HELP! New System Overheating!

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

I was there when they tested it. I bought form them several times and they sell good stuff. The cpu in my personal system was bought from them and they tested it the same way. However, they just plug it into the board and show you the boot screen with the specs. Nothing on the screen shows temp nor would anyone suspect and ask about it. Everything like I said is brand new. Even if it has a low temp on boot up, the temp doesnt rise until you launch an app.

 

I have my driving theory test tomorrow and I haven't even began to study. This is more important and its taking up my mind here. How can I even concentrate on my test with this on my mind? I just hope the thermal paste works as well as the cmos reset. I'll have to buy the paste in the morning. But do you guys think I should unplug the extra 4pin plug from the board? The board has a 24pin power socket. The PSU supports it by adding an extra 4pin plug along with the regular 20pin P1 connector. I plugged both in correctly. But I'm wondering if I should unplug the 4pin one? Do you guys think its giving extra power to the board and thus causing the overheating?

 

The heatsink is secure tight & alright. That's a really good idea to boot in Safe Mode and leave it idle. I might even try launching a few apps and see what happens. Thanks for the tip and thanks to everyone so far for your support and advice. I'll try resetting the cmos and booting in safe mode and let you guys know what happens. Keep 'em coming in the meantime.........

 

Hi Neil

What a coincidence - you have my middle name and live in the birthplace of my father!

Anyway, back to the problem. If you have confirmed that it isn't the thermal paste, then given that a CPU cannot spontaneously generate heat, it has to be the power fed to it by the motherboard. I would go with the mobo being faulty.

 

Eh? If you remove the heatsink like this then you *must* clean it carefully and re-apply new heat-sink paste.

I have to ask - and I don't mean to be nasty - but what are you doing building stuff for other people if you don't have some paste in your toolkit?

Note that allowing a CPU to get to such high temperatures can damage it.

Have you checked the various voltage rails for correct values? Overvoltage can cause an alarming heat rise.

Harry.

 

NO! A 24 pin motherboard should be fed with a 24 pin PSU.

The extra 4 pins are a separate supply from the 20 pin - check the voltages on this.

Harry.

 

My motherboard only has passive cooling on the Northbridge and nothing on the southbridge my case is crap and not well ventillated so I had an extra fan and an exhaust fan but my CPU temp has never gone above 51 degrees as I know the cpu has the paste covering it correctly so all of the heat is drawn in the heatsink.

In my opinion you either have a paste issue which I think is the most likley or the heatsink is knackered.

I doubt it would be both.

 

If the heat sink itself is hot bad thermal paste cannot be a part of the problem. The fact that the heat sink is hot means that heat is transferring from the cpu like it should.

The fact that the heat sink gets very hot to the touch says one of three things: 1.) It is not adequately sized for the job. 2.) It isn't getting the amount of airflow it should. 3.) Air temperatures inside the case are too high.

If it wasn't pulling the heat away from the cpu it would be cool. If it was rejecting the heat at a high enough rate it wouldn't be hot, nor would the cpu.

Heat transfers by 3 means: conduction, convection, and radiation. Conduction is how the heat transfers to the heat sink. The two objects touch. Convection is moving heat from one object to another by means of a third medium, air, water, etc.... Radiation is sensible heat that you feel when you put your hand near a fire or the heat you feel from the sun.

Heat transfer is also dependent on temperature difference. The rate of heat flow is directly related to the temperature difference between two objects. The higher the temperature difference, the faster the heat flow from high termperature to low temperature. Thus, if the air temp in the case is too high heat transfer will be slowed enough to cause an overheat in the cpu and a hot heat sink. Also, if there isn't enough air flow over the heat sink, or there isn't enough surface area for the air to hit, then the heat transfer will also be slow and both heat sink and cpu will be hot.

The first place I would look is for a failed fan on the heat sink.

 

Kool!

However it seems as if the temp goes in rhythm with the cpu performance in task manager. Definitely not the mobo.

 

That's it.......the manual doesn't go into detail.......just shows how to connect it. It doesn't give any recommendations or useful info. The manual is just plain pretty lame.

 

Yea after doing a bit of reading taking the time period into consideration, I guess I should've cleaned it and reapplied heat paste. Well I always meant to get a tube of thermal compound, but since the heatsinks always came with its own patch (which was always sufficient), I never bothered. Now I'm stuck without thermal paste.

