Hello experts, I am currently working on a tesla coil design, it involves a half bridge inverter and I ran into some problems, so I though I'd come here for some advise. Below is the half bridge I made, it's driven by a gate drive transformer(powered by a gate drive IC), it has 320VDC as the input, 2 IGBTs as transistors, a LC as the load, and a switching frequency of around 154kHz, I also added a RC snubber for transients.

My first question has to do with soft switching, I wanted to decrease the power loss and heat so I did some research and added a soft switching circuit(the red part in the picture below, I believe it is called the "passive lossless snubber"), but the circuit doesn't seem to be working as there are still overlaps between the voltage and current during turn on and off(second picture below, green is the voltage and red is the current, see the nodes on the first picture for referrence), and the power loss can go up to a couple thousand watts which is scary. It might also be the values of the soft switching circuit not being correct but I cannot find an article on how to calculate the values, so if anyone is knowlegable please help, and if the implementation is just wrong please tell me also.

*also these are all done in PSpice simulation




And here's the second question, from what I see on other half bridge projects on the internet, their voltage and current graph is pretty different from mine, their voltage and current all show a square wave-like form, but mine just looks really wierd(below is a clearer picture of the full result). You can see that it starts of with a constant 160V, then the voltage across the IGBT has a lot of wierd spikes, it also doesn't fully reach 0 when it's turned off, then only after a while will it finally become normal. As for the current it's just wierd and I dont even know what is happening. I believe if all is operating normally then both current and voltages should both look like square waves, and when V is high I should be low and vice versa. If you know please explain what causes these wierd behaviors and what I might do to fix it


Finally I have one last question, for most tesla coils I see online, they have a peak current of hundreds of Amps going through the primary coil, but I seem to only be getting 10-20 or even lower current(below is a graph that shows current(red) and voltage(green), refer to the nodes), could some components be limiting the current? Please comment if you find something.


*
IGBT: AFGB40T65RQDN
Gate Drive IC: UCC21520
Also I'm pretty sure nothing is wrong with the gate drive, I tested everything regarding that.
*

Sorry if this is a lot, I am quite new to all this so there are definetely a lot of things that I missed or overlooked, your help will be much appreciated, thank you in advanced, also if you need more details please let me know, thanks!

 

I can’t see what IGBT you are using so I would question the 12V Zener across the gate and emitter. Also the snubber values of 8uF and 40 ohm. How did you come to those values??
What dead time are you using between the two drive signals ?

 

the IGBT is AFGB40T65RQDN, dead time is around 0.25us(about 3% duty cycle) and the snubber values were initailly calculated(initially 50uF, 300m ohm) but through some testing I made some tweaks(these values are already the best among all, the waveform looks the cleanest). Also I used pulse signal instead of GDT and added the zener to test if it was the problem and I find the two to be the exact same, so I don't think that is the problem(I believe the gate driving part is good, the problem is only the main inverter). I hope these imformations are enough, let me know if you need more info, thanks !

 

Switching 300V+ at 154kHz SiC seems like a better fit for this application. From your IGBT datasheet, switching 400V @ 40A at temperature, the switching losses only (no conduction or snubber losses) are around 3mJ which at 154kHz mean 460W of switching losses. That alone is more than the maximum power dissipation for that D2PAK.
Also the snubber caps seem orders of magnitude off.
Hope this helps.
JP