Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Reapplied Thermal paste on 2010 13" MBP

Wow is that really how hot your computer gets?

I have the 15" i7 and idea around or below 90f, Flash player, Itunes, and Chrome running i run 120f. That seems really warm for a core2 but i guess its because of the size of the inclosure and only having a single fan. Thats pretty good though to see a 10 degree decrease.
 
vant said:
Do you have any accessaries like a palm guard or a keyboard protector?
Click to expand...

First of all, why does it matter? There are no vents in the unibodies under the keyboard or on the palmrest. Second of all, I have a Bestskinsever palm guard.

I just re-reapplied thermal paste, because I was not happy with the results. In my original reapplication, there was too much paste. On my new reapplication, there is a very thin layer of paste. Thinner than a sheet of paper. I am getting about 110-115ºF CPU temps and 105-108ºF heatsink temps (when surfing with chrome and 2 tabs). I will post screenshots of iStat Pro with 2x "yes >/dev/null" processes and when idle, later.
Here is a picture of my new reapplication. Sorry for the red tint. I accidentally covered the camera flash with my finger.
DSCN0803.jpg
 
I still find it interesting that the Core2 are running so hot in the 13" I just did 4x yes > /dev/null and can't break 181F. Its been running for about 10 mins now and the fans aren't on full blast, only 5,500.

Don't worry about what people are saying about your warranty, the whole process isn't very hard and there is no way of them telling unless you did something extremely stupid.
 
AlphaDogg said:
No, not on a bed. I live in Denver, CO. My ambient temperatures are always around 67-72ºF.
Click to expand...

Maybe a dumb question- are you certain iStat is set to Celsius? my 2.4 13" MBP idles at about 34º C /92ºF CPU in a room that's about 67ºF. Heatsink B reports about 5ºC less.
 
AlphaDogg said:
First of all, why does it matter? There are no vents in the unibodies under the keyboard or on the palmrest.
Click to expand...
The computer housing itself serves as a heat sink too. It seems unlikely that a palm rest cover would have a big effect on temps, but I'd be surprised if it didn't have any.
 
Reelknead1 said:
Don't worry about what people are saying about your warranty, the whole process isn't very hard and there is no way of them telling unless you did something extremely stupid.
Click to expand...
Thank you for reassuring me. The only difference between the original paste any my paste is the color. The original paste was grey. My paste is white.

chrfr said:
Maybe a dumb question- are you certain iStat is set to Celsius? my 2.4 13" MBP idles at about 34º C /92ºF CPU in a room that's about 67ºF. Heatsink B reports about 5ºC less.
Click to expand...
I have it set to fahrenheit, and am using a converter widget to convert it to celsius for you.

chrfr said:
The computer housing itself serves as a heat sink too. It seems unlikely that a palm rest cover would have a big effect on temps, but I'd be surprised if it didn't have any.
Click to expand...
You do realize that only the battery and hard drive are under the palmrests, right? These do not generate any heat (ok, the hard drive generates a very minute amount of heat). The temperatures that I am watching are the CPU temps and the Heatsink B temps.


Stay tuned for idle and 2x "yes >/dev/null" temps :)
 
devilstrider said:
How do you do this 2x mode thing? You know make it run at full load?
Click to expand...

Open up terminal (/Applications/Utilities/Terminal.app) and type in (or copy and paste) the following (without quotes): "yes > /dev/null". Then open another tab in terminal. This can be done with the keystroke: Command+T. Then type in the "yes > /dev/null" command again. This is for a dual core processor. For a quad core processor, you do this with 4 tabs open.

Now onto the iStat Pro screenshots (set to º Celsius) with the new reapplication of thermal paste (not the initial reapplication).

This is a screenshot of the iStat Pro window after 15min of idle.
newappidle.jpg


This is a screenshot of the iStat Pro window after 15min of 2x "yes > /dev/null" processes.
newappyesdevnull.jpg
 
AlphaDogg said:
You are not doing it correctly. You are not fully stressing the CPU. You probably have a quad core. Do you have an i5 or i7? Your temps would be higher if you did it correctly.
Click to expand...

