OT: Oh god Please let it be fixed!

[Overkiller]

Hello everyone,

As some of you might remember....I have been fighting a LOSING battle against heat w/ my new P4 system...IT is the Perfect beast..well it should be at least:

Abit -IC7 MAX 3 w/ Otes Cooling setup
1 GB Dual Channel PC-3700 Corsair XMS
P4C 2.8 Ghz w/ a stock voltage of 1.525V (1.48V ACTUAL!!)
Radeon 9700 PRo
Antec Sonata Case w/ 2-120MM fans
Thermalright slk-947U w/ Smart fan 2
Enermax 465Watt PSU
36 GB 10K Raptor HD

Ideally this should run very cool (<50C) Cpu temp and overclock like mad..well For over a month i have tried EVERYTHING and the cpu still hits 80C under full load/IDLE...w/ System temps being around 20-25C

SOB was running (b4 the recent Bug) @ 520K cems/S...which is WELL under spec for this system (due to throttling)...

RMA time passed and i still hadn't found a solution...its either being read wrong (its not b/c its throttling), something was up w/ the IHS, or i'm an idiot

Well i've reseated for the 3rd time, applied a dead thin layer of Ceramique...and am hoping..hoping so bad that this system is going to just WORK as it should...this was my major 3 year upgrade and I've been lukewarm...


well right now i'm getting 650K cems/s and rising..and that is w/ the NEW client which hinders performance on P4's....Hope for me its fixed TA...lol, i've never tried so hard to get one thing to work, nor have i ever ever encountered this problem w/ all the pcs i've put together!

:heart:

[edit] dropping from 650K cems/s to 620K cems/s and continuing the downard trend... i guess i just got a Dud b/c i dont believe i can do anything else...the HSF cannot be attached any tighter and snugly than it already is....:brokenheart:

 

[deerslayer]

Your rate will drop over time that the client is open, especially if the client can't download work.

I wouldn't be concerned with a 30k drop.

 

[Overkiller]

Yes, but the chip is back to 80C :/

Also the Drop is continuing.

(throttling is OFF yet the heat is causing the processor to screw up i assume)

Pcmark 2002 scored ~400 points lower than similar systems, testing 3mark '01 now.

 

[jarsoffart]

Are you sure the heatsink is properly contacting the CPU? That is a common problem that causes high temperatures. mechBgon has a nice tutorial in his signature.

 

[Overkiller]

There is no way it isn't, but i will read that tutorial and 2x again tomorrow

 

[Mardeth]

Are you sure the mb isnt broken?

It might be broken and show to high tempratures tat will slow your computer....

 

[Coolkid]

I believe the H/S is very hot to touch aswell

 

[Allio]

How many different heatsinks have you tried?

 

[Overkiller]

Lol,

I am not rich here. Just a poor college student paying for it piece by piece

I have a zalman ALCU but i am at a loss as to how i cann try that out b/c the SLK-947U attaches w/out the retention cage...of which i no longer have (not as if that would've mattered b/c of the plastic screws that held the cage to the motherboard basically have to be pried/cut off to take the retention cage off).

If anyone can tell me how to attach the Zalman ALCU w/ out the retention bracket i would much appreciate it (posting in the Zalman forums and emails have led nowhere).

And yes, i have directly talked w/ Thermalright and have checked about 10 sites on how to assemble the HSF (and i've done this before!) but it still runs @ 80C. If i overclock to 3.0 ghz it runs @ 90C :Q (but it still runs fine..odd, huh?). I am about to give this system to my family (taking out everything except ram/mobo/proc.) and get a barton :evil:

LIke i said, if anyone has any ideas..let me know...maybe i should just pry off the IHS (i've read a few complaints of the IHS being improperly installed in a few processors...taking it off solved similar problems)..

And finally, yes the heatsink is burning to the touch and the IHS has left a scorched mark on the HSF Base (made it nice and smooth though)

 

[YellowRose]

Hey don't feel bad. According to the report my stock P4 2.4 is running at 126.7 degrees C.

 

[titanmiller]

And finally, yes the heatsink is burning to the touch and the IHS has left a scorched mark on the HSF Base (made it nice and smooth though)

I was going to suggest that maybe the senosor is way off but if the above is the case then I guess I would be just as frusterated as you. I really wish I had an answer for you, I know very well what it feels like to have a wacko problem with no apparent cause and no obvious solution, when it comes down to several hundred dollars, even more stress is thrown into the mix.


oh and...

Hey don't feel bad. According to the report my stock P4 2.4 is running at 126.7 degrees C.

