Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
15" MacBook Pro (SB) Thermal application/guide
- Thread starter fr4c
- Start date
- Sort by reaction score
My MacBook Pro 15'' 2011 2,2 GHz never had any freeze issue even with very high loads. But I decide to use AS5 just to see if something changes.
Ambient temp is 20~21°C
Before (Intel HD 3000) :
After with AS5 (Intel HD 3000) :
Idle = iTunes playing music, web browsing
Full load = in terminal 8x yes > /dev/null &
All temps were measured after 10 min from /dev/null processes start.
When I start my Mac this morning, CPU temps stays at 29°C for about 30 minutes then it goes at 30°C for about 1 hour. When I start to use it, it goes from 32 to 36°C.
I aslo run Smallluxgfx withc is very heavy for 20 minutes : CPU * 8 + GPU
There are not big changes. But the original thermal compound looks fine for me. I never see more than 93°C with the Mac out of the box. Of course there were a lot of compound around the dies but between die and copper HSF, it was realy good. I only get a few °C less but I'am very glad the fans speed is down from 4600 rpm to 4300 rpm with 8 times yes > /dev/null &. So mine is just quieter with medium to heavy load. That's fine for me.
Edit : Today, ambiant temps is 23~24°C and I run Smallluxgpu again. Temps is 89 to 90°C for CPU and 76°C for GPU. I thinks it's nice according to +3°C ambiant. But what will happen in summer ?
Ambient temp is 20~21°C
Before (Intel HD 3000) :
- Idle : 36~40°C - fans @ 2000rpm
- Full load : 87C - fans @ 4800rpm (auto)
After with AS5 (Intel HD 3000) :
- Idle : 32~36°C - fans @ 2000rpm
- Full load : 85C - fans @ 4300rpm (auto)
Idle = iTunes playing music, web browsing
Full load = in terminal 8x yes > /dev/null &
All temps were measured after 10 min from /dev/null processes start.
When I start my Mac this morning, CPU temps stays at 29°C for about 30 minutes then it goes at 30°C for about 1 hour. When I start to use it, it goes from 32 to 36°C.
I aslo run Smallluxgfx withc is very heavy for 20 minutes : CPU * 8 + GPU
- Before : CPU 92°C - GPU : 78°C @ 6200 rpm
- After : CPU 89°C - GPU : 76°C @ 6200 rpm
There are not big changes. But the original thermal compound looks fine for me. I never see more than 93°C with the Mac out of the box. Of course there were a lot of compound around the dies but between die and copper HSF, it was realy good. I only get a few °C less but I'am very glad the fans speed is down from 4600 rpm to 4300 rpm with 8 times yes > /dev/null &. So mine is just quieter with medium to heavy load. That's fine for me.
Edit : Today, ambiant temps is 23~24°C and I run Smallluxgpu again. Temps is 89 to 90°C for CPU and 76°C for GPU. I thinks it's nice according to +3°C ambiant. But what will happen in summer ?
Last edited:
Just replaced my MBP's standard thermal compound with Antec Formula 7 and seeing about a 15 degree celcius drop in load temps. They had put way too much on and it was horribly thick and hard stuff.
All good now.
All good now.
hey, i'm about to do this.
what's a good app that i can run so it makes my computer run hot. i wanna do a comparison...
no handbrake (don't have dvd's to rip)
what's a good app that i can run so it makes my computer run hot. i wanna do a comparison...
no handbrake (don't have dvd's to rip)
Open like 4 terminal windows and run the following command in each one.
yes > /dev/null
That will completely max your CPU. When done just kill all the terminal windows.
Just did this myself on my 15" MBP with MX4. My old application actually wasn't too bad and I had a hint it was not too bad cause my temperatures did not spike at all during the 2 weeks I've owned it. Still, nerds like to dissect computers so I had to check it out.
I must say, you guys make it seem easy. There were a couple of cables that were confusing as to how they are pulled/popped out. It is certainly not for the faint of heart to perform this surgery. Hope I didn't do any permanent damage.
