Is my PSU Degrading?

[ajgreen]

New build in Nov 2012: ASUS P8Z77-V LX, I7-3770K, 8GB Corsair RAM, Samsung DVD-RW, 212 EVO CPU Cooler, Zalman Z11 ATX Case.

I installed a Zalman ZM500-GS PSU. I've been able to overclock the CPU to 4.7, run all kinds of stress tests, etc and the rig has stood up well.

In April I also installed XFX HD 7970 Core edition and until recently the system has behaved itself.

In the last month or so I've been getting system crashes where the mobo detects power supply surges and anti-surge kicks in. I can replicate the crash with 3DMark11 Basic. GPU-Z shows GPU temps at 75C and high VDDC currents.

Reading forum posts suggests either mobo, GPU or PSU at fault here. I was always a little unsure that the ZM500-GS was enough for my rig. Has it degraded and is not now good enough?

Any thoughts?

TIA

 

[mactronix]

Yes its quite possible that your PSU was just enough but now is struggling.
Many things can effect the demands on a PSU even a simple driver change ca make a difference, they do naturally degrade over time anyway.
I would recommend a good quality Corsair or other equivalent makers PSU of 650 -750 Watts.

Mactronix

 

[JJ1217]

Your particular PSU has had a reputation for having bad 12V rail voltages. If it dips too low, it will cause a system reboot. It could definitely be the issue. I would never trust a Zalman PSU in my rig. The other models of the ZM-500 series are atrocious, I don't know whats exactly different about this one. 500W in general is low for an OCed 3770k and a 7970, let alone a low end 500W..

Its up to you what to do here. Even if it doesn't fix your issue, getting a better power supply will be something you'll have to do eventually anyways. Degradation of the power supply is also a possibility, I doubt it would output much more than 400W on the 12V rail, and if you were running an OCed system with a 7970, you could definitely be above 90% of the output of the PSU, which can lead to degradation.

 

[ajgreen]

Thanks to you both for your replies. I'm not OCing anymore. The crashes are all happening at stock speed. GPU 12V remains at around 12 until just before the crash when it typically dips to around 11.6. Is this low enough to cause system crash?

 

[mactronix]

Thanks to you both for your replies. I'm not OCing anymore. The crashes are all happening at stock speed. GPU 12V remains at around 12 until just before the crash when it typically dips to around 11.6. Is this low enough to cause system crash?

Hi,
The official atx power supply voltage rail tolerance fro a +12V rail is +/- 5% or about 0.6 so it shouldn't be enough really.

Mactronix

 

[ajgreen]

Update: I set a power meter on the PC input and ran 3DMark11 again to replicate the crash and power readings never got above 270W. I would have thought that the ZM500 should be able to handle this easily, even if it has degraded.
Maybe the PSU is OK and there is a fault with the GPU or mobo. How do I find out without swapping in similar components which I dont have.

I have now disabled anti-surge on the mobo and it is still crashing.

 

[ajgreen]

OK so I've installed a new Corsair GS700 and the rig powers up fine but as soon as I put any load on the GPU, the screen goes blank. The PC doesn't seem to crash - everything else is powered up - but the only way to get it back is to reboot. Sometimes I have to reboot a number of times before I get anything on screen.

Does this suggest the graphics card is bust?

 

[smeezekitty]

High end XFX GPUs (7870/7950/7970) are known for VRM overheating.
Does GPU-Z give you VRM temperature?

 

[ajgreen]

High end XFX GPUs (7870/7950/7970) are known for VRM overheating.
Does GPU-Z give you VRM temperature?

Yes it was getting up to 75C before the crashes