E6300 Still Running Hot

Bealer

Junior Member
Oct 30, 2006
15
0
0
#1
Ok, I'm stuck my C2D is still running hotter than I'd like. Here's the progression since I got my E6300.

Firstly, I started off with the Arctic Cooler Freezer 7 Pro, using Arctic Silver 5. I mounted and remounted it a dozen times, and tried using different amounts of AS5, but the best I could get was around 44-45C idle and 58C load (this is without it overclocked and the fan at 1500rpm).

I then decided to lap the IHS, as upon checking the how flat/level it is, I could see light coming through. There was a slight lip all the way around the edge of the IHS. Anyway, I set to lapping it, and got it nice and flat. This only dropped the temps by 1-2C (42C idle, 57C load).

So, I've now bought a Scythe Ninja revision B. It's currently mounted in (just!) with the 12cm fan spinning at 1200rpm, and again using Arctic Silver 5. I now get idle temps of around 40C and load temps of 55-56C. I still find this far too high.

I had been toying around with overclocking it. With an FSB of 400Mhz (running it at 2.8Ghz) it idles at 47-48C and under load goes to 64C.

All my temps are read using CoreTemp and Intel TAT.

Anyone got any ideas?

I'm gonna re-lap it slightly when I get a chance, just make sure it's perfect.
And also remount it again (it's bloody tricky trying to mount the Ninja).

Apart from that I don't know what to do. My case temps are fine and there's plenty of ventilation. My old Athlon64 3500+ used to be in there and was idling in the low 30's. Could I just have a hot Core 2 Duo?
 

AstroGuardian

Senior member
May 8, 2006
844
0
0
#3
yes, i think the same. Eider you are using you AS5 wrong or the senor is faulty. I also encountered a case where the PSU was partially bad and heated the components more than normal
 

gerwen

Senior member
Nov 24, 2006
312
0
0
#4
How is the airflow in your case?

If your case isn't getting rid of the hot air from the HSF properly, nothing you do will drop your temps.

If you open up the case and point a household fan inside, do the temps drop much? More than a few degrees means you could really improve your case's cooling.

My case has good airflow through it with two fans, but just cleaning up the wires inside so that they didn't impede airflow got me a 3-4 degree temp drop.
 

vassalle

Junior Member
Jan 2, 2007
3
0
0
#6
Mine is idling around 46c. Overclocked to 3.0ghz @ 1.25v. But im staying in malaysia where its 30c all year lonngg..
The temp should be fine I guess.
 
May 16, 2002
13,360
0
81
#7
That is about normal for a C2D cip - mid-50's is normal for them. People over at XS have those temps all the time. 60's seems a bit high but still I've seen those reported before as well (also it seems that intel TAT will make the CPU run warmer than anything else) - I'd check your HSF base just to make sure it is flat as well.

I can see why you'd be expecting lower temps since you came from an AMD but the intel chips do run warmer than AMD's and it is also dual core vs your old single core AMD.
 

n7

Elite Member
Jan 4, 2004
21,306
0
81
#8
You're worried about nothing.

C2Ds use a different temp measurement system, so it's not actually at all as hot as your previous A64 or Pentium 4/D, it just reports the real temps, unlike the older CPUs.

I swear we need a sticky explaining the fact that measurement of temps is NOT THE SAME ON C2D CPUs as previous CPUs, so yes, temps will report "higher".
 

lopri

Elite Member
Jul 27, 2002
12,651
50
106
#9
Originally posted by: gerwen
How is the airflow in your case?

If your case isn't getting rid of the hot air from the HSF properly, nothing you do will drop your temps.

If you open up the case and point a household fan inside, do the temps drop much? More than a few degrees means you could really improve your case's cooling.

My case has good airflow through it with two fans, but just cleaning up the wires inside so that they didn't impede airflow got me a 3-4 degree temp drop.
 

lopri

Elite Member
Jul 27, 2002
12,651
50
106
#10
OP: This might just well be a silly question, but what is your case? I've seen suprisingly many people with Lian-Li case (those with rear-side intake fan) not noticing the conflicting air-flow.
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#11
Thanks for the responses so far. To answer a few of the questions.

I'm applying the Artic Silver 5 as per the instructions on their site. I've tried varying amounts (no more than a grain of rice though). I've tried smaller amounts too as I've lapped my heatsink so it effectively shouldn't need so much.

The airflow in my case is good. It's a Lian Li PC60. Two fans blowing in at the bottom, one blowing out at the back. The air coming out of the back is never that hot. Neither is my Ninja actually.

Can the temp sensor be faulty? It's reading from the actual cores, not a motherboard sensor.


Now, voltages. I've recently upgraded to the F9 BIOS (for the DS3) which for me isn't very good. I'm going to drop back to F7 when I get a chance.

