Electronics > Projects, Designs, and Technical Stuff
TESLA COIL Hbridge problem PLEASE HELP
(1/1)
justin66:
Hello guys i cant seem to get my TC working
Im using universal driver UD 2.7c
And im using this hbridge design
https://github.com/profdc9/DRSSTC-PCB-Pack/tree/master/full-bridge-transistor

Project photos
https://ibb.co/1n6z77S
https://ibb.co/Bqc8YrV

When probing my GDT (gate drive transformer) output. Notice at the waveform there is a spike that is not supposed be there. Why is that present?
https://ibb.co/rfPjchR

This is the bridge output using fgy75n60 igbts (its supposed to go to the LC primaries
https://ibb.co/dpCZfTB
What is this mess and how do i sort this out? Btw, connecting halogen light bulb does nothing when connected to this mes
dmills:
On what planet is that monster a sane gate driver transformer for a couple of TO247 IGBTs?

Also 400+KHz, with IGBTs? really, are you sure?

From the look of that scope plot you are probably measuring between something like local ground and the high side gate?
Yea, not going to get sane measurements that way (Quite apart from maybe blowing up the scope if you get it wrong). An isolated differential probe would be your friend, as with the whole dogs dinner up at rectified mains I don't see a reasonable way to get a trace of the actual gate drive voltage (as in the gate-emitter voltage, which is what matters).

You might also wish to consider the potential between the low side gates and the high side ones, and the possibility that two strands of cat 5 signal wire just might not be insulated for all that voltage, just a thought.

Can you get your schools technicians (Generally more useful the the lecturers for this kind of thing) to help you sort it?
This is impossible to debug remotely (Especially as you clearly don't really know how to measure what you need to measure) and you really need someone who has high voltage, high frequency power experience on site to try to unfuck that thing. Expect there to be new IGBTs (and possibly new IGBT over voltage protection silicon) in your future together with a sane gate transformer and probably a current transformer to protect the IGBTs, also expect there to be some sheet metal and an isolation transformer involved, together with some much better wiring.

You have clearly bitten off WAY more then you can chew here.

Regards, Dan.


justin66:
Thanks for the reply. The calculated resonant frequency is 270-330khz depending on topload.
I was driving feedback pin with 300khz 5vp-p sinus. Im not sure why osciloscope measures 400+khz.
Im not sure about 300-400khz application but i have seen several igbt datasheets describing resonant zvs application above 200khz
I dont have, neither my coledge has diferential probe handy, my measuring methood was isolate osciloscope grounding pin at the power plug and use osciloscope that way.

I know i bit a bit too much for me on the electronics side of things. Im capable of assembly and im always open to learning.

The bad thing is none of the lectures want to help either because they dont know shit about tesla coils the ones who want to help dont know enough to help. Im kind of stuck with it

As for measuring i was measuring gate drive transformer directly. It is 1:1 ratio. I previously used 1:2 but i think it had too much voltage, because my driver board is 24vdc and it is directly switching 24v

I used already known working design from the steve ward design website. The UD2.7c is just an improved version of his design
jttps://www.stevehv.4hv.org/DRSSTC1.htm
alsetalokin4017:
You are floating the scope? Not really a good idea, but...

dmills has it right. Dog's breakfast, more than you can chew, etc. I can't even think of how to start troubleshooting except what we've already said.
Maybe there is something in this video that will help.

dingo27mobile:
Hi, i did some DRSSTC back in the days with IGBTs as well, and from the design -
1) i would swap the gate resistors from 10 to something like 1-2 ohm ones, make sure they are SMD or low induction ones.
2) i never did the lan wire, just used some kind of small high frequency ferrite core (2cm dia max., dont have datasheet or PN anymore) and thin enameled wire, that worked well enough. Altough the lan wire seems to be popular choice.
3) try to use shorter wire connection from GDT to driver board / use coax wire.
4) I never used the "bridge controllers", a simple setup of two pairs of PNP+NPN sink/source common emitter like here http://www.talkingelectronics.com/projects/TheTransistorAmplifier/images-P3/PushPull_5v.gif worked well. You can drive it with easily with any N mosfet and pullup resistor (it will twist the signal 180 degrees though!)

The "swing" indicates you have some inductance somewhere.

However all this was like 7 years ago, i guess things changed rapidly since then. I also had a around 400kHz TC. IGBT of choice was IRG4PC50W, and i was able to get like 15cm sparks. Never did a scope sampling of output though. Also, i may be wrong, i am no wikipedia.
Navigation
Message Index
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod