Making a Inexpensive Thermoelectric Chiller....Brains Needed

[hquintanilla]

I came up with this idea for a inexpensive chiller...I don´t know if it works, but it seems it will be good to explore the idea.
I wonder if you can adapt a igloo kool mate cooler (http://www.igloocoolers.com/products/active/active.html)cooler as a chiller. I imagine you can drill holes in it an fix two bulkheads and drive water through the inside of the cooler containing those plastic bags with artificial ice. When the correct temperature is reached I´m figuring out you can diverge the water flow or turn of the coolers. I think this type of thermoelectric devices can produce heat as well, so maybe you can use it also as a heater...
This coolers run at $150 bucks in amazon, so maybe with a couple of hundred bucks you can outperforme a 700 chiller...So if it works, well it will be a great idea...
Any comment will be great so replies are appreciated...
Best Regrads...

 

[cerreta]

Hmmm. good idea. i too am looking for an alternative to the expensive units, but I think I am leaning more towards a drop-in unit. This igloo is promising, but at $150.00 you will also need to purchase an AC/DC device since the igloo is designed for DC current. That is the basis of allowing it to heat or cool. In a DC circuit, switching polarity will allow the unit to heat up. I also think the unit is too large for my system.
Here's a couple more odeas I have run across:
1. If you have a refrigerator with a cold water spigot, you may be able to use this to cooling aparatus for your aquarium. The tank needs to be fairly close to the fridge. Not too practical.
2. There are some small table top water coolers that have a freon system that can be incorporated fairly easily into an aquarium. Again, this device is designed on cooling the aquarium water, which means the internal components may be made of metal and won't last long with saltwater.

If you have any more ideas, especially for a drop-in (immersion coil probe) system, let me know.
Cheers,
Scott

 

[mikey98]

I have looked in to chillers a lot. I am starting up a 60 gallon reef tank and because I have high room temp. and plan to use MH lights I am most likely going to need a chiller. Of course I don't want to spend $700.00 on a chiller. Unfortunately the type of DIY ideas do not put out the BTU's necessary to cool a tank. If you only need to pull it down a degree or two it these ideas might work. However when you need to pull down 5 or more degrees like me these small refrigerator type units are insufficient. My only advice is to look up the formulas to figure out how many BTU's you will need to cool you tank adequately. Also be sure that the cooler is not over worked. Having extra power is always better than not having enough. HTH

Michael


If you give me some time I think I can find the formula.

[ February 26, 2002: Message edited by: mikey98 ]</p>

 

[mikey98]

Found it!

Btu/Hr.=gph x Wt. x SH x *T °F

gph= Flow rate of liquid being cooled

Wt.= Weight of cooling liquid per gallon

SH= Specific heat of cooling liquid

*T °F= Difference between cooling liquid inlet and outlet temperature


For example: Let's say the flow rate (gph) of the liquid you need to cool is 120 gallons per hour. Now let's say the liquid you're cooling is water. Water has a weight (Wt.) of 8.34 lbs. per gallon and a specific heat (SH) of 1. Finally, let's say you've determined the difference in temperature (*T °F) between the inlet (supply) liquid and the outlet (return) liquid is 6° F. Now let's plug all of this information into the formula: gph (120) x Wt. (8.34) x SH (1) x *T °F (6) = Btu/Hr. (6004). You will need to select a liquid chiller with at least 6004 Btu/Hr. to remove heat from your process.

if you have any questions let me know I may be able to help more.

<a href="mailto:[email protected]">[email protected]</a>

 

[Anonymous]

hi.
That's an awfully big chiller. 6000 btu/hr is more than 2 horsepower, or 1750Watt.

 

[mikey98]

Hey i know it was just an example to show the formula. I know that the in temp and the out temp only change by a fration of a degree. I think. This formula is used to determa the power needed to cool inustrial cuting liquids and thngs. Mostly used in production cases. But I belive that this formula will work for an aquarium. If not I would love to know. I want to know how big a chiller I am going to need befor I buy one. If any one has a list of how much diffrent size chillers will cool a givin amount of water I would like see it.

 

[mud]

Has anyone investigated peltier junction devices for chillers? I seem to recall thes devices can be daisy chained to produce significant cooling.

mud

 

[DJ88]

A peltier device would only work if it was placed on the outside glass of the tank or sump. If it was placed inside the water it would not be able to draw the heat away from it's one contact plate. So you wouldn't have to worry if it had copper in it.

I have been looking into peltier devices for cooling/heating as they can do both..

haven't had a chance to try it yet but it is a very viable option. IMO

setting up and integrating circuit you can easily control the DC voltage supplied to the peltier to heat/cool the water.

