I need to hold a 60x60cm aluminum panel at a fixed temperature of 35 degrees Celsius. I thought about using Peltier elements but I'm not sure how easy it is to maintain a constant temperature with them. (Does anybody know of an off-the-shelf product that can do that?)

Another possibility is running heated water through brass pipes underneath the panel, but again I don't know of a good mechanism to keep the water at such a temperature.

Any thoughts or suggestions?

  • \$\begingroup\$ It also depends on the environment. If that can exceed 35C you'll need both heating and cooling. \$\endgroup\$
    – user16324
    Jul 2, 2021 at 11:03
  • \$\begingroup\$ @user_1818839 the environment is held fixed at 23 degrees Celcius \$\endgroup\$
    – Oren
    Jul 2, 2021 at 11:32
  • \$\begingroup\$ How thick? And is it OK to epoxy power resistors to one side with active temp control? \$\endgroup\$ Jul 2, 2021 at 11:45
  • \$\begingroup\$ How close do you need to hold the temperature to 35 deg C? What is your cost target? Are you trying to heat or cool something that is attached to the aluminum panel? \$\endgroup\$
    – SteveSh
    Jul 2, 2021 at 12:29
  • \$\begingroup\$ Consider aluminium tubing as aluminum will also corrode where its exposed surface contacts the brass plate because brass is more cathodic. The design of the plate and tubes will have a big impact on how close you can control the temperature as will your temperature controller. You need to define your problem in more detail. With the tubing will it be laminar or turbulent flow? Since you are heating you could use Power Resistors connected to the back of the plate. \$\endgroup\$
    – Gil
    Jul 2, 2021 at 18:13

1 Answer 1


Peltier is the best bet, you will need quite a few of them.

It is important to have good thermal non-rigid transfer, the best is to use thermal pads.

You will need a bigger base plate underneath the Peltier.

If you want something solid, regulate the Peltier with current (like a PWM would do) each independently and have a temperature sensor near each of them, have an MCU that manage the whole with a PID loop for each Peltier, this will give the best uniformity of temperature.

This method should yield a uniformity of temperature of less than 0.1°C.

The problem with the water, other than the mess to deal with leaks, is that your input temperature will be different than the output temperature, thus uniformity is difficult to achieve, also it will have a very slow regulation response as you have to heat the water, and to cool it it's even more trouble.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.