STP19NM50N SPICE Model Complete

Great post! Unless my eyes deceive me, it seems that many of your images are a composite of a simulated waveform superimposed on a black V versus C grid, which itself is superimposed upon a oscilloscope photo.

_SPICE - Practical Device Modeling_ (Kielkowski, 1995) contains a whole chapter, "Chapter 6 Modeling the Power MOSFET," which is devoted to MOSFET parameters. But it may be too dated for a STP19NM50N. Anyhow, keep up the great posts. :)

Thank you,

Where'd you get the SPICE model?

Mouser has a factory lead time of 38 weeks; worth the wait I guess?

Digi-Key has 307 in stock, $3.55 ea qty 10. The TO-220 stp19nm50n (or TO-247 stw19nm50n) make more sense to me for high-power applications, also lots of stock. The 1nF input capacitance is a pain, but these are superjunction mosfets, which is helpful for low Coss, etc. Coss drops from 2200pF!! at 5V to 40pF at 100V, helpful for high-freq switching. If you look at what's available for 500, 600 and 650V parts, there are many good choices. Like the 650-volt 230-milli-ohm STP16N65M5, $3.30 at Arrow.

Tim, I've forgotten, did you ever take a look at my RIS-724, a 1.4kV MOSFET Eoss testset?

Think I had... it's a lot of work though. No need when I can extract that, and dynamic and static characteristics, from just a couple waveforms. Simple breadboard setup, then the engineer's favorite, spreadsheets.

I do need to build a fixture to test reverse recovery (the last item to touch). Breadboard has too much stray to do that. :)

Your other question: I started with the STW21NM50N model, adjusted DC parameters to meet measurements as best I could, and fixed up dynamic parameters with my own capacitance model.

Tim