DIY Mini-chiller

[Keyoke]

what's the best way to detect a short in a sensor? I'm not so much worried about the temp readings themselves, as I am about the wires wiggling a bit in the potting compound and allowing water to get 'sucked down' via capillary action into the sensor. -- that's why I first thought about the silicone filled tube first. But it's awfully bulky and not as sleek looking as I'd like it to be. (course, this whole project is turning out to be a little bulky)

I did heatshrink everything as well - individual wire leads to the pins on the DS18B20, then all the wires, then the whole DS18 and the wires.. then I smeared some silicone on the whole thing. before I either embedded it in epoxy or the tube.

I suppose, since I've hooked this sensor up backwards w/ no ill effect (that I can see) it won't be a big issue, but I don't want to fry the MCU.

 

[Keyoke]

Just a quick update - it's in the tank sans the temp controller while I work out some bugs on the thing. For some reason, when my app gets above 10K in size, it starts doing strange things.

Anyway, it appears to be doing an OK job, but I'm wondering if I need to move to a bigger TEC based on just how the water "feels" coming from the return. I've got a *really* small pump pushing water though it, hoping that the low flow will cool the water down more.

I'll be running tests tomorrow with the fans off to see what kind of difference it makes temp-wise.

 

[Pbrown3701]

Keyoke - cool project!

There are formulas you can use to calculate the optimal water flow through this thing. It would require you to know some info about the heat transfer properties however. There could be some simple experiments you could do to find out this info - or you could just trial and error it...

Following along

 

[MarkS]

Any updates?

 

[widmer]

Tried skimming over the tread, and sorry if this has already been stated, but:

I was looking into building one of these a couple years ago because the peltier plates are super compact, simple, lightweight, reliable, and all that other good stuff I'm leaving out.

But...Their one big drawback in my opinion is the fact that they're something like 1/6 as efficient as compressor based cooling systems as far as what they can do with the electricity you feed them. I'm sure some people aren't as concerned with efficiency, but for me it was a dealbreaker...

 

[BeanAnimal]

Yes... that is about right. Add in the losses due to poor thermal transfer to the water and it may be even worse (up to like 1/20).

On the other hand, for a very small tank with a very modest heat load, they are a simple solution.

 

[Keyoke]

No updates yet. I've got the unit hooked into the tank sans controller like I mentioned before, but I've been on vacation with the family for the last week and a half or so and not had time to work on it. I've got one of the kids next door watching the tank for me right now, and he's telling me that the temp is under control right now. Since I'm in Florida at the moment and not there to monitor, the 150W MH on the tank is off, and I switched my lighting to dual 10K 32W PC's .. it's not the best solution, but I'm trying to cut back on evap and heat while I'm away and can't monitor. Thankfully I don't have anything in the tank except for some GSP that benefits from the high lighting.

I've been kicking around the idea of changing the setup somewhat though. I found some Ti tube that i think I'm going to try and bend into a tight little set of S curves, radiator like, and sandwich it in between two peltiers maybe.. I dunno yet.

anyway, my kids are bouncing around wanting to go to Disney. I'll get another update in whe I get back to town.