CPU Temperature Poll

Message boards : Number crunching : CPU Temperature Poll
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · Next

AuthorMessage
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 912263 - Posted: 28 Jun 2009, 9:39:04 UTC

I will also pop in here to mention that you only need enough thermal compound to fill in the [nearly] microscopic texture on the CPU's heat spreader and the base of the heatsink.

Whether you're using Arctic Silver 5 or not, the guides they have are really good, although I deviate from them a bit. For the CPUs that have the heat spreader, they say to just put a BB-sized glob in the center, and apply the heatsink, do some gentle twisting, and then strap it down.

I follow the old rules of the K7 exposed core method of using a razor blade and spreading that BB-sized glob across the entire surface of the heat spreader. I just swapped out a 95w dual-core with a 119w dual (Opteron 2210->2222 SE), and spent a good 5 minutes making the tiny bit of AS5 cover the entire heat spreader. I didn't think I could actually cover the whole thing with the tiny bit I used, but I finally got it, seated the heatsink, and began the stress testing.

The heatsink I was using was only intended for the early single-core K8's (so I would say less than 90w), and after an hour of Mass Effect, FEAR2, and DiRT, Everest reported no more than 50C at any point. The 2210 would get up to 50 unpacking a large compressed archive, but I may have had too much AS5 on that one.

The point is, more is not always better. For thermal compound, too much is just as bad as not enough.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 912263 · Report as offensive
Ianab
Volunteer tester

Send message
Joined: 11 Jun 08
Posts: 732
Credit: 20,635,586
RAC: 5
New Zealand
Message 912266 - Posted: 28 Jun 2009, 9:50:50 UTC - in response to Message 912253.  

Probably the best is to install a couple of exta case fans, especially a side vent one blowing fresh cool air down on the CPU.

That can actually reduce the efficiency of the CPU cooler. The CPU fan isn't able to blow as much air over the heatsink if it's air supply is turbulent (ie moving).
Having a duct in the side of the case though does make a significant difference. Make it so that the duct is centred over the CPU fan & the edge of the duct goes down to at least the top of the fan (even better if it can be lower, but that makes it very difficult to fit the side panel back on). That way the CPU fan is getting air at the ambient temperature and not at the internal case temperature (which can be 5°c or more above ambient).


Possibly true.

Does the positive pressure of a fan blowing into that duct make up for the turbulance on the CPU fan?
Does the turbulent air cool better than a laminer flow?

I dont know, but generally more air is better.

Some Hi-tech server designs etc may in fact cool better with all their covers and ducts in place. But they are relying on the ducted air to cool all the various components. Disrupt that and the disks, voltage regulators, Northbridge etc may get hot. CPU is fine. but the machince still crashes.

Or the CPU is actually cooled by a passive heatsink with air supplied by the case fans.

Ian
ID: 912266 · Report as offensive
Ianab
Volunteer tester

Send message
Joined: 11 Jun 08
Posts: 732
Credit: 20,635,586
RAC: 5
New Zealand
Message 912268 - Posted: 28 Jun 2009, 9:52:29 UTC - in response to Message 912256.  

had my Q6600 go into thermal slowdown due to dustbunnies in the heatsink after about 8 months. Standard heatsink and clock speed. It still ran OK, just clock speed was 1.4 instead of 2.4 so RAC dropped like a stone.

Right now PC Wizard says cores are around 72C which is well within spec.

Ian

Really? I thought the specs called for it to be lower. Mine stays below 60 most of the time, except on the really hot days here. Guess they're doing better than I thought.


Reading the spec sheet again it does seem a bit warm

Case cooling is good, maybe I need to check the dustbunnies again :-)

Ian
ID: 912268 · Report as offensive
Profile Virtual Boss*
Volunteer tester
Avatar

Send message
Joined: 4 May 08
Posts: 417
Credit: 6,440,287
RAC: 0
Australia
Message 912269 - Posted: 28 Jun 2009, 9:56:37 UTC - in response to Message 912260.  

Hi, except for the HP LT (60-65C) the Pentium 4 DUAL820 runs hot 85-87C, but my Q6600 (@2,8GHz), even now, runs too hot 95-105C! Then it throttles back.(108C)
The other 2 QUAD's run @ 65-75C.

Even for the Lowlands it's hot, again, since a week.

Don't know how trustworthy these readings are. I'de like to use a 'thermo-couple'. But they aren't easy to get and should fit between CPU and cooler.


Ouch that's HOT.

