Peltier Cooler

[Rain_Frog]

Some of you DIY folks recommend using epoxy to bond to acrylic. How strong is it really? Has anybody actually used it to stick to acrylic? (basically, some first hand accounts). Will it be strong enough if you "yank" it off? I've sanded acrylic a ton before, and silicon will still peal right off with your own hand strength.

I've only had experience sticking epoxy to epoxy (when making coats for backgrounds).

I'm planning on cutting a square in acylic and patching it with a square of titanium sheet metal, in which a peltier can attach to the other side. All I need is a leak or a seal failure and then I cause a fire or electrocute the frogs. :shock:

[defaced]

I'm pretty sure all silicone sealants won't stick to certian polymers. My DAP Aquarium Silicone says it won't stick to methylmethacrylate, polycarbonate, polypropylene, polyethylene and polytetrafluoroethylene. I'm almost postitive acrylic is a polycarbonate.

What about a mechanical link, say like some stainless hardware to hold the ti to the acrylic? Does the peliter work on DC? If you have a drawing of your idea I might be able to help more.

[Rain_Frog]

You know, I was thinking about the "sump box" being made of stainless steel, but cutting through metal I have no knowledge of. I don't know if I can get a large enough sump box out of grade 321 stainless, the kind used for beverages. Obviously, the only advantage of using a stainless steel sump is because you can easily bond epoxy or silicone to it. However, I would have to insulate it with styrofoam.

Another possibility, but less efficient, is creating a heat exchanger box or water bucket. My other, compressor chiller has a water tank where the cooling coils sit in and a titanium heat exchanger comes into contact with it. I did this because aluminum would corrode coming in contact with acidic water. Another idea is to make a "fridge box" around the cold coils to keep things cool.

Possibly, I could hook up a "fridge box" (beer cooler peltier idea) to a peltier so I wouldn't need to worry about bonding epoxy/ silicone to acrylic, but it isn't as efficient. However, at least I could take advantage of the cheaper Ti tubing (not to say that's much cheaper either).

Problem, if water flows out through the sump, through a titanium heat exchanger in the fridge box, how would it reenter the vivarium? I have no problem with the big chiller, as a temp controller powers my Rio pump whenever the water needs to circulate through the water sump, and back to the tank.

Still, epoxy is a great way to go, but its so expensive and taking the time to dry it and cutting wood for a homemade sump is time consuming. It took me several weeks to finish my DIY epoxy background.

Anybody used the Tough as Tile or the bathtub paint (by Liquid nails)? Its a two part epoxy paint that sells for $25 for coating your bathtub.

[defaced]

Right now the image isn't showing, I'm not sure if that will change later. If it's still broken I'll fix it. Anyway, what I'm picturing is an insolated box with a small heat conductive area to which a peliter will be attached for the purpose of cooling water. Is that correct? Where will this water eventually go?

If that is the case, I can see some problems from a theoretical standpoint. A peliter does a great job of cooling, infact too great for use with water. At standard conditions your peliter will quickly turn your chill box into an ice box if you don't have a large enough flow. If for some reason water flow stops, and your peliter keeps working you could end up with a busted chill box because of the water-ice transisiton.

As you said, SS is a bit of a pain to work with and it's a conductor so you won't be cooling your water as much as you could. Have you considered Gorilla Glue, with the addition of some SS hardware at the corners of your TI plate? Perhaps a plywood box coated in polyuretane wood finish.

[Rain_Frog]

If that is the case, I can see some problems from a theoretical standpoint. A peliter does a great job of cooling, infact too great for use with water. At standard conditions your peliter will quickly turn your chill box into an ice box if you don't have a large enough flow. If for some reason water flow stops, and your peliter keeps working you could end up with a busted chill box because of the water-ice transisiton.

Yes, that is a problem...if you don't have two temp controllers. My other chiller has two temp controllers, one cheap one for beer, and one fish type kind to monitor the tank temp (on the big tank). However, I haven't incorporated the "chill box" yet, instead, I used a rubbermaid container filled with water with the cold coils submersed. Probably though, it is more stable and less strain on the dehumidifier than a chill box surrounding it. However, that a chill box would be certainly less messy than a big tank full of water and probably better for the cooling coils.

