Jet impingement with larger cooling area?

[Aleck]

Is there anyway of utilizing jet impingement in a wb to cool down a larger area ?

For instance 40mm x 40mm (size of peltier )

[Cathar]

Yes.

[msv]

Zum Beispiel

http://forums.procooling.com/vbb/sho...5&pagenumber=1

regards
Mikael S.

[Aleck]

Did the block made by MMZ_TimeLord performed..?

Could it beat blocks that utilize other concepts such as Rotor.?


Wondering why there is know feedback on the thread posted by MMZ_TimeLord..

[Aleck]

Well... it seems nobody is interested in this topic ...

Or maybe it has been discussed before...?

Btw ..I'll drop in a few questions here that i hope will get the attention of some one...

In order get a jet impingement wb to cooldown a larger area, we may choose 2 aproaches.

The first is to add more nozzles but shrink the diameter of the holes of the nozzles so that the cross sectional area of the new holes are the same with the initial holes on the nozzles of the original cascade design( to maintain the same velocity of the fluid)
This approach may reach a limit as to how small can we drill the holes in the nozzles. The disadvantageous of this approach is that as the nozzles get smaller the resistance of the wb will become larger, implying greater preasure drop

the 2nd aproach is to stick with the same number of nozzles and spread out the jet array so that it covers a larger area. This approach may have some disadvantage of the cooling area being localized only at the spots where the jet is positioned.



In the case of spreading out the jet array in a cascade to accomodate a larger cooling area, is there need to widen the cavity at the baseplate to get maximum area that reacts with the boundary layer created by the jet impingement.
Or just leave the cavity diameter as it is and just spread it out according to the position of the spreded out jets.

[leejsmith]

the ID of each tube directley effects the size of cup it can cool.
if you keep them the same size and spread them out it will leave hot spots that are not cooled enough.

if you look inside the cascade you will see the ID is already very small and going any smaller will also reduce the cooling effect of each tube so you could end up cooling only a fractionaly larger area.

to cool a larger area i would try larger tubes with larger cups but designed to be used with a bigger pump.

at least thats what i think anyway

[bigben2k]

You do have a few options. If you're trying to cool a pelt, you may do well simply by making fins in a radial pattern, around a Cascade's jetted holes.

If you're just shooting theory, and looking to optimize the design for a pelt, I'd stick to the same exact jet tube pattern, but consider giving the coolant more room to exit upwards. In other works, make the top sit a bit higher.

[Aleck]

Originally posted by leejsmith

to cool a larger area i would try larger tubes with larger cups but designed to be used with a bigger pump.

at least thats what i think anyway

So you are saying that to have a bigger cooling area , a bigger pump is needed

The problem when we start using bigger things is that other things would have to be bigger in order to get an optimized configuration. This is what we call proportion.

If we simply change only one feature of a proven design without looking into other features that have direct effect of the change we will start getting diminishing returns

What im trying to say is that if you use bigger holes with bigger cups, we will need a larger inlet to suply the nozzles with enough fluid without accelarating it (if the velocity at the inlet is to high, preasure drop will occur before the fluid goes through the nozzles)

We will also need a larger chamber to direct the spent fluid out

To make it simple , what you are trying to do here is just blow up the scale of the cascade

[bigben2k]

Gotta agree here, to this extent: the walls between the cups act as fins, which are essential to extract the heat in the high density of the area over the core.

With a pelt, I'm pretty sure that the heat density is not going to be the same, but I would certainly bet that it's not uniform. 1/4" spacer? (aka cold plate)

[Aleck]

Originally posted by bigben2k
Gotta agree here, to this extent: the walls between the cups act as fins, which are essential to extract the heat in the high density of the area over the core.

With a pelt, I'm pretty sure that the heat density is not going to be the same, but I would certainly bet that it's not uniform. 1/4" spacer? (aka cold plate)

So you are telling that a flat base plate is suficient for peltier cooling using jets..?

[bigben2k]

No.

With a pelt, you'll need a cold plate, which would typically br 1/4" thick, and should help disperse the heat, right before it goes through the pelt itself. Once the heat hits the water block, the area should be much larger that the die.

How big it turns out though, I really don't know. I'd venture a guess at up to 1/2" extra, all around the original size of the die, at best.

[Aleck]

Originally posted by bigben2k
No.

With a pelt, you'll need a cold plate, which would typically br 1/4" thick, and should help disperse the heat, right before it goes through the pelt itself. Once the heat hits the water block, the area should be much larger that the die.

How big it turns out though, I really don't know. I'd venture a guess at up to 1/2" extra, all around the original size of the die, at best.

I know a pelted block needs a cold plate....

Im refering to the base plate ...where the jet impinges on....
Is there a need for the dimpled cavities..?

or a flat plate woud be better...? (in a pelted block)

[Cathar]

Originally posted by Aleck
Im refering to the base plate ...where the jet impinges on....
Is there a need for the dimpled cavities..?

or a flat plate woud be better...? (in a pelted block)

Better heat transfer simply means better heat transfer, no matter what scale it is. If something that works well for a small area can be scaled up to a larger area, then that will work well and be preferable there too. Need to be careful about managing flow velocity and flow resistance when one scales up designs though, but basically the same principles will apply, no matter the scale.

[rocketmanx]

Cathar, would the jet velocity not also have to go up with cup and jet size? Seems a bunch of very tiny jets with very thin walls between cups would be ideal.