Anyway, all the voltage settings (cheers QuietRiot) are set to Auto apart from the DIMM voltage which I've added +0.1v. I seem to remember people mentioning about Auto settings with the DS3 before. Anyway, I've set my vcore at 1.25v, and temps have dropped loads, plus it's stable.

I now get 42C idle, and 54C load. Tempted to try and drop the vcore slightly more, is that ok? I guess it'll eventually not be stable as there won't be enough power to the cpu.

I'll need to look at the advanced settings too, and maybe set them to the default values myself, rather than leaving them on Auto.
 

tcG

Golden Member
Jul 31, 2006
1,199
0
76
#12
Originally posted by: Bealer


The airflow in my case is good. It's a Lian Li PC60. Two fans blowing in at the bottom, one blowing out at the back.
This is the problem. My load Opteron temperatures decreased 21C when I moved my parts from that garbage case (PC60) to an open testbed. I'll never buy Lian-Li again until they stop putting 80mm fans in their otherwise awesome cases.

Open testbed all the way. :)
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#13
Even with the side panel off the temperatures don't improve much. The air coming out of the back is barely warm. I can't see it being the case. It's been fine with every other build I've had, even a Prescott.
 

tcG

Golden Member
Jul 31, 2006
1,199
0
76
#14
Originally posted by: Bealer
Even with the side panel off the temperatures don't improve much. The air coming out of the back is barely warm. I can't see it being the case. It's been fine with every other build I've had, even a Prescott.
Oddly enough, I had the same issue. Side panel removal decreased my temperatures by exactly 3C.

Trust me, I couldn't see it being the case either, that is, until I removed the motherboard tray and ran open testbed (with a fan blowing directly onto the motherboard).
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#15
An open testbed isn't exactly practical though. I need it in a case. What's frustrating is that my ambient temp is ok, and all my other parts (hard drives etc... are running cool). It's just the C2D with a Ninja on it (with 120mm blowing across it) that isn't.

The only things I think it could be is either an inaccurate reading, or the IHS isn't transferring the heat as efficiently as other C2D's out there. Maybe I got one from a poor batch.
 

tcG

Golden Member
Jul 31, 2006
1,199
0
76
#16
Lay a razor blade across the top of the IHS to test for levelness.

I've found that inaccurate readings are usually very rare... I would doubt that.

I'm still sure your problem is a combination of poor airflow and [probable] high ambient temperatures... out of curiousity, what ARE your ambient temperatures; those can have a huge effect on temperatures.
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#17
I used a straight edge for checking the IHS when lapping it.

Ambient is around 28C. Everything inside is running nice and cool. Hard drives at 34C (No fan over them). The Ninja never gets very hot at all, nor the air being pumped out at the back.

Strangely I've been playing around with the vcore voltage, and currently got it at 1.2250v instead of the stock 1.3250v and it seems fine. This is overclocked with a 400Mhz FSB (2.8Ghz).

I may try checking the contact again, maybe take the chip out re-lap it a bit, check it makes decent contact with the bottom of the heatsink. I may just have a hot C2D. I have read some others with quite high temps.

I can understand what you mean about airflow. I'll maybe look at other cases. It's just I've seen loads of people with far worse setups regarding airflow, but temps maybe 10C less than mine.
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#18
Ok, a few more developments.

I've placed a 92mm fan over the north/south bridge area as both were getting very hot, to the point where I could burn myself. That's made quite a difference. I've drilled some holes in the side of the case so the fan can pull in air properly. I'm thinking of just cutting a huge circle out and placing a grill across it. That way the air would flow more freely.

Doing the above has helped my temps hugely.

Anyway, is this normal. I've been dropping my VCore down and down and it's currently at 1.175v (the stock is v1.325). This is overclocked, currently at an FSB of 425Mhz (roughly 3Ghz). I may keep dropping it so how low I can go.

Temps are absolutely fine now. 37C idle, 50C load.

Edit - meant 1.175v not 1.75v.
 

Bealer

Junior Member
Oct 30, 2006
15
0
0
#19
Got my Vcore down to just over 1.15v and it runs stable overclocked to 2.8Ghz. I've got the 12cm fan on the Ninja at about 600rpm, and both case fans at 5v, and idle temps are 40C, load 52C.

If I take the side of the case off, and up the fans to 7v, it all drops another 2-3C. So all in in all happy now. Looks like it was just the Vcore that needed dropping.

I did some research into undervolting, and the E6300 should be able to run with a Vcore of roughly between 0.8v and 1.35v (on Intel's website). My motherboard obviously sets it to the highest part of that range, so dropping the Vcore to v1.15 has got it running nice and cool. The fan over the north/southbridge also helped instead of cooling them passively.
 

gerwen

Senior member
Nov 24, 2006
312
0
0
#20
Wow, that's low voltage. I thought i was doing well when i stopped at 1.275V. I'll have to try and push that down further, now that i see that there may be more room.
 


ASK THE COMMUNITY

TRENDING THREADS