Ice Probes...

Due to certain heat problems in my own tank, Ive been looking at Iceprobes.

All they are is a chunk of metal with a heatsink built in that sits in your tank and provides heat transfer right? Anyone thought of building one? Youd need a peice of corrosion resistant metal, like titanium, and a computer heatsink and fan.

If you could put some kind of sleeve on the probe in fact, you could get away with a more corrosive material, but one that is cheaper, and hopefully a better conductor of heat. Unfortunatly, most plastic sleeves would function as insulators. A condom maybe?

Any ideas on this? Just wondering if theres something in an Iceprobe's $104 pricetag that Im missing.

There are a looooooooooooooot of threads on this. Basically, the answer is no, you can't build one for less than what they're selling it for.

sorry yes you can. all you need is a few heat sinks from CPU. An acylic dow rod. and a few cpu fans. In "theory" you would use rpm adjustable fans and the acrylic rod would act as the titanium probe would. No where near as eficient but it would work. The heat the acrylic rod would ransfer is no where near what a titanium rod would do.

so basically you can't do it then salty. lol. if it isn't as efficient at titanium, then you're probably gonna waste more money than just bucking up and paying the hundred bucks for the iceprobe, right? i doubt that the acrylic could even drop the temp of a 10 gallon tank one degree. the conductive properties of acrylic are just crap compared to titanium and what you would have on your hands is money wasted that could have gone to buying an iceprobe!

but don't let me mislead you...i think that you should make this acrylic iceprobe and test out your theory. if it works....awesome! if not, i get to laugh at you.

Ive been looking into this heavily with a few structural and aerospace engineer friends of mine, and weve come to some conclusions on materials that could be used.

It seems that galvanic corrosion resistance and thermal conductivity are pretty much mutually exclusive... Even Titanium has comparitivly poor thermal conductivity.

The best material would be pure silver, but thats prohibitivly expensive, so Im on a search for a chunk of inconel or monel.

Bronze or copper would work, but would wipe out the invert population. It would be an option for a heavily planted freshwater tank though methinks. Im also looking at some exotic stainless steels that I might be able to get my grubby little hands on. Updates coming.

You guys are all leaving out the key componet that makes the Iceprobe work... the Peltier. They are not just a block of metal and a CPU fan. The probe and the heat sink are isolated from each other with a Peltier sandwiched in between. The heatsink severs to cool the hot side of the Peltier.

For those of you that have not heard of a Peltier, it is a solid state cooling device. Basicly it works off a modification of the Seabeck effect. If you join to dissimilar metals at their ends to form a loop, a current is generated in the loop proportional to the temp difference between the two junctions (this is how a thermocouple works). However, if you apply a current to the loop you can generate a temp difference between the two junctions. A Peltier has many of these junctions sandwiched between two ceramic plates in such away that when a current is applied one side gets very hot and the other very cold.

Jay

How about a copper heat exchanger with an epoxy coating? (The epoxy would protect your inverts from copper poisoning.) The coating would reduce the thermal conductivity, but maybe not as much as some of the other options.

P.S. Awesome avatar DongShenYin!

The Sisterhood thanks you.

http://www.thsengineering.com/surplus/tec.html

The march of progress continues... Still need a decent, cheap, non-corrosive, heat exchanger.

Great, MrYenko. Now you just need to come up with how to power it. Even the smallest Peltiers draw about 4 amps continous, which is much more then any plug-in type inverter can produce. I would suggest looking some place like AES for a small power supply. Thats where I got mine (though it is not small).

Jay

The other thing I would suggest is that if you are going to put the trouble in to build something like this, you take it a step further. And use a water jacket/radiator system to cool the hot side of the Peltier. Unless you can effectivly move the heat away from the hot side, Peltiers become in-effective at best and heaters at worst.

You may also want to consider a multi-phase type configuration. Where multiple Peltiers are stacked in series. This allows for greater temp differentials between the hot and cold side. Of course this assumes you can provide enough current to drive it.

Jay

I saw the rated current draw and damn near fell out of my seat, but Ive got a few converters laying around that should take care of it. Got a few blanks of possible material that Im gonna try. First I have to go get them IDed the old fashioned way...Asking the guy who dropped them off what they are, and why the hell he didnt tag them...

The funny thing is that mounting this thing is gonna be the hardest part for me. I didnt anticipate something like this when I built the tank and hood. :|

As for liquid cooling, I COULD use a CPU glycol setup, but why complicate it. If it actually works in the end, sure, I might try it, but at the moment, Im building for a price point. Namely, less than what the online stores want for an ice probe. None of this is driven by cool factor. Instead, its driven by my inherent cheapskateness. More to come...