TEC controller for laser cooling

[I_R_Machinist]

Hello again!

I am building a laser projector using a red DVD diode, a blue diode from the casio projector, and a green module. Because I want to be able to run as long as possible without turning the projector off I decided to use a TEC for every laser in the system. The TEC will electrically isolate the red diode and green module as well so they will serve another purpose. The TEC will also stabilize the green module, hopefully enough that I will be able to run it with full analog modulation without it changing modes or any other weird things! Also, the red diode decreases in wavelength the cooler it is run making it appear brighter which will help even the color balance out nicely.

I need some kind of controller for my TECs, I can't direct drive them as condensation would likely be a problem. Also, as I mentioned above, each diode/module has a temperature it will work best at, for example the green module will most likely work best at around 25c, the red diode would be best as cold as possible, and the blue diode should be held stable at 15-20c. The base of my projector is going to be the "hot" side of the TECs as well so I can't go too crazy as I would have no way to get rid of the heat from the TECs.

Is there a simple solution for an adjustable temperature dependent TEC driver? From my limited experience with electronics I thought about a transistor and a thermistor but I do not know if this setup would be stable enough for my project. Does anyone have any insight or experience with something like this?

Here are pictures of my progress so far...

This is the setup, all three lasers going foward to flexible mirror / dichromatic filter stands.

http://img594.imageshack.us/img594/3327/wholepo.jpg

This is one laser assembly, the TEC (30x30mm) is sandwiched between the base and the top block. The top block was machined to a specific depth to keep the laser perfectly level with the mirrors of the galvos. The mounting bolts are sleeved with nylon bushings to keep the laser unit thermally and electrically isolated.

http://img853.imageshack.us/img853/3087/tecj.jpg

Currently I have no laser drivers with any modulation so there is no "blanking" which makes things look like a garbled mess. I did manage to draw and scan a simple smiley face though!

http://img853.imageshack.us/img853/4645/facef.jpg

Thank you for looking and sharing!

[solutions]

Might want to try a high powered LED driver circuit to run the current control for the TEC.  You can fake out the current feedback resistor so you can control the AMPS you need for the cooler. A PID to measure temp and control the driver if you must keep the temperature fairly tight. 

The other option is to just put it all on a cold plate, water cool it to room temp, and forget about your red aspirations....the red you should be able to gamma correct in s/w, assuming it gives you the gamut you need already.

You also might want to order a white cane from the catalog now....while you can still see....
 

[DarkClover]

Temperature controlling is one of the major problems in electronic industry.
Well there are thousands of paths to the same goal but which one fits this device?

I'm not that familiar with temperature stabilizing but wouldn't digital thermometer
and a controller with MOSFET do a pretty good job?
You can control everything and even put some reference curves into the memory.
If you do a little testing and accurate calibrating I think this might be a solution.