Computer Shuts Off

[galbicake]

This has happened to me a few times when I play computer games. The screen would turn to glitchy colors and then black out and the computer would shut off. I would actually have to turn off the PSU manually and then restart the computer. I smell a slight burn smell.

I have a Radeon X800. I checked the my internal fans and they are allow running at fast speeds, the northbridge, zalman CPU fan & the other one...I have an Antec Sonata case that has 3 fans blowing.

Does anyone have any experience or thoughts as to what my problem may be and how to fix it?

 

[dfuze]

If you are smelling something burning, that is definitely a bad sign. Even though the fans are running, I wonder if any heat sinks are not correctly seated? Have you recently changed any fans/heatsinks on your video card or cpu?
After giving it plenty of time to cool down, I would start it up and monitor your system temps (I use Everest Home Edition) and see if that tells you which component is overheating.

 

[galbicake]

I built this computer about a year ago. One problem I have had since the beginning was that when I shut down the computer, it powers off and then it would power up again anywhere between 30 seconds to 3 minutes after a proper shut down. However, that went away for the most part for some inexplicable reason. I have noticed it acting up like this again this past week.

I have not altered with the Zalman Heatsink fan since installation. However, I did recently install a Hauppauge PVR 150 and a Audigy ZS 2.

My PCI-E card runs really hot and the fan blows right on top of one of those PCI cards.

Could it be my PSU? I am using the stock Antec PSU 350W that came with the Sonata case.

 

[acole1]

are you overclocking?

 

[galbicake]

Negative.

I am just running it at default settings. I am not smart enough to OC.

Is there a place I can go where it tells you what proper temperature settings should be for CPU & GPU?

 

[CraKaJaX]

Originally posted by: galbicake
Negative.

I am just running it at default settings. I am not smart enough to OC.

Is there a place I can go where it tells you what proper temperature settings should be for CPU & GPU?

Download Everest, good program for temps.

 

[galbicake]

Thanks

But if the temps are running too high on either the CPU or GPU, how can I fix this problem?

I guess I can take another look at the Zalman that is mounted on top of my CPU.

However, what should I do for the GPU?

 

[galbicake]

Anything out of the ordinary on this reading?

Field Value
Sensor Properties
Sensor Type Winbond W83627THF (ISA 290h)
GPU Sensor Type National LM63 (ATI-I2C 4Ch)
Motherboard Name MSI MS-7046 / 7100 / 7125

Temperatures
Motherboard 34 °C (93 °F)
CPU 35 °C (95 °F)
Aux 20 °C (68 °F)
GPU 49 °C (120 °F)
GPU Ambient 39 °C (102 °F)
Seagate ST3160023A 34 °C (93 °F)
Seagate ST3160023A 36 °C (97 °F)

Cooling Fans
CPU 2220 RPM
System 1663 RPM
North Bridge 7584 RPM

Voltage Values
CPU Core 1.39 V
+3.3 V 3.26 V
+5 V 4.89 V
+12 V 11.67 V
+5 V Standby 5.09 V
VBAT Battery 2.94 V
Debug Info F CB 98 59
Debug Info T 34 35 20
Debug Info V 90 CC C0 B6 1D 12 02 (01)

 

[Bobthelost]

Was that when you were at idle, as the GPU temps are rather high.

Try running RTHDRBL for five mins or running 3d mark and then doing it again.

 

[galbicake]

This was when I was running idle.

What is the temp I should expect when idle vs active?

 

[AMCRambler]

Nothing in your temps looks out of the normal ranges. If you're smelling a burning smell though I'd say 99% chance it's your power supply. The voltage values look ok, but I had the same thing happen with mine. I had this funny electrical burning smell for a a day or 2 before my psu died. You should be able to to determine if it is the power supply, just sniff by the exhaust fan opening in the back of the power supply. On mine the smell was noticeably strongest right there. I doubt it would be the cpu or ram making the smell, since if either got hot enough to burn they would be permanently damaged. If the psu isn't the source pull the cover and look for burn marks on the board, the cpu, the ram, anywhere you've got bare contacts.

 

[galbicake]

Thanks for the insight. I have noticed that the PSU has been pushing out really hot air but I thought that is more common since it is the PSU.

Just strange because I have only had this for less than 1 year and Antec are supposed to be the better than the typical brand. I have not OCed my rig.

I have to wait until my PSU totally craps out before I can do an RMA with Antec. However, their warranty kind of disturbs me Antec Warranty

It does not cover normal wear and tear. This makes me wonder what do they cover?

I just purchased an Ultra PSU, after all the slamming people gave on this forum...I am not sure if that is a long term solution.

Anyone else can shed some thoughts on this?

 

[Bobthelost]

I disagree, those temperatures are unusualy high

GPU 49 °C (120 °F) @ idle is not something i'd be happy with. A quick google found this http://www.overclockersclub.com/reviews/ati_radeon_x800pro_11.php where the X800 runs 20*C cooler at idle, run 3d mark and see if the temperatures don't rise a lot.

