View Full Version : climbing temps?
My 1000 Athlon has been oc'd since I got it- 1400mhz, and originally had temps of 48c ( all temps are @ 100% load) or so...case temps around26 to 28c.
I changed heatsinks a few times, fans, etc and got temps to 41/42c (cpu)and ran for a while- but every few months temps seem to jump up 1 or 2 degrees c!
I now have bought a full tower, with more than ample cooling (case is less than 1c above ambient!) and once again temps are going up: I have been @48c for about 5 weeks, but its starting to hover around 51/52c!!! WAY TOO HOT for my tastes!
Has anyone seen this before? I have redone the AS2, reseated hsf and put the 2 LOUD YS Tech fans on this puppy (supposed to be just 1!)
Is my cpu burning out and getting ready to fail, or what?
OK folks, BUMP!!!
I've been trying to bump my other pcs up a bit as well, and my xp is now stable only at default! The only oc that still works is the 900@1071!!
I am going to try the overclocks again in a while, but I do NOT understand this behavior.... they ran for quite a long time, stable, and the only one that has done anything strange has been the 1000.
Any guesses?
I am at my wits end here!
Direwolf
11-10-01, 06:38 PM
What kind of heatsink you using!?!
I would prefere a Sk6,Alpha PAL8045 , or a swifttech with a delta 68cmf blowing on the heatsink for (best results)
perhaps check your case and make sure cords and what not are not geting in the way of the airflow. Plus back your case away from the wall if it is near it (give it a good 12 inchs) so it has room to blow out the excess air/heat
plus make sure the room temp is not to hot. usually keep it a good 69/70F :D
hope any of that helps
I'm not using anything great for a heatsink, and the pc is shoved in a cabinet with some, but not much airflow; most of the case mods I have done are to compensate for this.
But the thing ran 1400 mhz for almost 2 months (?) with cpu load temp of 44c @1.825v before staring the slow climb to 51c-WITH NO CHANGES TO THE SYSTEM!
My first thought was that the AS@ was "leaking" out, but I've redone that...
I am starting to think that this cpu is getting ready to let the "magic smoke" out!
I'm running at stock speed now, voltage at 1.625 (bios) 1.73 MBM5.
Possible power supply glitch? Its a fairly new 350w no name unit.
As stated above, make sure that you've got good airflow in and around the case. If you have your case inside the compartment of your desk, I would suggest that you make sure that you have an opening in the rear, at or near the top of the compartment that your tower is in. If you don't provide an exit for the hot air, it will be recycled within this compartment and be drawn back into the case as "preheated" air and limiting the effectiveness of your hsf.
Another thing that you may want to look for is obstructions at the case fan intakes. Depending on the local environment, you may have an accumulation of dust and lint restricting the airflow into the case.
And lastly, check the surface of your heat sink. It doesn't seem to take very long for a film of dust to accumulate on the fins. This "film" of dust will act as an insulator thereby preventing the transfer of heat from the heat sink.
Good Luck!
Case runs cool, 26c or so, airflow around the case sucks, but that is why I have fans and ducts: room air from the front of the case is ducted to the cpu, and drawn out by the case exhausts. This is proven effective by the case temp NEVER having gotten above 28c in this pc!
HSF was cleaned yet again today, new AS2 was applied.
And all of this does not change the simple fact that this system was running EXACTLY as it is for a while with decent temps! Usually I expect a slow climb like this to be fixed by cleaning of the hsf, but this has had little or no effect each time the temp has increased!
All of these are things I have thought of (so far:D !) and did not work: case fans are added, second fan on the heat sink added, cool air duct added, etc! THESE are things I've had to do each time to get temps back down! All that's left (with air cooling) is to run a duct from outside the house!
Keep the basics coming though: I'd rather find out I missed something simple than find out that yes, in fact, I have screwed up this cpu or mobo and this is it's death rattle!!
ok, as a reference:
if temps on a cpu (especially when running at high volts) climb over a period of time and the hsf is NOT dirty, the chip may be dying: I swapped my 900 and 1000 and the 900@1119 runs cool-39/40c in the abit board but the 1000@1300 runs at 49c in the AOpen mobo.
It is the chip!
Time to start looking for a new one for when this chip craps out!
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.