jeffed1969
Mechanical
- Aug 7, 2001
- 3
Hello:
I am a design engineer working on thermal management of IGBT modules.
As most of you must know, when mounting these components on a heatsink it is always helpful to use some sort of interface material to fill the insulating air gaps arising from irregularities in the surfaces.
In comparing various commercially available products I am finding that manufacturers report different data; thermal conductivity, thermal resistance and thermal impedance.
I prefer to work with conductivity, power/(length*temp) or conductance, power/(area*temp) but heatsink analysis and electrical folks use resistance.
Logic tells me that the resistance should be the inverse of conductance, (area*temp)/power but many suppliers call this thermal impedance. These data sheets often refer to the ASTM standard above. I have also seen what I call "specific thermal resistance", C*in^2/W/mil.
In most heatsink literature, thermal resistance = deltaTemp/power which leaves out any reference to inches or meters, squared or not. I assume this is because each object has discrete dimensions so if one reports dT/Q for a certain sink, a term in the full equation may be omitted. The coefficient of convection is left out for similar reasons as thermal resistance is plotted against the volumetric flow rate (which in turn relies on the discrete dimensions of the object to obtain flow velocity).
When testing per the standard, what is the correct term & unit and are these results to be plotted against applied pressure?
General feedback and discussion?
I am a design engineer working on thermal management of IGBT modules.
As most of you must know, when mounting these components on a heatsink it is always helpful to use some sort of interface material to fill the insulating air gaps arising from irregularities in the surfaces.
In comparing various commercially available products I am finding that manufacturers report different data; thermal conductivity, thermal resistance and thermal impedance.
I prefer to work with conductivity, power/(length*temp) or conductance, power/(area*temp) but heatsink analysis and electrical folks use resistance.
Logic tells me that the resistance should be the inverse of conductance, (area*temp)/power but many suppliers call this thermal impedance. These data sheets often refer to the ASTM standard above. I have also seen what I call "specific thermal resistance", C*in^2/W/mil.
In most heatsink literature, thermal resistance = deltaTemp/power which leaves out any reference to inches or meters, squared or not. I assume this is because each object has discrete dimensions so if one reports dT/Q for a certain sink, a term in the full equation may be omitted. The coefficient of convection is left out for similar reasons as thermal resistance is plotted against the volumetric flow rate (which in turn relies on the discrete dimensions of the object to obtain flow velocity).
When testing per the standard, what is the correct term & unit and are these results to be plotted against applied pressure?
General feedback and discussion?