inexpensive chiller for a nano

[psteeleb]

I found some inexpensive thermoelectric chillers at an electrical surplus / discount place. I'm going to buy a couple of them and see if I can come up with an inexpensive chiller for my seahorse tank. The thermoelectric chiller/heaters are the type used in those 12v powered fridge / ovens coolers. They are also used in automobiles to chill a drink or heat up a sandwich.

I may just mount a couple to the back side glass and let the glass be the cold side heat sink, or, I may look into adding a probe of some sort and immerse the probe directly, maybe glass or titanium. If I go titanium, I’m thinking about tearing apart a titanium heater and incorporate the tube somehow (immersed or flow tube). Anyone have any thoughts on this or seen anything?

 

[mylittlesquishy]

Pete,

That would be great for us nano tank owners that battle the temp swings. I am affraid to try seahorses in my 12g nano due to this.

I have one of those small coolers that you plug in to the car? Is that what your talking about? If so, we never use it, so I would tear it apart.


Gail

 

[psteeleb]

@mylittlesquishy wrote:

that's the type

heres what I'm looking at:

the thermoelectric chillers are about $15 without shipping etc. this price does not include any heat sinks

there is another one with heat sinks (and I'll need at least one on the hot side) are $35

If I go without the heat sink I may be able to find some cheap heatsinks used for computer cpu's

my initial thought is to just stick the element to the glass or plastic back of the nano and add the heat sink (and maybe a small pc fan). I'd need a 12v transformer, those are usually less then $10, and I know I have at least one laying around somplace.

the only think left would be a controller to turn it off. But if it works and doesn't get too cold I may just leave it on all the time in the summer and let the heater kick in if needed.

So the cheap / easy version I would be looking at;

$15 for the device
$10 for a hot side heat sink
$10 for a 12v transformer
$35 bucks (assuming it works)

If sticking this to a glass back doesn't work, I'll have to go with some sort of cold side heat sink to get the cold into the water. This is where I'm thinking of a titanium tube from a heater may work

not sure when I'll get to this, as it's just something else on my list

when I get all the part togehter I'll test it on a small glass tank filled with water and see how much it brings down the temp.

 

[Firewolf4]

Sticking it to the glass wont work. Glass is a thermal insulator. If you can mount it to a titanium tube from a heater then pump water through the heater you should be in good shape. Keep in mind that the water flow through the titanium tube will have to be low as you're dealing with thermal mass vs temp. I would use a minijet or microjet to pump the water and use thermal epoxy to glue the tube to the heat exchanger/peltier device.

 

[HSLee]

i agree that glass is very poor thermal conductor. also, check the requirements on the power. most of these units draw at least 2-5A DC power. you may not be able to get such PS for $10.

there is at least one commercial nano chiller that uses this technology. i forgot the name (Ice Probe?) some online aquatic supply stores carry the line.

 

[LEDfreak]

For 12V DC try a computer power supply. They usually output somewhere in the 10 to 20A range on the 12V rails. I use one to power 19 HO 3W LEDs on my 12 gallon nano. (Also something to consider to save heat rather than chilling) Mine is a little overkill but I am growing SPS in it.

-Brad

 

[psteeleb]

thanks for all of the feed back

I don't want this to be an expensive chiller and don'r mind experamenting with $35 bucks or so. But if I have to spend allot more, I'll just buy one of the Iceprobes"

I'm just looking for some cooling capacity and will sacrifice the poor thermal conductivity of thin glass to test this. If mounted with thermal paste and insulated for a couple inches around, the cold has no place else to go but through the thin glass and into the water. the question is how much cold is generated by these and how much of that cold will penetrate the 1/8" glass.

The point made on the AC/DC transformer is a good and probably can't be avoided. At first I was thinking about using a surplus transformer like those used for printers etc. But now I think I'll test this with my heavy duty electric cooler transformer and if it works then look into testing with a cheaper surplus unit. Or, I do have a couple extra computer power supplies . good idea

thanks

 

[dmarkham]

@Firewolf4 wrote:

What about Stainless Steel?

 

[Ashlar]

Stainless Steel.. isn't actually stainless. I'd go titanium.

