MacBook Pro temps drop with thermal paste makeovers

“Any PC enthusiast knows that thermal paste is something to be applied in precise amounts. Maybe Apple could take a lesson from all of us out there who have tacked a cooler onto a CPU. Apparently, the recent rash of toasty MacBooks can be attributed to excessive amounts of silicon thermal paste applied between chips and heat sinks. A number of MacBook users have, in heated desperation, resorted to cracking open their MacBooks only to find minuscule amounts of thermal silicon between the chip and sink, but massive amounts of silicon paste spilling out around the edges. Now, if the silicon thermal paste had been applied properly all these MacBooks wouldn’t be cooking themselves to death,” J. Micah Grunert reports for Neoseeker.

“Some MacBook users had cried that their computer was running at just over 50 Degrees Celsius. But once a better thermal paste replaced the silicon, temperatures dropped to roughly human body temperatures, some 35 to 39 Degrees Celsius,” Grunert reports.

Full article here.

Advertisements:
• Get the new iMac with Intel Core Duo for as low as $31 A MONTH with Free shipping!
• Get the MacBook Pro with Intel Core Duo for as low as $47 A MONTH with Free Shipping!
• Apple’s new Mac mini. Intel Core, up to 4 times faster. Starting at just $599. Free shipping.
• Apple’s brand new iPod Hi-Fi speaker system. Home stereo. Reinvented. Available now for $349 with free shipping.
• iPod. 15,000 songs. 25,000 photos. 150 hours of video. The new iPod. 30GB and 60GB models start at just $299. Free shipping.
• Connect iPod to your television set with the iPod AV Cable. Just $19.
• iPod Radio Remote. Listen to FM radio on your iPod and control everything with a convenient wired remote. Just $49.

58 Comments

  1. Ummmm… doesn’t a cooler case mean toastier chips??? I mean, if the case was so hot, it means the thermal paste was doing it’s job and transferring heat from the chips to the metal case. So, now, if the case is cooler, then is the heat trapped on the chips? Maybe these guys should check their chip temps before they get too excited. Or maybe I’m missing something. First post?

  2. I’m glad MDN posted this article, apparantly quality control is dropping considerably at Quanta Computer, Apple’s (and just about everyone else’s) laptop computer maker.

    Apple is having a hard time demonstrating their hardware is better when they are made as cheaply as cheap Dell PC’s.

    Get cracking Apple, your reputation for quality is going downhill in a hurry.

    1: Mac OS X security

    2: Whine, speaker and heat problems on MacBook Pro’s.

    3: Aperture

    Perhaps the Apple folks are getting pushed to hard lately?

  3. The first two posters are wrong.

    If there is too much thermal paste, the paste will act as a insulator, thereby preventing heat transfer to the copper heatsink, creating the high default temperatures on MBPs.

    Apple should have used a thin film of thermal paste — less than 1mm thick — on the MBP chips. This is the proper way, and allows unencumbered heat transfer from chip to heat sink, resulting in cooler temperatures.

  4. is this what the new commercial will be about?

    “The Mac Book Pro: the best notebook in the world can even make your breakfast” (Show two eggs sunny-side up sizzling on a flipped over MBP)

    ::sigh::

  5. I noticed that many MacBook Pro users reported in the Apple forums that resetting the PMU fixed their heat issues.

    In fact, it may well be that taking your MacBook Pro apart enough to re-apply heat sink compound has the side-effect of resetting your PMU, which actually fixes the heat issue.

  6. I doubt that a tiny bit of extra paste will matter that much in this case. This is a Core Duo, not an Athlon64. When they apply the heat sink, it squishes the paste pretty thin anyways, and the paste around the chip doesn’t matter at all – only between the CPU and heatsink.

  7. I’m a mechanical engineer who’s designed plenty of power supplies and other printed circuit boards. HERE’S HOW IT WORKS:

    1) Thermal paste has MUCH poorer conductivity than solids such as ceramic, metal and plastic.

    2) Thermal paste is applied between plastic or ceramic chips and metal heatsinks to improve thermal conductivity between them.

    3) Isn’t this an oxymoron? No. Because the interface between the two flat surfaces is imperfect. It is neither perfectly flat nor is it perfectly smooth. Consequently, there are areas that touch (fabulous heat conduction), and areas (both macro and micro) that have air between them. Air is a much poorer conductor of heat than thermal heat sink grease (“silicone” grease) so the compound improves conductivity only if it replaces air without diminishing solid-to-solid contact area.

    4) You only want a small, nearly transparent coating of the thermal heat sink compound before mating the surfaces. You want the compound that would otherwise interfere with the perfectly touching areas to easily squeeze out of the way and only stay in the areas where there are macro-size debonds and micro-size (surface roughness) spots.

    5) If one has applied too much compound before mating the parts, you can float the heat sink atop a couple-mil-thick layer of “thermally-conductive” heat sink compound. Again, the stuff has better conductivity than air but it’s much worse than solid-to-solid conductivity. So floating the part is bad.

