Weird Primary Voltage Waveform / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Weird Primary Voltage Waveform

Move Thread

LAN_403

MRacerxdl

Sun May 22 2011, 01:00AM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Hi all,

My DRSSTC is running fine, or, when it works, it runs fine for very long times.

Last time my DRSSTC was running with 1m sparks from a Full-bridge of IRG4PC50UD and 200A Primary current. (I think, I have two or three peaks with 230A, on the current waveform). It works good, with no problems, but since last time it blowed by a spark on the bridge:

It blowed two IGBT's (the two where the spark got), 4 TVS (two in each IGBT), 4 Zeners, and for good, the BUS Cap over the IGBT's survived (the spark didnt strike at it). And the gate short circuits blowed the TC4429 at GDT Driver.
The other side of the bridge is working good, but for safety, I changed the IGBT's too.

So I rebuilt the bridge, same specs like before, but replaced the blowed parts (and even the two IGBT's that was working. Who knows, it maybe isnt working well, I didnt tested it at all). The coil isnt working good now. At about half input voltage (220VAC mains, so half is about 150VDC at bus) the arc got unstable size (like if it was failing to oscillate), but the waveforms are normal, and the Frequency is at 97kHz all the time.

I suspected that the arc affected the primary cap (4x5 bank of 47nF 1600V, totalizing 3200V 117nF) but I measured the capacitance, and its the same as before. Any ideas that what can causing the problem? I has this problem before (the coil was working fine on a day. On the other not. After I spend much time looking for what happened to the circuit, I realize that is nothing wrong, and the circuit works o.O )

Also, I always measure everything with my scope, and the primary voltage (it was ALWAYS like that, even when it working very good with long arcs. And I runned the coil for more than 10 minutes at max power), and its VERY strange:

And at the end of the cycle, is even worsed:

And here, is a pic of the TC4429 Output (is one output in reference of ground, so the effective voltage at GDT is the double, so 24Vpp)

The yellow one is the current at primary. As it increases, the primary voltage waveform gets worsed. Some people said about scope probe interference, but I dont got anything strange at gate waveforms (the gate waveforms is very good, about the same that TC4429 waveform).

For specs, I use 2.7uF 400VAC Polypropilene Caps at each side of the bridge (as the igbts are a little far from each other, I put one for each two), there are 5R Gate Resistors with 33V Zeners (I realized that when the switch off is not synced with the fres, I get a really high ring [ about 40V ] at gate, So I put the zeners for safety), running at 24V, One GDT for two IGBTs with 15 turns (and primary is two wires in parallel) . I'm testing with 220VAC Mains, with no OCD (I dont need it here, I think because the primary surge impedance, I get the current limited. Here it doesnt get more than 230A)

Any ideas what can causing this strange voltage waveform? As I said, it was that waveform when it works perfect, and when it works bad.
Also I dont know why the coil isnt working, if someone has any sugestion to where I can start looking :D

Sorry for the long post, I know that is bad reading much :S

Thanks!

MRacerxdl

Tue May 24 2011, 05:44AM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Tested the coil today, and solve few things. Increasing the GDT cap from 300nF to 4uF (with an eletrolityc cap in parallel) makes the wave better. Also I noticed that the average voltage at IGBT gate was 10V. So I made the GDT's as 1:2:2 each, now the average is 20V the peak 24V.

The waveforms are still the same, but the coil I think is working better, hooked up today, 60% of Variac at 220VAC Mains and I got about 75cm sparks:

Goodchild

Tue May 24 2011, 08:41PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

It looks to me like your feedback phasing is reversed. Try swapping the CT 180 to change the phasing.

[EDIT]

Forgot to mention another tell tail sign of reversed phasing is faster than normal switching.

So faster than normal switching and a flat topped current waveform like you have.

MRacerxdl

Thu May 26 2011, 05:30PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Hey, thanks for the tip :D

But if I reverse the feedback, the coil doesnt oscillate.
I revised a few things and got better Gate Wave, but the primary wave is the same.

The strange thing is that the coil is working perfectly now:

That is 60% power input.

My driver isnt the normal ones, but its not very complex, since I dont have access for UCC, I did only put 74HC08 And Gates to make the Enable. So the driver is simply a 74HC14 -> 74HC08 -> TC4429.

In the changes, I made a 1:2:2 GDT instead 1:1:1 GDT, that was because the average voltage was 10V at IGBT gate, and I think that is no good for DRSSTC. With a 1:2:2 GDT I Got 20V Average (24V Peak) so I think its ok.

The only thing that I didnt got the maximum power is because I get racing sparks between Primary and Secondary, but all other things are good.

My Scope Isnt grounded, someone said that can be noise. But it only happens at bridge waveform. Even the primary current waveform is correct.

Goodchild

Thu May 26 2011, 07:25PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Oh well you were running 10V? Yeah you where probably running the IGBTs in the linear region, not good at all!

MRacerxdl

Thu May 26 2011, 08:27PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Yeah, the peak was 12V, but the average was 10V, the minimum conduction voltage. So I made 1:2:2 GDT. Now the average is 20V and Peak 24V.

But even with the 1:2:2 GDT, the waveform still the same. The curious thing, is that the coil is working PERFECTLY (I think 1m sparks with IRG4PC50UD FullBridge is a good result), the coil runned yesterday for about 1 hour in audio modulated mode. The Bridge didnt get more than 25ÂºC temp (21ÂºC ambient). about 200Apk on primary. Nothing wrong, just that bridge waveform.

EDIT:

Arc Pics:

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob