Water cooling booster.

[Unseen]

Ok so here is our new project in outofspecs.gr . I know its not too extreme but we wanted to try sth like that. I cannot call it chiller because it is not.Watercooling booster suits more i think.

Check the pics:











You can see more pics here -> http://outofspecs.gr/forum/showthread.php?t=3014

As you can see i am still testing the prototype and at the moment i am using 3x135w tecs @ 9.5v (the energy meter shows ~200w in total). There are many things that we can improove of course but we had to start from sth.

Waiting for your suggestions/comments

Chris

[Serpentarius]

i like the way they symmetry design of it
it's very uniform-like

[zipdogso]

I cannot call it chiller because it is not.Watercooling booster suits more i think.

Looks to me like a perfect example of a TEC coolant chiller.

Why do you say it's not a chiller what's your definition ?

looks like it is working quite well too if the data is to be believed.

[cooling storm]

looks nice give us the temps.

[lalPOOO]

Looks pretty cool, I bet you could get a bit more efficiency if you insulated the cold side loop.

[demonkevy666]

Looks pretty cool, I bet you could get a bit more efficiency if you insulated the cold side loop.

needs shoulder washers too on the cold block too.

[flak-spammer]

That is definitely a water chiller, I would suggest that you insulate as much of the cold block as you can because that's where a lot of your losses are going to be.

[cooling storm]

jes it is a smart thing to insulate it also to prevent condens and frost on the coold side

[cooling storm]

u shod use thise tec http://cgi.ebay.com/ws/eBayISAPII.dl...m=280419123557

[zipdogso]

u shod use thise tec http://cgi.ebay.com/ws/eBayISAPII.dl...m=280419123557

Why should he ? Oh yes 545w...

It's the biggest rip off this century.....

That super duper extra high quality mega wonder actually only has a maximum cooling power of 266w. It is a 12730 and a general 12730 is only rated a 266w. That's only at max. input power with a Dt=0 i.e. no cooling what so ever. At 12v (It is actually a 15.6v unit.) you are probably looking at a shade over 200w with good TIM and the correct compression. Less if your thermal transfer inefficiencies are high.

Ebay seller's marketing hype.....looking around for the nearest mug....

and if they used 4 of those 12709 in that unit it doesn't look big enough to have 4 of these in it. 12709 are generally 40mm square, these are 62mm square units.

Granted they will be better than the TEC's in that unit but there are better TEC's than those rip offs around if you really wanted to go better.

[lalPOOO]

Check out the hands holding that tec in the picture. Those are either some really girl mans hands, or some really manly girls hands... definitely not the person you wanna buy a tec from!

[Unseen]

Hi guys

I am in a middle of tests at the moment. I will give you an update soon.

[cooling storm]

i wish there was a 1000w tec

[zipdogso]

i wish there was a 1000w tec

As a single TEC....Not in a million years, not for general retail anyway....you'd need one mighty hell of a PSU for it and with potentially 2Kw heat coming of the hotside a super duper cooling system.... No, the extra bits neccessary to run something like that, and the cost of those bits, would be far to prohibitive to make it worth while.

So long as you could cool the hot sides and it would have to really good you could potentially build a 1Kw chiller. The input power need not be a great problem because you can join units in parallel and series combinations BUT as I said your cooling would have to be good to cope with a potential 2kw heat from the hotsides.
i don't know how he's getting on but someone on here was at one point going to try this.

[Ultrasonic2]

Round off the internal corners

Make it out of COPPER

Make the flow lengths the same ( ie a re-design )


Nice job

[lalPOOO]

Round off the internal corners

Make it out of COPPER

Make the flow lengths the same ( ie a re-design )


Nice job

I don't know if the switch to copper would really be worth the increase in costs if you've already machined one out of alu, you're talking about a pretty sizeable peice of copper...

[Ultrasonic2]

I don't know if the switch to copper would really be worth the increase in costs if you've already machined one out of alu, you're talking about a pretty sizeable peice of copper...

Copper has almost twice the conductivity than aluminum ..

So if you want to make it better than this the simplest thing to do .. also it solves any galvanic corrosion issues you may have with ALu and Cu in one loop

I know copper costs more but making TEC products Cheap has ALWAYS been there down fall ..

[Buckeye]

Very nice idea Unseen

Just a dumb question here if you don't mind. I see you have 2 of the blocks used here, with the Tec's in the middle. In the video and pictures you showed you are only using one side of this. Do you have plans for the other side or is it not needed if you only use 1 side ?

I do not use Tec's at all, but I find this very interesting to use with Phase/Cascade/LN2 setups for cooling NB & GPU's.

[flak-spammer]

Well since it's a chiller both sides would be needed. If it was converted to direct die (I doubt it could be...too big) then it wouldn't need the cold side it would just need the cooling loop side to dissipate the heat.

[zipdogso]

Very nice idea Unseen

Just a dumb question here if you don't mind. I see you have 2 of the blocks used here, with the Tec's in the middle. In the video and pictures you showed you are only using one side of this. Do you have plans for the other side or is it not needed if you only use 1 side ?

I do not use Tec's at all, but I find this very interesting to use with Phase/Cascade/LN2 setups for cooling NB & GPU's.

I watched the video....

Unfortunately you slightly mis-understood.
Both plates of the chiller are in use - one side of the chiller is against the cold side of the TEC's and gets all frosty - the other side of the chiller is against the hot side of the TEC's and they are water cooling the hot side of the TEC's to keep them cool.
The temperature of the hotside, to some extent, determines the temperature of the cold side so generally the cooler the hotside is the better.
The difference between the temps of the hot and cold sides of the TEC are determined mainly by the input power.
The point to remember here is TEC's are merely electrically operated heatpumps and not chillers per se.
They remove the heat from an object one side BUT you have to remove that heat plus that generated by the TEC from the other side - hence the water cooled hotside.

Whilst TEC's are useful where Phase is expensive, UK for instance, they may not help you much Buckeye, as a seasoned phase builder. Whilst TEC can have a COP higher than 4 a phase unit operating at the same COP will have a far greater cold temperature. So your better with phase....

...Oh sorry, I mis-understood . yeah you could use a chiller like this for your NB & GPU's but you yourself might find it a tad cheaper and dead easy to get a small AC and convert it for slushbox. You will get lower temps more easily, I think so anyway, I've never had the luxury to try it !

TEC's lowest possible is -50ºC but it's one hell of a lot of work to get it there ! Your unlikely to get a chiller down to this. Well the chiller might but the coolant wont be

[malkiewicz]

I'll buy one.....

[Unseen]

Hi guys, sorry for the long delay. The project is still active.
I will do some tests in the weekend.

Some more pics













Unfortunately I didn't have antifreeze liquid, so the water froze.

[Duh]

Lovely.. I love that craftsman .. I was looking the pics into the greek forum but I have just realised that you already posted the hottest ones

[=[PULSAR]=]

Whats the power draw from that chiller setup?

[Boulard83]

What about using only a single ~130W tec with the Hot side on a TRUE and the coldsid on a block for the loop ? this can work and you only need 1 loop. I think this can work and you can still acheive sub ambian Water temp. Im not hoping sub zero wiith this but at least sub ambiant. Lets hope ... 4-5°c for the water temp.