Substituted New IGBT for Old CM300 - ISSUE

[Jamespie]

I built a coil identical to Loneoceans DRSSTC 3 (except I laid out a better, lower inductance, laminated bridge) https://www.loneoceans.com/labs/drsstc3/ . I found some genuine CM300s on eBay, I laid out my own UD2.7c's PCBs in KiCad, got them fabricated, everything was a breeze. Running the coil from my Variac at 1/2 max voltage yielded fantastic results.

Long story short, I blew the CM300s and everytime I ordered more from eBay, they were fake/already used/blown. I decided to find some IGBTs from Digi-key with similar specs. I found a great alternative BSM150GB170DN2E3256HDLA1-ND. I bought the rest of them as they're at their end of life. They can handle higher voltages and their gate capacitance is 1/4 of the CM300s, same pinouts.

I put these new bad boys in and adjusted phase lead (everything else stayed the same). The coil is working, but performing horribly. Maybe a few inch streamers from 90VAC input while my old coil would have been putting out multiple foot streamers. My Variac is humming/vibrating and it's chassis is slightly shocking me! Everything on the power splitter that the Variac is plugged into is shocking me via exposed metal. Old coil did not do this. I'm at a loss. Any help is much appreciated!

SCOPE SHOTS:

I put my high voltage differential probes on the inverter output and on the primary current sense transformer

[davekni]

My Variac is humming/vibrating and it's chassis is slightly shocking me!

I'd guess that your AC line incoming bridge rectifier is fried.  Half-wave rectification would be a likely reason for Variac to hum/vibrate.

Everything on the power splitter that the Variac is plugged into is shocking me via exposed metal. Old coil did not do this.

Is the exposed metal grounded to line safety ground?  One possibility, especially if metal isn't grounded, is that the rectifier or some other item is conducting current to heat sinks and/or other metal structure.  Do you get shocked if interrupter is off and power is applied to rectifier (and therefore to H-bridge)?  Or, is shocking only when coil is making small sparks?

[Jamespie]

My Variac is humming/vibrating and it's chassis is slightly shocking me!

I'd guess that your AC line incoming bridge rectifier is fried.  Half-wave rectification would be a likely reason for Variac to hum/vibrate.

Everything on the power splitter that the Variac is plugged into is shocking me via exposed metal. Old coil did not do this.

Is the exposed metal grounded to line safety ground?  One possibility, especially if metal isn't grounded, is that the rectifier or some other item is conducting current to heat sinks and/or other metal structure.  Do you get shocked if interrupter is off and power is applied to rectifier (and therefore to H-bridge)?  Or, is shocking only when coil is making small sparks?

It's not the diode bridge, I already checked it and plus diodes typically fail short. I have a FLIR and checked all components while running after a while, none presenting any abnormal heat. Good question about shocking with/without interrupter, I'll check that later today

[Jamespie]

I'd guess that your AC line incoming bridge rectifier is fried.  Half-wave rectification would be a likely reason for Variac to hum/vibrate.

I just double checked this. Diode bridge is fine. I even checked from pins to ground pad. See my response about checking via FLIR. The vibration is very mild and light. It's not audible until I'm above 40-50 VAC, even then, it's quiet.

Is the exposed metal grounded to line safety ground?  One possibility, especially if metal isn't grounded, is that the rectifier or some other item is conducting current to heat sinks and/or other metal structure.  Do you get shocked if interrupter is off and power is applied to rectifier (and therefore to H-bridge)?  Or, is shocking only when coil is making small sparks?

Just checked this. The exposed metal on the Variac is tied to safety ground. This is slightly shocking me, but only at higher outputs. What's shocking me more noticeably is my little fiber optic signal generator I bought from AliExpress (see attachment). It gets 24VDC from a power brick which is plugged into the same power strip that the tesla coil is plugged into. The generator has metal knobbed rotary encoders used to set frequency and pulse width. These metal knobs are shocking me. I can see teeny tiny arching to my finger. This shocks me, the safety ground shock me, and the variac hums,  ONLY when the interrupter is on. When it's off, everything is normal.

One more thing to note:

The coil has a strange intermittent "cracking, crackling" sound. Here's a video of it: https://youtube.com/shorts/Q5-mH2lG2kE?feature=share

[Jamespie]

After watching my video again, I noticed one more thing that's different between this coil and the last coil. My breakout Rod is not the same. My last coil had maybe 8 to 9 in sticking out the edge of the toroid. I just quickly grabbed a rod that was on my work desk which doesn't even stick past the widest point of the toroid... I assume breakout Rod length directly influences breakout voltage. I'm not sure if this is significant enough to make this big of a problem... Won't be able to test this until tomorrow morning or late tonight

[davekni]

After watching my video again, I noticed one more thing that's different between this coil and the last coil. My breakout Rod is not the same. My last coil had maybe 8 to 9 in sticking out the edge of the toroid. I just quickly grabbed a rod that was on my work desk which doesn't even stick past the widest point of the toroid... I assume breakout Rod length directly influences breakout voltage. I'm not sure if this is significant enough to make this big of a problem... Won't be able to test this until tomorrow morning or late tonight

Seems possible that breakout difference is the cause, making top load voltage higher.  Would require higher primary current to reach higher voltage, perhaps enough to make Variac humming.  (Also, previously arc sound may have drowned out Variac sound.)  Higher voltage capacitively-coupled to your body will make your body voltage higher, so make bigger shocks when touching anything grounded.

