WB-design for direct pelt cooling *revised*

[ZaVkE]

new design:



This is the new design.
The small cilinders in the waterarea don't restrict much but still put high pressure on the pelt.

So this time i have a large (practically unrestricted) waterflow that make contact with the entire hotside of the pelt (except for a small edge) so the pelt should be cooled fine.
There is al lot of pressure on the pelt wich can be adjusted by the 8 holes. The pressure is evenly spread over the entire pelt so he should function good.

Could you please give some comments or suggestions about this design.

thnx

[bigben2k]

That'll work best in a high flow rig. Reach for 200 gph.

[ZaVkE]

i have a eheim 1250.
But i don't know the flowrate in gallons, it's 1200 l/h.
Will that be enough?
I don't know how mush liter goes in a gallon, so im unable to calc it.

[bigben2k]

Hum... it's a simple design, in that it has little flow restriction... You might be able to reach 200 gph with a 1250 (rated to 317 gph @ 0 head)

(3.76 liters per U.S. gallon, I know, I'm from Canada...)

[ZaVkE]

Thnx

[bigben2k]

What's your rad setup (which one(s))?

[ZaVkE]

For the moment it goes from pump -> CPU -> bong -> friged rad (old car rad) -> pump
but im planning to make a split to the chipset and my GF3 ti 500.
CPU is 2100+ xp

[jtroutma]

Your bong cooling then going to a fridge?

Either you want absolute best temps or you dont want to risk thawing out your beer.

[ZaVkE]

its the first one

[Miss_Man]

Nice. The few mm should be lesser than the thickness of pelt so you can clamp and have space to seperate cold side and hot side.

[bigben2k]

Given that the bong can reach sub-ambient temps (in theory) and that the rad can't, I'd send the flow through the rad first, then the bong.

[ZaVkE]

@Miss_Man: I was already planning to do that.

@bigben2k : Don't forget that the rad is installed in a small fridge 'fridged rad', so it will cool more below ambient then the bong.
So the loop is correct or isn't?

[Dix Dogfight]

ZaVkE:
If the rad is in a fridge then yes. It should be after the bong.
Otherwise it's like BB2K said, before the bong.

Here is another thought:
The bong is a ecellent way to loose heat but what about it's ability to add heat? My guess is that if you run water through a bong that is cooler than what the bong itself can cool it down to. The bong will heat it up to its lowest temp. And because it's much more efficient than a rad it will win, the cooling/heating process, over your rad in the fridge.

To sum it all up:
The rad in the frige will perhaps be able to cool the bongwater a degree or two but when it comes back to the bong it will heat up again, (if its colder than what the bong can do). What good will that do in a WC-setup?
Maybe give you a few MHz more but more importantly your fridge will run 24/7 at max trying to cool the room where the bong is. And that costs electricity and your fridge will die a premature death.

just my 2 cents

[ZaVkE]

Very intresting thougth.

But i don't think i will have this problem because the water cools down 2 172W pelts , a GF3, chipset and 2 Hdd's.
Under normal conditions the water will reach the bong pretty hot, so the problem you mentioned will not happen i think.

[Dix Dogfight]

ZaVkE
I don't wan't to ruin your day and be a pessimist but have you done any calculations on this?

I did a rough calc and came up with this.
(all numbers are fabricated but represents a average of WC systems Iv'e seen on the net, take them with a grain of salt OK)

Lets say you run a normal wc setup with the water 5C warmer than ambient, with a 100W load. And you have a pump that gives you a flow of 5L/s. When the system runs at equilibrium the rad dups 100W into the air.

Since the rads ability to dump heat energy into the air is dependant on the temp difference it's ability to dump heat is much bigger in a fridge due to the bigger temp difference.

The same setup with the rad in a fridge with a 50C temp difference (lets go to the extreme). The same rad will then be able to loose 1000W of heat instead, (OK I'm extrapolating the rads behaviour linearly and this is not the case in the real world but it will show you what I'm talking about).

5C => 100W
50C => 1000W

A water flow of 5L/s and loosing 1000W of heat gives you a temp difference of: 1000/(4180*5)=0.05C

To sum it all up: This rad in the fridge will produce a tempdifference of 0.05 degrees celcius from the time the water enters the rad untill it exits.

If you build a bong that is so good at cooling, that it is able to lower the temp of the water to the same temp indifferent of what heatload you throw at it then the fridge is a bad idea.

However if the bong is totally worhtless then the rad in the fridge will take ower as main cooling device and really lower the temp. As a normal rad does when it operates in the room. Yours in the fridge will only have a slight advantage due to the colder air.
This of course puts your fridge to the ultimate test, and I don't think it's good enough to stand for 2 172W pelts , a GF3, chipset and 2 Hdd's.

But hey I might be wrong after all.

cheers