TEC on TO3 red diode

I am trying to get the built in TEC to work on a TO3 red diode, I used sam's laser faq to correctly identify the pinout. When applying 3v and 100mA of current in the correct direction of the TEC I get the resistance of the themistor to rise like it says it should in Sam's laser faq. But this causes the diode to heat up, and if I reverse polarity on the the TEC it does the same thing, except it causes the resistance of the thermistor to decrease. Any help?

Can you point me in the direction of where does it talk about TEC on Sam's Laser FAQ?
It's very weird indeed, I should do some research before answering to this..


EDIT: Ok this is what I found:

http://www.xtremesystems.org/forums/showthread.php?t=67927

http://www.xtremesystems.org/forums/showthread.php?t=38367

Last link is a guide, REALLY helpful.


It seems what's happening to you is that you've got a TEC that can't fulfill the job you're trying to use it for. You may need more power.

Here are a few examples with difference watt TECs cooling a modern processor which usually generate around 60+ watts of heat. Practice the formula on at least one of these examples before calculating Delta T for your own project!

*) A 40watt TEC cooling a modern processor generating 60-watts of heat.
Delta T = -34C (negative 34C). That’s an INCREASE in temperature on the cold side, so the processor would have been warmer than before! Using an under powered TEC can destroy the processor. Cooking your CPU with your TEC is NOT a good thing, avoid this common mistake.

*) A 60-watt TEC cooling a modern processor generating 60-watts of heat.
Delta T = 0 (no change in temp). Congratulations, you’ve accomplished nothing!

*) A 176-watt TEC cooling a modern processor generating 60-watts of heat.
Delta T = 46C. That’s a decrease in temperature on the cold side, making the cold-side 46C cooler than the hot-side. Not bad.

*) A 226-watt TEC cooling a modern processor generating 60-watts of heat.
Delta T = 51C. That’s a decrease in temperature on the cold-side, making the cold-side 50C colder than the warm-side. Even better, but it took a lot more energy from the TEC (28% increase from 176-watt TEC) to get an additional 5C (only 11% increase in Delta T).

Click to expand...

nikokapo, GOOD JOB, plenty of good info on TEC's. I didn't know there was a hot side of a TEC. Plenty of reading to do...

Also, 3V@100mA won't do much for a TEC. I recommend hooking the TEC up to it's own full-range LM317 driver and adjusting the pot until you get the best results.

The tec will require over an amp to keep the diode cool. I've found a good way to monitor the temperature is get a cheap little digital thermometer (ones withoutdoor probes are easy to work with)

Chop off the probe or remove the thermistor inside, and connect the wires to the thermistor pins on the diode. This will give you a reading of the temperature inside the diode.

An easy way to run the tec is to hook it in series with the laser diode. I've monitored the diode temperature while doing this and it runs the tec enough to keep the diode cool. However tecs generate a lot of heat and for runs longer than 30 seconds or so you will probably need to heatsink the TO-3 laser diode.

Also, TECs are polarity sensitive. They do have a hot and cold side, but if you reverse the polarity, the hot and cold sides swap.

Here is the pinout for the TO-3 laser diodes from boston laser.

Also, I have this picture of a diode I popped open:


Click Image to Enlarge

Sorry all, but let me state a thing, before you continue (not for teach you something, just cause i've heard so much this thing about TEC's ..... ) ..... TEC are not cooling devices, are heat pump devices

Some peoples, also that i know personally, always do this mistakes, reading about TEC's ..... cause they are presented as "cooling" units, they think that is enough to connect it to a power supply, and get cold ..... forgotting some thermodinamic laws, perhaps

So, saying it in the more simple way that i can think with my bad english, think to the TEC plate as a PUMP that transfer the heat from one side to the other ..... but, at the end, you still need to dissipate the heat that is transferred, PLUS the heat that the device itself generates in the working process (cause it's not a perfect machine) ..... in these applications, your TEC plate keep the temperature of the diode stable for long period of working time (in conjunction with the thermistor and a thermostatic regulator), but only if you continuously take away the heat transferred with a good heatsink, from the hot face.

If the temperature difference becomes too high (and, basically, a TEC have a maximum ideal difference of temp of approximatively 70 C), you loose efficency ..... example, if you keep the hot side at 40 C, you can go under 0 C easily, on the cold side, but if you left the hot side reach, like, 80 C, then you can't go under 10 C (ideally, cause in reality are also more) on the cold face, and so on ..... just a poor example, but can give you the idea ..... if you don't take away the heat from the case (that is connected to the hot face), and it rise and rise, til, like, 120 C, also at the maximum current run of the TEC, you still fry your diode running it at 50 C

Yes I thought a TEC was just a cooling device, now I know how it actually works.
I can't run the diode in series with the diode because I'm going to be using this diode in a scanner. I don't want the TEC pulsing on and off like the diode, I'm going to have to figure out a way to power it and control it. Thanks so far for the help.

you still need to dissipate the heat that is transferred, PLUS the heat that the device itself generates in the working process

Click to expand...

^ THIS. MISTERWILLING, you need to know that TECs are used to transfer all heat to just one spot and disipate it there with something else, they can cover a larger surface...well...everything's explained on the links I posted..


You can call them cooling devices, it depends on how you look at them.

@misterwilling, you can use a proportional pwm thermostat, also commercial types, if you don't want to have on/off cycles ..... or also just improvvisate one with a 555 in pwm configuration, with a mosfet like buz71 or similar, depend from the current of the peltier cell ..... just put the thermistor in the duty cycle loop, and set it for the temperature you want (not professional, but works, LOL)

That is an awesome pic Mariomaster! The server that hosts your picture is very quick serving up large pics too. Thanks

why not use a wavelength electronics laser diode driver like those on ebay with a TEC controller?. The TEC will need about 1.5V at about an amp or so of current.

I am powering this diode with a flexmod for analog modulation. For the TEC I am talking with Robin from DIE4LASER to get a TEC controller that will work for this diode. His current model has a 600mA ripple and he doesn't think this will work well with such a low current TEC so he may be modifying it for me.

Did you receive the to-3 module I sent?

No not yet, hopefully comes in this week.

I got it in, I'm glad its all american Standard threads. The lenses doesn't seem to be placed correctly to focus the particular diode I have. It looks like an ashperical lens with a cylinder lens after it. I will need to get the diode correctly mounted and heatsinked before I can do any more testing with it.