Thank you sir this is brilliant. This RU-vid channel will be part of your legacy. Keep them coming though, I hope to see many more in the years to come 😊
Many thanks! Comments like yours keep me going. We will keep crossing our fingers that the RU-vid's servers will last at least as long as the pyramids.
Hi , Thanks for your presentation . Basic Question: How did you create the " Cauer Netwrok" from Zth curves ?. The manufacturer do not provide the thermal circuit and only Zth curves are in hand . however with these curves only the Tjmax is possible to be calculated
Thank you for the all of the your great videos! These thermal models are usually from junction to case or from junction to mounting base (nexperia). Now if you want include ambient temperature to simulations/calculations and to get junction in different ambients pcb layout etc starts to affect. Do you have any suggestion which would be the best way to get estimate for junction in different ambients?
The key is of course the thermal resistance between case and ambient. There are simulators on the web that can help to estimate this thermal resistance for heatsinks and PCBs
Is there a easy way to change the displayed unit on the scale from V to C ? this is mostly for ease of reading the data. This may be on a tangent BUT the FET is rated for 1200V but safety wise what margin is there or what should i use when designing? can i switch 1200V and call it safe? or should i derate and only go for 1000-1100V? Asking since for lower voltages i often add 12-15V and call it a day but here at the high end that amount is potentially in the measurement tolerance
V to C? please indicate the relevant minute of video. The safety margin largely depends on the expected voltage over shoot. Without a snubber the voltage peaks could be very high.
Depends on the author of the spice model. If you use a spice model, self heating, level3, then you can model linear mode as well. Manually, you can always estimate the power loss in linear mode, and force an external current source through the Cauer Model. I might publish this technique.
