Power outage causing high temps

[razortonguekiss]

Last night the power went out in my house - I made sure to turn off the power strip, but forgot to unplug it. Now my computer idles at 135 F (58 C) when it used to idle at 95 F (35 C) ... and if I bother running anything at all, even IE, the temperatures jump to spontaneous levels - IE 162 F, 154 F, 148 F, etc... and I can no longer run boinc, as my temperatures start going up to 180 F +. This is quite frustrating/distressing since this is a BRAND NEW rig, and I've had it less than a month so far. The components are are follows:

Intel Q6600 G0 @ 3.00 Ghz (http://www.newegg.com/Product/Product.aspx?Item=N82E16819115017)

Gigabyte GA-EP35-D3SL - FSB @ 1333 Mhz and RAM at 1000 Mhz (http://www.newegg.com/Product/Product.aspx?Item=N82E16813128337)

Rosewill Stallion 600W Power Supply (http://www.newegg.com/Product/Product.aspx?Item=N82E16817182150)

Rosewill Case (http://www.newegg.com/Product/Product.aspx?Item=N82E16811147042)


I've never experienced these temperatures before - my max full load under BOINC was 150 F (which is the max it has ever been period). Anybody have any suggestions? I've checked all the fans - they are all working.

 

[yomamafor1]

You should double check if the fans are intact, and heatsinks if they are as hot as reported.

Theoretically speaking, a simple power outage shouldn't increase temperature like that.

 

[razortonguekiss]

I checked the BIOS settings for my OC - the CPU voltage is set to 1.55 volts, which is WAY over what it should be, even for a 3.0 Ghz overclock in my opinion.... is it common that a power supply would be corrupt in only one part, and not in anything else? My 1.8 V RAM is at 1.92... but everything else seems well within reason.

 

[razortonguekiss]

I've lowered the CPU Voltage to around 1.30 anbd my temperatures have gone back down to what they used to be... could this really be the cause? Seems simple enough... just looking for a second opinion.

 

[nukemaster]

more voltage = more heat

Was speed step and C1E off as well?

 

[grieve]

Run the Q6600 @ stock and see what the temps are.....

 

[yomamafor1]

I've lowered the CPU Voltage to around 1.30 anbd my temperatures have gone back down to what they used to be... could this really be the cause? Seems simple enough... just looking for a second opinion.

Sounds like you found the culprit. Why did it suddenly jumped to 1.55V is beyond me though....

 

[razortonguekiss]

Speed step is on, C1E I'm not sure.... even at stock speeds the temperatures do not change, and neither does the V-Core (I'm using automatic settings that used to work just fine before) ... I'm really beginning to think its the PSU. Thanks for the replies so far btw.

 

[razortonguekiss]

Is it possible it could be the motherboard? Although I would imagine there would be several other problems besides the CPU V-Core alone if that were the case.

 

[nukemaster]

well with 1.5+ your temps where to be expected, you should be ok at 1.3, I run at default(1.225, under load as low as 1.17) without porblems

 

[mike99]

It's rare, But power outage can corrupt BIOS settings. If power cut by damage to supply line there are spikes on the line before it dies. Get UPS or power strip with surge protection.

Mike

 

[razortonguekiss]

Well its at 1.28 V right now - I've never had this problem before. The funny thing is, even if I adjust it in the BIOS - its never the value I set it to - its always way under - the only way I get 1.55 is when I set it to AUTO voltage in which I think it defaults to one of the highest values!

 

[razortonguekiss]

I guess I will try two things:

Flashing the BIOS to see if that fixes anything, and then I will get another PSU - if the voltages are normal under the automatic settings, its gotta be the PSU. If the voltages are bad even with the other PSU, and after a BIOS flash, it has to be the motherboard.

More replies are welcome of course!

 

[nukemaster]

Auto on many boards increases voltage to maintain stability, the problem is it does it way too much. If its running cool and can pass prime95 @ 1.3, leave it there, or try for less.

The reason your bios and computer never show the same is something called Vdroop(and it may also be off due to the sensors in use). its normal and only has an effect if you are pushing very high clocks.

 

[jimmysmitty]

I would suggest you try to set the voltage lower. The stock voltage for a Q6600 is 1.35v but I have a 3GHz OC too and run it on 1.225v stable. I am thinking of lowering it again to try to lower the temps again but 32c idle is not bad as well as 55c 100% Prime95 load.

As I said try to lower your Vcore a bit more to.

 

[razortonguekiss]

I changed the PSU with the same results - the V-Core starts skyrocketing again.

This wouldn't be an issue/concern if it hadn't just happened after this storm - I can guarantee that this has never happened before. Something is different and I'm skeptical to keep something that could potentially harm my system.

I guess that the motherboard might have errors? I tried flashing the BIOS - nothing changed.

 

[razortonguekiss]

I set my V-Core to 1.3000 V and the BIOS reports it at 1.2180 ... this is where my concern lies. The adjustments are VERY innacurrate - and nothing seemed to change with a different power supply. I will try one more time, and if its not the PSU, I think I will RMA the board.

 

[razortonguekiss]

Er well this sounds stupid but I just noticed something... there are 4 tiny LED lights on the motherboard, labelled "Power Phase" ... two are green, one is orange, and one is red... I imagine this means that two of the power phases on the motherboard are bad... in which case I need to RMA the motherboard. Any comments?

 

[mike99]

Don't GUESS!

RTFM!!! Read The F***ing Manual!

Mike

 

[nukemaster]

Odd i send a reply that never went, trying again

Your board is supposed to be able to turn off phases to save power(this saves power), maybe thats what the led's are for. The manual has ZERO mention of the leds and what the colors mean.

Try to load the system and see if it auto kicks in extra phases.