't trust ECS anymore)

 

[Meghan54]

I'm betting the board has bad traces that only show when the board is "stressed" by screwing it down to the case's mb tray mounts, probably around the cpu socket and affecting power delivery circuitry.

 

[txlonghorn]

I'm betting the board has bad traces that only show when the board is "stressed" by screwing it down to the case's mb tray mounts, probably around the cpu socket and affecting power delivery circuitry.

Yup. That's gotta be yet. The board is warped when resting on flat surface, and feels very cheaply made as compared to my GigaByte board. I am shopping for replacement CPU/mobo combo right now. Thanks for all the inputs that steered me in the right direction.