IGBT signals problem / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

IGBT signals problem

previous 1 2 3 4 next

Move Thread

LAN_403

Physics Junkie

Thu Feb 14 2013, 08:43PM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Thank you both for response. I just compared with the 5.1ohm.. what I found is that the 10ohm completely eliminates overshoot but at the cost of 30-50ns or more, hard to tell exactly. I will indeed be going back to the 5.1ohm. Hopefully the weather clears up so i can fire up the coil this weekend

Physics Junkie

Thu Feb 14 2013, 09:22PM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

I thought I would add this for people's reference, it might be helpful. This is the waveform with the 5.1 ohm as well as the rise and fall

Its hard to tell, but the rise and fall is slightly faster with an 'acceptable' amount of overshoot, whereas in the above posts you can see with the 10ohm resistor that there is no overshoot but the rise and fall is unreasonable.

Also, I believe the culprit was the 33V zeners. There were two that were dead, one on each side. I forget exactly but I think It might have been the ones on the emitter side. How they got destroyed? I have no idea lol.

Graham Armitage

Thu Feb 14 2013, 09:47PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

Here's a thought - what if the GDT output was +-35V and the zener clamped it at 30v. Should prevent any overshoot that both you and I are experiencing? Wouldn't that give a very square output? I think that might be why EVR suggests a 1:2:2 GDT winding in the design.

Physics Junkie

Thu Feb 14 2013, 10:11PM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Bushman wrote ...

Here's a thought - what if the GDT output was +-35V and the zener clamped it at 30v. Should prevent any overshoot that both you and I are experiencing? Wouldn't that give a very square output? I think that might be why EVR suggests a 1:2:2 GDT winding in the design.

That's an interesting thought, I really have no idea but it makes sense what you say. In the design book I have, the full bridge shows 10:10 GDT and the measurements are all +-20V with barely any overshoot. I know the minibrute is slightly different though but not by much. The cores I am using are the heavy duty 5000 type, I wonder if that has anything to do with it.

Graham Armitage

Thu Feb 14 2013, 10:21PM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

I am using the EVR toroids - (Fair-Rite 5977006401 Type 77 Material). The design does call for a 10:20:20 winding, which when I tried, gave me a 30v output. Not sure what the extra voltage provides, other than reduce overshoot? Seems higher voltage would mean a steeper rise/fall angle it required in order to keep total rise/fall time sufficiently low. I am using a 15:15:15 GDT right now. It's working so I don't want to mess with it. Will do some experimenting at some point with my experimental SSTC.

Physics Junkie

Thu Feb 14 2013, 10:28PM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Yea I think you're right

Physics Junkie

Sat Feb 16 2013, 12:47AM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Through searching around other forums I found how to adjust the add function on my DSO (since the user manual was wrong) so i went ahead and did some differential measurements of Vge. In Dan's design book, it shows that it should be +20V with rise and fall times less than 250ns. Hopefully I am taking the measurement right.. what I'm doing is inverting CH2 and using the add function. Both probes are set to 10x on the scope and the probe itself, and these are just regular scope probes nothing fancy like the 400$ diff probes. And here is what I get:
Below is +-2V rather than 20. Rise and fall is fine, and overall it appears to be a good waveform.

Above is what i'm scoping on the emitter only, about 10V, and it is the same on all emitters.

Due to the fact the all the gate wave forms look great, I'm not all too worried about it. But I did find it odd that It's reading +-2V instead, and I dont think there is anything wrong with the way I'm taking the measurement. If anyone thinks this is an issue please let me know.

Graham Armitage

Sat Feb 16 2013, 02:12AM

Registered Member #6038 Joined: Mon Aug 06 2012, 11:31AM
Location: Salado, TX
Posts: 248

Not familiar with your scope, but your invert+add technique is correct. Sure you have accounted for the 10x probe setting on the scope (2x10 =20). Why do you see both traces if you are adding them? I know on my CRT scope I have to set both channel voltages to the same setting else it gets messed up, but by adding I only get one output trace.

Physics Junkie

Sat Feb 16 2013, 02:49AM

Registered Member #7267 Joined: Tue Oct 16 2012, 12:16AM
Location: Detroit, Michigan
Posts: 407

Bushman wrote ...

Sure you have accounted for the 10x probe setting on the scope (2x10 =20).

Yes, this was my first thought that I had hit the slider on the probe with my thumb or something, but I checked and everything was still set to 10x, scope and probes.

Why do you see both traces if you are adding them? I know on my CRT scope I have to set both channel voltages to the same setting else it gets messed up, but by adding I only get one output trace.

That's just the way to DSO is. probably so that you can overlay the Math wave on the other channels to compare and whatnot. You can barely tell in the first pic but CH1 is way at the top and not visible, while the Math trace appears yellow at first glance and may look like CH1, its actually green. Anyways, I initially had CH1 voltage set to 20V but when I put the probe on the emitter I had to go down to 5 to get a good view, changing the individual channels has no effect on the outcome of the Math trace.

teravolt

Sat Feb 16 2013, 04:45AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

usually The transorbs will start clamping a lot earlier so a 33v transorb will clamp around 28v. If this happens and it clamps it will look like a short or a load to your transformer's winding. whats in the secondary will be reflected to the primary and your driver chips. usually when a transorb fails it is a dead short which can destroy the driver chips. If you look at the specs for the ixgn60n60c2d1 the continuous Vges is +-20v and you shouldn't need more than that so if your transformers don't exceed 20 you should be ok. maybe some of your earlier scope shots that were puzzling might be from shorted or clamping TVS

previous 1 2 3 4 next

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