Push vs. Pull Tested

[ojas]

@amuffin, interesting experiment. Was wondering, was the pull config quieter than the push? I'll be doing some testing over the next week on this, have to wait for some parts to come in though 😀

Interesting discussion has arisen from this...

B is wrong I am sorry if I was coming off as rude earlier but you are not being helpful since this is a science project.

The fan blow the heat off of the heat sync this the air coming off of it is warmer than it was upon intake *truth

I'm inclined to agree with spentshells.

To the best of my knowledge, this is how a heatsink + fan setup works:

hot component-> heat-> needs to b transferred away-> direct dissipation in air not a good idea because of low thermal conductivity-> transfer heat as fast as possible away from the component via conduction -> use high conductivity material -> material can only absorb so much heat, because you can't really have a perfect sink (that absorbs heat forever). Heat continues to get transferred to the heatsink till the temperature of the heatsink is the same as the component. after that you reach equilibrium. -> to solve the problem, you need to dissipate heat absorbed by heatsink into surroundings-> for air cooling, you must dissipate it into air -> larger surface area => more dissipation as you've more area for heat transfer -> thus we use fins -> fins transfer heat to surrounding air -> must avoid same equilibrium problem -> must move hot air away from heatsink and provide it with cooler air -> hot exhaust air now in the case, must remove it-> exhaust case fan throws this stuff out.

Nice diagrams:
http://www.seymourcomputers.com.au/images/CPU%20Cooler%20Diagram.jpg
http://www.activecool.com/technotes/images/TEC_JC.jpg

 

[13thmonkey]

hot component-> heat-> needs to b transferred away-> direct dissipation in air not a good idea because of low thermal conductivity-> transfer heat as fast as possible away from the component via conduction -> use high conductivity material -> material can only absorb so much heat, because you can't really have a perfect sink (that absorbs heat forever). Heat continues to get transferred to the heatsink till the temperature of the heatsink is the same as the component. after that you reach equilibrium. -> to solve the problem, you need to dissipate heat absorbed by heatsink into surroundings-> for air cooling, you must dissipate it into air -> larger surface area => more dissipation as you've more area for heat transfer -> thus we use fins -> fins transfer heat to surrounding air -> must avoid same equilibrium problem -> must move hot air away from heatsink and provide it with cooler air -> hot exhaust air now in the case, must remove it-> exhaust case fan throws this stuff out.

This is all perfectly correct.

I think that the previous discussion would make sense if you replaced cool with not-warm, moving air feels cool because its moving and the latent heat of evaporation being liberated from you hand cools it further. The air in reality would be warmer than ambient. Anything that is warmer than ambient should not be described as cool. It was almost a semantic argument.

 

[FinneousPJ]

B is wrong I am sorry if I was coming off as rude earlier but you are not being helpful since this is a science project.

The fan blow the heat off of the heat sync this the air coming off of it is warmer than it was upon intake *truth

No, B is not wrong. Obviously the air is going to be warmer. I never said anything to the contrary. I'm trying to say if the air coming out your heatsink is hot, like 100 centigrade for example, then your cooling is insufficient. I think you're trying very hard not to understand my point.

 

[spentshells]

No, B is not wrong. Obviously the air is going to be warmer. I never said anything to the contrary. I'm trying to say if the air coming out your heatsink is hot, like 100 centigrade for example, then your cooling is insufficient. I think you're trying very hard not to understand my point.

Think you are simply just missing the point of what is supposed to happen when trying to do a project that involves being accurate

 

[FinneousPJ]

I do have a reasonable amount of experience with this kind of thing.

 

[BVKnight]

The point is, for the purpose of this guy's experiment, he has to delineate the types of air going into and coming out of computer components. There are only two types for this experiment, cool and warm. Cool air is taken in, and warm(er) air is pushed out. Spentshell's objection to your idea, rightfully so, is that labeling the air that comes out of the heatsink as cool would be technically incorrect and seriously jeopardize the point of this project.

I think we all understand the point you made about heatsinks that exhaust very hot air being a warning sign that something is not right. But for the purposes of this scientific project, that air can only be labeled as "warm/hot" for the system to make sense.