DRSSTC Approval

[evaniscool]

Hi, I'm 16 and I have been creating a drsstc plan for the past few months in hopes of finishing one for college credit. My parents want me to get feedback on my project from experts before purchasing the parts and creating it. My plan closely follows the one featured on Kaizer Power Electronics DRSSTC design guide and will look somewhat like the one featured here with a few extra tvs diodes and bleed resistors. Some important details about the project are listed below:


- IXGN60N60CD21 IGBT fullbrige
- 942C20P15K-F 0.15uf capacitor x12(3 in series and 4 in parrallel)
- 22 inch by 4.5 inch secondary with around 1800 turns with topload resonating at 120khz
- UD2.7 driver by loneoceans
- OCD at 500A
- Two ALS70A242KF450 450v 2400uf capacitors in voltage doubling configuration
- My busbar will be made of 1mm copper sheet
- I plan on running the DRSSTC at a maximum of 350v at 10% duty cycle.
- As for my interrupter I will be assembling a similar one to the one featured here.
- More details are shown on the whiteboard.


My parents main concerns about this project is saftey. I have had previous expereince with high voltage circuits and have made my own ZVS driver and flyback transformer along with a smaller spark gap tesla coil but my parents and I would appriciate feedback and tips regarding saftey and regarding my project as a whole. 

[flyingperson23]

The only thing I can see in your schematic that you definitely shouldn't do is having any feedback transformer between the MMC and the secondary. The voltage there is too high, you need to put feedback transformers on a direct output from the IGBTs.

As for safety, the main danger is the bus and anything connected to it. Make sure any capacitors are discharged before going near the coil.

General advice:
Minibrick IGBTs aren't great. Full bricks can be found for quite cheap on eBay (i.e. https://www.ebay.com/itm/392904008730?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=N3501plZTCG&sssrc=4429486&ssuid=crhscpv5ses&var=&widget_ver=artemis&media=COPY) and are more robust than minibricks. Just make sure you get 1200v bricks for any bus voltage >400v. Buying power electronics from ebay generally isn't a huge safety concern and can get you much better parts for your money

It would be better to put an MMC drain resistor across the bridge output instead of directly on the MMC so that it won't see the super high voltage across the MMC

For bricks at least, TVS diodes across the bus are unnecessary and can be harmful. Bricks' internal diodes can act as tvs diodes to an extent and you shouldn't be reaching that voltage anyways. I haven't used TO-247 enough to give any advice if you use those

Put a reverse diode across the gate resistor so the IGBTs turn off faster than they turn on, preventing any shoot through

If you're not planning on powering it from a variac, you might want some form of precharge

Good luck!

[Mads Barnkob]

Hi Evan

Welcome to HVF and it is great to see a young man that has done his homework. That is one crazy white board, maybe try to go for excel for the formulas next time, makes it much easier to adjust the design when you basically have collected all the stuff already.

C17 can result in quite a lot of leakage, which is better than a blown bridge, but the this also means that if you put the coil in a metal enclosure and touch it when powered up, you can feel it / close to getting zapped mildly, it's not pleasant.

I will also advice against the 60N60 minibricks, as I actually blew that bridge up (minibricks are just beefed up TO-247 devices) and its now using a SKM200 brick bridge instead. Even some SKM100 or similar bricks would better than spending 110$ on IXGN60N60. If you can not find any, I got some for sale in sales section, we can make a deal.

If you are not going for bricks, then use TO-247 IGBTs instead and use the UD2.9 DRSSTC pack which has a nice dual full bridge PCB layout in it as well: https://highvoltageforum.net/index.php?topic=353.msg2401#msg2401

The voltages that you use in calculations, 230/400V, is that AC or DC? It should be DC.

[evaniscool]

Thank you both for the feedback, I have since corrected the issue with the feedback transformer and will look deeper into the issue with the TVS diode across the bus. To clarify some things I am going to use a variac and the voltages in my 230/400V calculations are DC.

As for my choice of using 60N60 minibricks, I have chosen these mostly because when I started this project back in December I found a great deal online for these and bought 8 of them in fear that they would have sold out before I had completed planning the project. Now that I am more knowledgeable on the topic I wish that I had purchased a brick IGBTs instead of the 60N60s.

I would be willing to sell/trade my 8 60N60 minibricks in order to get some IGBT bricks. Which IGBT would you recommend?

Again, thank you flyingperson23 and Mads Barnkob for the feedback I appreciate it a lot 

[flyingperson23]

Which IGBT would you recommend?

Look for 100, 150, or 200A half bridge bricks. 50 or 300A bricks would probably work but would be less ideal. Definitely look for half bridge bricks, single transistor bricks are much harder to design for. Any brick with a name starting with 'CM', 'SKM', or 'BSM' will definitely be from a good manufacturer. Other manufacturers make suitable bricks as well, but those three series (from powerex/mitsubishi, semikron, and eupec/infineon respectively) are the most commonly used for drsstcs.

[evaniscool]

Would the 2MBI200N-060 IGBTs from https://ham.brugtgrej.dk/annonce/192531 work? If so I would be happy to purchase them 

Also, my parents are concerned about my saftey while constructing the drsstc. I have told them about my bleed resistors, the one hand rule and how I would test the voltage across the bus before touching the circutry. However they(understandably) would like an expert opinion on the topic of staying safe while constructing the drsstc.

[flyingperson23]

If you look at the datasheet for the IGBTs, it says they have an internal device that limits current to 600A, so as long as you have no plans to target an OCD >600A they should work fine.

The main danger is the voltage in the bus caps, so if you're just building the coil and you don't charge the caps and have bleed resistors on them, you're fine. The only significantly dangerous act is to mess around with the coil while the bus caps still have a lot of charge. If you're really worried, maybe put neon lamps across the capacitors for a visual indication of whether or not they're charged. Of course all the normal construction safety concerns still apply, like use eye protection with angle grinders, don't touch a hot soldering iron, etc. but the only thing within the drsstc itself that can seriously hurt you is capacitors keeping charge after a run.

You probably also want a discharge resistor across C17. Unlike the bus caps, it's unlikely to be lethal but can still shock you pretty good if it builds up charge.

[evaniscool]

My target OCD is 500A so I think I should be fine. I will also put a bleed resistor across C17 to stop myself from getting a shock.

Thank you so much for your help, I appriciate it a lot