Sign in with
Sign up | Sign in
Your question

PC keeps restarting

Last response: in CPUs
Share
October 13, 2007 7:26:39 PM



Hey, today my pc shut off by itself. Its all pluged into the same power bar and while it was off my speakers were still on.

since then it has restarted twice on me.The first time while surfing the net, the 2nd time while I was playing COD4 demo. How can I fix this. I have a 600 Watt PSU its called OCZ slientStream. q6600, 680i mobo, Corsair ram 2 GB DDR2 and a x1600 Pro

Could it also have something to do with over heating

More about : restarting

October 13, 2007 9:28:08 PM

PSU or it is possible to be a heating problem. Especially if oyu OC it.
October 13, 2007 9:28:54 PM

not ocing it at all. How do I fix the Power Supply or test it
Related resources
October 13, 2007 10:12:43 PM

help
October 13, 2007 10:23:09 PM

Im using speedfan

here are my temps. What temp should I want?
system 21 C - CPU 45C - HDD 36C - core 0 to 3 are 35C. These are the temps when im running nothing but the browsers.

oh man, I turned on a game demo and I looked at the CPU temp.It said 51 C with a flame next to the icon
October 13, 2007 11:36:10 PM

can someone help me
October 13, 2007 11:52:35 PM

Is speedfan still off by 15c? If so your getting hottttt. You can buy a psu tester for $10.00. Wouldn't hurt to run a memory test either. If you are getting hot look into a better heatsink and use as5.
October 14, 2007 12:00:27 AM

Does this mean my CPU isn't working right? It shouldn't be over heating from just running a a web browser. I am NOT OCing
October 14, 2007 12:16:54 AM

I would check the CPU voltage in BIOS first. If this is correct, I would check to make sure the CPU cooler is still properly mounted. The only real way to do this is to pull the board, and hold it vertically, like it would be mounted in the case. Then look down where the cooler meets the CPU. You can also check the back of the board to see if the mounting tabs are uniform.

The Q6600 and the 680i board do run a bit hot normally. Depending on room temp, and how good your case cools, this could vary from day to day.
October 14, 2007 12:18:40 AM

I have an Antec P180 case.
How many volts should it read in my BIOS?
October 14, 2007 12:19:36 AM

70C is bad. It is possible that your heatsink isnt installed right or to little or to less compound between cpu and heatsink. At this time I'm leaning towards memory problems. I need more info. And you need to run memtest86.
October 14, 2007 12:52:33 AM

1.100V to 1.372V depending on load. Just running in BIOS it should be closer to the 1.100.
October 14, 2007 1:00:21 AM

psu!

open your door and if you runs longer its the psu
October 14, 2007 2:58:55 AM

I do not remember ever putting thermal paste on. Would this have already been stuck to the metal heat sink? In my mind thermal paste looks like gel. I don't know what it really is.

I looked at my bios and here are my results

CPU CORE: 1.24 V (AUTO)
CPU FSB: 1.2V
Memory:1.850V

Memory timing

tCL 5
tRCD 5
tRP 5
tRAS 18
CMD 2T

Corsair XMS2 TWIN2X2048-6400C4 2GB 2X1GB PC2-6400 DDR2-800 CL 4-4-4-12 240PIN Dual Channel Memory is my ram
October 14, 2007 4:26:37 AM

Voltages look good. I assume you used the stock Intel Cooler. If so, it comes with pre-applied thermal paste so you should be OK there.

One thing about the way the Intel coolers install is, you need to make sure the mounting pins fully seated through the motherboard holes. If not, it may have worked itself loose over time which could cause the overheat.

Here is the Intel video on heatsink install. http://www.intel.com/cd/channel/reseller/asmo-na/eng/24...
a c 172 à CPUs
October 14, 2007 4:35:14 AM

I think specifications from Corsair say run the memory at 2.1 volts.
October 14, 2007 4:57:50 AM

jsc said:
I think specifications from Corsair say run the memory at 2.1 volts.


Actually I believe 2.1 is what you can over-volt it to in order to achieve the 4-4-4-12 timings. I think the "out of the box" default is 1.8v and 5-5-5-15. That's the way mine worked initially on my DS3L MB. I saw this in the fine print over at the Corsair site.

I was more concerned with severe over-volting when I asked for his readings. The would have indicated a PSU problem.
October 14, 2007 5:27:37 AM

You have no thermal past on your heat sink!?!!?!?!

If you put your rig together your self the you really should have put some thermal paste on the heat sink. However some heat sinks do come with thermal paste already on them, it's generally a grey, sticky compound on the stock cooler, and i know some after stock coolers come with small syringes of thermal past that you need to apply before installing the heat sink.

If you don't have thermal paste on, you are very likely overheating, go buy some arctic silver now and follow the instructions on how to apply it.

If you do have thermal paste on, and it's not you're CPU overheating, then I'm with the other guys, run a memtest to see if it's your ram.
October 14, 2007 9:08:01 AM

I just solved a similar problem on my pc.I had Vista transformation pack installed and that caused my pc to keep restarting.So although your considering hardware problems you should also consider software problems too.Did you install anything new just before the problem started?
October 14, 2007 10:27:03 AM

How do you use memtest. I downloaded it but nothing happens when I open the programs. Do I somehow need to boot from this?

