Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

Simple TEC driver?

The BC337 does not need to dissipate power, is only for invert the driving signal of the 555 to the mosfet ..... the mosfet can require some heatsinking, but if you use a high current one, probably it does not need too much of it (after all, it's drived mainly in on/off configuration, it dissipate power mainly in the transition times) ..... no, resistors does not need to be high power ones, i used all 1/4W, and yes, i powered it at 12V.

Just be sure that the power supply can give enough current without "sit" when the TEC is in "on" state ..... or, also better, you can decouple the two power lines, for avoid problems, or at least for try to reduce them ..... the positive for the TEC can be hooked directly, where instead you can place 3 or 4 diodes (1N400X, no matter the exact type) in serie to the positive line that power the 555 part, and a big enough capacitor or bank of them (like, 470uF25V, or more) on it, so the switching cycle of the TEC does not influence too much the 555 part.

About the mosfet, i used a PHD78NQ03, a 60A one, but just cause they was easy to find here ..... anyway, more ampere they hold, more low is their internal RdsON, more low is the power dissipated from them ..... there are so much types of them, that i cannot imagine a part number to suggest you, cause this mainly depend from the availability of them in your electronic shops ..... also, mine was in SMD, but for hook it on an external heatsink, you probably prefer to find one with TO220 case.

Last, the NTC resistor, i used a standard model, miniature, with a common value of 10 Kohm at 25C, that is one of the more common types available around ..... if you use a different value, you may need to change also the trimmer value.
 





TECs are not as sensitive components as laser diodes are.
However, a bit more complicated drivers with multiple outputs exist in professional laser systems, they also incorporate MCU controlled temperature, in 473nm lasers, LBO crystal is anything but temperature-ignorant.

However, since I don't think you gonna be colling LBO crystals, any applied power to it will help keep the temperature of your components down.

Keep in mind however, that TEC as a components is not thermally conductive, means that if you mount the diode on it without TEC powered, it's worse than just simply mounting the diode on a heatsink - if the TEC would be thermally conductive, it would have a hard time keeping the cold side - cold.

IIRC, TECs are around 20% efficient - hence, calculate the temperature dissipation of your laser diode - for blue diode driven @1A, it'll be around 3.2-3.5W of heat - and set your TEC accordingly, 5x more power consumption, so it'll provide aproximately needed cooling power.

Maybe I'm wrong about some facts, though...

Gawd I wish I had some MCU programmer boards, I'd be making temp-controlled laser systems...
Why are the programmer boards so damn expensive? :thinking:


That's why you mount a plate on top of the TEC or put thermal (silver epoxy on it) then your heat sink with the diode.

No, you'd better have control over the temperature or you'll end up with a dehumidifier instead of a lab laser or a really crazy looking ice maker.
TECs can get down to -20c!
 
That's why you mount a plate on top of the TEC or put thermal (silver epoxy on it) then your heat sink with the diode.

No, you'd better have control over the temperature or you'll end up with a dehumidifier instead of a lab laser or a really crazy looking ice maker.
TECs can get down to -20c!
Oh yeah that sounds like fun.
I know, Bill once told me, with fully heatsinked hot side, he managed to freeze a glass of water into ice.

Wish I had some bench PSU capable of supplying enormus currents...
 
The BC337 does not need to dissipate power, is only for invert the driving signal of the 555 to the mosfet ..... the mosfet can require some heatsinking, but if you use a high current one, probably it does not need too much of it (after all, it's drived mainly in on/off configuration, it dissipate power mainly in the transition times) ..... no, resistors does not need to be high power ones, i used all 1/4W, and yes, i powered it at 12V.

Just be sure that the power supply can give enough current without "sit" when the TEC is in "on" state ..... or, also better, you can decouple the two power lines, for avoid problems, or at least for try to reduce them ..... the positive for the TEC can be hooked directly, where instead you can place 3 or 4 diodes (1N400X, no matter the exact type) in serie to the positive line that power the 555 part, and a big enough capacitor or bank of them (like, 470uF25V, or more) on it, so the switching cycle of the TEC does not influence too much the 555 part.

About the mosfet, i used a PHD78NQ03, a 60A one, but just cause they was easy to find here ..... anyway, more ampere they hold, more low is their internal RdsON, more low is the power dissipated from them ..... there are so much types of them, that i cannot imagine a part number to suggest you, cause this mainly depend from the availability of them in your electronic shops ..... also, mine was in SMD, but for hook it on an external heatsink, you probably prefer to find one with TO220 case.