Anyway, still trying to see if I can notice a significant temperature drop. So far it seems to be about the same, but my ambient temperature is also higher in the house than 2 weeks ago.
I must say, you guys make it seem easy. There were a couple of cables that were confusing as to how they are pulled/popped out. It is certainly not for the faint of heart to perform this surgery. Hope I didn't do any permanent damage.
Anyway, still trying to see if I can notice a significant temperature drop. So far it seems to be about the same, but my ambient temperature is also higher in the house than 2 weeks ago.
Quick question: So after the upgrade, my CPU temperature has not gone over 84, even w/ Handbrake running. However, I am noticing that the temperature difference between the CPU and the heat sink is at around 30 degrees where as before it was closer to 10 degrees difference.
Did I not seal them together tight enough or is this what you guys are seeing too?
Did I not seal them together tight enough or is this what you guys are seeing too?
i just did this with AS5.
took 2 hours total, wanted to make sure everything was cleaned.
seems to be running a couple of degrees cooler, between 40-45. compared to 45-50 before with just safari and adium open.
there wasn't that big of a mess, picture below.
took 2 hours total, wanted to make sure everything was cleaned.
seems to be running a couple of degrees cooler, between 40-45. compared to 45-50 before with just safari and adium open.
there wasn't that big of a mess, picture below.
What are you using to report the temps? A 30C (I'm assuming celsius) jump is a big difference, since there should be more of a 5-10C difference in temp. What's the temperature reading of the GPU during Handbrake when your CPU is pegged at 84?
I'm using iStat Pro to monitor the temps. Let me see if I can replicate this and snap a picture.
This thread is so awesome. I kind of enjoy the satisfaction of sucessfully taking apart a machine and applying thermal paste. Just ordered a spudger and wrist straps just to be safe although never had a problem before... anyway good stuff, keep it up with any numbers and thoughts. I will be following.
Take your time when taking the MBP apart, especially with the ribbon/cables. Let us know if you have any questions, and definitely post results when you finish.
This thread is simultaneously great and depressing. I'm impressed by how intrepid you all are, and disappointed in Apple's QC. I'm about to buy a MBP, and if there's anything that has made me fear doing so, this is it.
Two questions: 1) does this issue affect 13" MBPs? (I saw that someone asked but didn't see an answer), and 2) will Apple let you return an MBP that runs hot after the 14 day window? (I imagine not.)
Two questions: 1) does this issue affect 13" MBPs? (I saw that someone asked but didn't see an answer), and 2) will Apple let you return an MBP that runs hot after the 14 day window? (I imagine not.)
Keep in mind for every MBP that does have an insane amount of TP and overheat issues, there are machines out there that are running perfectly fine. In the end it's up to you to decide whether if it's worth the risk of taking apart your $2000 computer, but for some it's worth it to perform this DIY task on a otherwise wonderful machine.
The issue with too much thermal paste has been an issue for many years, and exists on machines from the MBA to the MacBooks. In fact I have done this on every single machine that I have owned, from the Santa Rose/Penryn MBP's and the first generation MacBook Airs. I expect the new 13" MBP to have similar issues with the amount of TP I've seen Apple put on them, but again that will vary between machines as some users have not experienced any overheating issues.
Im not sure if this is normal, but after running yes > /dev/null in 8 terminal windows for around 15 minutes to gauge temps (remained steady at 84C with stock paste and 4200 fans), my computer literally got faster:
opening applications was faster, strangely internet sites opened up faster, youtube videos got ready to stream a heck of a lot faster, overall it was as though having all cores maxed got the "juices flowing" so to speak, and so made it easier to run simple tasks like that. I know close to nothing about these sorts of things, but is this normal?
opening applications was faster, strangely internet sites opened up faster, youtube videos got ready to stream a heck of a lot faster, overall it was as though having all cores maxed got the "juices flowing" so to speak, and so made it easier to run simple tasks like that. I know close to nothing about these sorts of things, but is this normal?
