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 > [Pic] How Apple applied the thermal grease on MacBook

[Pic] How Apple applied the thermal grease on MacBook (Page 2)
Thread Tools
Senior User
Join Date: Nov 2002
Location: Bay Area
Status: Offline
Reply With Quote
May 19, 2006, 03:36 PM
 
Reminds me of the Bagels w/ Cream Cheese at Noah's :-)
     
Mac Elite
Join Date: Sep 2000
Status: Offline
Reply With Quote
May 19, 2006, 04:51 PM
 
You know it might be useful (or at least informative) if all the MacBook and MacBook Pro users could list their CPU temperatures while running idle and while running at maximum load. Maybe from this information we could discern which ones might have to much thermal paste and which ones might not.
     
JAR
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 19, 2006, 11:26 PM
 
Good idea, Gabriel. I'm new to Macs; is there a specific tool that people are using to measure the temps? I tried iStat and it does not display any temps. Someone else told me there are no tools out yet that can measure MB temps... but then how are other people here reporting temps?

Also, some people say the pics of the MB chips and heatsinks look normal, i.e. not too much thermal paste applied. Check this closeup from kodawarisan:
http://mactree.sannet.ne.jp/~kodawar...image/1417.jpg

I dunno about you guys but to me that looks like too much. Not ridiculously too much like Apple's instructions, but still too much. Look at all the gunk oozed on the sides of the heatsink, and the thick patches of paste on the heatsinks - that's bad.
     
Forum Regular
Join Date: Jul 2002
Location: Los Angeles, CA
Status: Offline
Reply With Quote
May 20, 2006, 12:02 AM
 
Stupid question: How does one remove thermal paste efficiently without damaging the chip? Is it okay to just scrape it off with a plastic ruler or something similar?
I'm a bad...motherf%#!ing DJ
     
Junior Member
Join Date: Nov 2002
Location: Chicago, IL
Status: Offline
Reply With Quote
May 20, 2006, 12:04 AM
 
Originally Posted by Seamus
Stupid question: How does one remove thermal paste efficiently without damaging the chip? Is it okay to just scrape it off with a plastic ruler or something similar?
Rubbing alcohol and qtips... it will loosen up and start to wipe off. No need for something harsh like a ruler or anything metal.
     
Mac Elite
Join Date: Sep 2000
Status: Offline
Reply With Quote
May 20, 2006, 03:19 AM
 
Originally Posted by puffarthur
Good idea, Gabriel. I'm new to Macs; is there a specific tool that people are using to measure the temps? I tried iStat and it does not display any temps. Someone else told me there are no tools out yet that can measure MB temps... but then how are other people here reporting temps?
Welcome to the Mac, puffarthur! Perhaps you've hopped on at the most opportune time in recent history, with the Intel transition and all. On the matter of temperature reading: I don't know much about that either, but check out what this guy says on the Ars Technica MacBook review (p. 4):

Originally Posted by Clint Ecker
To test out the temperatures, I loaded up an application called CoreDuoTemp that loads a kernel extension and then monitors the machine's temperature, processor speed, and CPU loading. I then put together a python script that looked like the following:

import math
while 1:
for x in range(10000):
y = math.cos(x)

I then ran two instances of this script in order to peg both processor cores.
     
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 20, 2006, 03:23 AM
 
Originally Posted by puffarthur
Good idea, Gabriel. I'm new to Macs; is there a specific tool that people are using to measure the temps? I tried iStat and it does not display any temps. Someone else told me there are no tools out yet that can measure MB temps... but then how are other people here reporting temps?
You could try CoreDuoTemp.
     
JAR
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 20, 2006, 01:15 PM
 
Gabriel, Simon, thanks for the welcome and the links. I just installed Coreduotemp and I'll be testing the temps under different CPU loads. Any ideas on what programs I can run that will max out the CPU?
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 20, 2006, 01:27 PM
 
A lot of these issues have been discussed in detail in the Apple MBP forums. I am a MBP owner that had a very hot computer and decided to do something about it myself. I didn't get a chance to measure the temps before the operation but using the postings of others as far as temps go, I decreased the max temp my processor reaches by about 30C. It really was a major difference in my case. My idle temp now is about 47C and max is about 57C.

