First real DRSSTC / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

First real DRSSTC

1 2 next

Move Thread

LAN_403

Dalus

Mon Oct 20 2008, 07:31PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

After building my coil for what seems like forever I'd like to show you the almost finished coil.

*The secondary has an 95cm winding length and is 20 cm in diameter. (low Fres easy to drive

)
*The primary which now is being extended with another 5 turns is made out of 12mm copper pipe,
*The MMC consists out of 30 cde 0.15 qF caps that can be arranged in any configuration from max 5s 30p
*The whole is powered by an h-bridge with IXGR40N60B2D1 igbt's and is protected by 440V transient voltage suppressors.
*The drive schematic is basically Steve Wards DRSSTC 2 schematic

Here are some coil winding and some pics in which the coil has just been varnished.

The whole on the base with the MMC in there.

And last but not least the electronics board.

After some cross conduction problems that where hard to discover (bridge failed after running 15 min @ max Vbus(vid with the crossconducting eating all the power

) we finally got it to work with a new gdt that could only do +8V / -8V gate we where able to test the coil @ 50A Ipri max with these results.

After i added a few more dc decoupling caps to the gdt drive circuit the voltage on the gates increased to +28V / -28V so we can use the full rated 200A of the igbt's in the next run. Hopefully we'll also be able to tune the coil better with the added turns on the primary. With a 5T primary we weren't able to tune the coil in the first run, we had to take the toroid of to get some output out of the coil.

I hope to upgrade the H-bridge which is obviously way too small for this coil with something befier, atleast after i get it to work propperly.

Dalus

Thu Oct 23 2008, 12:13PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

I tested the coil yesterday and got 60cm groundstrikes at 110Vbus. During these tests the coil eventually stopped. After I started measuring i noticed that the negative side of the signal from the gate driver is missing. So I replaced all the UCC's and it still doesn't produce the negative side. Any ideas why?

(edit) It is something in the bridge or gdt, I replaced the gdt with a dummy load and I did get the negative drive signal

The bridge isn't shorted either

I do have the feeling that it's going to take a long time to find the cause :(

One other thing i noticed was that the coil produced four or so normal discharges and every fifth one was very intense, is this normal? If not what can be causing this?

Linas

Sat Oct 25 2008, 07:26AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

i advise you don't use just UCC in GDT driver, I never killed UCC, (even in short_circuit) because i have mosfet buffer, it give very high speed, and security of driving semiconductors
Use UCC3732X>>IRFZ24N and IR9Z34N
I can drive GDT even with the one winding

Use two to drive GDT

Dalus

Sat Oct 25 2008, 10:52AM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

The four ucc's suffice, your gate driver is more usable for large igbt's like the cm300 I traced the problem to one igbt that has all the pins shorted together causing the negative gate drive to dump all the energy via a diode in this short.

Now, how did it blow up? The igbt's can handle 200A peak, the ocp was set to 100A this rules overcurrent out. The igbt's where protected by bipolar transient voltage suppressors, so overvoltage wasn't the problem either. Most likely it was interference causing the igbt's to switch of resonance and thus blowing them up.

I'm going to build a Faraday cage around the electronics to try solve this issue, hope it works. I only have 12 Igbt's left

MRacerxdl

Sat Oct 25 2008, 01:20PM

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

My IGBT's blown every time that occurs ground strikes after a few sec. Some say that is ground fault, but I dont know what really is, but I'mt using Current Control, for only sparks to the air, I have no problems.

Dalus

Sat Oct 25 2008, 05:58PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

It can handle the groundstrikes :) It failed during some large arcs into the air. So it's most likely caused by interference.

Here's the new H-bridge under construction. A class mate accidentally tripped over the cable and destroyed the old one

But this one has a way better layout

(edit) I really need to think when i build something

I connected the two wrong igbt's. . .

fixed

Dalus

Sun Nov 02 2008, 08:15PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Today I tested the new H-bridge and I can say it works

It's working better than the old one that made the bus voltage go up by it's large conductivity.

Here's the youtube link. It's running at 140Vbus and 100A Ipri max

High power testing will be done in the next weekend

MRacerxdl

Sun Nov 02 2008, 11:08PM

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

Wow, I cant get white sparks as so, I think it because I dont have OCD, and I cant get the duty cycle on the interrupter more than 14% without blowing all things, with 10% I get orange sparks.

Dalus

Mon Nov 03 2008, 09:07AM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Youtube changes the collour quite a bit, irl it looks more like this

Hope to test the coil at full power soon

I'm also designing a new H-bridge for it with CM300 igbt's But i still need to find a good gate driver for them.

Wolfram

Mon Nov 03 2008, 11:25AM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

This

seems like the simplest way to gate-drive CM300s.

Anders M.

1 2 next

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