:Q:Q:Q:Q:Q:Q:Q:Q:Q:Q

 

[Allio]

Originally posted by: Overkiller
Lol,

I am not rich here. Just a poor college student paying for it piece by piece

Believe me, I know the feeling. Still using a GF4 MX

To me it really sounds as if the only explanation is that you got a bung chip. Trying it in another motherboard would help narrow it down, but I realise that not everybody has that option - and if it really is dangerously hot, it might not be worth dropping it into a friend's motherboard.

I'd look to RMA.

 

[Overkiller]

Well...i do not have another motherboard to test it in and neweggs RMA for processors is 7 day or 15 day w/ a 15% restocking fee...I did not even get a chance( b/c of school full time + work full time :Q) to build the system around a week after i received it and then i've spent a ton of time trying to fix it thinking i did something wrong but i cannot find any signs that i have erred. So now its 45 days after i received the product, so that's 30 days too late to RMA @ newegg. Is there any way i can send it to intel and get the $$ or a replacement w/in say a week? (doubtful)
Right now i'm willing to pay for shipping and perhaps a lil something extra to send it to someone in AT so that they can fix it and make me feel like a dumbass for forgetting to do something (although i assure you, no stone has been left unturned)

 

[Freewolf]

If you could get someone to drop it in their MB to test it at least you would know if it's the cpu or the MB. I don't have any intel products myself or I would test it for you. The cpu should still be covered by intel but how long depends on weither it's retail or oem.

 

[titanmiller]

I was thinking and the best thing I can come up with is that the heat spreader isnt attached very well to the core.

On second thought it must be attached well if you heat sink is gettin very warm.

I guess I dont know what to say.

 

[SlowBones]

Grasping at straws...

Have you verified HS Fan speed?

Dumb q here: if fan plug reversed, would fan go backwards?! e.g. some air draw but grossly inefficient? perhaps you could visually confirm on shutdown. Fan mis-manufactured?

Good luck.

 

[Overkiller]

The CPU Fan reads @ 0 RPM in the bios even though it is attached (properly) to the CPU FAN header on the motherboard. The Fan appears to be running @ full speed as it is audible (granted a Full-speed Thermaltake Smart fan 2 shoudl be very loud @ full speed?). Could the Fan running @ say 2000RPM vs. 4000 RPM really make a 40C difference in temperatures? (just trying to get a better understanding)
[edit] and yes, the fan is pushing air INTO the HSF as it should be.
I thank All of you for all the help that i have been given. This has, hands down, beeen the weirdest problem i have ever experienced while working w/ pcs...

 

[dmcowen674]

OK, been seeing this thread for a few days now. Is the Computer shutting down or having problems when you detect the temp has risen or is the temp staying steady?

If the temp does not stay at one temp reading and continues to rise then you have a thermal runaway problem that indicates that either the heatsink and/or fan is not sufficient for the heat generated by the CPU.

If the temp remains steady at or below the rated spec for the CPU then you are fine.

PS - Should have this thread either in the CPU and/or Motherboard area and resident expert MechbGon chime in as well.

 

[lobadobadingdong]

I'd send you a different heat sink outa my collection, but they are all for socketA/370.

it's also possible the heat spreader on the cpu core has seperated, I've seen that a couple of times, and the only way to fix it is to replace the cpu.

on other thing, that you may have looked into, is that air isn't moving though your case enough, do have any case fans, the ps fan will blow out, but you may need another fan or two. if you get one, it too should blow out, toward the back, if you get two, one should blow out the back, the other one should blow in the front. my 1700+ had heat problems until I got a few 80mm panaflows going, the problems haven't been back since.

 

[Overkiller]

Case airflow is NOT a problem. That i can guarantee:Q

I have an Antec Sonata which some ppl report airflow problems but that is because they only use 1 -120 MM fan instead of 2.
Right now i have a 120 MM fan pulling air into case from front, over CPU and directly out the back 120 MM fan. There are no deadzones outside of right behind the dvd drive (but athat's all the way to the top). As if the 120 MM fan on the same level as the processor wasn't good enough, there i s an intake fan for the psu right above processor and then it spits out out the back end. On cool days the pc (i live in a finished basement) can have an ambient temperature of <20C, sometimes as low as 16C :Q

oh, i almost forgot! there is the otes cooling apparatus at the back end of the case which keeps all the capacitors & Etc. about 20C cooler than they normally are so those dont generate negative heat anymore.

WIth the above stated...the ambient temp can be 20C yet the processor will be 80C and the mobo about 25C.

And the temperature does not rise, it stays @ a constant 80C, even after a week of 24/7 DC. If turned off it very quickly cools down but w/in 2 minutes of restart will be back up to 80C (sometimes close to that IDLE)...

And i've tried in the overclocking forums..they are at a lost, i posted here b/c i thought i had solved it