I think one problem with this design, unlike my freon based chiller, peltier's hot side will most likely traverse to the cold side, so a standard temp controller will not work (you would cause the device to stop, then start up again, thus having to recool itself after some heat migrates to the chill box). A special temp controller for peltiers is needed (like the one for the iceprobe).

Probably the best idea is to make a carbon copy of an iceprobe. I have drawn out plans to make one, but when I have calculated the cost for all the individual parts, it turns out to be no more expensive than the real thing, and STILL no temp control included in the price. Instead, you would slip a piece of Ti tubing through a bulkhead and seal up the gap between the Ti tubing and bulkhead inside. Inside the titanium tubing itself, could be an aluminum rod that connects to a small aluminum block, attached to the peltier itself.



[/url]

[defaced]

I've got a few ideas, but I'll need the basics to run the math.

Ok, so what are your design paramaters?

How much water do you need to cool and by how much?

For what purpose is this water being cooled?

Where is your ideal location for the chilling device?

What materials/tools do you have available?

[Rain_Frog]

Ok, so what are your design paramaters?

snip: How much water do you need to cool and by how much?
For what purpose is this water being cooled?

I'm going to start with the firebellied toads. I'm only needing to cool a couple of gallons for my 10 gallon tank, for starters. They are most important to keep cool. They tend to darken and become more nocturnal when temps reach above 75, not easy to maintain in my room. I don't want another frog tank in my crummy, dirty basement.

snip: What materials/tools do you have available?
If I find a place for cheap epoxy, and an easy way to cut wood, all I need is a titanium plate, a powerhead, and a peltier/ heatsink/ temp control combo. A cheap acrylic box is what I had in mind, if I put screws in to tighten it (using silicone like an O ring) but the drawback is that acrylic warps. I'm afraid that if it warps, then I'll spring a leak and start a fire or electrocute the frogs... that's why an epoxy box i think is safer because you can bond epoxy or silicone to the metal, that way. There is another option, using an actual glass tank. Perhaps a glass place can drill a hole that large for the peltier plate? (about 3-4") That'd be a good idea.

Where is your ideal location for the chilling device?
Two scenarios. The ideal scenario, is a sump. Ten gallons aren't easy to drill, or at least drilling glass. Plus, then I need to remove another frog tank because the bottom shelf must be used for a sump. My mantellas are on the bottom shelf, and a sump underneath, but often the water splashes on the iron stand and for sometime (thank goodness I caught it in time!) rust and paint was getting into the sump water. A sump that narrow in height isn't recommended either for mounting peltier devices.

Then, there's the hang on box idea. My powerhead/ filter should be on all the time, unless it breaks. I guess you might be a little concerned if the powerhead fails, and the chill box overcools, it could burst, but at least I wouldn't need to worry about drilling a 10 gallon (However, the microchiller has a peltier inside of it) A sump is safer.

Perhaps I should hook up a cheap reptile temp controller inside the hang on box, just as a safeguard if the pump fails? (so it will override the other temp controller that monitors the water temp)

Another idea, is to drill directly a large hole into the tank, but not sure if you can do that cheaply. But then there is no need for epoxy/ acrylic, etc. Just be careful that the cold plate is always submerged and sectioned off (perhaps behind some rocks) so the frogs will not come into contact with it. It is probably the most efficient, and is like how the iceprobe itself works.
If this was used for dart frogs/ mantellas, perhaps allowing wet soil to come in contact with the cold plate would be sufficient, like heating cables are for growing seedlings. (perhaps a titanium tube should be placed under the soil as well to spread it out).

At this rate though, it seems that its best just to go searching on Nanoreef.com and ask if anybody has used Iceprobes. You can get some good deals for them and the temp controller.

It either that, or I ask my local glass shop if they can actually drill into my tank (or a glass tank sump) a 3" hole. How much would that cost? I have an old 5 gallon that I hate that I should get rid of anyway some how. Perfect for a sump.