Here is one of my posts on the subject:
http://discussions.apple.com/message...305896#2305896

Here is a temp graph screenshot after the surgery:
http://www.pbase.com/eclecticphoto/image/60056125

This is what was used to create the graph (Install and run CoreDuoTemp before running Hardware Monitor):
http://www.bresink.de/osx/HardwareMonitor.html

Here is the correct method to apply thermal paste which I followed religiously:
http://www.arcticsilver.com/arctic_s...structions.htm

By the way, I just wiped off the existed paste really well with paper towels. I used no chemicals. Someone previously asked about it.
     
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 20, 2006, 02:07 PM
 
clbell,

I've created a new site that allows you to easily submit temperature information and view the entries in a chart with different filtering options available. We'll be able to see if there are any trends in serial and manufacturer dates, MacBook Pros vs MacBooks vs iMacs, compare similar configurations, factory vs modded thermal paste and a lot more.

Take a look at http://www.intelmactemp.com and submit your temperature readings. Read the Howto before doing the temperature measures.

Feel free to mention the site whenever you see fit.

-ivc
( Last edited by ivc; May 20, 2006 at 09:43 PM. )
     
JAR
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 20, 2006, 02:13 PM
 
Quick update after using CoreDuoTemp:

At first boot it idled around 50C, which only made the bottom slightly warm. After a while, idling around 5-10% CPU the temps were around 65C. If the CPU went to 25-50%, even briefly, the temps shot up to 70C easily. That is frightening to me because on my Athlon 64 desktop, I have it automatically shut down if the temps exceed 65C to avoid damaging the CPU and motherboard.

ivc, I will submit my info on the database as soon as I have time.
     
Addicted to MacNN
Join Date: Jan 2003
Location: ~/
Status: Offline
Reply With Quote
May 20, 2006, 03:38 PM
 
My MacBook Pro, 2.16GHz (W8612... serial #) idles at about 55°C, and peaks at about 78°C under full load. Not bad for a mobile processor. High for a desktop part, but acceptable for a mobile CPU.
     
baw
Mac Elite
Join Date: Jun 2005
Status: Offline
Reply With Quote
May 20, 2006, 05:53 PM
 
( Last edited by baw; May 22, 2006 at 08:11 PM. )
     
Posting Junkie
Join Date: Nov 2000
Location: in front of my Mac
Status: Offline
Reply With Quote
May 21, 2006, 03:38 AM
 
Nice one.
     
Dedicated MacNNer
Join Date: Feb 2006
Status: Offline
Reply With Quote
May 21, 2006, 03:48 AM
 
LOL, that's to funny!!
MacBook Pro 2.5 with 4 GB Ram, 250 GB 5400RPM, iMac 20" Intel Dual Core 2.0 with 2 GB Ram
http://star-fire.deviantart.com/gallery/
     
Mac Elite
Join Date: Jan 2003
Location: 127.0.0.1
Status: Offline
Reply With Quote
May 21, 2006, 09:40 PM
 
Apple Legal should be here shortly to demand that those pics be taken down. Just remember what happened to SomethingAwful for posting pics from the official Apple MBP manual.

I, for one, will be sticking with my PowerBook for a long time to come. Then again, us PowerBook G4 owners also seem to have CPU sandwiches as well (Link).

iBook owners have nothing to fear as thermal pads were used.
     
Fresh-Faced Recruit
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 22, 2006, 07:03 AM
 
Originally Posted by Gee4orce
Hello, the voice of reason here:

Apple is one of the worlds biggest computer companies, and their machines are manufactured by some of the biggest fabricators in the world.

Do you not thing they know what they are doing ?
From my experience as an engineer in electronics manufacturing there are many simple explainations for such mistakes. Typically development engineers write a tehcnical specification, that goes to the manufacturing plant. The technical specification is then put into a manufacturing specification which is what the techs on the floor follow.
One simple mistake could have been the technical specification from the design team said to use 2cc of thermal compound, when infact they really meant 0.2cc. Or it said 2g which got interpreted as 2cc by the manufacturing engineers trying to convert it to something an operator on the manufacturing floor could more easily follow, forgetting that although 1cc water = 1g, thermal compound is denser.
Then typically you have a trainer, who may or may not be involved in the actual content of the manufacturing spec, who takes the pictures according to the wrong specification, and trains the techs on the floor. Even if the techs are experienced and know its the wrong amount of compound, they are still forced to follow it. It's also possible they communicated back up their chain of command, but since the manufacturing is probably subcontracted, even the manufacturing engineers and supervisors can't enact changes without permission from the customer. And customers don't like being told they are wrong.

