I have recently been working on a power H-Bridge meant to drive an inductive load in hard switching mode. I calculated the maximum operating frequency limitation due to maximum power dissipation of the IGBT that I have chosen. I am not very experienced in this topic because all my knowledge comes from reading and I would greatly appreciate if someone could verify the accuracy of my calculations and correct me if something is wrong. I would also be happy to hear any general advice regarding this topic if you have some on your mind :) Thank you for your time and have a great day!

First page of the document

Second page of the document

enter image description here

  • 1
    \$\begingroup\$ To it sounds strange that busbar voltage doesn't come into play, it rules switching losses just as much as peak current. Also, I believe you should account for thermal resistance, impedance applies to peak power, but what you reckon is average power instead. \$\endgroup\$ – carloc Feb 23 at 18:54
  • \$\begingroup\$ Ive nevner heard of anyone hard-switching IGBTs at 180 kHz, more common would be 20 kHz. According to your marks in the graph, shouldn't Eon be more like 1-1.2 mJ? Also, where is the 1/2 coming from in 1/2(Pcond+Pswitch)? Correcting those two terms gives 35 kHz, which is more reasonable. \$\endgroup\$ – ocspro Feb 23 at 19:54
  • \$\begingroup\$ @ocspro Thank you for your response! You got me to rethink it and clearly my extrapolation was wrong. At 25*C Eon drops down to only 1mJ which is way bigger than my estimation. I corrected both of these. The 1/2 however is there because I interrupt the high frequency waveform with a 50% duty cycle square wave so I averaged the losses over the period of this square wave. That is why I put the 1/2 in there. Now the maximum frequency of ~100kHz seems a bit more reasonable. (This is an FGA60N65SMD fast IGBT) \$\endgroup\$ – PaulW Feb 23 at 21:09
  • \$\begingroup\$ @carloc Thank you for your response! Well the busbar voltage would only affect the total power loss because of the leakage current (250uA for this IGBT) which is very low and can be omitted. I plan to use rectified EU mains voltage (325 VDC) which is almost 2x smaller than the breakdown voltage of this IGBT. And in terms of accounting for the thermal impedance I have noticed that its value for 50% duty cycle is very close to the normal Rth value and I just prefer to underestimate a bit rather than overestimate \$\endgroup\$ – PaulW Feb 23 at 21:19
  • \$\begingroup\$ @ocspro is my reasoning right? \$\endgroup\$ – PaulW Feb 23 at 22:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.