My Q6600 at stock was running 87-88C under load till I undervolted it to drop temp back to 77-78C on a hot day (ambient up to 40C).
Flying high with Team Sicituradastra.
ID: 912269 · Report as offensive
Iona
Avatar

Send message
Joined: 12 Jul 07
Posts: 790
Credit: 22,438,118
RAC: 0
United Kingdom
Message 912273 - Posted: 28 Jun 2009, 10:26:39 UTC

I've got two PCs running here...Intel Core 2 Duo CPUs on MSI P35 Platinum boards - E6600 and E6550 CPUs. During the last few days, the ambient temp in the room where the PCs are, has been about 30C - 34C (its upstairs...warm air rises). The E6550 (nild O/C to 2.4 GHz) has been running at a temp of 48C with a system temp of 36C, while the E6600 has been running at 52C with a system temp of 37C. The E6550 has a Scythe Ninja 2 cooler and the E6600 has an OCZ Vindicator.
Don't take life too seriously, as you'll never come out of it alive!
ID: 912273 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 912278 - Posted: 28 Jun 2009, 10:56:48 UTC - in response to Message 912273.  

Hi, in the meantime I spotted some dustbuns (again) so I'll have to ckean it out and refit the COOLER, add some Arctic Silver and reseat it again.

It'll be safer and I can UP my OC on this one ;)

ID: 912278 · Report as offensive
Profile Rick B

Send message
Joined: 6 Mar 01
Posts: 299
Credit: 1,532,791
RAC: 0
Canada
Message 912279 - Posted: 28 Jun 2009, 11:07:02 UTC - in response to Message 912273.  

Morning All

Thanks for the input. Yes I definately put to much goop on when reset the heat sink so that will be one of my first things to do ... after coffee. I did however not use the stuff when cleaning out the dust bunnies a while back and was careful not to disturb what was there. So now the question is; Do I scrape off the old (and the newly applied stuff) right down to the bare metal first?
Rick
**************************
ID: 912279 · Report as offensive
Profile gizbar
Avatar

Send message
Joined: 7 Jan 01
Posts: 586
Credit: 21,087,774
RAC: 0
United Kingdom
Message 912283 - Posted: 28 Jun 2009, 11:19:01 UTC
Last modified: 28 Jun 2009, 11:27:21 UTC

Hi guys,

Still following this one with interest. Bear in mind that Arctic Silver 5 does need to thermal cycle a few times to cure it. Basically this means after fitting, running the machine up to temp, then turning off to cool completely. (Check AS5 instruction for full details, as I haven't actually used it myself). Repeat a few times and your temps should drop another couple of degrees when done.

Update - now 27c upstairs with my main rig, at 46c with all 4 cores running seti, and a 9800GTX+ doing cuda. I don't turn this one off, and it only gets reset to install updates, etc.

Edit - sorry if I'm teaching anybody to suck eggs here. Just information for those who don't know.

@Rick - Yes you do need to clean all of the old stuff off first. There are several ways of doing it. You can get a proper TIM (thermal interface) cleaner, some people use isopropyl alchohol, some use Acetone (used to be nail-varnish remover).

regards, Gizbar.


A proud GPU User Server Donor!
ID: 912283 · Report as offensive
Profile Rick B

Send message
Joined: 6 Mar 01
Posts: 299
Credit: 1,532,791
RAC: 0
Canada
Message 912299 - Posted: 28 Jun 2009, 12:56:01 UTC - in response to Message 912283.  
Last modified: 28 Jun 2009, 12:56:33 UTC

Success

I removed the outer cover prior to shutting down the computer as I wanted to watch the cpu fan stop to ensure that the factory installed fan was indeed blowing down on the cpu as I had assumed. I assumed wrong! It was indeed blowing up and away from the heat sink. The only other time I removed the heat sink for cleaning I did not remove the fan and assumed it was correct.

The next part of this goes to the heat sink compound. The original factory installed material was hardened on when I removed it the first time and I did not add any new material. Yesterday I got some and applied liberal amounts of it on top of the factory material. I now know that was wrong but the upside is that it softened the original material and all of it came off quite easily with a dry cloth.

After re-assembling and restarting I now have Temps at 51-53 and quiet fans again. Ambient temps are at 21. Its not quite back to the 47-49 range when I got back into Seti with this computer but its a far cry from the 61-62 temps that it did get up to.

Moral of the Story . . . Don't ASS-U-ME anything!

Thanks to all of you for your help on this one.