i think it affects any, and every mass produced computer.
they don't have time to "properly" put on the paste, so they slap it on and put it together.
btw.. when i was putting on the paste yesterday, i used a plastic spoon to apply it. it was very easy... i wish i had a plastic knife to do it. but regardless, i was able to evenly spread it out. before this method, i spread it evenly with my finger wrapped up with plastic wrap. i found that to be way too messy
I'm experiencing the exact same thing. I tried running the test to see if my macbook pro would freeze (it didn't), and the temperature of the CPU got up to about 83*C, but the heat sink only hit 53*C.
Is this a signal that I did something wrong?
Any help would be greatly appreciated, thank you.
This is interesting. With a better thermal conduction path, there should be a smaller temp delta between the CPU and its heatsink.
In fact, i'd say that the better way to determine how well your thermal paste job is is to examine this delta. A high delta means poor thermal conduction so you might want to improve it. This delta is also insensitive to ambient temperature so it'd make comparisons before/after and with others more meaningful.
In fact, i'd say that the better way to determine how well your thermal paste job is is to examine this delta. A high delta means poor thermal conduction so you might want to improve it. This delta is also insensitive to ambient temperature so it'd make comparisons before/after and with others more meaningful.
Hi guys, just wanted to share my experience =), i've got an 15" 2010 i5 MBP.
I bought the Collaboratory Liquid Ultra (the difference with the pro is that ultra is more thick and easy to install). so, i followed ifixit and got this D:!!!!
![]()
So i cleaned with isoprophilic Alcohol and applied the product as told in the manual and youtube guides. The result was the following
![]()
Nice huh? =)
What did i get?
MUCH MUCH MUUUUCH less fan noise (in mac they don't start even using parallels).
For temps in bootcamp (wich i use to play) the temp went from 70 to 62-55°C without load and from 95 to 80°C after playing, also, the temp goes down impressively fast after gaming for a while, the fans shuts down in less than 30sec.
I'm very happy, the liquid costed me 13bucks in amazon so very very very nice.
Happy dissasembly Maforum readers!!!!
I bought the Collaboratory Liquid Ultra (the difference with the pro is that ultra is more thick and easy to install). so, i followed ifixit and got this D:!!!!
So i cleaned with isoprophilic Alcohol and applied the product as told in the manual and youtube guides. The result was the following
Nice huh? =)
What did i get?
MUCH MUCH MUUUUCH less fan noise (in mac they don't start even using parallels).
For temps in bootcamp (wich i use to play) the temp went from 70 to 62-55°C without load and from 95 to 80°C after playing, also, the temp goes down impressively fast after gaming for a while, the fans shuts down in less than 30sec.
I'm very happy, the liquid costed me 13bucks in amazon so very very very nice.
Happy dissasembly Maforum readers!!!!
My delta has dropped to about 10-15 degrees instead of the 20-30 degree difference. The only two things I could think of that may be "wrong" to cause this difference is if we didn't screw the heat sink close enough to the CPU and/or if we applied too little paste - though neither seems likely, honestly.
8-15 degrees seems like it falls within normal parameters though I can't recall for sure. Maybe give it some time and see if yours normalize after some usage? On a side note, my 2011 MBP runs cooler than my 2008 MBP now. So take that quad-core heat!
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.
Old paste wasn't too bad and I didn't run into huge temperature issues.
Cleaned.
Reapplied (with a toothpick)
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.
Old paste wasn't too bad and I didn't run into huge temperature issues.
Cleaned.
Reapplied (with a toothpick)
^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.
Hi everyone
The thread is now a month old -- has anyone been able to check if the most recently-bought MBPs have less TP applied?
Basically I am wondering if this issue may sort itself out over time since I am planning on buying an MBP later.
/p
The thread is now a month old -- has anyone been able to check if the most recently-bought MBPs have less TP applied?
Basically I am wondering if this issue may sort itself out over time since I am planning on buying an MBP later.
/p