    6) If the chip runs hotter, it heats up its local area of the printed circuit board and the nearby case rather than shed it’s heat into the heatsink (which, in turn, sheds its heat into the fan-forced air).

    7) If you have good chip-to-heatsink conductivity, the chip is cooler, the case is cooler, the heatsink is hotter, and it means the air coming out of the ventilation ports is hotter. This is good.

    8) Proper heat sink compound application is the most poorly understood thing on the world’s assembly lines and even some line supervisors still think that if “some’s good, more’s gooder”.

  8. Jon:

    Interesting question. Perhaps Apple will be able to utilize the heat energy produced by the chips to also power the CPU. You know, like cold fusion, so battery life can be extended for weeks and weeks. What we need is a little civil disobedience and violate the laws of thermodynamics.

  9. Oh, and notwithstanding some pooh-poohs from others about hot chips aren’t bad, they’re wrong.

    If an engineer added a heatsink, it was done to keep something from overheating (most likely the chip). There is a logarithmic relationship between chip life and temperature. An over-capcity heatsink costs more money and takes up more space (a serious constraint in a laptop computer) so it’s not like there is excess heatsink capacity; engineers almost NEVER get excess heatsink capacity. If the heat isn’t getting into the heatsink at the “advertised” rate (under ideal conditions with the proper application of heatsink grease), then something’s getting too hot and that’s not good.

  10. Jon: See Greg’s post – hot laptop case = poor thermal dissipation.

    MacBook Pro owners – go get some Arctic Silver thermal paste, apply thin transparent layer between heat sync and CPU. Then see temperature drop 10-20 degrees as a minimum.

  11. If this was happening to Dell laptops MDN would be jumping all over them for their crappy laptop heaters because Apple has a great vertical business design model that gives them superior design, engineering and control. And they would complain about the shoddy work and components in the Dell.
    But Apple has defective products and its a learning experience for us all.

    I believe if anyone attempts this procedure their warranty is voided on their Apple heater….laptop.

    Now you know what´s delaying the next iBooks – they are too damn hot!

  12. anyone visit the Apple support forums lately? They’re beginning to look like a Dell forum. There some real quality conrol issues mentioned.

    Also, it appears from reading the comments that Apple is deleted users who post negative comments about problems they are experiencing with MacBooks. Buyer Beware.

  13. I’ve had several aluminum PowerBooks, and my new 1.83Ghz MacBook Pro runs cooler than any PowerBook I’ve ever used. I’ve even played Quake 4 on it for a while, and the fans stayed really quiet.

    As far as I can tell, the MacBook Pro is the coolest and quietest professional Mac laptop yet.

    The 12″ PowerBook, on the other hand, is another story. Run a 3D game on it for 20 minutes and it will be sizzling hot with its fans screaming like banshees. Sucks big time.

  14. Nick says: “I’ve had several aluminum PowerBooks, and my new 1.83Ghz MacBook Pro runs cooler than any PowerBook I’ve ever used… As far as I can tell, the MacBook Pro is the coolest and quietest professional Mac laptop yet. The 12″ PowerBook, on the other hand, is another story. Run a 3D game on it for 20 minutes and it will be sizzling hot…”

    Maybe you’ve gotten a recent MBP; one that has the firmware patch, and maybe better thermal compound application. The reason I say this is that your statement runs counter to my experience – The MBPs I’ve been around are just as hot as the worst 15″ PB I ever owned (a 1.33Ghz model). Plus, I don’t think the 12″ PB had worse heat problems than the other PB sizes. In fact, with it running at slower clockspeeds, it should run cooler by definition. Again, from my own experience, 12″ PBs in general have only been just a little hotter than equivalently sized iBooks (which they are in fact base on), and the iBooks have never had a heat issue.

    I think the problem here results from a combination of things: First, Apple rushing a product to market in order to get bragging rights about introducing Intel’s Yonah before any of the Windows crowd (and to show how swimmingly the Macintel transition will go). And second, that Intel’s new generation of CPUs aren’t as cool and efficient as Apple (indeed, the world) expected. The latter has been noted by many in the blogosphere – how a 65nm die shrink has netted little if any gain in power efficiency, even when comparing the old Pentium M to the Core Solo. That seems to indicate that much of the space gained from going to 65nm was taken up by architectural changes – i.e. necessary improvements in gaming and multimedia – which in turn uses more electricity.

    Granted, the Yonah is more capable than the Pentium M, and thus you can look at the increased computing power as sort of an efficiency improvement, theoretically. However, using the dual core G4 could have achieved that over the last PB too (and the DC G4 uses LESS power than it’s predecessor). Plus, the reality of bigger batteries in MBP netting no improvement in computing time, and really high temperatures, aren’t theoretical at all.

    Unless you do this, of course ” width=”19″ height=”19″ alt=”shut eye” style=”border:0;” /> Then it’s all good.

  15. Odyssey67:

    I might have gotten lucky with this one… I did reset the PMU first thing, since I heard it fixed overheating problems. I applied the latest firmware patch first thing too, like people are pretty much supposed to do.

