Help! Newly assembled e6400 system shuts off randomly.

[Whimper]

I just put together a new system (my first ever home build).

Intel Core 2 Duo E6400
2GB DDR2-667 RAM
GeForce 7950gt gpu
Antec Solo case
Asus P5NSLI mobo
Seagate Barracuda 320GB HD

Everything went well during all the steps right up to Windows XP installation. I did all my Windows Update stuff, installed new drivers for everything, etc etc.

But now, several times today, the system has just shut off and gone dead right in the middle of doing something. It stays powered off for a few seconds, not responding to the power button, but then it will start up again.

The first time it did this was in the midst of copying a large batch of files from an external hard drive. The second time was after installing Quicktime. The third time was during a graphics demo.

I have a feeling that this is some kind of overheating problem, because each time it's happened during some kind of CPU strain. But nothing inside my computer is overclocked, and the BIOS says the CPU's temp is around 70degrees, GPU around 50. What could be going on here?

 

[NotAPimecone]

Overheating is certainly a possible cause, but I'm not sure you'd get as far as installing Windows if that were the case. Then again, 70 is really quite hot. You are using the stock HSF? make sure all the clips/pins are clipped in right.

It could be some kind of instability, have you tried testing with prime or orthos while monitoring temps in CoreTemp or TAT? I'd try to get the temps lower before trying this, though...

Also, what kind of power supply do you have?

 

[chogo]

It seems your CPU is overheating as 70C is very hot and that is after it has started to cool down.

You should ensure that the heat sink fan (HSF) is properly attached to the CPU. Are you using the stock HSF that came with the CPU or an aftermarket one?

Turn the machine off and unplug it. If you are using the stock HSF, undo each leg and press down on each leg firmly while reattaching each leg to the motherboard.

You can download and install Intel TAT to monitor your CPU temps.

 

[yeason]

If it's 70C then yes that's hot, if it's reporting in F then that's quite a nice temp. A little more detail in this area might be helpful. Also faulty power supplies can cause this type of intermittent shutoffs.

 

[AMDThunder]

First of all, 70F or 70C? 70C ain't so good. Money says your HSF isn't installed correctly.

*edit* Don't let your message sit for a few minutes before hitting reply. Talking about me here) Then we won't all say the same thing in 3 different threads.

 

[Whimper]

Sorry if I didn't reply fast enough, AMDThunder. Was busy installing CoreTemp.

Some more clarification:

1. I'm using the stock heatsink fan that came with the e6400 processor.
2. I didn't add any thermal paste, because the stock fan said it already has "thermal conductive material" on it.
3. I'm posting temperatures in Celcius.


I checked again with my system sitting at idle, and got the following temps:

CPU (from the BIOS): 70C
MB (ditto): 36C
Nvidia says my GPU is at 49C.

Then I installed CoreTemp and checked again, and CoreTemp gave me a warning message saying my system was overheating, shut down immediately.

CoreTemp reported:

Core0: 127(?)C
Core1: 82C

(I may have these reversed, but either way, they're high!)


So basically, I have to take off the fan, and reinstall it, and try everything again? If it's not the fan, could it be a problem with the CPU itself?

(Also, because I love pain, I tried running 3DMark2005. It shut my system down in about 10 seconds into the first test.)

 

[Guest]

Recommended way to seat an Intel heat sink preferred. Have motherboard out of case and place on a piece of soft foam. Ensure all of the black center pegs are full pulled up. Make sure that slot on the black plastic part points towards the heat sink. Place the heat sink with grease the processor and gently wiggle the heat sink left and right. This is to set the pegs into the holes. Next is the part that most people get wrong. Take two opposite black posts and press down firmly until you hear 2 clicks. Repeat with the opposite pegs. Next flip the board over and verify that all 4 black pegs are through and even with the 2 white stacks. If the white stakes have not gone all the way through the board then heat sink is not seated properly and you need to repeat the procedure.

This can be done with the board in the chassis but you will be unable to verify that you have the heat sink properly seated.

You can also refer to this thread when I had the same problem

http://forumz.tomshardware.com/hardware/modules.php?name=Forums&fi...viewtop

 

[AMDThunder]

I was digging at myself, not you.