@ Grant and Byron ... That cooler reference was definately overkill and a little bit of my humour but If I had a free one I would definately tried to attempt it. Im not afraid of the mechanical installation parts of a computer its making it all work together (software) that I have a hard time with.
Rick
**************************
ID: 912299 · Report as offensive
Profile Rick B

Send message
Joined: 6 Mar 01
Posts: 299
Credit: 1,532,791
RAC: 0
Canada
Message 912309 - Posted: 28 Jun 2009, 13:31:00 UTC - in response to Message 912299.  
Last modified: 28 Jun 2009, 14:20:00 UTC

OK. Maybe not a complete success. (There I go assuming again) First 30 minutes back on temps at 51-53. Then a jump to 54-56 in the next 30 minutes. Its the same work unit as its got 1:30:00 done and 00:30:00 to go. Will continue to monitor.

Edit: That wu completed and temps dropped to 50-51 range on the start of the next wu. All seems good now.

Thanks again to everyone here for your help.
Rick
**************************
ID: 912309 · Report as offensive
__W__
Avatar

Send message
Joined: 28 Mar 09
Posts: 116
Credit: 5,943,642
RAC: 0
Germany
Message 912322 - Posted: 28 Jun 2009, 14:26:36 UTC

Hello from Germany,
if you clean up your cpu don't use Acetone! It's very agressive against many kinds of plastic. Trie an alchohol, isopropyl work's fine and ethyl alcohol(methylated spirit) will do it too.
And if you can't get it, use high percentage vodka - helps for everything ;-).

Greetings WOODY
ID: 912322 · Report as offensive
Iona
Avatar

Send message
Joined: 12 Jul 07
Posts: 790
Credit: 22,438,118
RAC: 0
United Kingdom
Message 912370 - Posted: 28 Jun 2009, 17:07:52 UTC - in response to Message 912322.  

Agreed. However, I think I'd use a good Vodka, in a quite different way.



Don't take life too seriously, as you'll never come out of it alive!
ID: 912370 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 912381 - Posted: 28 Jun 2009, 17:42:55 UTC - in response to Message 912206.  


Its the Pentium Dual Core E2180 @ 2.0 Ghz. The processors data sheet says the Max Tc at 65W is 61.4 Deg C. ... Granted I'm not really sure what that means other than the fact that when the cpu hit the low 60's it froze up.

Hmm, the spec I see says Thermal Specification: 73.2°C. Early E2140 and E2160 CPUs had stepping L2 with 61.4°C but AFAICT the E2180 only has M0 stepping like later E2140 and E2160.

I'll also note that the E2180 could probably be undervolted, the default voltage settings are usually considerably higher than most samples really need. It looks like you have the cooling improved enough, but you might be able to reduce the power used and heat produced by 10 or 20 percent.
                                                               Joe
ID: 912381 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 912389 - Posted: 28 Jun 2009, 18:04:22 UTC - in response to Message 912266.  
Last modified: 28 Jun 2009, 18:05:13 UTC

Probably the best is to install a couple of exta case fans, especially a side vent one blowing fresh cool air down on the CPU.

That can actually reduce the efficiency of the CPU cooler. The CPU fan isn't able to blow as much air over the heatsink if it's air supply is turbulent (ie moving).
Having a duct in the side of the case though does make a significant difference. Make it so that the duct is centred over the CPU fan & the edge of the duct goes down to at least the top of the fan (even better if it can be lower, but that makes it very difficult to fit the side panel back on). That way the CPU fan is getting air at the ambient temperature and not at the internal case temperature (which can be 5°c or more above ambient).


Possibly true.

Does the positive pressure of a fan blowing into that duct make up for the turbulance on the CPU fan?

Not usually.
One option is to put a large fan in the case chimney & remove the CPU fan. But to do so requires the chimney to be a very good fit over the heatsink.

Does the turbulent air cool better than a laminer flow?

Not sure.
In this case the problem is with air supply to the fan. A fan drawing air from a turbulent source is unable to draw as much as from still air, so temperatures rise.
I found this on one of my systems. Removed side case & CPU temp dropped 5°c. Put a large fan on the side directly over the CPU, temp rose 5°c. Moved the fan so it wasn't blowing dirtectly onto the CPU fan but moving the air around & out of the case, temps dropped by 7°c.

Some Hi-tech server designs etc may in fact cool better with all their covers and ducts in place. But they are relying on the ducted air to cool all the various components.

Yep- they've been carefully designed to make full use of the air flow to cool the system as best as possible in such a small space. BTX cases were designed to improve cooling through better airflow in PC systems, but they never took off.
Grant
Darwin NT
ID: 912389 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 912392 - Posted: 28 Jun 2009, 18:05:48 UTC - in response to Message 912309.  