Everybody knows what they are doing, just a simple mistake or poor communication can result in big problems.

Now that the issue has been discovered, why hasn't it been fixed? Bureaucracy. First Apple has to fix the the technical specifications, then communicate it to the manufacturer. Then new manufacturing specs need to be written up, new pictures and training done, then a new round of testing done to ensure quality and reliability (make sure they didn't go too far the other way and not have enough grease). Once all those things are done, they can implement it on products for customers.
     
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 22, 2006, 11:07 AM
 
Originally Posted by servognome
Everybody knows what they are doing, just a simple mistake or poor communication can result in big problems.
Very well said. Communication is a terrible thing to waste!

Glenn -----OTR/L, MOT, Tx
     
Dedicated MacNNer
Join Date: Oct 2001
Location: Washington DC
Status: Offline
Reply With Quote
May 22, 2006, 11:09 AM
 
Heh. This is now on the front of Powerpage, with pix and this thread referenced by name.
     
Mac Elite
Join Date: Oct 2004
Location: Downtown Austin, TX
Status: Offline
Reply With Quote
May 22, 2006, 11:44 AM
 
We've seen in the past that Steve has often preferred quiet operation over heat dissipation in his Macs, even if it meant system instability.

I wouldn't be suprised if the MacBook/Pro is one of those cases.
     
Junior Member
Join Date: Nov 2002
Location: Chicago, IL
Status: Offline
Reply With Quote
May 22, 2006, 01:21 PM
 
Originally Posted by jamil5454
We've seen in the past that Steve has often preferred quiet operation over heat dissipation in his Macs, even if it meant system instability.

I wouldn't be suprised if the MacBook/Pro is one of those cases.

Somehow I doubt Mr. Jobs would sanction unstable machines just to cut down on a little noise pollution.
     
Mac Elite
Join Date: Jun 2004
Location: Nashville
Status: Offline
Reply With Quote
May 22, 2006, 02:54 PM
 
Do you guys really think SJ personally sanctioned the amount of thermal grease applied to the MacBooks?

I can just picture the scene:

-Engineer: "Sir, here are the assembly specs for the MacBook line..."
-SJ :" WTF? MORE THERMAL GREASE SLAVE!"
-Engineer "But Sir too much and it won't cool properly..."
-SJ : "THERMAL GREASE ABSORBS SOUND! MORE MORE MORE!"
-Engineer " Yes sir right away sir."
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 22, 2006, 11:26 PM
 
I was re-reading this thread and I came upon an opinion I have often seen. The opinion that it's possible to just apply some paste and squeeze together to get the correct amount of paste. For instance I found this on the previous page:

"The "correct amount" is just enough to get squished out and cover the actual chip surface, with maybe a tiny bit oozing out over the edges. That applies to ALL CPUs, by the way. Just enough so that a thin, thermally conductive layer exists between the chip and the heat sink, no more."

Now, anyone can and probably will choose not to believe me but I'm a mechanical engineer and have taken a fair amount of coursework dealing with forces, heat transfer, etc. (NC State, '98 btw). So I have a little better than average understanding of forces than the average person.

Just squeezing the 2 surfaces together with paste inbetween is something Arctic Silver would not recommend. I would say, if anyone knows the absolute best way to apply thermal paste it is them. There is a reason a sharp edge is needed to remove paste to arrive at a very thin layer. I'm arguing that the reason is that in practice, there is not nearly enough force to create a paperthin layer interface. The reason it is not done this way in the manufacturing assembly processes is because the correct method is hard to do quickly and cheaply.

A couple of important things to keep in mind. The paste Apple is using, (at least what I found on my machine) is quite thick. It's consistency is not that different from the cream in an oreo cookie. It's very high in viscosity. Also, the force applied by tightening the screws is very close to perpendicular to the interface. A larger surface area means more force is required to squeeze out all the paste.

I'm not going to do the math here and bore anyone. But believe me, If someone angers me, I'll do it! Instead, I'll suggest a scenario to illustrate how much force is required to squeeze out a necessary amount of paste. The correct amount of paste is about .003" thick. That's about the thickness of a sheet of paper.

