Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > Hardware - Troubleshooting and Discussion > Mac Notebooks > better thermalpaste for last gen MBP?

better thermalpaste for last gen MBP?
Thread Tools
AltecXP
Junior Member
Join Date: Jul 2009
Status: Offline
Reply With Quote
May 3, 2010, 11:22 AM
 
Would putting AS5 or ICD on a last gen MBP help the temps any? the bottom of mine gets really warm even if the 9600GT is on and I'm not even putting any load on it.

Are there any guides to doing so if I decide to try it? I haven't torn down a unibody MBP before, but I have many other laptops.

thanks
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 3, 2010, 12:40 PM
 
I have yet to see hard evidence that re-applying thermal paste to Apple's notebooks really does what some people hope it will do.

There are a few anecdotes on the web, but by large there is no reason to assume Apple is systematically applying the wrong stuff or applying it the wrong way. And hence re-applying it will not automatically improve anything. In fact, you're just taking the risk you'll screw something up during the process.
     
ghporter
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 3, 2010, 08:12 PM
 
I recall a very long and extended discussion of this. Some people thought Apple didn't know how to assemble their own machines, while others felt that if you could detect thermal past between a CPU and heatsink then way too much had been used. Wasn't the end point in that thread a consensus that "if it isn't really broke, don't try to fix it?"

Glenn -----OTR/L, MOT, Tx
     
imitchellg5
Posting Junkie
Join Date: Jan 2006
Location: Colorado
Status: Offline
Reply With Quote
May 3, 2010, 08:56 PM
 
If the 9600M GT is on, you're going to heat up a lot faster. It's not the most efficient GPU ever. Mine gets warm just idling sometimes.

The bottom line with this discussion is that Apple's been manufacturing computers for a very long time, and I'm guessing most of us haven't.
     
AltecXP  (op)
Junior Member
Join Date: Jul 2009
Status: Offline
Reply With Quote
May 3, 2010, 09:28 PM
 
That doesnt mean much. HP has too and they still use crappy thermal pads instead of heatsinks.
     
AltecXP  (op)
Junior Member
Join Date: Jul 2009
Status: Offline
Reply With Quote
May 3, 2010, 09:30 PM
 
Originally Posted by ghporter View Post
Wasn't the end point in that thread a consensus that "if it isn't really broke, don't try to fix it?"
I subscribe to "if it aint broke, make it better."
     
imitchellg5
Posting Junkie
Join Date: Jan 2006
Location: Colorado
Status: Offline
Reply With Quote
May 3, 2010, 09:34 PM
 
Well, then take it apart. If you eff it up, don't whine to us.
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 4, 2010, 02:42 AM
 
Originally Posted by imitchellg5 View Post
Well, then take it apart. If you eff it up, don't whine to us.
QFT.
     
sudowork
Fresh-Faced Recruit
Join Date: Apr 2010
Status: Offline
Reply With Quote
May 4, 2010, 05:44 AM
 
Not 100% sure how the heat transfer works in your generation of the MBP, but from what I can tell... you're going to be limited by your heatsink's capacity for heat and not by the method of induction. Because it's a laptop, the airflow is limited, so you're not going to be able to draw the heat away from the GPU+heatsink. For more info: Info: How does a heatsink work? - Overclock.net - Overclocking.net

If anything, I would try manually increasing the RPM on your fans rather than try to induct more heat into the heatsink. I actually replaced the stock thermal paste on my early 08 MBP with AS5 because I replaced the lower case (and had to remove the logic board). I don't remember the exact temperatures pre-replacement, but now I'm getting around 65 degrees celsius doing normal work on my GPU diode (from iStat).

What temperatures are you getting exactly?