I was thinking the same thing. I have the 15in i5 just got it on Oct 2nd.
 
chrfr said:
The i5/i7 MBPs are dual core, not quad.
Click to expand...

Physical cores yes, but with hyper-threading it sees it as 4, when running this terminal command if you only put "yes > /dev/null" twice you will only be running 50% of your processors. You must open 4 tabs and run all 4 at once to get a run the computer at all 400% of the CPU.

Do that and then let it run for 10 mins and report back.
 
Hi Alphadogg

Your new paste application looks much better than the previous, you have got 5 C temperature difference at idle which is what I get with "stock" paste, I note chrfr also gets the same value, I assume he has "stock" paste as well. I've not seen anyone post on here with a value much less than 5 C, so your paste application is probably optimum.

Re the full load stuff, thanks a lot. But it looks like you terminated the "yes" processes before taking the screenshot - the CPU was mostly idle - so it will have been cooling down. You need to take the screenshot while the 2 x "yes" is still running.

But if we're just looking at the paste quality you don't need to run for 10 mins. Just start the "yes" processes off, wait a minute or so and look at the temp difference between CPU and heatsink. The absolute temperatures should be rising as the CPU heats up, but the difference between CPU and heatsink should very quickly stabilise to a constant value. I bet you see low 20's C as the difference. Again that's consistent with lots of C2D MBPs with "stock" paste I've seen post on here. (i series MBPs tend to have a higher difference because they have a higher TDP, 35 W vs 25 W for the C2Ds. Also Intel has set a lower temperature limit for the i series because of the integrated graphics processor, so the max CPU temperature for an i can differ from that for a C2D in quite a complicated way depending on whether the CPU and/or IGP are working hard)

So well done - sounds like you've fixed your Mac. If you really had 62 C at idle it must have been a real Friday afternoon paste job at the factory. Perhaps the heatsink wasn't seated properly. Those slave kids, eh?

(At the moment my C2D 2.26 13" is idle and all seven of my temperatures are within 1 C of the iStat you posted after the 2nd paste installation!)
 
Pax said:
Hi AlphaDogg
Re the full load stuff, thanks a lot. But it looks like you terminated the "yes" processes before taking the screenshot - the CPU was mostly idle - so it will have been cooling down. You need to take the screenshot while the 2 x "yes" is still running.
Click to expand...

So how long do you want me to run the "yes > /dev/null" processes?

Oh, and now with one tab in Chrome open, I am at 104ºF CPU and 99ºF Heatsink.

Oh and I forgot to mention that the fans don't go above 2000rpm while watching flash videos.
 
AlphaDogg said:
So how long do you want me to run the "yes > /dev/null" processes?
Click to expand...
As long as you've got patience for!
My point is, the difference between CPU and heatsink will settle down very quickly, say 1 min. But the CPU won't reach its max value for 10-15 minutes because of the way the fans work.

So if you take an iStat screenie after say 1 minute we should be able to see the difference & check the paste.
 
Pax said:
As long as you've got patience for!
My point is, the difference between CPU and heatsink will settle down very quickly, say 1 min. But the CPU won't reach its max value for 10-15 minutes because of the way the fans work.

So if you take an iStat screenie after say 1 minute we should be able to see the difference & check the paste.
Click to expand...

celsius.jpg


This is after 1-2 min.
 
Looks good - you have about 30 C between heatsink and CPU which is in the same league as most of the "stock" MBPs I've seen post on here.

Well done for fixing it - it must have been really bad. And thanks for your help. It adds weight to my view that there is nothing wrong with the "stock" paste in most Macs. The stock paste is usually about as good as well applied Arctic. But as you've shown there may be a few "lemons" out there as well.

Of course old Macs from 2006 or whatever may be a different story - the paste may have been wrongly applied, or have dried out, or whatever.

cheers now
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back