Supermicro PWS-0060 *failure*, please advise

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

Guys, again, I added a LOT of stuff but still the same thing happens. So, finally this is my setup:

1 Raptor, 1 IDE, 2 SCSI, 4 Optical 1 FDD.

As soon as Prescott's VCore is set to 1.525V bam, reboot.

[tiresias]

Re: Supermicro PWS-0060 *failure*, please advise

As soon as Prescott's VCore is set to 1.525V bam, reboot.

Have you checked what the motherboard's doing to the PSON signal though?

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

I didn't quite understand that. Should I check between any GND point on the MB and the PSU GND?

[tiresias]

Re: Supermicro PWS-0060 *failure*, please advise

I didn't quite understand that. Should I check between any GND point on the MB and the PSU GND?

Remember that the signal for ATX or EPS PSU to "turn on", as it were, is the setting of PS_ON (that's the green wire on the 24-pin connector) to ground, so zero potential.

So basically, you measure Vdc between the green wire (PSON) and a ground, anywhere, such a black wire on a molex connector.

A potential of zero means the motherboard has turned the PSU on. A potential of 5V or so means the PSU has is turned off. This is normal operation, it's how the motherboard is able to turn on the PSU when you press the chassis power button.

What I and some others were saying is that perhaps your motherboard is, for some reason, releasing the PSON signal for a while, thus setting the PSU effectively out. This would explain your drop to zero on the 12vdc rail.

If this were the case, you would see PSON develop a potential with respect to ground just before the 12V rail plummets.

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

Mate, I'm checking that out. However, I may have a very simple solution i.e. my overclock just crapped out!

What I would like to know is, if setting VCore = 1.525V @ 250 (works fine) FSB draws the same power as VCore = 1.525 @ 265 FSB (craps out)? What I may be witnessing is that the CPU has just been maxed out and can do no more on air. I did turn off thermal throttling (EIST), but maybe the BIOS is buggy and this causes the system to shut down anyway?

Any other point I can tap into the PS_ON signal? The ATX 24 pin is really jammed up pretty tight and the Green wire almost completely occupies the hole. I can't sqeeze the DMM probes into it.

[tiresias]

Re: Supermicro PWS-0060 *failure*, please advise

What I would like to know is, if setting VCore = 1.525V @ 250 (works fine) FSB draws the same power as VCore = 1.525 @ 265 FSB

Not the CPU, although your MCH probably will - shouldn't be much though, but I'll leave that up to the Intel experts to confirm.

the BIOS is buggy and this causes the system to shut down anyway?

That is possible. The only (sane) way for the BIOS to shut down the PSU is, of course, indeed via that PSON signal.

Any other point I can tap into the PS_ON signal? The ATX 24 pin is really jammed up pretty tight and the Green wire almost completely occupies the hole. I can't sqeeze the DMM probes into it.

Good point - if it's anything like mine, even the green wire is 18AWG. All the other lines, molex included are 16AWG - it's these things that make these units so solidly built.

There isn't an easy place in most motherboards where you can catch the PSON as far as I know. What you could do is to insert a very thin wire alongside the green one, and attach your DMM to that. You'd have to first check with PSU in the "soft" off state for the 5V signal to be sure that you have a contact, of course....

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

The VMCH needs to be bumped up only after 325 MHz+. I'm getting blasted at 265 MHz. I'm still looking for a thin wire I could use.

I was speaking with a my friend who is familiar with Intel stuff (I've only used AMD thus far) and he said that my CPU has completely topped out on air. That would explain why everything is normal even when I bumped up the VCore ~ 1.58 V (bad for the chip, but it didn't die yet) at the lower FSB (255MHz) without topping it out.

So all this may be because I'm just an Intel N00b.

[Galvanized]

Re: Supermicro PWS-0060 *failure*, please advise

I looked at Google images of your board. There is a 4 pin ATX connection between the I/O plate and the CPU socket. There is a 4 pin Molex right next to the 24 pin ATX connection. Are both of these hooked up? Did you read the motherboard manual for power hook-up?

I am not being silly. Some DFI boards have a floopy power connect and guys miss this hook-up.

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

I didn't read the manual, but the first thing I looked for was an extra power connector (like in the DFI, there are 4 connectors). I'll go back and check on this. Thanks dude.

I can only see 2 power connections. The 24 pin and the 4 pin AUX. Where is the other one? It will not power up without the AUX, so that is in the clear.

[Galvanized]

Re: Supermicro PWS-0060 *failure*, please advise

Well...What's up Nade.? Did you get the ballasts I sent.?

Has this PSU issue been sorted out.? :-D

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

Sorry Galvo, I should have sent you a pm.

Got the resisitors. I didn't need them. The Presshot was crapping out. I saw the same stuff happen with the OCZ Powerstream (a quick test I did). As long as I don't exceed the CPU's potential (stupid thing throttles even when I turn of EIST in the BIOS), it is handling everything I throw at it. So far I have 5 HDD's, 2 Opticals, 1 fully clocked up x1800xt 512Mb, 2 120mm fans and few other crappy 80 mm fans.

I'm in the process of moving everything back to my lab and figuring out probe points on the board to see what component is doing what. Everthing is super thus far.

What I'm pissed about is the damn pertol prices. > $3.15 a gallon. WTF?

[gonzo0815]

Re: Supermicro PWS-0060 *failure*, please advise

What I'm pissed about is the damn pertol prices. > $3.15 a gallon. WTF?

He he he , i wish i could buy some petrol for that low price.
One litere is about 1.40€ wich will give about 5€ per gallon.

[tiresias]

Re: Supermicro PWS-0060 *failure*, please advise

He he he , i wish i could buy some petrol for that low price.
One litere is about 1.40€ wich will give about 5€ per gallon.

Tell me about it! Still around $6.00/US gallon here. Over half of that is taxes.

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

Tell me about it! Still around $6.00/US gallon here. Over half of that is taxes.

Wow! That is expensive dude! But atleast you guys have good public transportation. Public transportation does not exsist in Indianapolis. Bus service is just crap.

[arneson]

Re: Supermicro PWS-0060 *failure*, please advise

I need a new motorcycle, bad!
I gave up on the old one, it's been parked in the bushes, for years.

[gonzo0815]

Re: Supermicro PWS-0060 *failure*, please advise

Yea good public transportation is for shure better, but if you are livin not in the top most biggest town, they only are driving untill 00:30PM. There is not mutch in going disco or drink some beer late in the night without a car.

[Super Nade]

Re: Supermicro PWS-0060 *failure*, please advise

O.K Guys, finally I can confidently put the 12V going to 0 V issue to rest. This particular board does this regardless of the PSU being used.

EDIT#
---------
Listed as a MB issue here:-
http://xtremesystems.org/forums/showthread.php?t=110193