0
$\begingroup$

Setup: I am running a simulation of two heat sinks connected by some wires. I set the top heat sink to a constant temperature, and the bottom one to a constant heat flux.

Results: According to the simulation results, changing the thermal conductivity of the wires does not affect the steady-state temperature of the bottom heat sink - it stays the same regardless of the wire conductivity.

This is curious to me. I think that a higher conductivity will allow heat to transfer faster from the top to the bottom heat sink, and that will raise the steady-state temperature of the bottom heat sink.

Do I have my physics right? The only other possibility I can think of is that the thermal conductivity affects how fast the system reaches equilibrium, but not the final temperatures, but I can't formulate this, so I can't be sure.

Please help.

Improve this question
$\endgroup$
3
  • $\begingroup$ Usually, when people say "heat sink" they mean an ideal, infinite, constant-temperature source. It is odd that it should be able to change temperature. But if you do have a heat sink that can change temperature (I would rename it from heat sink in that case), then the steady state will always be when the two sinks are at equal temperature. Faster heat transfer just makes you reach this equal-temperature steady state faster. $\endgroup$
    – Steeven
    Commented May 17, 2018 at 14:36
  • $\begingroup$ I haven't made myself clear, then. The top one is what you would call a heat sink. Let's call the bottom one a refrigerator (refrigerators can change temperatures, right?) The problem I'm trying to model concerns heat transfer into the refrigerator (via imperfect insulation) on one hand, and heat pumped out on the other hand. Does my question make sense now? $\endgroup$
    – psitae
    Commented May 17, 2018 at 15:00
  • $\begingroup$ Clearly something is going wrong with the simulation. What is the temperature is the constant-flux component? Does it change when you change the temperature of the constant-temperature component? If not, then maybe it’s not connected properly. Try debugging the simplest possible configuration (e.g., one wire, steady state). $\endgroup$
    – Chemomechanics
    Commented May 17, 2018 at 16:04

0

Reset to default

Your Answer

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

Browse other questions tagged or ask your own question.