 

[psteeleb]

@Ashlar wrote:

I agree with Scott. Even a high grade stainless will pit and corrode in salt water. Titanium is the best solution but is also a cost driver in all chillers. I bet the titanium coils are a good 25 - 30% of the chiller cost (maybe more)

I have a titanium heater that the controll unit went out on, so I have one tube available, so, can go that route if needed. But for the starting fresh DIYer this may be an expense that pushes the cost over the edge.

I'm still in the conceptual stage, looking for feedback and ideas/suggestions. Thanks to everyone that has provided their thoughts and ideas to date. Heck, I was even trying to figure a way to work in one of my surplussed CPU heatpipes as cheap nano chiller.

I hope to be able to get to this before the summer months

I may even try the version where you buy a under counter fridge and run a coil of tubing in it. the issue is you have to keep the water running all the time (basically over cooling) and control the temperature in the tank with a heater. Vinyl tubing also is poor at thermal transfer so you need a lot of it. But on a positive note; I guess you could use the fridge for other things (beer or fish food)

 

[dmarkham]

@Ashlar wrote:

It is the nickel in the stainless that determines the grade. There are also high nickel alloys that work better than titanium. All of the salt companies I deal with ( Cargill, Morton,and US Salt )use Stainless not titanium. If you have something to use already then go with it. I would like to see the final product.

 

[psteeleb]

@dmarkham wrote:

It is the nickel in the stainless that determines the grade. There are also high nickel alloys that work better than titanium. All of the salt companies I deal with ( Cargill, Morton,and US Salt )use Stainless not titanium. If you have something to use already then go with it. I would like to see the final product."]

thanks and that makes sense. I wonder how high a grade one would need and how expensive a short piece would be.

 

[dmarkham]

Tubing or Plate? I may have some drop pcs around here. :lol:

 

[LEDfreak]

I am trying something really similar. We should combine our efforts. I just started a thread about this today as well, and quickly realized the value of the thermoelectric modules. I think I bypassed this info when I first read what you were doing. What do you think about taking 1/4" titanium tubes and building a heat exchanger by drilling holes through an aluminum block 4" X 4" X 2" thick. using thermal grease or thermal epoxy to glue the titanium rods in and attaching a thermoelectric module with a heat sink and fan to each side of the aluminum block so I am building an ice sandwich. Pump water through the tubes and back into my tank. I could probably build a compact 80W thermoelectric heat exchanger in a 4X4X6" envelope. Just need to connect the tubes and add a micro power head to them. (Mini jet, Micro jet)

Any oppinions???

 

[LEDfreak]

Like this...

http://www.customthermoelectric.com/assemblies.html

 

[HSLee]

Check with Tanner's Electronics in Carrolton (?). Last year, they had a bunch of surplus units for sale. The units had the heatsink already mounted. Don't remember the price... maybe about $10.

 

[LEDfreak]

May have to swing by there and see what they have. I posted a ling to a site wher you can buy them on my thread. They have units that can remove as much as 200W of heat - no designing an efficient heat exchanger to go with this is the key. I still like the idea of a aluminum block with titanium tubes running through it surounded by thermal grease.

-B

 

[psteeleb]

that should work pretty good - but see now you got me re-thinkiing this :?

 

[LEDfreak]

I need some help picking a TEC. How much research have you done? What would be the best way to build the heat exchgr? Long straight bar, or short stubby block? I am leaning toward the second. I could also insulate any exposed sides of the heat exchanger with the rubber dip you use for hand tools. This would ensure that the exchanger only absorbs heat from the water running through it and not from ambient air. (Efficiency is key) Willing to meet you at a LFS to chat it over sometime.

-B

 

[psteeleb]

here's what I'm looking at

http://www.allelectronics.com/cgi-bin/item/PJT-10/775/LARGE_THERMOELECTRIC_DEVICE_.html

the easiest thing would be to set this in a tray with the cold side down and run tank water in and out of the tray through the cold side heat sink and add a fan to the hot side to move the heat away. This is a risky approach and I don't like the thought of exposing the aluminum long term.

My inital thought, and I may still try it, is to remove the cold side heat sink and thermally glue the unit to a thin wall glass aquarium and insulate around it. there wil be some cold themal transfer, but as stated not very efficent through glass. Reardless my thought is, even if it's not super effecient but works, presto a cheap nano chiller.

Routing water through a tube or set of tubes would be more effective, but add time and money to the project.