Sign in with
Sign up | Sign in
Your question

E6400 heat problem. 50c no overclock?

Last response: in Overclocking
Share
June 25, 2007 10:59:44 AM

I bought a E6400 with a Arctic Freezer 7 (or something) in January and I was very happy with it. The fan was spinning at under 1000 rpm and idle temps was down to 30C.

Cut to 6 months later and my fan is runnin 1500-1800rpm IDLE and the CPU temp is 45-50C.

The room temp has gone up about 5C, but I dont think that can cause so much problems.

Can it be bad paste? Broken Fan? Any ideas? Ive opend the case and it seems like the fan is stuck like it should be.

Maybe I should just take the entire thing off, clean it and put on new paste?
June 25, 2007 9:04:02 PM

It's probably the fan. Thats exactly what happened to me.

I was getting 20C higher than when I bought it, and when I peeked underneath the heatsink I found that the paste had hardened and been burnt. So I took a knife and and forcefully took off the dark burnt paste and applied a new coating. Temps went back down.
June 25, 2007 9:20:06 PM

thanks, ill give that a shot.
Related resources
July 2, 2007 10:21:11 AM

I took the fan off and added new heatpaste. I didnt see anything wrong with the way it was set up before.

After this the temp dropped about 10C for a day or two, but now its back to the way it used to be. High fan-speed (1500-2000) and 45+ish CPU temp.

Any other suggestions?

At what point should I start to worry? Sure, the sound of the fan sucks, but at what temp does the CPU start taking damage?
July 2, 2007 12:33:38 PM

Have you checked to see if there's any dust build up on the HS fins, cpu fan blades AND any case fans you have?
July 2, 2007 12:35:34 PM

yupp, did a very turrow cleaning when i took out the fan.
July 2, 2007 12:56:25 PM

Maybe it is a problem with case ventalation.
Are there any other fans in the case that are not working correctly or filters that are plugged?

Does you're VGA card exhaust at the back, and if so, is its exhaust blocked?

Have you relocated the case to an area that has less ventalation for the hot air to move away from the case, forcing it to get sucked right back in?

Try running it for a day or 2 without the side panel and see if the temp drops.

Just a few ideas I had.
July 2, 2007 1:03:04 PM

cpus start taking damage when they hit 90oC + on the core when active and about 145oC+ when inactive.(aka, when the computer is turned off and when the computer is turned on, not idle and load)
July 2, 2007 1:04:12 PM

3 120mm fans, front back and side. All working. Two sends air out while one drags it in. Should hav ebeen the other way, but thats how it is :) 

Its under my desk, but it has atleast 15-20cm of "air" around it on all sides, so I dont think thats the issue eather.
VGA sends air out the back, and it seems to be OK. Atleast I see no diffrence in GPU temps.

Ill try running it without the side panel ans see hwat happens.

Maybe I should just get a new cooler...
July 2, 2007 2:12:52 PM

Sounds like you should have plenty of case air flow. Maybe the heatsink mount is loosening or the springs are weak and the heatsink is pulling away from the cpu?

I don't know. I'd recommend re-seating the heatsink, but you already tried it. Third times the charm?
July 2, 2007 4:57:54 PM

As the last guy said the only other thing I can think of is that maybe 1 of the screws/clips securing your AC 7 to the mobo has worn out and isn't maintaining a tight seal to the backplate.

Can you wiggle your AC7 to see how much movement there is? Does 1 axis perhaps shift more than the others (indicating a loose connection on 1 side)?

If that all seems fine perhaps you could try using the stock cooler that came with the CPU. If the temps stay the same or go up that will help narrow down where the problem is.

Let us know how it turns out.
July 2, 2007 5:07:30 PM

Only "wiggle" movement I can feel is from the plastic itself, seems to be stuck.

I dont have a stock fan, i bought a bulk CPU and the fan seperate.

Will a E4300 fan work? I got one of those :) 


Also, been running without a cover now for a few hours. CPU temp is pretty much the asme, but "system temp" has gone up about 7C. Now 48C.

Im not saying it us, but CAN it be a software problem? OS related? Bios related? Or is things like this ALWAYS hardware?
July 2, 2007 5:08:46 PM

1 other thing, which of your fan positions is extracting/pushing air? If you have a case fan at the front as well as the back trying to extract air at the same time it can leave a void of air in your PC which will increase temps.