However, the deeper I investigate this thing, the more confusing it gets. Like I said earlier, the temp works like the task manager......when the cpu usage goes up (eg when opening a new program like media player), the temp rises from 51C to 72C then to 89C! I know that this seems to be artificial, but that's how its detected. Now when I touch the heatsink, its cool.......but the north bridge is usually hot. Sometimes when I close the programs, it drops back to 53C. I'm wondering if the sensors are bad. I also reset the cmos and even loaded the "Fail-Safe" settings for minimal performance and it still didn't work. I even started it in safe mode and it didn't make a difference. I'm completely stumped with this now.

 

Thanks for understanding that its not really the thermal paste. And amazingly the fan is in good working order. However, just now, the software showed that the fan was clocking 225,000RPM!!! Then it fell back to 2690RPM then raised and dropped back. The cpu temp also fluctuated during this time. Whenever I launch a program the temp rises to 74C then to 83C and I'll usually have to shut down. When I reboot, the cmos shows the temp as 65-68C and it continues to rise. But I knew that I only spent lest than 12mins on it so when i touch the heatsink, its cool or like room temp. So I don't know why the system is showing all those high temps whenever I try to launch a program. By default now, the system is using 236MB of memory on startup. When I start up I leave the system idle for about a min or 2, then when I check task manager, it shows that 236MB memory is being used and I have no background programs or services running except the default system resources. Then when I open a program, the temp rises to 74C then to the 80's and I'll have to shut down.

I'm thinking about taking the system to a pc repair shop and get some info. Even if I have to pay the guys to fix it, then I would. But it pays to learn and its all experience for me. I gotta get the system up and running by friday no matter what. I tried everything possible. Its out of my hands now, I'll let the store techs handle it and see what they recommend. And when it works, then I'll see to it that I never forget the solution!

 

THANK YOU GUYS so much for your assistance. I did everything I could. I'm frustrated and disgusted right now and I don't intend to waste anymore time with this right now. I gotta get some sleep and study for my exam early in the morning.......I haven't studied at all thanks to this. I'll take it to a pc shop and see what the techs there have to say. Once it works, I'll definitely post here and let you guys know for sure. Don't get me wrong guys, I'm NOT a quitter........but this just happened at the wrong time when alot of things are going wrong for me and when I have too much on my mind. I'll pay to get it fixed and in the meantime, learn what the problem is and how to fix it should I be faced with it in the future.

 

It sounds very much to me that there is a fault on the motherboard itself. I think it's getting bad information from it's own sensors and shutting the PC down as a precaution.

If as you have mentioned, the heatsink itself is not getting too hot to touch then that would support this theory. Unless ... it is a problem with the thermal compound as previously mentioned.

You could try altering the warning and shutdown thresholds in the BIOS (providing you are sure the heatsink is sat properly and has the required thermal compound). This would hopefully stop the system from shutting down when it thinks it's overheating. The only other issue with this is, if it where ever to properly overheat, then it wouldn't shut down also ...

I do hope you manage to get it sorted out in time.

Now to be a little pedantic ...

I don't want to burst your bubble mate, but the system is anything but high spec. That's quite low spec by todays standards really. Ram is cheap as chips these days so you shouldn't really be putting any less than a couple of gigs into the system. Especially with that onboard GFX card which probably nicks ram from the main memory pool. It's unusual for onboard video cards like the 6100 to have their own dedicated ram. Oh, and the 6100 onboard is dog slow, so I hope it wasn't meant for gaming. That particular motherboard is also a dirt cheap one (33 quid on ebuyer), so be aware that you do get what you pay for. I learnt a long time ago in system building, that if you try and increase your profit margins by buying cheap crap, then it will come back to haunt you later on. Sorry, I don't mean to rant as it's not going to help your current situation but it made me feel better to do so.

 

A few months ago I had some trouble with "odd" temps even though I was using a custom heatsink designed to improve the cooling. It had been fine for months though and then went bad.

I would learn from this experience, always clean and re-apply thermal paste on the CPU and clean the heatsink!

I would suggest trying another heatsink, they can be funny things. Failing this, I'd get the PSU checked, you can do this by using speedfan to check the voltages but it would be better to take it to a shop and have them use a voltage meter to check the flow.

Have you tried re-seating the heatsink? I would really go and spend the extra £4 on some good thermal paste and try it..

 

I disagree. You're even getting spontaneous fanspeed readings, if I've read your following posts correctly. Everything you've mentioned thus far points to a problem with the mobo, and is ultimately managed by the motherboard.

 

It does sound like it could be a motherboard problem... but could also be a PSU problem.

 

Ok guys, I took the system to the store and they said that they'll run their diagnostics on it and that I should get it back in 2-3 days. I'll just tell the customer that I'll spend a few more days testing it and doing diagnostics to test its stability. It wqs also the same store where the mobo was purchased. More on that when I get the info.