TEC questions, none PC related.

UglyChild · Oct 30, 2007

This has nothing to do with PC cooling, so i just need few simple answers about TEC cooling. And the sticky above doesnt have the info i need.

I need to cool a piece of metal thats about the size of 4"L x2"W x1"H, cool it to 4c. It doesnt produce any heat of any kind and its not attached to any heat source, it just needs to be cold.

I have 2 TEC's that are 85 watts each. And im going to use water cooling to cool two TEC's. I will have 2 water blocks, one for each TEC, one pump, one triple radiator and a couple of fans.

*Would this be enough to get constant 4c or lower on the cold side??
*Can i add a voltage regulator to TEC to keep it at 4C on the cold side, so it just doesnt keep cooling till it cant cool any lower??

Another question.

*Can i reverse the voltage on the TEC to make the cold side hot at 37c ?? Or can i just flip the TEC over to get 37C??

Thanks.

Moto7451 · Oct 30, 2007

IIRC a TEC works like a diode so you can't reverse the voltage.

If it's not producing any heat, you'll easily get lower than 4°C depending on environmental conditions (i.e. if the container it's in is 100°C, good luck, but room temp should be fine).

killermiller · Oct 30, 2007

Your watercooling will be more then enough. You could use a pot of some sort and adjust voltage along with fan speed and flow from the pump. 85 watt tecs should be more then enough as well. usa am AA battery and put one wire on each side. feel the cold side. Switch sides and check to see if the other side gets cold. That will tell you if you can reverse voltage.

UglyChild · Oct 30, 2007

It will be done at room temps.

Thanks guys.

aja · Oct 30, 2007

What is it for if I may ask? just curious...

NoL · Oct 30, 2007

Yes a peltier can be reversed to change the direction of heat movement.

torin3 · Oct 30, 2007

NoL said:
Yes a peltier can be reversed to change the direction of heat movement.

I'm pretty sure this is correct. I thought I remembered this when I was testing with a 1.5V battery.

NoL · Oct 30, 2007

It is, I have a few thousand watts worth of TEC's and PSU's under mybelt.

UglyChild · Oct 30, 2007

Helping a friend with PhD hes working on, and hes working on a project, thus some kind of a chemical has to be cooled, then heated while infrared light passes through. Space is extremely limited thus no large air cooling can be used to cool TEC's on the metal block it self, because its enclosed.

He wanted to use a NB Cooler to cool TEC's, and a 20watt PSU as power source.. he so silly.

ill try the AA battery first before attaching it to a PSU.

Thanks again.

Goonda · Oct 30, 2007

You would have to build a temperature controller to ensure that the TEC does not overshoot its set-point. You can build one to use a thermistor sensor or an AD590 sensor.

We use something like this (Using Diode Lasers for Atomic Physics by Carl Wiemann, Rev. Sci. Instrum., Vol. 62, No. 1, January 1991):-
http://tf.nist.gov/general/pdf/739.pdf

Scroll to page 7 for the schematic.

The stuff I built it for looks like this. Precision of 0.01 C can be achieved and is a must for our application. The TEC is in between the base block and the Laser diode mount (to the left of the small tissue). It is a 25W TEC, but this controller should be good upto 1A. With a few tweaks you can extend its range.

torin3 · Oct 30, 2007

NoL said:
It is, I have a few thousand watts worth of TEC's and PSU's under mybelt.

Well, I'm not quite up to a KW myself, but I've managed to lose a fair amount of equipment to failures with TECs. I know a lot of ways to NOT use them safely

.

They are fun, but it cost me a bit too much.

NoL · Oct 30, 2007

Tricks to build in relays and temperature sensors

aja · Oct 30, 2007

torin3 said:
I know a lot of ways to NOT use them safely .

That is the best kind of knowledge to have. We have so much more fun in life than all those do-everything-right-first-time-round kinda people :beer:

torin3 · Oct 30, 2007

NoL said:
Tricks to build in relays and temperature sensors

Well, then there was the pinhole in the conformal coating that when I took everything apart had water dripping out of the CPU socket, and several pins that had corroded off...

UglyChild · Oct 30, 2007

Goonda said:
You would have to build a temperature controller to ensure that the TEC does not overshoot its set-point. You can build one to use a thermistor sensor or an AD590 sensor.

We use something like this (Using Diode Lasers for Atomic Physics by Carl Wiemann, Rev. Sci. Instrum., Vol. 62, No. 1, January 1991):-
http://tf.nist.gov/general/pdf/739.pdf

Scroll to page 7 for the schematic.

The stuff I built it for looks like this. Precision of 0.01 C can be achieved and is a must for our application. The TEC is in between the base block and the Laser diode mount (to the left of the small tissue). It is a 25W TEC, but this controller should be good upto 1A. With a few tweaks you can extend its range.

Thanks dude, ill read the PDF and see if theres anything we can use.

Goonda · Oct 30, 2007

If you need more ckts on precision temperature and current controllers, let me know.

TEC questions, none PC related.

UglyChild

Member

Moto7451

Senior Something

killermiller

Member

UglyChild

Member

aja

Member

NoL

Member

torin3

Member

NoL

Member

UglyChild

Member

Goonda

Member

torin3

Member

NoL

Member

aja

Member

torin3

Member

UglyChild

Member

Goonda

Member

Similar threads