Doug

[defaced]

I'm really thinking either a completely different approach is going to be needed, or a iceprove is going to be it. Without drilling the tank I really can't think of a way to get this to work easily, unless the pump you plan on using is going to be inline, not an inexpensive submersible one. I've thought of replacing the bottom the tank with metal and putting the peltier uner the whole thing, replacing the back with acrylic to make it easier to drill, using your 5g as a sump. The list goes on.

One thought about your temp control problem though. Because of how a peltier works you should be able to just reduce the voltage it's getting and thus reduce the peltier effect. Now this could be done with a simple potentiometer if the current draw of the pletier is very little, or with a computer fan controller (20 bucks) that uses more advanced techniques to control the voltage. So maybe just getting your peltier to say 36F on the surface then going with a more simple water delivery design will be the ticket.

[Rain_Frog]

The other day, Marcos (blort) came up with a good idea, and sent me a drawing in a file attachment. Basically, a wide PVC pipe is capped with titanium and the peltier placed on the other side. The original idea did not have a true way to vent the heat, but you could connect a side pipe to the PVC, and a fan is placed on the top of the main pipe. Therefore, the fan draws air through the side pipe around the hot side of the peltier and out the main, like a chimney. I'm sure you can mount a small heatsink inside the wide tube of PVC. The larger pipe prevents the peltier's heat from potentially damaging it (if the pipe was narrower). And yes, I have seen peltiers offered on ebay WITH the temp controller for less than 20 bucks.

I could buy the peltier/controller, the power supply, and a small bit of titanium as well as a fan for probably less than $60. (the main expense comes from titanium, but with that much titanium, I could make several more small devices for probably less than $40 if i wanted). Epoxy would be used to seal the titanium to the PVC. You could just stick the end of the pvc into a sump.
I find this method very efficient, but what would the level of danger be?
The main flaw is the level of danger involved. Al it would take would be a small leak/ flaw, or when the device ages and water could rush in from the bottom of the pipe up, electrocute the frogs and start a fire, but perhaps i'm too worried about it?

Perhaps if you get a small, hollow, aluminum/copper pipe and attach it to the titanium, that could work. Therefore, the peltier device itself would be attached to the aluminum pipe/ bar (thus, would be above the water line), in case the device would fail.

One question though, what's the best way to attach the peltier to the metal? Would I need to attach the aluminum bar/ copper to the titanium? I'd assume gravity would be enough to have enough contact.

[defaced]

This is going to sound odd, but without knowing the resistance of the frogs, there's really no way of telling if a failure could be of danger to them. Generally anything below 40v is considered safe to humans, but since frogs most likely have more water in them, they will conduct electricity better, and that number could go down quite a bit.

What I would do is read up on GFCIs and see if they will work on a 12v DC circut, or just experiment (post your finding, I'm interested in it too, just haven't had the time to search or try). That's the only way to be sure your frogs stay safe. If something happens and a short occurs it will cut power to the peltier.

As for attaching your peltier, you will need a thermal paste http://www.newegg.com/Product/Produc...82E16835100007 used for computers and some way of solidly holding it to the Ti. Gravity won't be enough, and peltier's aren't heavy at all. The thermal paste will ensure good thermal conductivity to the Ti and your heatsink, which you must have as well or you will fry your peltier. You'll probably also find that you'll need a fan directly mounted to the heatsink to be able to move the heat effectively. Even then I'm not sure it will be enough because with the peltier inside a confined space it will be difficult to get cool air to it and thus it will overheat.

Making the pipe longer and putting it between two pices of Al or Cu would be a good solution, like you mentioned. Something like this:
Say and 1-2 thick piece on the bottom and a 4-6 inch long piece on the top, both big enough to cover the whole peltier. This has no moving parts and keeps the peltier above the water line. With the larger piece on the top the heat will disipate fine and with a thiner piece on the bottom more heat will be pulled from the water.

Here's a good deal on some Ti: http://cgi.ebay.com/6Al-4V-Titanium-...QQcmdZViewItem

This isn't bad either, but it's probably overkill and would be difficult to cut without a good metal bandsaw:

http://cgi.ebay.com/6Al-4V-Titanium-...QQcmdZViewItem
[/img][/list][/url]