TO-247 thermal

[strawberry]

How to cool down high power 5s pulses
AlO ceramic?
heatsink anodising coating is bad thermal conductor?

[ajb]

5s is plenty of time for heat from the device to flow into an attached heatsink.  Depending on duty cycle, a relatively large chunk of aluminum to provide some thermal mass may be just as effective as a proper heatsink.  Beyond that it really depends on the power levels and duty cycle in question.

The layer of aluminum oxide created by anodizing aluminum is a worse thermal conductor than metallic aluminum (something like 15-50 W/mK versus 200 W/mK), but the oxide is typically only 2-25um thick (anything over ~25um is considered "hard" anodizing).  The oxide and the increased surface area that anodizing produces also provide much higher surface emissivity, around 0.83 versus 0.05 for bare aluminum.  Increased emissivity means the heatsink is much more effective at radiating heat into the environment.  How much this matters compared to conducting heat into the surrounding air depends on airflow conditions and other environmental factors (like is it closed in a box?).  But generally anodizing does not make for a worse heatsink, and may make for a substantially better one, especially given the increased surface area.  AIUI, color of the anodizing has little or no appreciable effect relative to the other factors.

https://www.qats.com/cms/2010/11/09/how-heat-sink-anondization-improves-thermal-performance-part-1-of-2/

[strawberry]

2.4kW peak for 5..10s (~40W loss per MOSFET) MOSFET plastic case gets noticeably hot (~80C) silicon die could be even hotter
duty cycle 50, freq 100k
isolator 2mm AlO ceramic wafer
50x75x20 heatsink fan cooled

[David Hess]

The oxide and the increased surface area that anodizing produces also provide much higher surface emissivity, around 0.83 versus 0.05 for bare aluminum.  Increased emissivity means the heatsink is much more effective at radiating heat into the environment.  How much this matters compared to conducting heat into the surrounding air depends on airflow conditions and other environmental factors (like is it closed in a box?).

Thermal radiation is proportional to the 4th power of the temperature so even with ideal conditions, high temperature, and natural convection, it amounts to less then 1/3rd of the dissipated power.  A small amount of forced convection reduces the contribution from thermal radiation to insignificant levels.