Fired up my 10kW (to be) CWSSTC on the weekend. Big sparks then bang!....questions! / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Fired up my 10kW (to be) CWSSTC on the weekend. Big sparks then bang!....questions!

previous 1 2 3 next

Move Thread

LAN_403

swordfish

Wed Jul 31 2013, 11:33AM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Second time lucky, image didn't attach first time...

Sigurthr

Thu Aug 01 2013, 07:02AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Gate drive leads should be twisted bifiliarly as much as possible to reduce stray inductance. It makes a huge difference in gate signals.

You should try to trim those DC Bus lines to as short as possible, and then bring your DC Bus caps as close to the bridge as you can. Every centimeter counts. I eliminated 4cm from one of my earlier bridges and gained over 200W of output and reduced transistor temps by 15degC.

Also, I don't see any gate TVS diodes or snubber capacitors.... they're not really optional.

swordfish

Thu Aug 01 2013, 09:52AM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Sigurthr wrote ...

Gate drive leads should be twisted bifiliarly as much as possible to reduce stray inductance. It makes a huge difference in gate signals.

You should try to trim those DC Bus lines to as short as possible, and then bring your DC Bus caps as close to the bridge as you can. Every centimeter counts. I eliminated 4cm from one of my earlier bridges and gained over 200W of output and reduced transistor temps by 15degC.

Also, I don't see any gate TVS diodes or snubber capacitors.... they're not really optional.

Yeah I intended to make the final version much more compact and neater. Still, I was pretty chuffed that it seemed to handle a full 240VAC ok! I will definitely shrink the new version and make those leads shorter. The TVS diodes are there, they are just hard to see due to the black heatsink fins. I look forward to giving it all another go this weekend!

swordfish

Sat Aug 03 2013, 12:13PM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Fired it up again tonight. Didn't want to go past 240VAC as I am only using a 400V bridge rectifier. Will order some 1kv bridges and try 415VAC next week. Here is a short video:

Sigurthr

Sun Aug 04 2013, 02:02AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Nice work! Didn't you say that this was to be CW? That's clearly an interrupted streamer.

Dr. Dark Current

Sun Aug 04 2013, 08:43AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

It is probably "CW with noticeable 100 Hz ripple".

swordfish

Sun Aug 04 2013, 10:06AM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Thanks for the encouragement! Once I get it running reliably I will build a 6 pulse 3 phase supply running at 560v or so. For now, full wave rectified 240v will have to do! I don't suppose anyone has any thoughts on the questions I asked a few posts back? Or some suggestions for a better IGBT?

swordfish

Sun Aug 04 2013, 10:51PM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Any thoughts on these? Good current rating and high dissipation TO3 package. Fast rise and fall times and low input capacitance. Looks pretty good I think. The only problem is the 650v rating might be cutting it a bit fine for my intended operation with 590v rails. Any thoughts?
http://www.fairchildsemi.com/ds/FG/FGA60N65SMD.pdf

Sigurthr

Mon Aug 05 2013, 02:38AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Dr. Dark Current wrote ...

It is probably "CW with noticeable 100 Hz ripple".

Idk, those arcs are way more "frantic" than any I have ever seen on CW coils before. I've run mine with various filter capacitances between none, enough to prevent drop-out at zero crossings, and as much as I could cram in there so that Vripple was under 20Vp-p, and it is always a "calm" bush corona with sharp spikes like on a VTTC. These arcs the OP is getting look very much like those of a SGTC or DRSSTC, and there is a high pitched squealing in the video. Likewise the 50/100Hz hum is pretty well suppressed. This may just be an artifact of the sound recording equipment, but even so I would expect several kW of 50/100Hz to be VERY loud. The 120Hz 150Vp-p ripple on my 3kW coil is damn near deafening at only 177Vdc Bus voltage (minimal capacitance). I'm curious if there is any interference feeding back in to modulate the enable lines. Perhaps they could be tied to Vdd to ensure true CW operation?

Re: the OP; Sorry, my experience lies mostly with MOSFETs, not IGBTs.

swordfish

Mon Aug 05 2013, 08:42AM

Registered Member #3218 Joined: Mon Sept 20 2010, 10:42AM
Location:
Posts: 29

Well there may be some interference going on, but it's nothing to do with any enable pins cause my driver ICs don't have them. Sigurthur do you have any advice regarding my idea of driving the driver ICs directly from a Feedback CT?

Oh, and I know you said you don't know much about IGBTs, but based on your experience with fets, is a 650v rating enough of a safety margin for 590v rails? I'm guessing not?

Appreciate your advice!

previous 1 2 3 next

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