20KW DRSSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

20KW DRSSTC

first previous 3 4 5 6 next last

Move Thread

LAN_403

Mike

Thu Mar 29 2007, 11:13PM

Registered Member #58 Joined: Thu Feb 09 2006, 05:40AM
Location: Tri-Cities, Washington, US
Posts: 317

Wow , very cool coil.

Eat your heart out Steve Ward!!! :P
Just kidding, but I think it is time for a tesla arc war!

Finn Hammer

Fri Mar 30 2007, 06:12AM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

hvguy wrote ...

I am currently redesigning the full bridge mechanical layout. I think the design I have come up with is about as low inductance as is humanly, or mechanically, possible. I will post an update with pictures and test results later today.

And from an angle, where we can see that novel gate transformer?

Cheers, Finn Hammer

hvguy

Fri Mar 30 2007, 07:47AM

Registered Member #289 Joined: Mon Mar 06 2006, 10:45AM
Location: Conroe, TX
Posts: 154

I have spent most of the day experimenting with various snubbers trying to find the best solution for the new bridge assembly. Hereâ€™s what I came up with:

Not shown in this schematic are four parallel 20nF 1KV ceramic disc caps placed rail to rail on each half bridge section. All my testing was done with a buss voltage of 170V at about 2.3KW. With this snubber design I was able to reduce my transients to about 1.5X the rail. This is just fine by me as the most input voltage I had planned to run was 800V and these are 1.7KV parts; thatâ€™s 500V head room. The TVS strings are designed to begin conducting at ~1250V and clamp at ~1550V, just in case a spike does make in though the snubber. Iâ€™d still rather nuke a string of TVSs than an IGBT. Here are a few scope shots:

No snubber:

Snubber:

And, Finn, here is the GDT. The core has a 1.5â€ ID, primary is .125â€ flat Teflon wire secondary is 600V 14ga PVC. Nothing special really. Although I would HIGHY recommend anyone duplicating this shield the primary from the secondary, it will eliminate noise feedback though all the various capacitances in the gate driver.

Just check out this folder for all the new bridge pics:

I will set up for some high power testing either tomorrow or next week. Thereâ€™s a hamfest in Raleigh this weekend...

Steve Conner

Fri Mar 30 2007, 10:40AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Yay, that new bridge looks great! BTW, about shielding in gate drive transformers: on my last coil I made the GDT primaries from 1/8" miniature coax (RG-174 or whatever it's called) and grounded the braid at one end to act as a Faraday shield. The secondaries were just ordinary PVC insulated wire: I taped two wires together with the coax to make a bundle.

Marko

Fri Mar 30 2007, 11:59AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

The bridge looks just.. awesome

Beautiful low inductance design.

And those bricks can actually be unscrewed and opened just so easily.. they aren't actually hermetically sealed?

Steve Ward

Fri Mar 30 2007, 03:37PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

And those bricks can actually be unscrewed and opened just so easily.. they aren't actually hermetically sealed?

Yep. I have some CM300s like that, where i opened them to disable the "real time current limiter" for tesla use

. Just have to be *damn* sure you snip the right bonding wire!

Aron, your snubber is now a capacitor on the output of the bridge? Interesting... i tried that too (small C of course), and it seemed to make things worse, as its in effect increasing the parasitic C of the CE junction. My thought at the time was that it would slow down the ringing, allowing the main snubber to be more effective (didnt quite work out). I havent run the calculations, but be sure not to exceed the current rating of the little cap (it will surely see some very high peak currents). But, your waveforms certainly do look better. I also like the bridge layout

.

Something crossed my mind... what is the current rating on those electrolytics? Seems like an awful lot of power for those things!

hvguy

Fri Mar 30 2007, 05:55PM

Registered Member #289 Joined: Mon Mar 06 2006, 10:45AM
Location: Conroe, TX
Posts: 154

Ya, Iâ€™m pretty happy with this design. As low L as possible and a little more space in my enclosure, plus I had all the materials on hand so it went together quickly.

Steve C. itâ€™s funny you mention that. When I first realized a shielded primary was necessary I tried RG-174 in that exact configuration. The results were excellent but I needed bigger coax to accommodate the ~35A pk primary current. RG-58 worked wellâ€¦

Thatâ€™s a good question Steve. I have not been able to find data on these caps so I don't know the current rating. Chances are I would be exceeding it anyway. The good news is I pulled them from a 50KVA VFD. I would assume they can handle this type of operation having been installed a drive that large. Also, after running for several hours at ~2.5KVA (1.2KA pk on the pri) they were only slightly warm.

Yep, my snubber is nothing more than a cap on the low side IGBT. At first I assumed this would make things worse, and in fact, it did. Then I rebuilt the bridge and all the noise except the ~1.5MHz spike at turn off was gone. I though Iâ€™d give the CE cap another try since there was almost no L for it to ring with the new design. There is some ringing cause by the cap, but the amplitude is only 10% higher than the rail, much better than the 2.3X with out the cap.

The noise currents though the cap are high. With the tank at 1.2KA and the rail at 170V the peak current though that cap was 200A. This is definitely a lot of current, but the noise is so fast that the pulse falls within the datasheet spec for pulse current and the 1.6KV .22uF CDE film caps showed know heating after several hours of run time.

Tom540

Fri Mar 30 2007, 08:53PM

Banned on 3/17/2009.
Registered Member #487 Joined: Sun Jul 09 2006, 01:22AM
Location:
Posts: 617

Well Elgar wasnt much help but I did find something interesting on an old schematic for a 10KW power supply. Don't know if any HVers have tried this one.

Since I made today my last day since I was only a temp and wanted a week off in between job. I didn't get the pdf I made of the schematic but I did draw this up of the snubber that was used across the igbt bridge. The cap was the only value I remember. I'm sure values differ depending on bridge voltage and frequency.

There was also another snubber with a higher parts count I found but it was an old power ten product and there was no schematic availibe. Problem with Elgar is theyve had so many diffrent people in and out doing layouts plus new managment about a year ago that all thier boards and schematics are half in Pcad and Half in PADS. Anyway enough rambling. Here is the drawing. You'll probably have to do a search on how to calculate it.

ragnar

Fri Mar 30 2007, 09:38PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Whoa, being able to see their guts must be really handy if (IF, not when ;) it ever fails, for some diagnosis.

hvguy

Sat Mar 31 2007, 12:03AM

Registered Member #289 Joined: Mon Mar 06 2006, 10:45AM
Location: Conroe, TX
Posts: 154

Thanks for checking. That circuit is whatâ€™s known as an RCD snubber and is fairly common in full bridge inverters. I did try that arrangement but found just the cap to be more affective.

first previous 3 4 5 6 next last

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