Thermal paste

[comda]

Greetings!

A few weeks ago i posted a thread about my Macbook 3,1. Yes i know its not power PC but this post is still useful to the power PC community. last year i did the thermal paste on my macbook with Startechs metal oxide thermal CPU paste compound and it dropped temps. But now im thinking i need to do it again but better. Is silver arctic really better then this stuff? local computer shop says no, but whats your opinion?

 

[bunnspecial]

I use AS5 on everything, but honestly don't have any particular brand loyalty-I do so mostly out of convenience(locally available).

I think that the silver pastes are better than the white/ceramic type ones and gravitate toward them, but IMO they're pretty equal when you right down to it. If you're chasing absolute lowest temperatures maybe it makes a difference, but I think in general as long as your paste is in good shape, applied properly, and applied in the proper amount you get into a "how many angels can dance on the head of a pin" argument.

If your paste were 2-3 years old, I'd say to redo it more for the sake of age than anything. Since it's only a year old, though, I'd say don't worry about it.

 

[eyoungren]

AS5 is pretty much the gold standard. And on Amazon, it's cheap. You can get it with a cleaning kit (which is what I did) that gives you a cleaning chemical and a purifying chemical.

For GPUs I usually use a cheaper paste (but not too cheap).

Note, you can use coffee filters as lint free cloths for all the work. That's what I use.

 

[comda]

Well im worrying about it because i think i put too little on it. Right now im using Snow leopard and this is the only page open using firefox. Its at 50 degrees Celsius. thats fine. But if i scroll down any mac rumor page it jumps fast to 70 plus. I think i didnt put enough on the video chip. sadly there is no temp sensor on the video card. Facebook is impossible to use on the system and i have uninstalled flash to find a small change.

 

[bunnspecial]

Usually "too much" is more of a problem than "too little."

With that said, what temperature monitoring program are you using? I don't have a Macbook handy, but I'm pretty sure that I've been able to get TemperatureMonitor to read the GPU temperatures. iStat can be hit or miss on some of this stuff.

 

[oneMadRssn]

For applications where there is a larger gap between the chip and the heat sink, people seem to recommend Tuniq TX-4 over AS5. AS5 is great, but it's not as good where you need it to fill a wider gap. TX-4 is pretty thick, and will more easily fill a wider gap.

I can't speak to all macs, but from the ones I have taken apart, there is usually a wide-enough gap where I tend to use TX4 over AS5.

 

[comda]

Usually "too much" is more of a problem than "too little."

With that said, what temperature monitoring program are you using? I don't have a Macbook handy, but I'm pretty sure that I've been able to get TemperatureMonitor to read the GPU temperatures. iStat can be hit or miss on some of this stuff.

i have both istat and SMC fan control reporting CPU temps.

 

[eyoungren]

I have both as well. My 2006 MBP regularly goes north of 150º when fully loaded (I've had it around 168º or so). Not had any issues yet. The Intels seem to run as hot as the G5s.

 

[flyrod]

I think the installation is usually more critical than the brand of paste. When you pull off the heat sink you can look at the old paste to see if the contact was even. I've seen some of the heat pipes get bent slightly, or on powerbooks sometimes the spring loaded screws come unsoldered from the board.

 

[Cox Orange]

Heh, found this via the suggestion of the forum on the bottom of the page and while we had a discussion in the other thread, if it is ok to revive old threads, a test I once found popped up in my mind and I think it might be helpful, anyway how old the thread is. Here is a comparison of several thermal pastes and pads http://skinneelabs.com/2011-thermal-paste-review-comparison/2/

I was always told though, that given the conditions are the same and good (seating of the cooler etc.) the results only differ in small variations that can be seen as coming with the inaccuracy of testing methods.