Edit: On a side note, I wonder why the pre-unibody MBPs don't utilize a heatsink for the CPU. Is it just a space issue? I'm interested in the difference in temp there. I'm getting about 60 degrees Celcius on my CPU. Also, when I turn my fans up from 2000 rpm to 6000 rpm, my CPU and GPU temps decrease by ~10 degrees Celsius.
( Last edited by sudowork; May 4, 2010 at 06:18 AM. )
sudo work
     
ghporter
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 4, 2010, 07:55 PM
 
Originally Posted by AltecXP View Post
I subscribe to "if it aint broke, make it better."
That would be a valid standpoint, except that nobody has been able to conclusively prove that adding heatsink compound beyond what Apple does at the factory actually does make it better.

However, if you're willing to meticulously document both before and after thermal handling characteristics of your MBP (along with probably voiding your warranty in the process), please do post your results. I'm not really saying "don't do it." But I would really like to not get into the ugly "urinary olympics" we had here the last time the whole "I think Apple doesn't know what they're doing with heatsinks and CPUs" came up...

Glenn -----OTR/L, MOT, Tx
     
imitchellg5
Posting Junkie
Join Date: Jan 2006
Location: Colorado
Status: Offline
Reply With Quote
May 4, 2010, 08:58 PM
 
Originally Posted by Simon View Post
QFT.
I mean I'm not trying to be rude or anything. But Apple has constructed the unibody MBPs so "tightly" and they are so well put together that I can't imagine anyone actually trying to improve it without totally ruining it. The unibody MBP really isn't meant to be taken apart by the user at all except for replacing RAM, HDD, and (sort of) the battery. Going anywhere beyond that is asking for a world of problems. And unlike Glenn, even if the OP is successful, I don't think a writeup should be put on these forums because I don't want a technically incapable person Googling "fix MBP heat" and coming upon this thread then destroying his $2k MBP.
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 5, 2010, 02:59 AM
 
It didn't come across as rude to me. The benefit is doubtful so if the OP indeed decides to go in there he needs to be aware of the involved risk. That said, like Glenn I'm always interested in hearing other people's tinkering experience. So if the OP wants to risk his MBP and fiddle around with its guts, I'll be more than anxious to read a report about it.
     
AltecXP  (op)
Junior Member
Join Date: Jul 2009
Status: Offline
Reply With Quote
May 5, 2010, 09:18 PM
 
I havent found any posts here rude.

I'm planning to document my temps over the next week while doing norma usage on both GPU's and CPU while working and while doing a bit of gaming. Then after a week or 2 I'll I'll pull her open and change out the paste. I honestly never expected MUCH, but I'd be happy with a 1-2* lower temp on average. Even that much would make me feel like it was worth the time.
     
vmarks
Moderator Emeritus
Join Date: Apr 2001
Location: Up In The Air
Status: Offline
Reply With Quote
May 5, 2010, 10:53 PM
 
Originally Posted by ghporter View Post
That would be a valid standpoint, except that nobody has been able to conclusively prove that adding heatsink compound beyond what Apple does at the factory actually does make it better.

However, if you're willing to meticulously document both before and after thermal handling characteristics of your MBP (along with probably voiding your warranty in the process), please do post your results. I'm not really saying "don't do it." But I would really like to not get into the ugly "urinary olympics" we had here the last time the whole "I think Apple doesn't know what they're doing with heatsinks and CPUs" came up...
It's not adding, so much as the proper application of it.

The contention is that too thick an application acts as an insulator rather than a conductor between CPU and heatsink. Basically, the thermal compound is meant to fill the air gaps in the uneven surface of the heatsink and CPU. Thermal compound conducts heat much more efficiently than air.

It's really easy to apply too much. I've done it before. I've got a 12" powerbookg4 that overheats throttles back. It started happening after I did remove the heatsink when changing out the superdrive. I haven't had the energy to go back in and redo it properly.
If this post is in the Lounge forum, it is likely to be my own opinion, and not representative of the position of MacNN.com.
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 6, 2010, 02:42 AM
 
Wouldn't the thickness of that layer depend just as much on the applied pressure as on the amount of compound?

