No announcement yet.

painting a heat sink

  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    Originally posted by MattiJ View Post
    Especially in forced air cooling or liquid to liquid heat exchangers (or is it possibly even liquid-to-snow heat exchanger) where heat transfer coefficients are much bigger.
    With free convection the heatsink-air interface is almost always the weakest link and painting things does not effect much anything.
    It may be decisive where there is not much temperature margin, AND the "paint" is thick, such as powder coat.

    It comes down to what percentage of the total interface thermal resistance per unit area is provided by the paint, vs the thermal resistance of the basic interface. It will make the same difference as to percentage, regardless of the amount of area. But if the watts per unit area is low enough, there is sufficient margin to make the difference inconsequential.

    EDIT: That last bit is not entirely true, because the thermal resistance from sink to air depends on air velocity, and the effect on boundary layer scrubbing. The coating thermal resistance per unit area is constant, but the percentage may not be as a result of air velocity. Therefore coating the sink will have a different effect depending on the watts per unit area in convection, and forced air velocity if forced air is used.

    You could look at the coating as essentially added to the heatink's own resistance from source to air interface.
    Last edited by J Tiers; 10-31-2017, 11:04 PM.
    CNC machines only go through the motions


    • #47
      The company I work for does quite a lot of work on coatings for heat sinks / exchangers for various military and aerospace applications. At high altitudes and in space, the primary form of cooling is radiation since there are so few molecules for convection / conduction. There are also some ground applications where having a large amount of radiative transfer but very little conductive or convective transfer is actually highly desirable.

      Anyway, here is an interesting article about some of the cutting edge work being done on coatings for IR radiation:


      • #48
        I have heard of that approach before, mostly as a theoretical idea. Nice to see that the theory is actually workable to some extent already.
        CNC machines only go through the motions


        • #49
          Originally posted by J Tiers View Post
          Powder coating is sufficiently thick that it may actually do a decent job of insulating the sink, reducing both radiation and convection/forced air cooling.
          One particular machine was a Yamaha 2 stroke triple,that when the powder coating was removed the cooling was back to normal instantly.These exchangers are cooled directly by snow.