MOSFET heat sink calculation: I am designing H bridge to drive brushed DC motor . The MOSFET I'm using is IRFB4620PbF. I calculated the MOSFET losses which are 7.8W per MOSFET. I'm not sure if I am doing the calculations correctly for heat sink selection.

Tj = Pd (RthJC + RthCS + RthSA) + Tamb

Tj junction temperature : 150C Pd = dissipated power (RMS current through the part x voltage drop) + switching losses = 7.8W
RthJC = thermal resistance, junction to case (from the device data sheet) = 1.045C/W RthCS = thermal resistance, case to heat sink (from the device data sheet) = 0.5C/W RthSA = thermal resistance of the heatsink to air.

Tamb = ambient temperature immediately adjacent to the module in your application = 25C

Susbstituting the above values I get the RthSA as 14.48C/W.

Could you please let me know if the calculations are correct or am I missing something. Thank you for your help.

  • \$\begingroup\$ Looks about right but add some safety margin. (Tamb is over 25C here today!) Having worked out 14C/W I would be looking for heatsinks 10C/W or even 5C/W or better if they meet your space, weight and cost budgets. It can be a cheap way to improve reliability and equipment life. \$\endgroup\$
    – user16324
    Commented Jul 2, 2014 at 12:02
  • \$\begingroup\$ @BrianDrummond 25C would be considered pleasantly cool where I live: It's around 36C here today :-) \$\endgroup\$ Commented Aug 1, 2014 at 15:35
  • \$\begingroup\$ I might interject that all bets are off if the heatsink isn't exposed to free-flowing air - if it's surrounded by obstructions, it's not going to sink heat at all well. \$\endgroup\$ Commented Aug 31, 2014 at 16:08

1 Answer 1


Assuming your heat loss calculations are correct, this is right (don't forget that Rds(on) increases greatly at high junction temperatures).

However, Tj=150°C is pretty hot, and it's unrealistic to use 25°C as the maximum ambient near the heatsink in most applications. It does mean the MOSFET will not be immediately destroyed at ambient temperatures up to 50°C. It does not mean that it will be very reliable.

  • \$\begingroup\$ I'm very new to electrical engineering. Could you please let me know the typical values of Tj and Tamb. \$\endgroup\$
    – user46642
    Commented Jul 2, 2014 at 13:20
  • \$\begingroup\$ @user46642 One rule of thumb is that lifetime halves for every 10°C rise. You should not allow the TJ to get close to the absolute maximum. One indication is where the part is specified (125°C = Tj). Tamb depends on your application and environment. If it gets to 40°C in the room and the enclosure adds another 10°C then 50-60°C Ta is not unreasonable, more range is required for many applications. There's not really a hard and fast rule, but I don't like to see heat sinks running at more than 100°C. \$\endgroup\$ Commented Jul 2, 2014 at 14:01

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.