Imagine an oreo cookie with cookie halves made of steel. You place it on a hard surface and take a large metal plate and put it on top of the cookie. Now you have to push down vertically. No twisting or sliding is allowed. You have to push hard enough to push almost all of the cream out. I can't tell you how much force it is but it's quite a bit. Do you think a few tiny screws provide the required force? I can tell you they do not.

An interesting thing about this is to realize that the macbook has 2 chips versus 3 chips for the mbp. This equates to a reduced surface area. This is a possible reason the non pro macbook are cooler. Because it is true that a certain amount of paste will be squeezed out and it's possible that the reduced surface contact area equated to a larger force between the heat pipe and the dies.

I don't mean to denounce the poster I quoted above. I just think it's a misconception. Or at the very least it doesn't work as desired in reality. Perhaps it is done this way in production quite often. Just because Apple's vendors are choosing to squeeze a bunch of paste on doesn't make it the best way. There is no reason to put these people on a pedestal. They can make mistakes or make choices to save money that don't necessarily create the best product. For instance, I know there are production processes that allow a paper thin layer of paste to be put on. Some heatsinks have a preapplied paper thin layer of paste for instance. I believe this process could have been used on the heat pipe before installation. If I had to guess I would say it could be sprayed on.

Here's an example showing just how little is really necessary: http://www.heatsink-guide.com/preapplied_compound.jpg

That is a clean process that doesn't rely on gobbing lots and lots of paste.

my 2 cents...
( Last edited by clbell; May 22, 2006 at 11:38 PM. )
     
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 23, 2006, 09:05 AM
 
clbell, you make some good points, but with respect to your engineering knowledge, I think you missed something. That heat sink plate acts as a flat spring-when it's fully secured with the screws, it bows upward slightly. That allows it to apply a downward pressure on the chips through the thermal grease, which WILL squeeze the grease out fairly evenly. Some heat sinks (on desktop PCs) require a little bit of "rubbing" to distribute the grease properly before they're secured, but you don't have that amount of freedom in a laptop, so the heat sink has to push down somehow. This will continue as long as the heat sink is installed, gradually squeezing the heat sink compound down to a very thin layer-and in this case squeezing out a lot of extra heat sink compound.

And while you may believe that my assertion is off base, I think we're talking about basically the same thing. If I put a dot of grease on a chip and it gets squeezed out enough that a bit goes over the edges, that's great. The rectangular chip interacting with the heat sink will translate that pseudosphere of grease into a very thin layer with the excess being what oozes off the edges. It's important to make sure the entire surface of the chip gets some compound to prevent hot spots in the processor, and there's almost no way to avoid some ooze in a production environment.

A LONG time ago I was working in a commercial 2-way radio shop when Motorola introduced a new line of radios. We installed a bunch of them, and then started having to work with Motorola because a lot of them quit. Motorola had to do a LOT of research to find out what was wrong, and it turned out to be their manufacturing process: they used a mylar insulator between the lead-side of the circuit board and the aluminum case, and their machines didn't quite trim the leads off short enough. The long leads poked through the mylar and shorted all sorts of things, but when you took the cover off, you'd NEVER see where things poked through because mylar "heals," hiding the pokes. The Motorola field engineer spent a lot of time handing out flush-cut wire cutters and showing us which parts of the circuit board to concentrate on. I think that this sort of manufacturing "oops" is what we're dealing with right now-something that will shake out as units hit the consumer and give them problems.

Glenn -----OTR/L, MOT, Tx
     
Fresh-Faced Recruit
Join Date: Feb 2006
Status: Offline
Reply With Quote
May 23, 2006, 09:59 AM
 
[QUOTE=Simon]I think we agree. I'm not trying to suggest they're right. Actually, if they really apply as much paste as on that pic I know they're wrong. I'm merely pointing out that there might be more to the full story. There are many pretty skilled people working for Apple (and Asus for that matter) and I find it hard to imagine that they're all simply idiots that can't even read a Wikipedia page.