I used the stock heatsink that came in the box with the CPU. Someone said it already had thermal paste on it. I can try and put the heatsink on again.

Here are my idle temperatures with core temp

here are my results from core temp

core 0:54
Core 1:54
Core 2:59
Core 3:57
October 14, 2007 11:05:34 AM

Memtest should have come as a .zip file. You need to unzip it. If you downloaded the floppy version, there will be 4 files. If you downloaded the boot CD version there will be one file called memtest.iso. You will need to use a program to burn this image to a CD such as Nero. This is not a copy operation, this is a burn ISO image operation.

Let me know if you need help with ISO.

I assume you checked the heatsink already and it was fine.
October 14, 2007 11:20:35 AM

you need to run memtest from DOS. But before you do that, pull your CPU cooler off and reseat it. That sounds like a more probable cause.
October 14, 2007 11:46:03 AM

what i recommend is you get one of those thing which tell you how much power a socket is drawing from the mains, in watts, although i doubt you'll even be getting close its a good tool.

your problem sound likely to be memory or a PSU (with it being a premium company PSU at a high wattage i do not think that it will be the latter.
October 14, 2007 12:17:36 PM

what should my idle and load temperatures be guys? What is the ideal range im looking for
October 14, 2007 1:01:37 PM

googleads said:
How do you use memtest. I downloaded it but nothing happens when I open the programs. Do I somehow need to boot from this?

I used the stock heatsink that came in the box with the CPU. Someone said it already had thermal paste on it. I can try and put the heatsink on again.

Here are my idle temperatures with core temp

here are my results from core temp

core 0:54
Core 1:54
Core 2:59
Core 3:57
Those Idle temps are too high. Get an aftermarket HS and throw that crap Intel HS in the trash. It's not mounted properly due to the crap mounting pin design. Do not get an aftermarket HS with the same push pin mounting design, it is flawed.

Post a budget and you will get some recommendations I prefer the Thermalright Ultra-120 Extreme CPU Cooler with the Scythe S-FLEX SFF21F 120mm Fan, but that may be out of your price range. There are alternative heatsinks in this test AnandTech: Thermalright Ultra 120 Extreme: Is More Better?. After you clear up the heat problem then check your ram etc. Even if it isn't shuttung down due to thermal overload you will shorten the life of the CPU needlessly.
a b à CPUs
October 14, 2007 1:30:41 PM

You say you do not remember putting on any thermal paste, did you peel off the little plastic cover that is over the thermal pad on the stock heat sink before mounting it? That will definetly cause overheating problems if you did not remove it.
October 14, 2007 2:25:00 PM

I ran memtest. I did 2 passes and I had no errors. It took me 1 hour and 30 minutes, testing both ram sticks at the same time.

Ram is not the issue I guess. I have not reset the heat sink. I will look to see if I took the plastic cover from the pad too when I do. Im going to do this right now If I can.

What temperature am I wanting to have though. I don't know what is low and what is mid range
October 14, 2007 2:50:17 PM

Idle should be between mid 20s and mid 30s, nothing more aT IDLE.

If you check the heat sink you better have some thermal paste to reapply. Never try to reuse what is on there. You have to clean both surfaces with alcohol before reapplying.
October 14, 2007 6:39:25 PM

I did not take the heat sink off. It looks like the pins are pushed down all the way.

People on the Anandtec forum are telling me the temperatures are normal I am getting.

According to the post above me, my temps seem to be in the safe to cool zone. What gives?

Scale 4: Quad

Q6600: Tcase Max 62c, B3 Stepping, Tjunction Max 100c, Vcore Max 1.372, TDP 105w, Delta 10c

-Tcase/Tjunction-
--60--/--70--70--70--70-- Hot
--55--/--65--65--65--65-- Warm
--50--/--60--60--60--60-- Safe
--25--/--35--35--35--35-- Cool
October 14, 2007 7:08:19 PM

The best way to make sure the HS pins are through, (I know it's a pain) is to take the MB off the PC case MB plate. And you should see the pins like this:



But, seeing your speedfan temps for 3 cores are at 35C (speedfan is 15C cooler) seems 50C is your correct temp, which is safe.

So I wouldn't say your temps are causing the reboots. Perhaps it is the memory voltage. Just look up the spec's on the modules you got, and set them accordingly in your bios. Using CPU-Z program can help tell you more info on everything while your in windows.

CPU-Z

Might want to be sure perhaps if you have the latest bios (check to see the changes if they list them) or drivers for all your components.

Hopefully you'll eventually find the problem in your trouble shooting.
October 15, 2007 12:34:42 PM

If you have a good handle on how to measure your temps by reading Computronix's guide then try stressing the CPU with Prime95 25.5. Check the maximum temps because your Idle temps should be in mid the 40C range not mid to high 50s. Here are the temps on my girlfriends machine. EIST enabled, voltage lowered in BIOS to 1.275, it shows lower in CPU-Z. Ambient temp ~20C. Ultra120 HS.






October 15, 2007 1:58:24 PM

Ooopph.. ya, 50C at idle is abit high. Dunno what I was thinking... Perhaps I wasn't thinking. :sleep: 

Ya, I'd have to agree with Zorg, that HS may not be on correctly, the more I think about it... now. :lol: 
!