I'd bet a ton of money that your HSF is NOT installed properly. Take it out and try again. Your temps shouldn't be anywhere NEAR 70 at idle.

 

[yeason]

One thing to try is touching the bottom of the heatsink with your hand, if it feels warm to the touch but not overly hot then your mobo's thermal sensor is most likely whacked out. Also the "thermal material" included with many heatsinks is garbage. If you have some I would recommend using Arctic Silver 5 or the like. Those temps are definately high, and something to take care of immediately.

And as previously suggested try remounting the Heatsink. Sometimes when you build your first PC you don't get it on right due to nerves, etc. Another thing is to make sure the cpu itself is properly seated, if it's out of place that would throw off the thermal sensors and the heatsink mounts.

 

[NotAPimecone]

Heat is absolutely the problem. It's not supposed to go over 85C, ever. The motherboard reports a temp from a sensor on the board, CoreTemp and TAT report a temp from the sensor built into the hottest part of each core of the Core2 CPU. Usually there's not this much of a difference (most people seem to see ~10C difference), but this is clearly an extreme situation.

You might not need to take the HSF off to get it clipped on right, but if you do, remember to clean the thermal gunk off the heatsink and CPU with rubbing alcohol and replace it with some new stuff like AS5.

As for the intel thermal material being crap, I've heard it performs very well with the stock HSF, on par with stuff like AS5, if you can get the little bastard seated right.

 

[yeason]

lol

<-- AMD fanboy =P

Anyway I've heard complaints about the HS being hard to mount. As far as the thermal goop most of what I see is junk, but I wouldn't be surprised to see some nice stuff here or there.

 

[NotAPimecone]

yeason, have you ever built a system with the stock socket 775 HSF? The first box I ever built was with a Pentium 630. It's a nightmare. Trying to get the little clips in, it felt like I was going to snap my mobo in half with the amount of force that was required. I had a similar problem to this too, only my machine only stayed on for 15-20 seconds before shutting down. Lousy Prescott.

Even after getting the thing on right (and getting it off and on again was agonizing), I ended up afraid something would go wrong with it, and replaced it with a Big Typhoon.

 

[AMDThunder]

have you ever built a system with the stock socket 775 HSF?

I have. It was a breeze. Stock thermal pad and all. Sorry you had a bad time with it, but not everyone will. Aftermarket coolers are great if you plan to push your OC, but not necessary for everyone.

My 6300 is OCd to 3.0 on stock cooling. Max temp under dual prime 95 in torture mode hit 64C twice, but stayed 62-63. And how often does a pc stay under a max load during normal use? Never.

 

[NotAPimecone]

Hey, everything's better with Core2!

I think the stock thermal gunk was different back in those days too (not that it necessarily made that much of a difference), at least, it was a big ol' square instead of the 3 stripes making up an octagon that were on the one that came with my 6400.

 

[SpudTECH]

The way he is describing the randomness of the shutdown and the fact he was able to install updates says somthing about it not being a PROSSESSOR OVERHEAT. But the Key is: the system REFUSING TO RESPOND TO THE POWERBUTTON AFTER shutting down. I smell a bad powersupply...Possably one that is overheating and it tripping the thermal switch (by-metal switch) inside of it.

Check your powersupply fan.

Remove the powersupply from the case, But leave the motherboard and drive connections connected, then hold it to feel for the fan to spin up when you turn it on... BE CAREFULE NOT TO GET ZAPPED!!!!

If you don't feel any vibration then the Cooling fan in the powersupply is DOA and so is the unit itself.

 

[NotAPimecone]

Yeah, but the temps he posted say that CPU heat is, without a doubt, at least part of the problem, and a big part too. And it seems not-so-random, but based on CPU usage.

My DS3 does a similar thing when I try overclocks that fail to POST - it shuts itself off for a few seconds, and seems to not respond to anything, then turns on again - like it's doing a hard reboot instead of a soft one, and giving itself a few moments between off and on.

 

[Whimper]

I removed and reinstalled the heatsink fan, and that seems to have done the trick. I just ran through a whole bunch of graphics demos and stuff, and the cores never got above 55C. At idle, they're happily sitting at about 47C- nearly forty degrees cooler than before.

No problems so far- thanks everyone for your help!