Maybe Apple's using too much paste to allow for (1) relative motion of CPU and heatsink and (2) mis-mating of the CPU and heatsink due to "unavoidable" misalignment in heatsink installation. Why (1) is plausible: the heatsink is below the keyboard, and gets a lot of force; when the force is released, an air gap could form between CPU & heatsink (think of two thin flexible boards placed against each other: when we flex the pair, they squeeze together, but when the flex is released, there's less force between the pair and a gap could form.) Why (2) is plausible: perhaps all the screws on the heatsink weren't tightened the same, or there's some sort of large scale warp to the heatsink. Such macroscopic imperfections (as opposed to microscopic imperfections that the standard thin-paste method overcomes), though relatively rare, need tons of paste to overcome. My guess is that Apple is optimizing for the worst case scenario (the cpu smoking out (as I've seen myself when I misapplied a heatsink)) instead of the average case scenario (poorer conduction due to thick paste). Given how PR saavy Apple is, I figure that a few widely-publicised cpu burnout accidents would be far worse than everyone dealing with a 20-degree increase in CPU temps. So we all suffer so Apple doesn't get embarassed.
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 23, 2006, 03:39 PM
 
Originally Posted by ghporter
That heat sink plate acts as a flat spring-when it's fully secured with the screws, it bows upward slightly. That allows it to apply a downward pressure on the chips through the thermal grease, which WILL squeeze the grease out fairly evenly. Some heat sinks (on desktop PCs) require a little bit of "rubbing" to distribute the grease properly before they're secured, but you don't have that amount of freedom in a laptop, so the heat sink has to push down somehow. This will continue as long as the heat sink is installed, gradually squeezing the heat sink compound down to a very thin layer-and in this case squeezing out a lot of extra heat sink compound.

I don't think we have enough information to know this. Basically what you are saying is that there is a constant force and that constant force continuously pushes out the thermal paste until there is a paper thin layer. This is not a given. As the layer decreases in thickness, the amount of force necessary increases due to increasing shear friction *. It really depends on the composition of the paste and the amount of force involved. We don't have all of this information.

But experimental evidence (my own reapplication of paste in my MBP) shows that there is more (much more in fact) than a recommended amount of paste on the chips after being clamped together by screws.

* think about it in terms of force vectors. As the layer decreases in thickness, the force required to move the paste in one direction or another becomes more and more perpendicular with the applied force. All the little particles have fewer degrees of motion and as a result interact with each other more aggressively. Take wet sand. It's a heterogeneous solution not so different in structure from thermal paste. If you put a bucketfull of it on the road, put a sturdy piece of plywood on top and stand on it, you are not going to arrive at a layer of sand that is the thickness of the largest grain of sand no matter how long you stand on it. At a certain thickness, the frictional forces will start to counteract and overcome your weight.

Does that make sense?
( Last edited by clbell; May 23, 2006 at 04:02 PM. )
     
Junior Member
Join Date: Nov 2002
Location: Chicago, IL
Status: Offline
Reply With Quote
May 23, 2006, 03:41 PM
 
This is getting a bit out hand here, people. Go outside, enjoy the weather. If you're losing sleep over how much thermal paste is inside your laptop, there are more serious problems at hand.
     
Mac Elite
Join Date: Sep 2000
Status: Offline
Reply With Quote
May 23, 2006, 05:23 PM
 
Originally Posted by Draco
This is getting a bit out hand here, people. Go outside, enjoy the weather.
I agree it's gotten a bit complex, but if anything it's become interestingly educational in the science of thermal grease!
     
Fresh-Faced Recruit
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 24, 2006, 03:17 AM
 
Originally Posted by thejam
Given how PR saavy Apple is, I figure that a few widely-publicised cpu burnout accidents would be far worse than everyone dealing with a 20-degree increase in CPU temps. So we all suffer so Apple doesn't get embarassed.
The chips will throttle down before a CPU burnout incident. In fact, that is what is happening on some of the chips that are overheating because of too much thermal grease.

Originally Posted by ghporter
That heat sink plate acts as a flat spring-when it's fully secured with the screws, it bows upward slightly. That allows it to apply a downward pressure on the chips through the thermal grease, which WILL squeeze the grease out fairly evenly.
True, thats why in high volume manufacturing typically you err to the high side. Since you get a distribution of wapage, a "little" more thermal grease is added to account for variation. However, the gross amount shown used is counter productive. As you mentioned the plate/mounting assembly has spring-like properties at high force, given the large amount of material, the force on the plate will reach the threshold before enough material is pushed out. This leaves you with a thicker layer than if you started with a lower amount of material to begin with.
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 24, 2006, 07:07 AM
 
If the PCB is acting like a spring, and we don't really know this for sure, then it's application of force actually decreases as the two surfaces come closer together. That's what happens when springs reach natural equilibrium. A rubber band is easy to pull apart at first and gradually becomes harder and harder as it stretches out.
( Last edited by clbell; May 24, 2006 at 07:19 AM. )
     
