15" MacBook Pro (SB) Thermal application/guide

[Nikonian]

I've since replaced the thermal paste as well with a small amount of Arctic Silver. Long story short, you will NOT see a decrease in temperates at their peak.

In windows, I was reaching 100C, and I still do.
However, my idle and general operating temperatures are noticeably lower. 40 to 60 when not streaming.

 

[legreve]

Dire straits...

Man... that has to be the scariest thing I've done in a long time :S

There's several things I'd like to make people aware off.

First of all, I haven't been on the inside of a computer for 6-7 years... so opening this was indeed like a first time again.

Apart from all the small cables and jumpers the OP highlights (and which you really really really need to treat with silk gloves) there's the fact that you have to take out the battery cable as the very first thing.
I was slobby... eager, and didn't. This resulted in a small spark when my screwdriver touched the logic board just next to that miniature cable (which I'm not yet sure wasn't a short :´( ), I quickly removed the battery cable after that.

The cable in the right hand corner with the metal clamp... be careful with that as well, both when removing it and reinserting it. My butter fingers put slightly too much pressure on the cable when reinserting, resulting in the clamp jump of the cable :S It's a nasty one to get back into place....

The thermal compound was very similar to what others have posted. Too much on the side and uneven on the chip.

I recommend preparing well... it took me several (10-15) digi cloth to remove the compound properly.

To reapply I took notice on people mentioning that you don't need too much to make it cool, so I did a thin worm on both chips and spread it out with a plastic straw that I had surfaced cleaned before use. It seemed to make it nice and even.

But well, let's see if I still have a computer that works after that small spark Right now it's installing OSX on the new harddisk I installed at the same time. Will report back later with temperatures and what not...

Cross your fingers that I didn't f up too much

Update: Finished installing everything basic. Here's how it looks in the temp department while on one Safari session.
Edit: Updated the screenshot to todays idle temps... a bit better still.

 

[kappaknight]

^its likely fine, but that application is heavy. Back when i used to overclock PCs a lot i would apply the paste and then scrape it off with a credit card. you barely need any on there. remember, you're filling essentially microscopic pockets/gaps.

No argument there. But I did use the recommended rice sized grain to start with.

 

[Andrmgic]

Just had this done by an apple authorized service center..

temps dropped from 47-49 idle in windows to 42ish..

gaming load temps (2 minutes of dragon age origins) went from 89-90C to 75C..

Cools off much more quickly, too.

 

[3lackwater]

Wirelessly posted (Mozilla/5.0 (iPod; U; CPU iPhone OS 3_1_3 like Mac OS X; es-es) AppleWebKit/528.18 (KHTML, like Gecko) Version/4.0 Mobile/7E18 Safari/528.16)

If anyone replaced their TTC on a 13" I'm curious to know which guide did they used. Is the 2010 teardown still valid or should I pay more attention to Ifixit's 2011 guide for the 15"?

Thanks.

 

[xxfury2xx]

I was hoping to get some advice from some of you here who are more knowledgeable than I am when it comes to CPUs, heatsinks and cooling.

Right now when my macbook is idle the difference between my CPU and the heatsink is minimal, but when I put it under full load the gap between the two is at least 30* celsius.

In your opinions, did I mess up the TP job? Should I go back in and re-do it all?

I've attached some screenshots from istat pro (one at idle and another under full load) in hopes that some of you can gleam some more information from it.

Any help would be appreciated.

 

[fr4c]

I was hoping to get some advice from some of you here who are more knowledgeable than I am when it comes to CPUs, heatsinks and cooling.

Right now when my macbook is idle the difference between my CPU and the heatsink is minimal, but when I put it under full load the gap between the two is at least 30* celsius.

In your opinions, did I mess up the TP job? Should I go back in and re-do it all?

I've attached some screenshots from istat pro (one at idle and another under full load) in hopes that some of you can gleam some more information from it.

Any help would be appreciated.

Sorry read your post wrong, thought you were comparing CPU/GPU

 

[stevejobswallet]

Also, wanted to contribute some pictures to this thread:

The tools I used to clean up the thermal paste. These worked well and polished my CPU to a mirror like state.

The display ribbon was tricky.

http://farm6.static.flickr.com/5149/5570601072_b5b7960699_b.jpg

Old paste wasn't too bad and I didn't run into huge temperature issues.

http://farm6.static.flickr.com/5184/5570594248_2fa630ca42.jpg

http://farm6.static.flickr.com/5137/5570006757_d9984ae910.jpg

Cleaned.

http://farm6.static.flickr.com/5267/5570594546_d99361d978_z.jpg

Reapplied (with a toothpick)

http://farm6.static.flickr.com/5025/5570007017_df7572082c_b.jpg

Thank you for posting pics of your tools! I just ordered a BTO MBP 15" HR AG. I stumbled across this thread & am both mortified by the heat issues & intrigued/impressed by everyone's skill!! Maybe mortified is a strong word but you get my drift.

 

[xxfury2xx]

Were you using the dedicated GPU or integrated Intel during the full load testing? That might explain the difference in temps.

I used photobooth along with the yes>/dev/null & command in the terminal so that it would both use the CPU and the GPU.

Also, I'm actually concerned about the difference between the CPU temperature and the CPU heatsink temperature (which can be seen in the screenshot I attached), not the difference between the CPU and the GPU.

 

[fr4c]

I used photobooth along with the yes>/dev/null & command in the terminal so that it would both use the CPU and the GPU.

Also, I'm actually concerned about the difference between the CPU temperature and the CPU heatsink temperature (which can be seen in the screenshot I attached), not the difference between the CPU and the GPU.