Last, the NTC resistor, i used a standard model, miniature, with a common value of 10 Kohm at 25C, that is one of the more common types available around ..... if you use a different value, you may need to change also the trimmer value.
Thanks about all these information...

I have few MOSFETS but these are max 8A, and 600-700V.
Can you recommend me the most common MOSFET for that use that I can find on ebay?
 
this is interesting... I posted a similar statement and it holds true here... we all have different results...LOL.

well i have a 1.5"x 1.5" tec hooked up to my ddl driver and i have about 2 hours on it. i have it set to 1 amp and the incoming power is 12v from a pc cpu. the tec stays well below room temp. the driver has not yet gotten any where neat "too hot".

michael
 
Yes, because voltage regulator doesn't have to transform unwanted voltage to heat.

@HIMNL9
How many amps can this circuit give?

On pin 5... is that 10nF capacitor?
 
Last edited:
Oh yeah that sounds like fun.
I know, Bill once told me, with fully heatsinked hot side, he managed to freeze a glass of water into ice.

Wish I had some bench PSU capable of supplying enormus currents...



PC PSU's make an excellent power source. I had bought some IR diodes on ebay rated 1W and for the hell of it hooked one directly to the 5v output.

Not only did the LD survive, it performed like a champ! I wouldn't recommend this type of hookup though, as I had an extra to kill, and just gave into the temptation ;)

Im not well versed in electronics yet, but using a PC PSU is the best power supply ive had yet. I highly recommend the use of one to experiment with.

Im sure there are some drawback as with anything, but so far I can only say good things.

Is there a way to change a typical PC PSU to extract more current? Maybe someone has a useful MOD for one of these that they can post for the power hungry members.


Another quick and dirty power supply that I have and use with TEC is an automotive battery charger that puts out 50A @ 14v on the high setting.

The power is not good for anything sensitive, but so far I have not wrecked any TEC's. My charger is 2-10-50A selectable. 50A seems overkill though for even my larger TEC's
 
Last edited:
Oh yeah that sounds like fun.
I know, Bill once told me, with fully heatsinked hot side, he managed to freeze a glass of water into ice.

Wish I had some bench PSU capable of supplying enormus currents...

yup. Done that. It works for larger TECs. The only ones I got here are 30x40mm.


just sink the hot side and watch the temperature swing into the -10s on the cold side.

You really need some way of controlling the current to the TEC.

By putting huge currents through the TEC you risk smoking it.
 
I like the idea of using this circuit. Just use a power supply of slightly higher voltage and current capacity, and then replace the fan with a TEC driver like the LM338.

Power Supply: 15V, 6.7A 100W POWER SUPPLY-MPJA, Inc.

Controller: http://www.coolcircuit.com/circuit/fan_controller/automatic__fan_controller.GIF

It even lets you set a temp. Its simple, and cheap at the same time. Try the 15V supply to account for the circuit's Vdrop. Should be close to ~12V for the TEC when the controller is on.
 
Last edited:
HIMNL9, you are so nice,
That's schematic is great thanks so much.
+ reps for you!
This is like the third time you have helped me figure out or draw a schematic.
 
Yes, because voltage regulator doesn't have to transform unwanted voltage to heat.

@HIMNL9
How many amps can this circuit give?

On pin 5... is that 10nF capacitor?

The current depend from the mosfet, not from the 555 ..... it just change the duty-cycle of the PWM signal (more exactly it's a "pulsed" driver, cause the frequency is low) ..... and yes, it's a 10nF capacitor.

BTW, also voltage regulators turns in heat the unwanted voltage ..... ;)



HIMNL9, you are so nice,
That's schematic is great thanks so much.
+ reps for you!
This is like the third time you have helped me figure out or draw a schematic.

No problems ;)
 
Sorry for asking all these questions but I'm not familiar with these circuits and I want to go to sea side in matter of minutes, I'm packing rucksacks, and don't have enough time to research about each component.

I ordered IRF3205 MOSFETs that can work up to 110A, I think these are fine.

One more thing.... this capacitor that is connected to gate... is that 47pF?
Can I use different capacitor?
 
That capacitor is mainly just for protect mosfet gate from high frequency spikes and disturbs ..... the mosfet is drived from low frequency, so anything til 10nF must not cause any problem.
 


Back
Top