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
May 24, 2006, 09:22 AM
 
I'm not saying that the standard process with too much grease and the heat sink pressing down will produce an optimum thickness of grease. I'm saying it will push out MOST of the excess. The real optimum amount is-as has been pointed out-a hair-thin layer applied by squeezing out a small amount of grease on the chip and then using a spatula to spread it REALLY thin. Of course that takes several minutes and is error prone, so the typical manufacturing process goes with the awful "the bigger the blob, the better the job" notion. It doesn't work in soldering, and it doesn't work well with heat sink compound either. It's just not as horrible as it might be...

Glenn -----OTR/L, MOT, Tx
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 24, 2006, 09:42 AM
 
Originally Posted by ghporter
I'm not saying that the standard process with too much grease and the heat sink pressing down will produce an optimum thickness of grease. I'm saying it will push out MOST of the excess. The real optimum amount is-as has been pointed out-a hair-thin layer applied by squeezing out a small amount of grease on the chip and then using a spatula to spread it REALLY thin. Of course that takes several minutes and is error prone, so the typical manufacturing process goes with the awful "the bigger the blob, the better the job" notion. It doesn't work in soldering, and it doesn't work well with heat sink compound either. It's just not as horrible as it might be...
That I will agree with.

There are at least a couple of alternate solutions.

1. Thermal pads - more consistent results and probably better heat transfer than paste with the method Apple has chosen.

2. Another paste application method. Spraying or machine spreading of some sort. Heatsink manufacturers have been doing this for a while. Here's a couple of examples....

http://www.heatsink-guide.com/preapplied_compound.jpg
http://www.techpowerup.com/reviews/A.../Freezer64Pro/

I don't know why Apple didn't go with one of these. I'm guessing that thermal pads are expensive and the process mentioned in #2 is also expensive. They can save money by putting the cheap Chinese labor to good use. Another possibility is that Apple was in too much of a hurry to get these to market to get the best process in place.
( Last edited by clbell; May 24, 2006 at 10:10 AM. )
     
Forum Regular
Join Date: Nov 2002
Location: Chicago
Status: Offline
Reply With Quote
May 24, 2006, 10:40 AM
 
I always thought that thermal pads only worked on relatively cool running chips -- like the G4. You'd never use a thermal pad on a Pentium 4 from what I understand.
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 24, 2006, 11:10 AM
 
Originally Posted by photoeditor
I always thought that thermal pads only worked on relatively cool running chips -- like the G4. You'd never use a thermal pad on a Pentium 4 from what I understand.
Well, the most important aspect of a interface material is it's thermal conductivity. Take a look at this one:

http://www.sidewindercomputers.com/shpc.html

Thermal Conductivity W/m-°C: 3.8

3.8 is actually quite good. It's not as good as Arctic Silver 5 but I can just about guarantee you it's as good or better than what Apple is currently using. With the amount they are using, they can't afford the good stuff.

Also, the pad is only .005" thick. That's just a bit thicker than a sheet of paper. They cost about 40 cents in large quantities.
     
Forum Regular
Join Date: Nov 2002
Location: Chicago
Status: Offline
Reply With Quote
May 24, 2006, 06:13 PM
 
Probably I should rephrase my question. Don't thermal pads kind of melt under extreme heat, while paste holds up?
     
Fresh-Faced Recruit
Join Date: Jan 2006
Status: Offline
Reply With Quote
May 24, 2006, 06:35 PM
 
Originally Posted by photoeditor
Probably I should rephrase my question. Don't thermal pads kind of melt under extreme heat, while paste holds up?
Yes they do melt, by design. The melting part helps it fill all the gaps.

"Shin-Etsu's phase change thermal pads soften at 48°C and flow into surface irregularities of the heatsink and CPU, reducing contact resistance."
     
Fresh-Faced Recruit
Join Date: Jul 2006
Status: Offline
Reply With Quote
May 28, 2006, 01:22 AM
 
Just a very simple observation here but...

Did it ever occur to anyone that perhaps the service manual
is shown to apply that much thermal compound because what
the folks in Cupertino are trying to do is save $$$?
What I mean here is that on the assembly line in Shanghai
(or wherever the units are assembled), they want to make sure the
workers apply enough compound (better safe than sorry right?),
but can't afford them taking extra time (say, 5 minutes or whatever) to apply
the stuff on evenly and efficiently (like the Arctic Silver 5 instructions stipulate)?

