1
\$\begingroup\$

I have a PCB with several power ICs. The PCB is connected to an aluminium plate as a heatsink on the back of the PCB.

How can I measure how much heat dissipates from the top of the PCB to the air and how much heat is conducted by the heatsink? I want to know it for the entire PCB and not only for the individual components.

\$\endgroup\$
7
  • 1
    \$\begingroup\$ The amount of heat produced is electrical power in minus electrical power out so, if your circuit doesn't have any down stream electrical connections of notable power consumption, then the total heat produced is the electrical power in measured in watts. I realize that this doesn't answer your specific question but it might help. \$\endgroup\$
    – Andy aka
    Mar 2, 2020 at 13:39
  • \$\begingroup\$ The temperature across a PCB made of typical FR-4 (there are lots of variants) can vary significantly as the thermal conductivity is quite low; this can be controlled (a bit) by using planes but for a power board the temperature across the PCB will typically vary quite widely. \$\endgroup\$
    – Peter Smith
    Mar 2, 2020 at 14:19
  • \$\begingroup\$ Essentially you can't. As Andy says you can measure power in... but you can also measure its temperature. So one approach is to measure its temperature while varying input power until you have a good picture of the relationship between them. If you can't control the power directly, bolt one of those metal-cased resistors to the heatsink and dissipate power in that. \$\endgroup\$
    – user16324
    Mar 2, 2020 at 14:46
  • 2
    \$\begingroup\$ Thinking out loud here...What if you were to mount the PCB into the wall of a small insulated container, like a styrofoam cooler, with the aluminum heatsink on the outside and the board on the inside of the container? Close up the container and monitor the inside temperature while the board is being powered/used. The temperature rise should be proportional to the heat being dissipated by the PCB into the container. Knowing the volume and specific heat of the inside of the container (air), you should be able to calculate the heat being put into the chamber. Sort of a poor man's calorimeter. \$\endgroup\$
    – SteveSh
    Mar 2, 2020 at 17:02
  • 1
    \$\begingroup\$ Continuing @SteveSh's approach: now that you monitored the temperature rise in the given chamber / container, you can put a power resistor in the same container and power it from a direct current (DC) source. Adjust the power to get the same temperature rise profile ( should be following a function like A( 1 - e^(- t/Tau)), where A and Tau are parameters, e is the Euler number and t is time. The power you have to burn in the known resistor P = V^2/R (with V being the DC voltage across it and R the resistance of the resistor) gives you the power that your original PCB dissipates. \$\endgroup\$
    – aknott
    Mar 2, 2020 at 22:57

1 Answer 1

Reset to default
1
\$\begingroup\$

You can measure the power in using standard methods.

So we need to measure the power dissipated into the heatsink on the reverse of the PCB. The best way to do this is to fully insulate the heatsink thermally, so only the top of the PCB is exposed. Now, if you know the mass of the heatsink, the specific heat of the metal, measure the temperature rise (with a thermocouple?) over a certain amount of time then you can calculate how much power is being dumped into it.

If possible replace the existing heatsink plate with something more massive so the temperature rise is not so fast. Remember, the heat sink now will not be dissipating any heat at all (ideally).

\$\endgroup\$

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.