Fan on Heatsink

leon · #1 26-11-2009, 03:13 PM

Hi guys, could someone please help with my question, when one uses a fan on a heat sink those one have the fan extracting or blowing into the heat sink.

Many thanks .

Leon

DavidU (Dave) · #2 26-11-2009, 03:17 PM

Blowing into the H sink.

Astrod00d · #3 26-11-2009, 03:43 PM

Blowing will focus the blast of air better and should result in better cooling. Bear in mind that you want the air to flow from bottom to top as well, otherwise you're working against natural convection. Vertical heatsink fins are more efficient than horizontal. On it goes... there's a lot of heatsink theory out there.

Cheers,

Rob

leon · #4 27-11-2009, 12:44 PM

Your all champions guys, many thanks for your response.

leon

leon · #5 27-11-2009, 04:35 PM

Hate to disappoint, guys but you are miles off, it has a 30% more cooling effect with extraction, rather than blowing into the H/Sink.

But thank you anyway, for your comments.

Leon

mithrandir (Andrew) · #6 27-11-2009, 06:54 PM

Quote:

Originally Posted by leon

Hate to disappoint, guys but you are miles off, it has a 30% more cooling effect with extraction, rather than blowing into the H/Sink.

But thank you anyway, for your comments.

I guess blowing causes reduced cooling because they are being compressed and compressed gasses get hot. Sucked gasses provide additional cooling via expansion into the space left by the extracted gasses.

Boyle's Law + Charles' Law + Gay-Lussac's Law at work? PV=kT

Pressure x Volume = Konstant x Temperature

leon · #7 28-11-2009, 11:26 AM

Don't quite know what all that means Andrew, but i expect your are right.

I tested it again today, and it proved right again, with extraction it cooled quicker and the temp went down from 21.5 degrees to 5.3 in five minutes.

Leon

kinetic (Steve) · #8 28-11-2009, 11:51 AM

Interesting Leon,
I have to try that myself!

It's not simply a result of ambient being different to when you tested
it on another day? Or have you done a fan swap today and tested
both directions?

Steve

Menthu Rae · #9 28-11-2009, 11:52 AM

I'm no expert, but I'm an engineering student and have completed a subject on thermodynamics and turbomachines...

You're dealing with an axial fan most likely - which causes negligible (if any) compression of air, as a result, PV = NRT doesn't apply...

What does apply is the thermodynamics of forced convection cooling. Without going into specifics and formulae, in general the effect of "blowing" or "sucking" air onto or from a heatsink will depend on the design of the heatsink, the volumetric flow rate of the fan and the ambient environment (including the enclosure) the heatsink is operating in.

Are you able to post any pictures of the setup? If you can then perhaps we can see why in your case sucking the air from the heatsink yields better heat transfer results than blowing.

Quote:

Originally Posted by mithrandir

I guess blowing causes reduced cooling because they are being compressed and compressed gasses get hot. Sucked gasses provide additional cooling via expansion into the space left by the extracted gasses.

Boyle's Law + Charles' Law + Gay-Lussac's Law at work? PV=kT

Pressure x Volume = Konstant x Temperature

Quote:

Originally Posted by leon

Don't quite know what all that means Andrew, but i expect your are right.

I tested it again today, and it proved right again, with extraction it cooled quicker and the temp went down from 21.5 degrees to 5.3 in five minutes.

Leon

Kal (Andrew) · #10 28-11-2009, 12:39 PM

Quote:

Originally Posted by leon

Hate to disappoint, guys but you are miles off, it has a 30% more cooling effect with extraction, rather than blowing into the H/Sink.

But thank you anyway, for your comments.

Leon

After reading this I would have to say experimentation finds the best answer

multiweb (Marc) · #11 28-11-2009, 12:52 PM

Interesting thread. Yeah I would havd thought that all heatsink would suck the heat off the blades, not blow on it. Will have to check the QHY8...

Starkler (Geoff) · #12 28-11-2009, 03:28 PM

The mechanism of heat transfer is quite simple.

Heat is transferred from the heatsink to the air surrounding it. This heated air needs to be removed by the fan so that cooler ambient temp air is adjacent to soak up more heat from the heatsink. Keeping a stream of air moving over as much of the heatsink as possible will give maximum heat transfer, therefore blowing should always be better.

DavidU (Dave) · #13 28-11-2009, 03:54 PM

That's what I thought.

Quote:

Originally Posted by Starkler

The mechanism of heat transfer is quite simple.

Heat is transferred from the heatsink to the air surrounding it. This heated air needs to be removed by the fan so that cooler ambient temp air is adjacent to soak up more heat from the heatsink. Keeping a stream of air moving over as much of the heatsink as possible will give maximum heat transfer, therefore blowing should always be better.

leon · #14 28-11-2009, 04:37 PM

Well I have had it on the bench and did the fan swap twice, it proved to me that extraction has 30%, or there abouts more cooling capacity than blowing it into the heat sink.

Anyway it worked for me and I suppose there are some out there that don't agree, so be it.

Thanks guys for your responses and suggestions.

Leon

Clayton · #15 28-11-2009, 08:09 PM

Interesting thread Leon,
it shows that experimenting can sometimes produce unexpected and valuable results.

Kal (Andrew) · #16 28-11-2009, 09:04 PM

When I was more into overclocking several years ago I remember doing the same thing on my CPU heatsink - measuring the different temps with the fan pushing or pulling. In the case of a regular CPU heatsink, where the fan points down to the CPU, when you have the fan push the cold air passes over the cooler extremities of the fins before it gets to the hotter base of the heatsink, while when you pull the air through you get colder air passing by the base of the CPU, creating the bigger temperature differential there. Depending on the heat transfer within the heatsink itself, either way could be optimal.

I really think it is down to the application, and the best way to work out what is best is to try both ways as Leon has done.

kinetic (Steve) · #17 29-11-2009, 04:55 PM

Did the test today on my setup and I actually found the reverse
happened. Setpoint of 3.0°C was reached with the fan blowing
down on to the heatsink...5 mins later all things the same, I couldn't
get the fan/peltier arrangement to reach setpoint.......
Same ambient, same 12V battery level, same setpoint etc.

so it seems that Leon's heatsink/fan shape possibly lends itself
for cooling efficiency to improve when reversed.
We should name this the ''Leon effect""

Steve

leon · #18 29-11-2009, 08:49 PM

Na Steve, I didn't know anything about this sort of stuff, and have just been experimenting, however I did experience a substantial difference with the time it took to cool and the rapid temp drop.

I might be completely wrong here but that is what I found.

Leon

citivolus (Ric) · #19 29-11-2009, 08:53 PM

One possibility to consider is that your fan speed may be somewhat higher in one configuration than the other, due to drag effects and pressure on the bearing/sleeve?

leon · #20 29-11-2009, 09:10 PM

Yes good point Ric, did notice taht effect when I had the fan laying on the bench buy itself, the speed did change some what depending which side it was laying.

Leon