Peltier cooling temp control- photographic application

The project I'm working on - several posts on this site - drives a stepper motor, camera shutter and illuminated lens cap (led array). Thanks to the help I've received on this forum all is working OK. To complete the project I want to drive a peltier cooler, to cool the camera sensor, quite simply, as cold as it should reasonably go, preferably -30C - 40C below ambient. The cooler will be on throughout the photographic session with temperature control.

There are two options - use a heat sink to the back of the sensor (about a 1.5mm gap between the sensor back and circuit board (use hot glue to seal the electronics from condensation), or put the camera in a cold box with the lens sticking out the side with a desiccant. Not as efficient.

I have seen it done both ways, but I want to regulate the temperature and keep the power demands low. So I think the heat sink is the way to go.

I will need a new 12v high amp power supply to handle the peltier current and an H-bridge circuit, I think. The idea is to have the peltier operate close to maximum cooling but control the temperature using a thermistor, which will be the most complex part. I read that peltiers can be controlled much like DC motors - PWM, but I'll have to work out how to do that.

You don’t need PWM or an H-bridge if you’re aiming for a 10 degree C range of temperature. Heat doesn’t move quickly enough to warrant that level of control; you could be switching this off/on once a second and still be way ahead of the accuracy of your temperature sensor. An H-bridge also suggests you’ll be reversing the current to the peltier and I don’t think you’ll be doing that. Finally, peltiers do lose a little efficiency if you PWM them so it’s not really an ideal way to go.

Do you know how many watts of heat your camera sensor generates (which would be closely related to how much it takes to power it)? Your peltier will need about 10 times that much power to cool that heat. While the peltier might accept a hundred watts of power it’s doubtful you’ll need to power it to that extent if you’re just negating the heat from a few LEDs and the ambient temperature.

Instead of a thermistor look into using a LM335 or DS18B20 (library). These are both very popular temperature sensors and won’t require the calibration that a thermistor would – and they’re cheap.

Thanks. Much simpler to control than I imagined.

Q: why do you want to cool the camera so much ? special effects? sensitivity? …

Q: why do you want to cool the camera so much ? special effects? sensitivity? ...

Sometimes when you are doing long exposures (really long) you get extra sensor noise from the heat that is generated from the process of capturing the image. Astrophotographers come across this problem quite a lot I believe, so maybe geoland is doing something similar.

Q: why do you want to cool the camera so much ? special effects? sensitivity? ...

That's correct a.d. Cooling the camera / sensor reduces thermal (dark) noise. A dedicated cooled scientific ccd camera is best, but good ones run into many thousands of dollars. I want to cool a Canon XS/1000D DSLR. - The hardware requirements are fairly straight forward.

So far, Arduino is controlling the imaging session almost completely, except for guiding. That will be an interesting project 8)

Right now I need to get the camera temp down to -30, ideally. It seems that most people use a variable power supply to the Peltier module to control temperature.

Ah , a star shooter, you have some amazing pictures

question comes up: - Do you precool the camera in the fridge? 0F / -18C ? and use the peltier to keep it cool ? or must the peltier do all the work? would speed up with half an hour or so?


The pictures on that page are impressive - I hope you achieve something similar!

Providing condensation is not an issue, fridge should be ok. Sealed plastic bag and desicant