I'd imagine that even if I were to dump 3 oz of goo onto my CPU, if I fasten the heat sink to the board with 10 Nm I doubt there would be that much goo left acting as an insulator between the CPU and the heat sink.
     
sudowork
Fresh-Faced Recruit
Join Date: Apr 2010
Status: Offline
Reply With Quote
May 6, 2010, 03:50 AM
 
Originally Posted by Simon View Post
Wouldn't the thickness of that layer depend just as much on the applied pressure as on the amount of compound?

I'd imagine that even if I were to dump 3 oz of goo onto my CPU, if I fasten the heat sink to the board with 10 Nm I doubt there would be that much goo left acting as an insulator between the CPU and the heat sink.
Not to be a smart ass, but if you're squeezing that much thermal paste to with 10 Nm force, the rest of it's got to go somewhere. In this case, it's the logic board, and (most) thermal paste compounds do conduct electricity.

Edit: I know you were using an extreme example, but I'm just saying.
( Last edited by sudowork; May 6, 2010 at 04:04 AM. )
sudo work
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 6, 2010, 03:55 AM
 
Sure. I'm also quite certain applying 10 Nm torque would bust something on a MBP's logic board.

As you guessed, those were just numbers I threw out. I have no idea about the details. Just interested in how important pressure is.
     
sudowork
Fresh-Faced Recruit
Join Date: Apr 2010
Status: Offline
Reply With Quote
May 6, 2010, 04:09 AM
 
I'm not sure of the physical properties of thermal paste, but my guess is that in an environment of dynamic volume, it is incompressible (density won't change).
sudo work
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 6, 2010, 07:14 AM
 
Hmm, I wasn't worried about the density in the compound. I was wondering how much the resulting thickness of the compound layer between CPU and heat sink depends on the applied torque when fastening the two together.
     
ghporter
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 6, 2010, 09:09 PM
 
vmarks is right-it's easy to put "too much" paste on. But I personally don't subscribe to the thought that too much paste reduces (at least not measurably) the transfer of heat from source to sink. What it WILL do is reduce how quickly peaks in heat produced by the source GET TO the sink to be dissipated. It takes a lot of excess to do this, and a few minutes of wiggling the heat sink around on a CPU generally gets the excess wiped off pretty thoroughly.

A good method for applying thermal paste is to put "way less than you think you need" in a small dab in the middle of the source's top surface, then use an anti-static spatula to spread it out in all directions. You should be able to detect the paste, but it should not be a "layer" as such. I don't think would be physically possible with any Mac laptop to "wiggle the heat sink on the CPI for a few minutes anyway, but this is a good example of the kind of task that you need to budget plenty of time for.

Remember, basic thermal paste is a fine grade of silicon grease with zinc oxide dissolved in it. Higher quality pastes include silver (Arctic Silver) or other metals as well-Arctic Silver can contain anywhere from 65%-85% "micronized" silver, depending on what version of that paste you get. Whether zinc or silver, these compounds will conduct heat in any thickness. But the time it takes to pass each calorie is dependent on the distance through the paste that the heat energy has to travel.

Finally, with properly applied torque, a factory heatsink should squish out almost all applied thermal paste. But what it leaves could easily be twice or three times as thick as the optimum thickness of paste. As mentioned above, the purpose of the paste is to get a continuous, uninterrupted thermal conduction surface between source and sink. It's not magic, it's just an interface medium.

Glenn -----OTR/L, MOT, Tx
     
Simon
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 7, 2010, 03:17 AM
 
Originally Posted by ghporter View Post
Finally, with properly applied torque, a factory heatsink should squish out almost all applied thermal paste. But what it leaves could easily be twice or three times as thick as the optimum thickness of paste.
[emphasis mine]

Thanks, Glenn. Great info. Exactly what I was wondering about.
     
   
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 07:41 AM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,