GPU voltage problem

Status
Not open for further replies.

apsoul

In Runtime
Messages
248
Location
Pennsylvania, USA
my friend's computer has been giving him some problems lately. It boots to windows, but then after 2 or 3 minutes, it starts beeping, and the beeping gets faster and faster. The system monitor program "Abit EQ" shows the PCI-E voltage steadily increasing up to 1.80v. Then the computer goes haywire.
I took the fan guard off the GPU HS and cleaned all the dust out, put it back together and now the PCI-E voltage seems to be hovering at about 1.66v.

Stats:
Windows XP professional SP3
AMD Athlon64 x2 4200+ @ 2.21GHz
1x1GB PC2-6400
Nvidia GeForce 7300 GS @ 550Mhz(core) 532MHz(ram)
mobo: Abit KN9(NF-MCP55P) NB/SB-nForce 570SLI

I was just wondering if there was anything else I should do/check before giving it back to him. All help is much appreciated. Thanks
 
It sounds like something is wrong with the video card, but I'm guessing.

When you cleaned the card it should of reduced the temps. But if the card was running hot it should of been reducing the voltage and clock speeds to stabilize itself, not the other way around.

It could also be a problem with the power supply, what make and model is it.

It could also be a problem with the mobo as some power is transfering thru the pci-e slot but it's unlikely.
 
its an Antec SmartPower 400w model SP-400
DC out: +5v@25a +3.3v@30a +12v@14a +12v(2)@15a -12v@0.3a +5vSB@2a

I've left it run for about 2.5-3 hours now, and the PCI-e voltage according to ABIT-EQ is 1.73-1.75

Edit:
I could try running it on my Corsair 650 tomorrow to see if that changes anything... If that works, then any suggestions for a PSU for this system? I know he's not looking to spend much money

EDIT 2:
system.jpg
 
Have you tried running stress test to see if you can replicate the problems? If it was heat, you should be able to get the same result from a program like OCCT. It seems to heat up my GPU more than any other GPU stress test. Carefully watch the voltages on Abit-EQ, and the temps with HWmonitor. Mine quickly rose to 90C with aftermarket cooling, and that's extremely higher than when I'm gaming, as it only reaches about 63C.
 
I replaced the antec PSU with my Corsair tx650, booted, ran a NES emulator (tecmo bowl) for about 15 min, and the pci-e voltage got up to 1.78 and started beeping like mad.
reboot
tried to recreate problem with HWmonitor open to view temps... PCI-e voltage hovers between 1.51 and 1.56
reboot
open up HWmonitor, tried to recreate, nothing, temps all good under normal conditions. Place my hand over the GPU fan while running itunes visualizer and playing mpg movie with VLC, both maximized, gpu temps got up to 62c. I check ABIT-eq, pci-e voltage up to 1.58, then drops down to 1.55.

during all of this i had a few random beeps from the computer. 1 time the DDRII VTT shot up to 2.20 for < 1sec (ABIT-EQ shows danger zone above .99v), another time the atx+12v jumped to > 13.20(i think it was this one, only caught the red text for a blink of an eye, and the danger zone atx+12v was > 13.20. might have been something else though). 2 times i did not catch what set the beep off. At this point my guess would be a problem with voltage regulators on the mobo.

Any thoughts?
 
No, I don't have any PCI-e video cards to try.

Beeping is from the computer itself. I don't have speakers hooked up to this computer. There is no discernible pattern to the beeping other than increasing frequency.

Just ran furmark 1.8.2 and the max temp i can get to is 60c. The voltage goes up to 1.65v (held hand over fan for a min or 2 and got temps up to 75c and voltage up to 1.73. None of this induced any beeping from the computer.

EDIT:
I just came back to the computer after letting it sit idle for about 30-45 min and the computer was beeping. It was alternating between 1 low tone, and 1 higher tone. I checked voltages and temperatures, and nothing seemed out of place. Rebooted and all is quiet.

EDIT 2:
I searched beep codes for this mobo and the 2 tone siren is a over-temp/over-volt alarm. so one of those 2 must have spiked high enough to set off the siren but then dropped back down by the time i got to the computer. Is there anything else I can do to determine whether it is the mobo or gpu that is acting up? My parents computer has a pci-e slot (uses onboard graphics), should I try to put the graphics card in that? If the problem is the GPU would I risk damaging my parent's computers motherboard? Their computer is an HP a6700f
 
So I tried to put the graphics card in the 2nd PCI-E slot to see if maybe it was just a problem with the slot.... Not it. Voltage slowly crept up to 1.80v while running furmark, temps did not go above 62c for the gpu. Then after restarting the computer, It started beeping again, I thought windows was finished loading, I turned the computer off with by holding in the power button and now the system startup file is corrupt. A wonderful time I am having here. He is contemplating whether or not he should just buy a new computer, and I'm getting to the point of saying go for it.

Before I go and reinstall windows to his HDD, should I try to put his GPU into my mothers computer with a PCI-E slot to see if that gets the same problems (thus indicating GPU is the problem) or do I risk damaging my parents motherboard by doing that?
 
Status
Not open for further replies.
Back
Top Bottom