Make sure the air is flowing in 1 direction in your case (usually front to back) to insure optimum air/heat transference.

Also can you fill out your PC's specs a bit so we know exactly what we're dealing with.
July 2, 2007 5:25:03 PM

that would be valid if it had been a issue from the start, but ive been using it for months or so before this heat problem came up.

Front sucks in, back and side blows out. PSU on top also sucks in i think.
July 2, 2007 5:34:28 PM

When at idle, what is you're cpu usage? is it going down to 0 or close to it?
July 2, 2007 5:36:43 PM

The higher temp is usually due to HS not fully clamped to CPU! It is very common when tightning the final clamp, that the one opposite comes loose.
If all seem tight, remove HS and check all lockpins are as they should be. Set them all to unlock, clean old thermal grease and re-apply and refit.

Mike.
July 2, 2007 5:37:32 PM

I just thought it was worth making sure.

The e4300 HSF should be fine to use, as I believe that the retail HSF is the same for all cpu's below the e6600. I could be wrong but my friend who got a e4400 got the same stock cooler as the 1 which came with my e6420.

Give it a try!
July 2, 2007 6:30:22 PM

How do you measure your temps? TAT or other utility, or it is just what your bios indicates? If it is the latter, maybe you updated your bios? Sometimes bios updates change the way the bios translates thermal diode indications.

If you haven't TAT, download and install it and check temperatures at idle and at 100% load. Also please tell us your ambient temps.

You also mentioned something about "adding" thermal paste when you removed your HSF. Did you thoroughly cleaned the cpu and HSF before "adding" thermal paste? What type of thermal paste do you use?

Sorry for all the questions but i think it is necessary to have the data in order to help...
July 2, 2007 6:33:25 PM

Quote:
I just thought it was worth making sure.

The e4300 HSF should be fine to use, as I believe that the retail HSF is the same for all cpu's below the e6600. I could be wrong but my friend who got a e4400 got the same stock cooler as the 1 which came with my e6420.

Give it a try!


The cooler is exactly the same even for the E6700. What makes the difference is the fan they use. In any case i agree it won't do any harm to try it.
July 2, 2007 6:59:59 PM

Ive been using the tool that came with my Motherboard, EasyTune5.

TAT shows IDLE temp at 45C and maxload 52-53C. Room temp is currently 23C.

It's not that the temp is so high that it "bothers me", the problem is more that the fan is running at such high RPMs that the sounds is pretty high and the fact that something definatly has changed. During the stress test the fan was up to 2600RPMs.

Yes, i cleaned it off using the alchohol napkins you get from IBM when you change CPUs. (Part of my job is to change/fix hardware) The paste I added was IBMs paste.

While I do have the E4300 fan, its a bitch of a job to change it over. Need to take out all the wires and take the motherboard completly out for every change :(  I cant get my fingers down to fasten the pins while its still bolted to the case.
July 2, 2007 7:35:26 PM

No OC but what about increased voltage? I could easily see that temp if the voltage were high enough for say a 2.7GHz C2D. Check your voltage or just try OCing with no voltage change and see how high it goes.
July 2, 2007 7:54:36 PM

Bios sais the voltage is at "auto", and CPU = Normal at 1.325v.

I dont know what that number should be. I have never changed it (atleast not intentionaly)
a b K Overclocking
July 2, 2007 9:19:11 PM

Very Good Warning - I agree. I came to the same concussion, had problems with orginal HSF, after about a month I replaced with a HSF with a backing plate. (Feb 07).

The OP stated that he had R&R(eplaced) the HSF and temps dropped, but only lasted for a day, or two. This is indictive of what you pointed out.

On CPU voltage. I have a simular system, as noted below. My cpu voltage is set to NORMAL in Bios, and BIOS reflects 1.250V. This voltage is also reflected in Easytune, Coretemp, and Speedfan (Note speed fan shows 1.22V).

As pointed out double check CPU usage. Three days ago my Temp shot up 7C, Spent a day R$R HSF, and redoing Fan Locations- Plextools was the cause, I had left a DVD in, and Plextools was running in the Background.