OK. Maybe not a complete success. (There I go assuming again) First 30 minutes back on temps at 51-53. Then a jump to 54-56 in the next 30 minutes. Its the same work unit as its got 1:30:00 done and 00:30:00 to go. Will continue to monitor.

Edit: That wu completed and temps dropped to 50-51 range on the start of the next wu. All seems good now.

Thanks again to everyone here for your help.

Glad you got it sorted.
Grant
Darwin NT
ID: 912392 · Report as offensive
Profile Byron S Goodgame
Volunteer tester
Avatar

Send message
Joined: 16 Jan 06
Posts: 1145
Credit: 3,936,993
RAC: 0
United States
Message 912401 - Posted: 28 Jun 2009, 18:22:55 UTC - in response to Message 912309.  
Last modified: 28 Jun 2009, 18:34:02 UTC

Glad to here it's doing better for you. Don't forget the break in period as mentioned by gizbar, which should also bring it down a bit in time.

This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.

ID: 912401 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 912408 - Posted: 28 Jun 2009, 18:32:15 UTC

I was doing a bit of gaming the yesterday and noticed that my game was getting all glitchy. I assumed it was my connection then remembered once a long time ago when my PC did the same thing.

I've run the Machine without a cover since then. However I am running a much bigger rig now. So I've put a small room fan blowing into the middle of the case. My temps have come down about 15-20 C and everything is much happier


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 912408 · Report as offensive
Profile TCP JESUS
Avatar

Send message
Joined: 19 Jan 03
Posts: 205
Credit: 1,248,845
RAC: 0
Canada
Message 912428 - Posted: 28 Jun 2009, 19:29:19 UTC
Last modified: 28 Jun 2009, 19:31:30 UTC

Specs:
-------
Xeon E5520 Core I7 on Thermaltake Ultra 120 Xtreme 1366RT using 2 120mm Fans (push/pull).

Since Summer is officially here and my computer room is HOT HOT HOT, this is the end result for me during crunching today:
.

.
I am going to Water Cooling this week, as I really don't want to subject my CPU to these kind of Temps 24/7. I would like to be well under 70c if possible, closer to 62-65c. With that temp range, 24/7 crunching will not bother me as much.

I have noticed that since (yesterday) switching to running optimized Apps (Intel SSE4.1) my Temps have gone up a few Deg C from before on standard BOINC installation.

Yesterday, I maxed out at about 74-75c. Today (as you can see) I am able to hit 78c peak during crunching for short periods of time.

Water Cooling - Here I come ...
I am TCP JESUS...The Carpenter Phenom Jesus....and HAMMERING is what I do best!
formerly known as...MC Hammer.
ID: 912428 · Report as offensive
Jörg

Send message
Joined: 10 Dec 02
Posts: 51
Credit: 1,547,286
RAC: 0
Germany
Message 912458 - Posted: 28 Jun 2009, 20:46:24 UTC - in response to Message 912428.  
Last modified: 28 Jun 2009, 20:46:59 UTC

Good evening,

I run a Q6600@3GHZ and use opt apps for Seti.

CPU cooler is a "Groß Glockner" in an Antec 9-Hundred tower (3 x 120mm and 1x200 cooler).


Am Ende ist nur Verwirrung
ID: 912458 · Report as offensive
Profile Rick B

Send message
Joined: 6 Mar 01
Posts: 299
Credit: 1,532,791
RAC: 0
Canada
Message 912491 - Posted: 28 Jun 2009, 23:06:56 UTC - in response to Message 912381.  


Its the Pentium Dual Core E2180 @ 2.0 Ghz. The processors data sheet says the Max Tc at 65W is 61.4 Deg C. ... Granted I'm not really sure what that means other than the fact that when the cpu hit the low 60's it froze up.

Hmm, the spec I see says Thermal Specification: 73.2°C. Early E2140 and E2160 CPUs had stepping L2 with 61.4°C but AFAICT the E2180 only has M0 stepping like later E2140 and E2160.

I'll also note that the E2180 could probably be undervolted, the default voltage settings are usually considerably higher than most samples really need. It looks like you have the cooling improved enough, but you might be able to reduce the power used and heat produced by 10 or 20 percent.
                                                               Joe


Joe you are probably right. I was looking at The Data Sheet on Page 75 Table 27.

@ Byron. I will do that break in. Looks like we have thunderstorms coming through tonight so it'll be a good time to shut down for a while anyway.
Rick
**************************
ID: 912491 · Report as offensive
Previous · 1 · 2 · 3 · Next

Message boards : Number crunching : CPU Temperature Poll


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.