Will be interesting to see how your breakout test goes.  Good luck!

[Jamespie]

After watching my video again, I noticed one more thing that's different between this coil and the last coil. My breakout Rod is not the same. My last coil had maybe 8 to 9 in sticking out the edge of the toroid. I just quickly grabbed a rod that was on my work desk which doesn't even stick past the widest point of the toroid... I assume breakout Rod length directly influences breakout voltage. I'm not sure if this is significant enough to make this big of a problem... Won't be able to test this until tomorrow morning or late tonight

Seems possible that breakout difference is the cause, making top load voltage higher.  Would require higher primary current to reach higher voltage, perhaps enough to make Variac humming.  (Also, previously arc sound may have drowned out Variac sound.)  Higher voltage capacitively-coupled to your body will make your body voltage higher, so make bigger shocks when touching anything grounded.

Will be interesting to see how your breakout test goes.  Good luck!

IT WAS THE STUPID BREAKOUT!! My god, after 8+ hours of debugging... Leave it to me to make the simplest mistakes and start debugging the most complicated, possible causes first... Yikes...

I mounted the rod a good 8 inches away from the toroid and all seems normal. I have some primary tuning to do now. I'm still getting some energy back on earth ground, but this is likely a grounding issue. I heated a brass 1/4" threadlet to melt and set into the base of the secondary PVC shell. But during the heating process, it got quite oxidized and is probably not making the best connection to my ground lug. Thanks for you help!!

[davekni]

Glad to learn that it is working again!  Many issues are simple in hindsight.  I've wasted many more hours chasing such "stupid" errors.  Comes with being human.

Nice laminated bus bars.  You have avoided many common problems that way, problems that could have required much more than 8 hours to find and fix.

[Jamespie]

Glad to learn that it is working again!  Many issues are simple in hindsight.  I've wasted many more hours chasing such "stupid" errors.  Comes with being human.

Nice laminated bus bars.  You have avoided many common problems that way, problems that could have required much more than 8 hours to find and fix.

That's very true. Oh the stories I have with debugging circuits that I thought I knew intimately....

Well thank you! 3D printed core with 1/8 copper plate I cut with a jig saw. Another thing that I will never do again! It is worth the $300 to get machined!

[davekni]

Well thank you! 3D printed core with 1/8 copper plate I cut with a jig saw. Another thing that I will never do again! It is worth the $300 to get machined!

Parallel copper planes is the important part, minimizing parasitic inductance.  Given TC duty cycles, copper foil is usually sufficient.  I built one with 100um scissor-cut foil.

[Jamespie]

Well thank you! 3D printed core with 1/8 copper plate I cut with a jig saw. Another thing that I will never do again! It is worth the $300 to get machined!

Parallel copper planes is the important part, minimizing parasitic inductance.  Given TC duty cycles, copper foil is usually sufficient.  I built one with 100um scissor-cut foil.

Yes I figured skin effect would allow a thin copper laminate; I wanted to use thicker to allow the bus to be structural and hold the whole assembly, no bracing. Also for heat sinking purposes. I hugely over estimated heat though. My FLIR shows a very small temperature rise running at 100us pulse width at 100-160ish Hertz

I love your idea though!! I never thought of copper foil! I didn't even think to look it up! What was your core material? How did it maintain structural rigidity? I may redesign the bus with this foil because it's a PAIN to assemble/disassemble

[davekni]

I love your idea though!! I never thought of copper foil! I didn't even think to look it up! What was your core material? How did it maintain structural rigidity? I may redesign the bus with this foil because it's a PAIN to assemble/disassemble

Mechanical structure came from heat sink (and two steel rods join the heat sink to an adjacent piece of polycarbonate sheet).  Mechanical design is a kludge - my weakest skill.  Insulation is 127um polyimide film.
https://highvoltageforum.net/index.php?topic=1324.msg11785#msg11785

This was made more as an example of an extreme in low parasitic inductance.  Used it for my low-frequency QCW experiment.  Will likely get used in the future for some other coil.

[Jamespie]

I love your idea though!! I never thought of copper foil! I didn't even think to look it up! What was your core material? How did it maintain structural rigidity? I may redesign the bus with this foil because it's a PAIN to assemble/disassemble

Mechanical structure came from heat sink (and two steel rods join the heat sink to an adjacent piece of polycarbonate sheet).  Mechanical design is a kludge - my weakest skill.  Insulation is 127um polyimide film.
https://highvoltageforum.net/index.php?topic=1324.msg11785#msg11785

This was made more as an example of an extreme in low parasitic inductance.  Used it for my low-frequency QCW experiment.  Will likely get used in the future for some other coil.

This is very impressive and something I would do! Thank you for sharing, I ordered some copper sheet