Edit, please see my response below. Read the post wrong.

 

[brentsg]

How tricky is it to secure the heatsink back to the logic board? Is it possible to screw it all the way down, or is some touchy-feely involved as to not over tighten?

Obviously alternate corners, etc.. just wondering how much guesswork is involved.

I've got a 13" that I might tear down.

 

[73CortinaV8]

another nooby observ here... what are those pill-shaped things on the logic board with a "+" sign on them?

i counted 24 of those on the 2011 mbp logic board, compared to only 14 on the mid-2010 mbp.

while i am at it, i am also curious about those grey-colored cubes right next to the cpu. user tilezzz has an "SCR" sticker on one of them. i counted 4 of those right next to the cpu on the 2011 mbp, compared to just 2 on the mid-2010.

i am just fascinated by these photos, btw.

any engineer care to answer?

these?

they're caps

 

[fr4c]

How tricky is it to secure the heatsink back to the logic board? Is it possible to screw it all the way down, or is some touchy-feely involved as to not over tighten?

Obviously alternate corners, etc.. just wondering how much guesswork is involved.

I've got a 13" that I might tear down.

Definitely do not over-tighten the screws, because I think you can strip right through the logic board. Not much guesswork involved, just go slowly and don't over-torque those screws.

 

[brentsg]

Definitely do not over-tighten the screws, because I think you can strip right through the logic board. Not much guesswork involved, just go slowly and don't over-torque those screws.

So you're tightening all the way, just don't want to damage the female side then.. Cool.

 

[xxfury2xx]

Would some of you others who've done the custom thermal paste job mind posting a screenshot of istat pro with your macbook under full load?

I just want to see what it is I should be looking for when I re-do this again next weekend, as obviously I've done something wrong with the application of the thermal paste the 1st time around.

Thanks.

 

[fr4c]

Would some of you others who've done the custom thermal paste job mind posting a screenshot of istat pro with your macbook under full load?

I just want to see what it is I should be looking for when I re-do this again next weekend, as obviously I've done something wrong with the application of the thermal paste the 1st time around.

Thanks.

I will take a few screenshots later tonight when I get home, hold tight.

 

[altecXP]

Maybe Apple fixed this?

I have Adium, Mail, and Safari with 4 tabs open, and I'm sitting @ 35-37* according to iStats widget.

My room temp is about 62*F

 

[lurrego]

^ mine is 40* with the same setup only with firefox 4, so that seems like a good sign

 

[altecXP]

Were you using the dedicated GPU or integrated Intel during the full load testing? That might explain the difference in temps.

The CPU will heat up faster than the heat-sink. Since the HS also dissipates heat more and depending where the sensor is located it will never read the same as the CPU.

 

[altecXP]

^ mine is 40* with the same setup only with firefox 4, so that seems like a good sign

When I first got it it idled @ 44 but it seems to have definitely dropped since now its almost 8* lower. I have noticed that Chrome also makes it idle in the 42-43* range. Not sure what its doing to make that happen, but it also makes the dGPU activate more, so I am sticking with Safari for better temps and battery.

 

[Andrmgic]

My difference is about 30C under full load.. I don't really want to pay the mac shop to open my machine up again.

It's a definite improvement over the stock situation, though.

I'll see what temps do after the AS 5 has cured.

 

[fr4c]

Would some of you others who've done the custom thermal paste job mind posting a screenshot of istat pro with your macbook under full load?

I just want to see what it is I should be looking for when I re-do this again next weekend, as obviously I've done something wrong with the application of the thermal paste the 1st time around.

Thanks.

Edited my previous post, but for some reason I thought you were comparing CPU and GPU temps. Please excuse my stupidity

But yes, I think the temps you're getting between the CPU and the CPU heatsink are normal. Under load/Handbrake, my CPU is running at 90C while my heatsink is 49C, fans at 6200rpm.

 

[anonymouse99]

Reading this thread makes me wonder if either the quad's are really ready for (mobile) prime time, or if this is an Apple design issue? I don't want my Macbook Pro turning into a cookie baking sheet/griddle when I use it.

I am completely lost as to why there are heat issues with quads? Or is this only like a miniscule fraction of the vast sea of users? Apple, what is the real story here? Silence is not golden on Apple's part here as it will eventually cause them more problems, especially if they don't rectify this .... we don't want another Toyota-gate here!

It would be nice to hear from folks that own dual core gen 2010 i7 machines. Did they heat up as much as these 2011 units? Just wondering .....

 

[xxfury2xx]

Edited my previous post, but for some reason I thought you were comparing CPU and GPU temps. Please excuse my stupidity

But yes, I think the temps you're getting between the CPU and the CPU heatsink are normal. Under load/Handbrake, my CPU is running at 90C while my heatsink is 49C, fans at 6200rpm.

That's a HUGE relief. Now I don't have to open up my macbook pro again.

 

[aznguyen316]

It would be nice to hear from folks that own dual core gen 2010 i7 machines. Did they heat up as much as these 2011 units? Just wondering .....

Yes, in briefly owning my 2010 i7 2.66Ghz I believe that got hotter than my 2011 2.3Ghz quad core. Using only SMC fan and just constantly monitoring temps, the 2010 hit 90C-93C often and took a minute to cool off. I've seen the 2011 hit 90C once and it cooled down to 86C and down very quick. Even running some benchmarks right now it hasn't gotten higher than 80C and when encoding and exporting in iMovie it's only about 88C, no higher. The fans are louder though in the 2011 but not greatly louder. I just feel like there's more of a higher pitch fan whirring.