This is of course speculation, but just an observation nevertheless!
     
Mac Elite
Join Date: Sep 2005
Location: Los Angeles, California
Status: Offline
Reply With Quote
May 28, 2006, 06:30 AM
 
Originally Posted by voyager99
Did it ever occur to anyone that perhaps the service manual
is shown to apply that much thermal compound because what
the folks in Cupertino are trying to do is save $$$?
What I mean here is that on the assembly line in Shanghai
(or wherever the units are assembled), they want to make sure the
workers apply enough compound (better safe than sorry right?),
but can't afford them taking extra time (say, 5 minutes or whatever) to apply
the stuff on evenly and efficiently (like the Arctic Silver 5 instructions stipulate)?
The flaw with this argument is that in saving the max $5 per Macbook that they would with this method only produces extremely hot MacBooks which:

a) Gives the MacBook Pro a bad reputation as an extremely hot laptop. I work at Fry's Electronics, and almost every time I bring up the MacBook Pro with various Mac users that come around to the store, they mention the heat issue. As a matter of fact, a co-worker of mine handled the return of a MBP because the wife of the customer complained about excessive heat.

b) Increases wear on internal components which in turn increases the likelyhood that one will fail, which, in (a second) turn, will generally cost Apple a whole lot more than $5 to fix under warranty, especially considering they'd most likely have to replace the entire Logic Board.

In addition, as stated several thousand times previously in this thread, applying that much amount of thermal paste results in far sorry-er than safer results about 99% of the time.
Linkinus is king.
     
Banned
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 5, 2006, 03:38 PM
 
clearly the people over at Apple haven't been sent this animated gif I made a few years ago about applying thermal compound to a processor.

     
Junior Member
Join Date: Nov 2003
Status: Offline
Reply With Quote
Jun 5, 2006, 04:39 PM
 
This is from the Arctic Silver website. Is it possible that the Macbook will run cooler as time goes on?

Important Reminder:
Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.
     
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Jun 5, 2006, 11:26 PM
 
Originally Posted by glhart
This is from the Arctic Silver website. Is it possible that the Macbook will run cooler as time goes on?
Only very slightly-note that Arctic Silver says there might be a 2-5 degree drop over this 200 hour operating break in period. I do not think anyone will notice more than that; what you get through this period is more thorough conformation of the compound to the surfaces, and the fill materials (not just zinc oxide as in most thermal compounds) align themselves properly after you spread and squished the stuff.

Glenn -----OTR/L, MOT, Tx
     
Fresh-Faced Recruit
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jan 23, 2007, 11:54 AM
 
Originally Posted by Simon View Post
[Honestly, no offense intended, but although what you say might sound reasonable, it does strike me that Apple seems to not know this and I'm doubting they employ 20'000 idiots and not one is capable of making the same conclusion you do...]
What people seem to forget is how Steve Jobs runs Apple. Sure there are boards, shareholders, etc. but Apple is largely a dictatorship with a boss who hates bad news or to have his idealistic visions/delusions contradicted. True "Yes men/women" are not inclined to tell someone like Steve that there's an issue with production. They're more inclined to sweep it under the rug until it manifests and they can blame it on manufacturing.

As for the amount of paste. I imagine Chinese manufacturing is well, Chinese manufacturing. The overwhelming responsibility is to speed production and cut costs. Corners are cut and you get what you pay for (Well, Apple does. We, the consumers, get to pay for their iPhone style 100% profit margins).

I like Apple's machines a lot but Apple really should ask questions first and not rely so heavily on customer service to show they jumped the gun and need yet another "Revision B."

With regard to thermal paste application, in general, my Shuttle SN41G2 XPC was very finicky. If I didn't use a razor blade to apply a perfectly thin layer of paste over the top of the CPU, it would overheat and shutdown.
I don't have much faith in our local Apple technicians and am truly tempted to void my warranty to ensure paste is applied properly.

If they keep churning out Not-ready-for-Prime-Time PC's, I should just become an Apple certified tech so I don't have to worry about dealing with an Apple Store or Apple "Certified" Support Center. Ever!
     
