You'll also need a pretty hefty power supply, peliter junctions are not very efficent. You'll expend about 5 watts for every watt of cooling power. That adds up to quite a bit of waste heat and supply current.

Mine run happily at 12 volts, and will draw around 5-6 amps. Of course you can run them at lower voltage to reduce the amount of power you will need.

One of the handy things about them is that simply by reversing the polarity of the supply you get a heater, so no need for a seperate heater, all you need is a sensor that will switch it whichever way you need.

I was planning on giving the microcontroller growing setup I've been working on the capability to run a peltier junction (switching it based on the input from a temp sensor in the chamber), but I wasn't sure if very many people would be interested in having a cooling system, since most seem to be interested more in keeping things warm.

Thats a minute amount compared to
the 350 to 2000 watts drawn by
standard refrigeration compressors.

Plus the smallness of the
Peltier Junction even with
a couple of heatsinks strapped
to its ass hardly takes any
space at all when ya consider
what a compressor and its
two sets of coils uses.

And for the power supply theres
always a cheapo battery charger.
At least untill someone will be
so kind to lend there expertise
and point out a good schematic
somewhere or even post one here
or on there webpage.

This device lends itself well
to cool chamber experimention.

Thanks for link ,,,ya can
never have enough resources.

:-)

quote:

---

At least untill someone will be
so kind to lend there expertise
and point out a good schematic
somewhere or even post one here
or on there webpage.

---

The power supply is pretty easy to make, a 12vac 6A transformer is usually less than $20 at surplus places, add a few diodes and you have a suitable supply.

I definately agree that anyone wanting to experiment with a cooled chamber ought to try them, unless you happen to have a spare 'fridge sitting around

I would like to try sticking a heatsink on the cold side of the junction though, mine just have aluminum blocks (like the picture in the link above), which seems less than optimal for heat transfer into air.

[This message has been edited by Code (edited January 22, 2000).]

Anybody got a link to a do it yourself
circuit to control thermoelectric devices
like a thermostat would ???

quote:

---

Something about it damages the Peltier units
by the on and off cycling of regular thermostats.

---

I hadn't heard that before, although with any small temperature control system you want to put some cycle time constraints on the system, otherwise they can tend to cycle very fast at the threshold.

Radio Shack carries a $20 digital LCD thermometer with programmable high and low temp alarm lines that you could use to drive a circut powering a heater and cooler. I have lost the instruction sheet, but I think it had a minimum cycle time built in.

Throw in a power supply, a few switching transistors, some buttons, power resistors, and a peliter, heatsinks and maybe a fan, and you've got yourself an automatic temp controller for your grow chamber.

I'm not exactly a wizard with a soldering iron, nor do I have much free time these days to learn, so if anyone is interested in putting together a little system like this for me I'd be interested in buying it.

So how about all you technical minded and practical motivated types ,,,how about a schematic which ,,,,

1. Gets the job done by controling the Peltier device useing temparture as an indicator without damage.

2. Be a simple circuit whose componants ,,at least most of them,,, can be had from Radio Shack which almost everybody has near them.

3. Be made available here or at someones web site in a clear readable form like a simplified diagram with " any luck " some pics. to aid in construction,,so that the novices and nontechnical of the readership will have a chance of putting it together.

The Nuts and Volts issue Aug. 1999 is yet to be made available on their website archives.
Maybe I will take the issue to kinko's and and scan it so it can be uploaded.

Anybody know what the best format is to save something with text and pics. for uploading ??

As Per. Hoyt

Power supply wise, you can buy 12V regulated power supplies for about $50.00 (5-10amp). You'd be lucky to build one cheaper than that. However, if you like just find a 120VAC to 16-12Volt AC transformer with a suitable current (5-10amps) and then put it through a bridge rectifier, then through a 12V voltage regulator and tada, regulated 12VDC power.

For a kit: http://www.electronickits.com/kit/complete/peltier/ck500.htm

General info:
http://www.gatewayelex.com/peltier.htm

In depth info:
http://trouble.ucsd.edu/~felix/peltier.html

Another source:
http://www.web-tronics.com/webtronics/pelmod4cmx4c.html

That should help a bit. : )

Article:

Thomas Penn recently suggested the use of Peltier devices (a/k/a thermoelectric coolers) to create a custom keg cooler. I had
the same brainstorm three years ago, and in fact played around with fabricating a prototype. Despite the assistance of several
friends who are electrical engineers, we could not produce a unit that would generate enough cooling power to keep a keg even
passably cool.

Why? Three big reasons:

1. Peltier devices are *incredibly* energy inefficient. It takes several times the watts per btu of cooling power when compared to conventional refrigeration, even at its theoretically best rating.

2. You will need humongous heat sinks (and forced air or liquid)on both the hot and cold sides, and a pretty stout DC power supply.

3. Getting a good junction between the unit and the heat sink -- absolutely critical -- requires very careful machining. You can't just slap a module against a scrounged heat sink (or the side of a c-keg) with some thermal grease and get decent heat transfer.

About the time I was ready to pull out my hair on this project, I read a comment by the editor of Nuts and Volts to the effect
that it was a total waste of time for most of us to try and use peltier devices to cool anything larger than a CCD chip for a lot of the reasons I had encountered.

Thus, if you're experimentally inclined, it's a interesting project to play around with (if anyone wants them, I can make you a great deal on some big heat sinks, DC power supplies,and 2 cm x 2 cm peltier devices). And if you are able to show us how to create a useful device that *actually* can keep
a C-keg cool for less than the cost of a second-hand fridge, great -- I'd love to hear how you did it. Just be aware that
you're facing a lot of practical problems that have stymied many experimenters.

My bottom line -- just scrounge up an old working fridge or freezer. Not real fancy, but they really work.

quote:

---

BTW Radio Shack no longer has that exact module. They now sell one that has only a pulsed output for 1 minute, sucks for what you want it to do.

---

Bummer about that. I called about those modules a couple of months ago and they said they didn't carry it anymore, but had a new model. I hadn't seen the spec sheet on the new one yet. And I've misplaced the data sheet for mine so I don't know what the part number is or what wires are what.

Its possable the new one can still work with a bit of extra cheap circuitry hooked up to it, I'll have to get one and check it out.

Ok, so what are the features that would be important to have?
Here are some options, roughtly in the order of what seems important. I'm not suggesting that all these things should be in it, but they are possabilities, so I want to get peoples opinions on it.

1: Temperature sensor (duh )
2: Ability to drive a peltier for cooling
3: Ability to drive heating element/light/air pump
4: LCD/LED display of the temperature
5: Graphing of recorded temps on computer
6: ??

quote:

---

Anybody know what the best format is to save something with text and pics. for uploading ??

---

You can use HTML Help Workshop, which is free from Microsoft. You can build your stuff all in standard HTML, and then Help Workshop will bind it up into a single HTML help file, which most any windows system with a recent IE will open (its the new help file format MS is using).

Certainly a bit more limiting than PDF, but not terribly so.