 

[AMCRambler]

Originally posted by: Bobthelost
I disagree, those temperatures are unusualy high

GPU 49 °C (120 °F) @ idle is not something i'd be happy with. A quick google found this http://www.overclockersclub.com/reviews/ati_radeon_x800pro_11.php where the X800 runs 20*C cooler at idle, run 3d mark and see if the temperatures don't rise a lot.

Well then my 6600GT must be effed up because at idle it runs 59C. I just recently overclocked it and checked the temps again and it's still at around 59-60C at idle.

 

[Bobthelost]

Originally posted by: AMCRambler

Originally posted by: Bobthelost
I disagree, those temperatures are unusualy high

GPU 49 °C (120 °F) @ idle is not something i'd be happy with. A quick google found this http://www.overclockersclub.com/reviews/ati_radeon_x800pro_11.php where the X800 runs 20*C cooler at idle, run 3d mark and see if the temperatures don't rise a lot.

Well then my 6600GT must be effed up because at idle it runs 59C. I just recently overclocked it and checked the temps again and it's still at around 59-60C at idle.

Yikes! I'd be rather concerned by that too.

 

[Howard]

Can you swap your PSU with something else? If you don't have one lying around, ask a friend.

 

[galbicake]

I have the GPU on the PCI-E slot. However, I have a PCI card directly under it in a PCI slot. It appears that the GPU fan blows directly on top of this PCI card and blows it back up.

Any experience on this?

I recently purchased an Ultra 500W PSU. But after reading all the bashing people posted on this site...I have serious doubts about it's reliability. I have not yet opened it. I was debating whether or not to return it at a $5 loss in shipping or selling it at some sort of profit after rebates.

I tried to do some google research on GPU temps and I cannot find much else to use as benchmarks...

Any other people here have experience on this?

 

[AMCRambler]

Originally posted by: Bobthelost

Originally posted by: AMCRambler

Originally posted by: Bobthelost
I disagree, those temperatures are unusualy high

GPU 49 °C (120 °F) @ idle is not something i'd be happy with. A quick google found this http://www.overclockersclub.com/reviews/ati_radeon_x800pro_11.php where the X800 runs 20*C cooler at idle, run 3d mark and see if the temperatures don't rise a lot.

Well then my 6600GT must be effed up because at idle it runs 59C. I just recently overclocked it and checked the temps again and it's still at around 59-60C at idle.

Yikes! I'd be rather concerned by that too.

Looking at the temps in that review I'd say his are pretty darn good. There's a better example of current GPU temperatures here. The temps he's seeing should not be causing any problems with his video card functioning at the very least.

 

[Luckyboy1]

Antec PSU 350W

Way too small Dude!

The 49 C GPU temp on a 800 card is fine as long as that's the temp you get under full load and NOT idle.

I've been posting crap here lately, so it's off to bed with me. Someone with the sense God gave chickens who's had more than 4 hours sleep in the last three days please expand on my answer.

 

[Bobthelost]

I dunno, what he's describing sounds like text book artifacting and reboots due to overheating. Run 3d mark or RTHDRBL and test the temps again. If it's got poor airflow then it could get much worse when at load.

PS. your link no workie.

The PSU should be enough as long as it's got 16A + on the 12V rail, that's not a power hungry system by any definition.

 

[Sean Maxwell]

Did you substitute mayonnaise for thermal compound?

 

[AMCRambler]

Originally posted by: Bobthelost
I dunno, what he's describing sounds like text book artifacting and reboots due to overheating. Run 3d mark or RTHDRBL and test the temps again. If it's got poor airflow then it could get much worse when at load.

PS. your link no workie.

The PSU should be enough as long as it's got 16A + on the 12V rail, that's not a power hungry system by any definition.

Friggin A. I must have clicked the ad header by accident. Fixed it. I didn't see the X800 in his sig there. I can't speak for ATI's cards I just know how my Nvidia runs. Stock out of the box 59C and it hasn't given me any problems so I haven't really worried about the temperatures.
I'm still thinking power could be the prob. If the card isn't getting enough juice at any point you could also get artifacts. But there is a case for both possibilities here. Conclusion: check em both. Swap a spare card in and test, then a spare power supply. The burning smell should be pretty easy to tell if it's coming from the psu though. Sniff away!

 

[zagood]

Anyone notice that his +12v reading is 11.67v? Yes, that's within the +/-5% standard, but with only 350W I think he's starving the GPU. If you've got a multimeter use that to check the readings for accuracy.

As far as that X-connect, there have been some pretty bad stories about them. To be on the safe side, sell it and pick up another. Search the forums, there's a lot of threads with PSU recommendations, also read the PSU sticky thread.

-z

 

[galbicake]

I do not have a multimeter, is that program or hardware tool? What exactly should I be expecting with it?

I just did the RTHDRBL test for about 5 min.

Temps

Idle = 49-50c
Active = over 70c

However, it did not crash when I did the test.

It did crash 5 minutes ago when I was playing off my DVD Drive Star Wars Episode III and downloading off Napster. WTH?!?!

Do you all think my rail V is too high? I do not want to purchase a need PSU unless I am certain that it is the reason. I will be out of town this weekend but please do keep adding to my post.

Thanks guys!

 

[galbicake]

Bump