Hi all members,

I am new to this forum and basically not an engineering student. Well, I'm a life science student, but I dare myself to create a simple device for heat or cold therapy for rheumatoid arthritis. The range of temperature is from 10 to 50' Celcius.

I want to heat or cool a pad (filled with water) by connecting it to parallely connected 6 peltier devices via heatsink (it will consume a lot of current for sure). I use thermocouple K as the sensor to feedback the temperature. The general idea is, if the user set the temperature below 25' Celcius (assumed as room temperature), then the current supplies the peltier will be reversed. But I really don't have any idea about the control system. One of my friend suggested me to use Autonics TC4S as the temperature controller, because it is relatively cheap and easy to get here.

Is it enough if I control it by using only TC4S? I have searched throghout forums, and some people mentioned about H-bridge module, LM18200, AD595, etc. Do I really need all of them? And what about the required power supply?
http://www.national.com/ds/LM/LMD18200.pdf
http://www.analog.com/static/imported-files/data_sheets/AD594_595.pdf

I desperately need your help, since the deadline is so close. THANK YOU SO MUCH FOR ANY HELP!! DD

Kind regards,
Yugiha

 

You say you want it "simple".
Well, that kind of went out the window when you decided to make it "hot and cold".

The TC4S looks like it closes the relay output to increase temp, open the relay output to decrease temp. Well, that's not going to work if the TEC devices are wired for cooling instead of heating; even via an H-bridge - you'd need to reverse the logic at the same time that the TEC devices' leads are reversed.

You can't bank on the ambient temp being 25°C, either, or you might wind up with a "runaway" condition.

You don't mention what your current requirements actually are.

Maybe you could start on your project earlier next time?

 

This is a tough one mechanically, but I think the folks here can probably help you with the circuit. You might find this project useful for a few ideas.

Some of your basic problems are going to be: 1) avoiding thermal shock to the TECs when switching from one mode to the other, 2) the simple fact that a TEC creates about 10X more heat than it moves. That means cooling is MUCH less effective than heating, and 3) You'll need whopping big currents to drive that much TEC. How are you supplying that power (safely)?