 

[bktocala]

I too am looking into the peltier coolers as used on PC chips. I think they may have the capacity to bring your tank water down a couple of degrees however I have not been able to find components without copper.

Another idea is idea is a liquid cooler. You would use a fan to blow ambient air (72 F) over coils the aquarium water runs through to exchange BTU's from the water to the ambient air. The problem here is you need a tight approach temp (difference between the cooling air the lowest temperature you can get the water)for it to do much good.

As far as figuring out BTU load you must figure on the amount of btu's to raise your tank water temperature from the temperature you want to the maximum it gets to. A BTU is the amount of heat required to raise one pound of water one degree. So if your 100 gallon tank (you want 78 F but your max temp is 84F it would take:
(100 gal X 8.34 lbs/gal)x 6 degrees = 5004 btus. Cooling equipment is rated in BTUPH. Mike98's equation ( Btu/Hr.=gph x Wt. x SH x *T °F)takes this into account but assumes a flowrate. I think if you measure how long it takes your tank to heat up to the maximum temperature you could get a good guess at the btuph. If it takes 3 hours of halides to get the water to 84 F then you need to size the equipment for 5004 /3 = 1668 btuph which matches up well with a Universal Marine Industries 1/5 HP chiller.

 

[bktocala]

You are right the peltier cooler can't toch the water, you would need to use a heat exchanger. The glass would not work well as it would act as an insulator. I have been looking at the heat exchangers used in PC cooling and they are all copper or aluminum. I found a promising heat exchanger at McMaster.
Model 35035K46

Cold Plate 12" X 3-1/2" X 1/2" Surface, SS Tube, Straight Fittings $ 68.51 You could fit sseveral peltier coolers to the plate portion of the exchanger. I am guessing a complete setup could be done for about $100 which is cheap for a cooler that works but expensive as an experiment.

 

[Anonymous]

hi.
The "Ice Probe" was available for a while, and that is the kind of device you guys/gals are look for. No need to DIY, since it is designed to be touching saltwater. Marine Depot used to sell them, but I can't find it on the web page anymore...

These kind of "chiller" is usually rated less than 50W, so it is not useful for anything larger than 20 gal. By daisy-chaining, you will be better of getting a real chiller instead, since it is not much cheaper than a real chiller.

 

[DJ88]

How much of an insulator do you figure it would be? If it was chilling it would pull more out of the sump. and if heating the water would draw that heat energy away from the glass quite well I'd imagine.

Yes there will be some insulation but I'd be curious to see how much.

This is something I am very interested in trying out.

 

[BORECKI]

Mikey98 and others;
The formula you came up with is ok IF you want to run the coller 24/7. Belive me, you don't!
You want to oversize the thing so that at maximum load the coller runs about 15 minutes/hour, no more. And less as the lights go off.
That's the reason that hobbiest that tried bar fridgs gave up. The compressors would break down from the work load.
Boris.

 

[saracd701]

I came up with this idea for a inexpensive chiller...I don´t know if it works, but it seems it will be good to explore the idea.
I wonder if you can adapt a igloo kool mate cooler (http://www.igloocoolers.com/products/active/active.html)cooler as a chiller. I imagine you can drill holes in it an fix two bulkheads and drive water through the inside of the cooler containing those plastic bags with artificial ice. When the correct temperature is reached I´m figuring out you can diverge the water flow or turn of the coolers. I think this type of thermoelectric devices can produce heat as well, so maybe you can use it also as a heater...
This coolers run at $150 bucks in amazon, so maybe with a couple of hundred bucks you can outperforme a 700 chiller...So if it works, well it will be a great idea...
Any comment will be great so replies are appreciated...
Best Regrads...

This is the RVCA Horton Dead Fish Playmate a special cooler collaboration with RVCA, The boundary-pushing, design-driven lifestyle brand.

Get Now

Get free shipping on order 75$ Igloo Coolers Coupons

 

[Czynot]

why not just float a bag of ice in the tank then to use a cooler with ice?

 

[Bobyboy]

The cheapest cooling option I’ve ever found was a use mini fridge off of Craigslist for $25, a cheap AC pump on a temperature controlled outlet and 100’ of 3/4” coiled hose.

drill 2x 3/4” hole in the fridge, run the inlet and output through the fridge.
Silicone it, connect one side to the screen pump and the other to the return area of the sump.

set the schedule pump to just always run, set the fridge to a temp controlled outlet and boom cheap cooler that can easily be modified for bigger or smaller setups.