[Noisebridge-discuss] N00b question - changing power supply voltage

[Corey McGuire]

This is the best back-of-napkin application I've seen for TEJ's, ever and a
24v power supply is the perfect supply for it.

There are two problems I see:

   1. When you cycle them to maintain a temperature, the heat you moved to
   one side will quickly conduct back to the other.
   2. In this process, a charge will be created and sent down the wire.

How you handle this is a mystery to me.

Here's what I would do.  I would NOT wire them in series.  Instead, I would
implement switching power supplies and use them to keep the TEJ's active at
16v when cooling or a lower voltage while maintaining the desired
temperature.  Then I would switch them relative to how far below the desired
temperature they are.  Any temp above desired, 16v; 1 degree below, 12v; 3
degrees below, 8v... or whatever.

I would NOT overdrive them because I imagine they just get even LESS
efficient.

1 arduino
1 temperature probe
a fist full of switching power supplies
a capacitor just to keep power going to the TEJ to help resist the heat
moving backward (I don't know if this is a problem, but it is a cheap
solution.)
whatever else (I am not even pretending to be an EE.)

YMMV





On Tue, Jul 6, 2010 at 9:35 AM, Sean Cusack <sean.p.cusack at gmail.com> wrote:

> Oh yes...I know they are terrible at efficiency...but they are also the
> only thing that I know of that can get you to sub-ambient temperatures
> without using a (comparatively) giant refrigeration system.
>
> I'm planning on using these to cool a few pieces of lab equipment.
> Typically, to get to sub ambient conditions, you have to use ice/water (gets
> you to 0C), or dry ice/acetone (gets you to -78C), or full on Liquid N2
> which gets you too cold for most practical applications. It would be
> *awesome* to hit like -20 or -10 or even 5C repeatedly and controllably for
> a million and one different chemical reactions.
>
> There is equipment that allows you to do this now, but pretty much its a
> standalone refrigeration system that pumps cooled silicon based oil through
> your reaction mixture. It takes up a ton of room on my bench, and since
> those refrigerators are on the order of $7k a pop, its tough to convince my
> boss to allow me to buy more than about 2 of them. In other words, longer
> hours for Sean in the lab = teh sux.
>
> So, I'm trying to use these doodads as a way to run a bunch of reactions at
> a controllably cold temperature. I agree there's problems, but given the
> application, it may just work!
>
> Sean
>
>
> On Tue, Jul 6, 2010 at 8:44 AM, Jonathan Foote <jtfoote at ieee.org> wrote:
>
>> On Tue, Jul 6, 2010 at 4:17 AM, Corey McGuire <coreyfro at coreyfro.com>
>> wrote:
>>
>> > Messy, messy stuff.  TEJ's are not efficient.  This is fine by
>> themselves.  When you stage them, their inefficiencies > become readily
>> apparent as they begin to compound.
>> >
>> > May I ask what you intend to do with them?
>>
>> Yeah, also curious. Corey is absolutely right: TEJs have terrible
>> Carnot efficiency --  way less than 10%. This means to move (not
>> remove) 5 watts of heat you have to put in 50+ watts of power, which
>> turns into heat you ALSO need to remove.
>>
>> So they are only useful in a few applications where the small temp
>> difference over a tiny scale is worth the waste. If they really were
>> the magic refrigerators people think they are, they would be in every
>> PC and laptop. And note that if you are trying to keep things cool,
>> there may be far better solutions.
>>
>> "In this house we obey the laws of thermodynamics!"
>>
>> -J
>> _______________________________________________
>> Noisebridge-discuss mailing list
>> Noisebridge-discuss at lists.noisebridge.net
>> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>>
>
>
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noisebridge.net/pipermail/noisebridge-discuss/attachments/20100706/a4a521e0/attachment-0003.html>