First DRSSTC Build Log

[Saattvik24]

This is my first DRSSTC. I've built most of the separate parts, now I just need to assemble them together.

Bridge - SKM100GB12T4 in Full Bridge
Bridge Supply - 0-260VAC, 35A Bridge Rectifier, 4700uF 400V EPCOS Electrolytic Capacitor
Primary Coil - Spiral Primary - 7 turns of 1/4 inch OD copper tube with 10mm spacing and an Inner diameter of 150mm
Secondary Coil - 110mm diameter, 540mm long, 1350 turns, 28SWG enameled copper wire
MMC - 3 strings in parallel of 3 in series of Cornell Dubilier 942C20P15K-F for 0.15uF at 6000VDC
Snubber - 2uF 1kVDC 827V/us dv/dt (KEMET C4BSNBX4200ZBLJ)
Top Load - Toroid -  Minor Diameter-127mm   Major Diameter-490mm 
Driver - UD2.7C Thru Hole
Resonant Frequency - Around 150kHz

I don't have access to a tunable inductor, so I was thinking of making my own. The plan is to 3D print a screw and a housing. The winding will wound on the housing, and there will a ferrite core inside the screw which I'll be able to move up and down by turning the screw. It will be much bigger than a traditional coilcraft inductor, but should work I guess? What are your thoughts? Perhaps when I figure out the inductance value, I can remove it and replace it with a fixed inductor.

While connecting the grounds together, should the 24v supply case be connected to the mains ground or the common rf ground?

I noticed later on, that the skm100's I've bought have slower on/off times compared to some other models. So I hope that wont be an issue. I was planning on setting the OCD to 300A during testing, and increasing it to 500A after proper tuning. Can these IGBT's handle 500A?


I've made some annoying mistakes while building my inverter bridge. The IGBTs are too close together resulting in there not being enough space for the bus capacitor, as you can see it covers up the igbt screw terminals and the snubber capacitors cover up the wire connections. This means the bridge has to be assembled in a particular order, and if I need disconnect anything, I'll have to remove the IGBTs from the heatsink. The only benefit of this layout is lower stray inductance I guess.


I made the toroid out of 5 inch aluminium ducting. I wanted to ask, is the way I've connected the secondary to the ducting good or bad? The top of the secondary goes to a lug connection, connected to a threaded rod, which comes out the top, which is connected to the ducting using aluminium tape.

[flyingperson23]

Your IGBTs can probably handle 500A for 100uS, probably not for 1ms. Most bricks can probably handle 500A as long as you have short enough on times and duty cycles and don't expect them to last forever. You mentioned you have a threaded rod coming out of the top of your secondary - any metal inside the secondary increases the likelihood of it arcing over and destroying your secondary.

[Saattvik24]

I built a proper gate drive transformer and turned on my bridge for the first time with a 10k resistor as the load, and 15V on the busbars and got the following output.


I think this is how its supposed to look like with a resistive load or are the voltage spikes too high? I think they should reduce once I reduce the GDT connection lead length.

[davekni]

I think this is how its supposed to look like with a resistive load or are the voltage spikes too high? I think they should reduce once I reduce the GDT connection lead length.

Looks reasonable.  GDT would be better with four parallel primary wires, one twisted with each secondary wire (along with shorter leads).