Thursday, June 16, 2005

Hot summer, cool mac

My mac is so hot that the fan kicks in so early and you know, I don't like to hear fan's noise, drive me crazy most of time.

So, I decided to get rid of noise of my mac. It really took me a while to figure out that the fan kicks in whenever the temperature of any parts in mac reachs 133OF ish, including cpu top, bottom temp, hard drive, video card die, battery outlet, etc. This is what I found for my 12" powerbook, might not be the same for everyone. And usually the first part to reach that limit is either video card die(Temperature monitor) or power supply bottom(Temperox), both reading came from different software, and their readings are very close, made me think they might be talking about the same thing, or these two parts might be just connected somehow inside.

Anyway, this is good. I know when the fan kicks in. And Whenever it kicks in, the speed of fan got several stage/mode: (1) relativly quiet one, the speed is around 4000rpm (2) bit noisy, the speed reach around 4400rpm (3) you don't want to get there, I've seen somebody's fan can go as high as 6000rpm. The noise won't let you focus on anything other than what do I do with this mac now.

I've searched around about this topic since I had my first laptop. I know you probably can buy those cooling pad, iCurve, iBreeze, iWhatever, but IMHO, though some are better than the others, they only can work to certain extent. The idea is pretty simple, you want to extract and dissipate as many heat as possible to the ambient as fast as you can. But I am not really conviced by the way they're dealing with this. I'll explain in the following.

The heat generated by cpu and graphic cards mainly will result in heat accumulation and therefore temperature increase if not cooled promptly. But it's usually a long way for heat to go to ambient 'cause there're several resistence it'll meet in the road. To name them, 1) is the materical of the case or fillings between case and cpu(for example), less the conductivity of the material will give you high resistence. 2), the resistence for heat transfer through metal to air layer which close to the case. this also applys to the part inside the case which are all immesed in air 3), without flow around case, the heat needs to diffuse away by conductivity of the air or natrual convection due to the air density dependence on temperature.

I hope you're still with me. Good, usually, the resistence of 3 is higher than 1 and 2. Which means, that if you add a fan or AC near your laptop, you always get better cooling instantly. Therefore theoretically those laptop pad with fans should work the best as long as the fans added are quiet. In a word, the resistence of 3 can be eliminated completed if you can keep the temperature of air on the case surface to be constant ambient temperature. If you work on a hot day, anyway you screwed unless you have a AC around.

Still with me? You need a cool room with good circulation, and problem solved. Not quite, I'm sure ppl will argue that their lap still get burned at a cool room. True. Let's see what else can I do.

Let's first talk about the resistance of 1. Actually we can't do much about it 'cause Mac made the case. And I notice the temperature gradient from the hottest spot to the coolest spot is huge, I feel it's huge I never meatured but I am 100% positive the difference is huge for an operational Mac. What I mean is the whole laptop is not made of a metal though the surface looks like a metal material to me, therefore the conductivity is pretty small and the heat can 't spread effectively to let more surface to cool it. This is bad, but as I said there won't be much we can do. I find a commerical Canadian company which have a workaround for this. Here's the link www.koolsink.com. They create another better conductive surface to accerlerate this procee. I found convicing because I bought one and tried it, I can feel the uniform warm surface. But I can be wrong about this argument.

The drawback is related to resistance of 2. Nomatterwhat, heat need to dissipate throught that interface between case and air and if you have another pad beneath it, you actually dealling with double resistance of type 2. I've seen some ppl use fluid bag/liquid ice bag to cool the laptop, their ideas is about decreasing resistance of type 2. For whatever reason, my gut tell me this is the heat limiting case here, though my background tells me air circulation is the key (werid). But anyway, my plan is to follow my gut and I plan to use some conducting tape to fill the gap between the case surface(hottest spot) to my cooling pad.

I 'll let you know how it goes next time. Happy with any comments

No comments: