GDT driver capacitor... / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

GDT driver capacitor...

Move Thread

LAN_403

Nicko

Thu Nov 13 2008, 07:38AM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

I'm building my second SSTC coil and was wondering about how to size the primary capacitor for the GDT. My first SSTC ran rather high at about 475kHz, but this one I want to move back down to 250kHz.

I'm using the UCC3732x driver pair, and was going to start with a 100n ceramic, but I've seen comments that "bigger is better". Is there any ways to accurately spec this cap, and are any particular types "better" than others...

Many thanks

Nick

101111

Thu Nov 13 2008, 08:06AM

Registered Member #575 Joined: Sun Mar 11 2007, 04:00AM
Location: Norway
Posts: 263

It should be about 10-20 times bigger then the gate capacitance of your transistors. Something between 470nF to 1ÂµF should do. I don't think you should use ceramic caps but polyester or polypropylene instead.

Anyone correct me if I am wrong

MRacerxdl

Thu Nov 13 2008, 08:39AM

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

Ceramic works for GDT, but its not the best, it have a little problems with high frequency. The poliester and polipropylene are best.
I use 300nF (Three 100nF caps in parallel) for the GDT, it works very fine, I used 100nF before.
But Remind that cap needs to be higher capacitance than the all mosfet gates capacitances.

Steve Conner

Thu Nov 13 2008, 11:39AM

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

That's 10-20 times bigger than the gate capacitances of all your transistors added together. If you use a 1:2 GDT, then multiply by a factor of 4 again.

I used 4uF in my last few coils.

Dieu

Sun Nov 16 2008, 06:53PM

Registered Member #1730 Joined: Thu Oct 02 2008, 08:12AM
Location: Poland
Posts: 35

I used polipropyl. cap of 4.7uF, is it too big? Will it be good for DRSSTC? I used TC4421 and tc4422.

Arcstarter

Sun Nov 16 2008, 07:58PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Dieu wrote ...

I used polipropyl. cap of 4.7uF, is it too big? Will it be good for DRSSTC? I used TC4421 and tc4422.

Well, the bigger the capacitance, the more current the ucc pair has to put out. With just .39uf my ucc pair got rather warm, after a minute or two. It was uncomfortably warm, as in i was uncomfortable with how warm it was, and you could not keep a finger on it for too long. Not hot enough to burn you, but i would not make it any higher. I was using a single output gdt to drive on mosfet, which was an irfp460, and the gdt was anywhere from 10 to 15 turns, and running at some 300khz.

Steve Ward

Mon Nov 17 2008, 01:25AM

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

Well, the bigger the capacitance, the more current the ucc pair has to put out. With just .39uf my ucc pair got rather warm, after a minute or two. It was uncomfortably warm, as in i was uncomfortable with how warm it was, and you could not keep a finger on it for too long. Not hot enough to burn you, but i would not make it any higher. I was using a single output gdt to drive on mosfet, which was an irfp460, and the gdt was anywhere from 10 to 15 turns, and running at some 300khz.

We need to clear this up before it spreads any further.

The capacitor needs to be large such that it presents minimal impedance between gate driver and the gates. Using a bigger cap is not a negative thing in most cases, because when operating properly, the capacitor has basically 0VAC across it.

In the case above, the capacitor was not the cause of gate driver heating, but rather charging the mosfet gate charge at a high frequency was. Using a smaller coupling capacitor limits the gate drive charging current so as to limit power dissipation in the driver. A better approach to this is to use discrete resistors in the gate drive path. Using a capacitor to lower the current will lower the gate voltage, which can be dangerous for the MOSFET/IGBT, while using a resistor simply slows down the gate drive and pushes the losses into the resistor (and not into the driver). It takes the same amount of power either way (with and without external resistance) so you can reason that the external resistor is reducing the power dissipation of the driver, at the cost of slower switching transitions.

Most recently ive used parallel groups of SMD type ceramic chip capacitors (MLCCs) at about 20uF total capacitance. I believe the capacitor should not add any impedance if possible. The only drawback to the larger cap is that if you pulse the driver at "DC" the current will surge for longer to charge up the capacitor, possibly increasing mosfet driver dissipation, but this is NOT the case when the drivers are switching at their intended high frequency.

Steve Conner

Mon Nov 17 2008, 10:14AM

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

+1 Steve

Dr. Dark Current

Mon Nov 17 2008, 02:22PM

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

I once tried 100nF but I had problems. Remember that the GDT draws a considerable amount of magnetizing current.

P.S. Steve I love your sig

I haven't laughed like this for a while...

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