reboot with enermax noisetaker 495

[tomati]

Hi,

I have just received the enermax noisetaker 485w and put it in my new pc , I OC the machine at max and connect the 9800pro oc at xt and run some 3d bench , sandra bench etc , for stressing the psu , what was my surprise to see that the computer rebooted imediately when I was running a memory bandwith test in sandra ,( I have succeed a 48h prime95 with my zalman 400w at this extrem OC) so I have decided to reconnect the zalman , and everything is allright , just my 12v rail on my zalman still at 11.46v/11.52v (when the enermax is at 11.83v/11.77v at the molex), so better voltage but total instability , may be not a good luck with this psu, and all I can do is to return it for a refund.

Now I wonder wich psu will be good for my set up (as I m affraid with the 11.52v on the 12v rail) the zalman have 18a on the 12v and the enermax have 2 rails of 18a on 12v , Is my setup incompatible with the new atx 2.01 ??? I want to try the s12 500w as it have a good press in here , so I think I will order one and try if...

[Devonavar]

Isn't it more likely that the problem is with your memory? Memory isn't exactly power intensive (not in comparison to the CPU / GPU). Try running Memtest86 and see what you get.

I realize that swapping the power supply seemed to help, but I'd say it's more likely that the change in power supply produced a change in your case's airflow, which may have caused the memory to overheat.

[tomati]

Isn't it more likely that the problem is with your memory? Memory isn't exactly power intensive (not in comparison to the CPU / GPU). Try running Memtest86 and see what you get.

I realize that swapping the power supply seemed to help, but I'd say it's more likely that the change in power supply produced a change in your case's airflow, which may have caused the memory to overheat.

My memory seems to be totaly stable with the zalman (same settings) , I have test it with memtest and all bench to found the good setting ,I have just to decrease the fsb for 5 mhz from memtest result for a complete stability under windows and increase the cpu voltage, but as you talk about the memory voltage I have notice that the 3.3v rail was weak on the enermax (didn't measure it with a multi but was report at 3.2 and below under load (3.15v)) I set my mem voltage at 2.8 in bios what seems to be 2.85/2.9 in reality don't know if that's the key .

As I'm still testing my pc for the stability ,my unit is out the case on the desk with one fan blowing over the memory , today was really hot (I got 33°c in my room) but after putting the zalman all reboot trouble have been desapeared (I have running 10 instances of mem bandwith test, where I was only able to run 1 or 2 with the enermax before the reboot) , I will check the 3.3v rail ,anyway thanks for your help.

edit: Devonavar you found the culprit , without any load (just make the strat trick wire between green and balck on the atx connector ) I got at the multi 3.08v (3.15v on the zalman)so I suppose when it come under full load my memory don't received enought juce that's explain why the pc reboot.