--------
E6400 @ 3.2
2 Gigs PC6400C4 (4-4-4-15)
ATI X1300 AIW (Not into Gaming)
4 HD, 2 Raid 0 Pairs + 1 removable HD + 1 USB HD + 1 Firewire HD
Modified side of case - One 120mm Upper Back side Blowing out and one 80mm lower Front of side blowing in
July 2, 2007 9:50:02 PM

I appreciate all the help fellas...

I have now broken my computer completly though... PSU is probably shot to hell.

I was trying to see if there would be a diffrence in temp\sound if I disconnected two of the fans in my case. (connected to a splitter that lets me connect several fans to one powercable).

It didnt, but when i plugged it back in again my computer did a reboot and wouldnt come on again. It powers on and I see the fans trying to spin, but dont even do one round. After 2 sec it powers down and up again.

Sounds like a PSU issue yes? Its only 6 months old, thing I can get a new one if im lucky and maybe not 100% honest? :) 
July 3, 2007 12:37:28 AM

Why would you not be 100% honest? You didn't do anything that would void your warranty. LOL. You don't need to RMA it! It's perfectly fine. It's the short circuit pertection of your PSU. You should NEVER NEVER NEVER NEVER connect or disconnect anything to your PSU while the computer is on with the exception of some hardware such as ESATA/SATA HDD, USB devices that doesn't draw great amount of electricity or devices meant to be hotplug/switchable

. Anyways.. try superglue the pressure clamp.... it seems like the problem is that the tension device isn't strong enough thus it let go of pressure after a few days.....

I would suggest you do a 2 day test since you said the temp went up after 2 days.

0.Reseat your heatsink and reapply thermal paste
1. Turn your computer on for 2 days straight
2. Record the temp every hour
3. Compile a graph to see if the temperature is raising steadily.
4. If it is, that means there's a problem with the tension clamp. and you need to find a way to secure the HSF so it doesn't lose tension after installation.
July 3, 2007 12:51:44 AM

Quote:
I have now broken my computer completly though... PSU is probably shot to hell.


Yes, it probably is :cry: 

Quote:
Sounds like a PSU issue yes? Its only 6 months old, thing I can get a new one if im lucky and maybe not 100% honest? :) 


Yes you can :D  It could have happened during a power outage anyway.


On the fan/temp problem: You have a gigabyte 965-DS3 (or S3 or DS4) mobo and you updated the bios recently, correct? The mobo is feeding your cpu with way too much voltage. It should be maximum 1.3525, 1.175 minimum, depending on your stepping. Check out Intel's web page for exact specs of your stepping.
http://processorfinder.intel.com/details.aspx?sSpec=SL9...
http://processorfinder.intel.com/details.aspx?sSpec=SL9...

You will have to adjust it manually. 0.1 Volts more can be accounted for 5 degrees Celcius more than normal. This is done from "MB Intelligent Tweaker", "CPU Voltage Control" in Bios. Also you should adjust the settings for your cpu fan in bios. This is done from "PC Health Status". You have to set "Smart FAN Control Method" to "Legacy" and "Smart FAN Control Mode" to "PWM". Fan speed will be dynamically adjusting itself according to temperature. When you have your PSU back from RMA, post again with results.

Good luck! :D 
July 3, 2007 7:23:52 AM

Quote:
You don't need to RMA it! It's perfectly fine. It's the short circuit pertection of your PSU.


So, how do i get it to work again then?

Ill try what you said about a twoday test and log temps once I get it up and running again.
July 3, 2007 3:11:43 PM

I think that jackluo923 is mistaken (no offense). Your PSU is dead....
July 3, 2007 4:15:22 PM

OK, new info:

I took the PSU out and hooked it up to another computer. It is working just fine. So we ruled that out.

It can only be two things then: CPU or Motherboard?

I CAN test the CPU on another computer, but im out of heatpaste so I wont get it back together again.

Im thinking motherboard this time around. The green light comes on so it gets power, but the backfan and the CPU fan only moves just as the computer powers down.
July 19, 2007 11:09:39 PM

it still could be your PSU, is the other computer as high end as this one. If not then it might not be pull anywhere near enough power to make it die like it is on this box.

I would look into trying a different PSU on this computer before you trying putting in a potentially bad CPU in a good board. I had a PSU fry a motherboard once, it died and since that point the motherboard ran fine,but it killed 2 more PS before i retired that mobo.
!