CPU possibly fried 6

[BenyG]

This thread refers to the problem that I mentioned towards the end of tihs thread: thread602-665831 Reading it is not required to be able to understand/help with this thread but I've included the link for completeness.

Before anyone decides to flame me I think it is a good idea to make a new thread as my problem has changed somewhat.

I have a AMD 750 Duron system.

I had a cooling problem with my CPU and vacumed clear all the dust between the fan and heatsink etc. as advised.

Before when at load the CPU temp rose to 69C before freezing. This was after 20 mins of running at full load.

However now after cleaning everything out the CPU rises to only 60C before freezing! The cleaning has helped as the fan now runs at 4115 instead of 3750 with all the dirt in it. But I cannot think why the system freezes after ~3 mins of full load now instead of ~20 mins when all the dirt was in there. Here is some data of the temp rise when at full load after the fan being cleared out.

Time Temp/C Time Temp/C
0 52 0 52
00:00:10 57 00:00:12 56
00:00:35 58 00:00:20 57
00:01:10 59 00:01:20 59
00:03:52 60 00:02:36 freeze
00:04:29 freeze

When checking the raised part of the CPU it was still sticky to the touch so I cannot see how that can be the problem.

Is 60C (70C inside) still to high for the CPU?

Why would the CPU now be less 'tolerant' to higher temps now that I've cleaned its fan out?

Is it possible that the decrased tolerance is a symptom of the CPU getting fried? The problem started in September but I have been careful with te CPU since October.

Thanks for any help,

Ben

 

[nlm9802]

1. re: installing a heatsink on the cpu.. one way to check that it is seated optimally would be put a very small drop of arctic silver in the center of the cpu, then clamp down the heatsink. remove the heat sink and observe the top of the cpu. the arctic silver should have *uniformly* spread across the entire surface. the less you used, the thinner the layer should be. If it did not spread uniformly there is a problem with how your heatsink is mounted, and you will need to adjust it or get a new heatsink.

2. your system shutting off. Have you checked the temperature on your hard drive? An overheating hard drive will produce the exact symptoms you are having. shorter and shorter times before shutdown each time it overheats. Next time it shuts down, quickly place a fingertip on a metal surface of your hard drive. If it is much warmer than 90-100F that is probably your problem. A temporary fix would be to place a normal fan (8-12&quot next to the case blowing room temperature air onto the hard drive and start backing up your data.

and I hear you on the asus boards.. I bought a p4c800 deluxe 2 weeks ago, and after 1 week of use it started getting the "system failed cpu test" message (caused by the heatsink mount being too tight). I put in a call with their tech support, who offerred me 8 suggestions apparently just copied and pasted off of some troubleshooting manual they have. of the 8 I had already tried 6, 1 did not work, and the other didnt even involve my motherboard. I thought I was upgrading from soyo to asus, but I won't buy from them again *rant mode off*

 

[BenyG]

SYAR2003,

Its only 52C idle, it goes up to 60C before it crashes, instead of making upto 69C before it crashed a while ago. I might have dislodged the pink-sticker paste that came with the heatsink.

Fog does not means its hot! It can be cold and be foggy you know


TheGoatMan,

Not that I particularly hate ASUS boards I can tell you that whenever I have tried to install Win2k on a non-Gigabyte board it always never boots up with stop error. - Gigabyte always seem to be the best (waits for flames

 

[paparazi]

BenyG
Hi I also live in England (Nottingham to be precise)
It does, by you description, sound like the heatsink is rather small, given that you say that it is cubed like and about the same size as the CPU, most new heatsinks overlap the CPU sucket on two ends , being more rectangular in shape, the increased mass of the heatsink, the larger number of fins, the often larger and more powerful fans and the common use two differant metals (copper based and alluminium alloy sink) all mean that newer heatsink/fan units have vastly improved.
I will add though, that some AMD OEM (original units supplied with retail boxed CPU's) are the small fine fin technology "scrived type" although very small they generally work well due to the vast number of fins, the only problem with these units is that they get easily blocked because of the compactness and tight spacing of the fins so their efficiency drops off very quickly.
You have to regularly and meticulously clean this type of fine finned heatsink
You really don't need to spend much to greatly improve your cooling, something like the arctic cooling copper silent at under £10, combined with a single 80mm case fan at around £3 (fitted under the PSU, exhausting out) should see your temperatures drop to under 50C
Don't worry too much about what you use to remove the old paste, generally 98% of the old pad comes off with you finger nails and the remainder will come off with WD40, petrol, thinners but only apply to a natural fibre cloth and not directly to the CPU (OK on the heatsink base)
Martin

Replying helps further our knowledge, without comment leaves us wondering.

 

[TheGoatMan]

I've always ran either MSI or Soyo boards becuase I love their dependability. I tried ASUS once.And my friends once die hard ASUS following has died due to unstable boards. My old girlfriend was German and wow was she hot. She had to move so that was that. British girls are by far the coolest so don't feel to bad. Anyways you should probably drop the duron and drop a few pounds on an Athlon.

 

[BenyG]

I think my CPU is getting very tired.

It has now crashed on normal and not just full loads! Now with it cleaned out the idle temp got down to 48C before it hummed louder and the temp rose to 51C. This is the same action that occurs when you turn it on after being off for a while.

The system now always hangs when playing movies and barely gets into games.

I'm going to get some paste but in the mean time does the fact that it freezes even when not at load and at 49C suggest that the CPU had been damaged?

I've also gone through and marked helpful posts

Ben

"To our surprise we have not been able to locate even
one scientific study on the proved harmful effects of addiction"
Dr George H Stephenson, speaking about Heroin at the University of British Columbia, 1956

 

[TheGoatMan]

Run prime95 if you can and it will tell you whats up. Just download it and run it. It takes forever though.

 

[paparazi]

Your first statement is interesting and maybe an insight into your problem "it hummed louder and the temp rose to 51C" it's true that some newer motherboards have fan speed control so to spin faster when needed but I dought your motherboard has this feature so, it leads me to believe that your power supply may be failing due to poor voltage regulation, just a thought?
Nice to see your temperatures coming down a little and well within "safe stable limits"
If you can possibly borrow a known good PSU? to try.
I'll keep my fingers crossed for you.
Martin

Replying helps further our knowledge, without comment leaves us wondering.

 

[SYAR2003]

Or you can disconnect one CD drive (if you have two)
This will free some wattage.

 

[BenyG]

Martin,

Thanks for crossing your fingers

I actually thought about this just after posting it. - Always thinking about posts *after I post them and realised that the noise seems to be coming from 1 of my HDs. This only happens when the PC has been off for a while. Additionllay it seems to be more prone to hanging when botting up from cold like this, especially at boot up.

Does this suggest that its PSU related?

The only thing against it is that it is 100% prone to hanging when the CPU is at load but sometimes it takes just 3 mins at load and other times as much as 13 mins. The system also hangs when not at load but does it about 2 times / day.


Ben

"To our surprise we have not been able to locate even
one scientific study on the proved harmful effects of addiction"
Dr George H Stephenson, speaking about Heroin at the University of British Columbia, 1956