Moderator Emeritus
Join Date: Nov 2000
Location: Illinois
Status: Offline
Reply With Quote
Jan 23, 2007, 01:21 PM
 
We don't generally like people to resurrect threads that are so old, but it does give a chance to look back. In the time since this thread was current, I don't think there have been any major problems with MacBook reliability that could be traced to improperly applied thermal grease. Maybe it wasn't as big a deal as people made it out to be.
     
Fresh-Faced Recruit
Join Date: Jan 2007
Status: Offline
Reply With Quote
Jan 28, 2007, 01:49 AM
 
hmm wonder if lapping these contact points would work (don't know how the retention interface works)
     
Fresh-Faced Recruit
Join Date: Jun 2002
Location: Berkeley, CA
Status: Offline
Reply With Quote
Feb 9, 2007, 03:58 AM
 
Originally Posted by icruise View Post
We don't generally like people to resurrect threads that are so old, but it does give a chance to look back. In the time since this thread was current, I don't think there have been any major problems with MacBook reliability that could be traced to improperly applied thermal grease. Maybe it wasn't as big a deal as people made it out to be.
A good thread is a good thread, regardless of age! I hope people say that about me when I get old.

The main concern originally expressed by most people seemed to be that it was uncomfortable to touch their overly hot MBPs, especially to put them on their laps--fewer people were concerned that this might also mean that their MBP would fail sooner. Apple originally marketed all the MacBooks and MacBook Pros as laptops, and quickly pulled all use of the word "laptop" from their advertising and their web site after these complaints started, and they began to tell people these weren't intended to be used on your laptop, but instead they were now "portables".

From what I've been told, processors that don't heat up past their maximum rated temperature are unlikely to burn out, but just half a year later isn't enough time to say whether the processors will burn out sooner than they would have otherwise. We'll see.
     
Moderator Emeritus
Join Date: Nov 2000
Location: Illinois
Status: Offline
Reply With Quote
Feb 9, 2007, 06:11 AM
 
Originally Posted by jonsaw View Post
Apple originally marketed all the MacBooks and MacBook Pros as laptops, and quickly pulled all use of the word "laptop" from their advertising and their web site after these complaints started, and they began to tell people these weren't intended to be used on your laptop, but instead they were now "portables".
Really? I was under the impression that computer makers in general had stopped calling their products laptops quite a while ago. I have trouble imagining an Apple ad that would use the word, but I could be wrong.
     
Fresh-Faced Recruit
Join Date: Jun 2002
Location: Berkeley, CA
Status: Offline
Reply With Quote
Feb 9, 2007, 06:42 PM
 
Further research reminds me that you're right, the term "laptop" has been used much less by manufacturers in recent years, including Apple, due to the heat (and not just the MacBook Pro), and Apple probably didn't advertise the MacBook Pro using the term "laptop" much, but at first sometimes the term "lap" did slip in--this is from Macfixit for June 15 2006:

"…Apple has been (in some cases) telling customers seeking technical assistance that the MacBook is not intended to be used as a laptop, but instead as a "portable computer" that should be placed on a desk or other hard surface away from indirect skin contact.

"However, Apple's marketing materials differ in their description. For instance, this education-related marketing page states:

"On your lap. In your dorm room: Organize your class notes and your music. Select models now feature a superfast dual-core Intel engine, a built-in iSight camera, and Front Row software."

After I read this article, I found others in which people reported doing a search on Apple's web site for the term "laptop", and finding quite a few references, and then doing the same search some days or a couple weeks later, and finding quite a few less. So I tried some searches myself, and found one page using the term "laptop" that referred to newer portables (I don't remember if the reference was to Powerbooks or Macbooks), and that reference later disappeared from that page when I later did another search. But a search on Apple's site that I did today turned up a lot of hits I'd missed before, but most of them refer to the iBook, or use the term in articles referring to school programs whose titles make use of the term, rather than Apple using it themselves. But here's a current page featuring a picture of a Powerbook G4 Aluminum, sprinkled with the term "laptop":

Apple - Education - University of Cincinnati, College of Design, Architecture, Art and Planning

So it seems that Apple's educational materials and web pages use the term "laptop" more often than their non-educational sales. Wonder why? Students have tougher skin?
     
 
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
Trackbacks are On
Pingbacks are On
Refbacks are On
Top
Privacy Policy
All times are GMT -4. The time now is 05:02 PM.
All contents of these forums © 1995-2014 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2014, Jelsoft Enterprises Ltd., Content Relevant URLs by vBSEO 3.3.2