    I also read that Apple is replacing MacBook Pros that run really hot, so that might be a better idea than tearing your MacBook Pro apart to re-apply heat sink compound (voiding your warranty).

    I’ve used 867Mhz and 1.5Ghz PowerBook 12″ models, and they were both super hot and had super loud fans that came on frequenty.

    I’ve had PowerBook 15″ 1.25, 1.5 and 1.67Ghz models, and they were all pretty toasty and the fans came on quickly when I booted up a game, but the fans are much quieter than the 12″ PowerBook fans.

    I’ve had a 1.67Ghz 17″ PowerBook, and it was the quietest and coolest of the PowerBooks I’ve had, but not as good as the new MacBook Pro.

  16. “Too Hot!”

    Yeah, I’ve done a similar test in a Mac store. The Intel-based Macs boot so much faster than the “old” G5 ones – largely due to EFI I’m pretty sure.

    My MacBook Pro boots into Windows via Boot Camp faster than my Athlon64 box at home boots into Windows. LOL

  17. @ Nick:

    Thanks for the link. That was a great shootout. Amazingly, the MBP running on Solo outperformed the 2GHz PBk on all but Rosetta-dependent tasks and iTunes.
    That bodes very well for the iBook, err, MacBook cominf out with Solo, as well as the mini.

  18. Yeah, I like how the 2Ghz Core Solo scored 312 in the CPU render, whereas the 2Ghz G4 scored 182. That’s pure floating point performance – hard drive speed, graphics card and memory performance are not very important.

    I did some testing with my home computer, looked up results at http://www.xlr8yourmac.com and http://www.barefeats.com, and here you go:

    Cinebench v95 CPU render scores:

    PowerBook G4 1.5Ghz ———————- 135
    PowerBook G4 2.0Ghz ———————- 182
    iMac G5 2.1Ghz ————————— 294
    2Ghz Core Solo MacBook Pro —————- 312
    2.6Ghz Athlon64 (Win XP 32 bit) ————- 385
    2.7Ghz G5 ——————————- 385
    2.6Ghz Athlon64 (Win XP 64 bit) ————- 423
    Dual 2Ghz G5 —————————- 521
    1.83Ghz MacBook Pro Core Duo ————- 535

    2.0Ghz MacBook Pro (Windows XP) ———– 528
    2.0Ghz MacBook Pro (Mac OS X) ————- 575

    Dual 2.2Ghz Athlon 64 (WinXP 32 bit) ——— 585
    Dual-core Athlon64 2.2Ghz (winXP 32 bit) —– 600
    2.16Ghz MacBook Pro (Mac OS X) ————- 610
    Dual 2.7Ghz G5 ————————— 721
    Dual 2.2Ghz Athlon 64 (WinXP 64 bit) ——— 757
    Quad-core 2.5Ghz G5 ——————— 1105
    Quad-core 2.2GHz Opteron (Win XP 64 bit) —- 1182

    Anyways, you can see that Mac OS X easily outperforms Windows XP on the same hardware, so this is not a Windows-biased test. You can also see that the Intel Core Solo beats the G4 by about 70% per Mhz, and that the Intel Core Duo beats the dual G5 by about 10% per Mhz.

    I especially like how the Core Duo beats the Athlon64 X2 at the same Mhz.

  19. Gibbering monkeys on meth could slather thermal goop better than the uninspired workers hired by Apple. Let’s get some motivated employees hired ASAP or replace them with more accommodating precision robots. Maybe the shift that produces the fewest mistakes gets a free Macbook Pro and a job. Damn, I’m a mean bastard, expecting folks to do it right the first time or at least to do it better the next time.

  20. Is it coincidence that George Foreman brings out the “G5” grill, and Apple responds with the Mac Book Pro Intel based grill, immediately rendering that old G5 based grilling system obsolete?

    I for one don’t think so…

  21. Also the Steve Jobs Grill is much easier to use and comes with many more bundled applications.

    Once George Foreman customers understand this they will immediately switch grills and never look back! The Steve Jobs Grill is poised for some Great Growth Opportunities!

  22. Also the design of the Steve Jobs Grill is much better. It is slimmer, sleeker and runs on batteries, and includes the same safety power connector found on Japanese deep friers and tea kettles. These features alone are worth the price difference

    Who wouldn’t want a Steve Jobs Grill instead?

  23. I just wanted to say how smart Steve Jobs is.

    Only a man like Steve and company like Apple could have produced a product like the Steve Jobs Grill. Apple is so far ahead of those puny competing grills it’s not funny.

    I want to have his babies…

  24. I took my MBP apart last night and that stuff was all over the place. It had also gotten hard. I guess because the excess had been cooking. I cleaned it up and applied some paste that I have (not AC5) and today the case is noticeably cooler. The bar above the F keys is no longer blistering hot and the bottom of the case is simply warm and no longer unbearable. The comparison is with it plugged in.

    What a big difference! It was fairly easy to do and fairly straight forward.

  25. @Ampar,

    Are you trying to say that only Mormons will be buying 1st gen Apple products from now on? ” width=”19″ height=”19″ alt=”tongue rolleye” style=”border:0;” />

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.