Help understanding and deciding on DRSSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Help understanding and deciding on DRSSTC

previous 1 2 3 4

Move Thread

LAN_403

HV Enthusiast

Sun Apr 20 2008, 05:16PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Josh Johnson wrote ...

Ken, Tonic,

Generally speaking MOSFETs turn on about as fast as they turn off and the gate resistor is there to limit large transient spikes. IGBTs though tend to have have a much longer turn off already, so a gate resistor is not needed to limit spikes and the resistor would just make the turn off even longer; so it's bypassed with a diode.
Josh

Actually, the resistor has little to do with transient spikes. The gate resistor is used to critically damp the gate so that it does not oscillate. The parastic L (usually from transformer) and C of gate form an LC circuit which can oscillate. The gate R is there to damp any oscillations out. It can also be used to control charge rate of the gate, but the hard limit on how fast the gate is charged is not due to the size of the resistor, but rather characteristics of the parallel diode.

Dan

Tonic

Sun Apr 20 2008, 08:14PM

Registered Member #528 Joined: Fri Feb 16 2007, 10:32PM
Location: Warsaw, Poland
Posts: 166

I see. But wouldn't lowering IGBT's turn-off time with diode cause bigger transient spikes? Or is it matter of compromise?

And, another question has came to my mind - you're saying that because MOSFETs has fast turn-off times, they need gate resistor without diode in order to get rid of transients. Okay, I understand it, but is it really important to get of transients of current that flows out from gate? Aren't those transients "going" to negative side of bridge? From what I see on H-bridge scheme, I can see that these transients coming from two upper transistors's gates would have to pass through lower transistor's drain-source, then to negative. Since those transients are about few voltage, they shouldn't hurt lower transistors. Summarizing this, I would say that adding those diodes to MOSFETs would give bigger benefit, even if there are transients. If I'm wrong, please to correct me.

Just to make sure - with turn-off time you mean turn-off delay time plus fall time?

Next question, regarding bridge's decoupling capacitors. What would happen if their capacitance were oversized (let's say, 15-30uF)? I know that the more power bridge pushes, the bigger capacitance.. but I still don't know what are limits. The one of reason I'm asking is that I'm trying to find any decoupling capacitors and I've found metalized polypropylene 10uF 400VDC and I'm not sure if they would hurt half-bridge consisting two 50A RMS IGBTs.

Josh Johnson

Mon Apr 21 2008, 03:26AM

Registered Member #793 Joined: Sun May 20 2007, 06:50PM
Location: Minneapolis, MN
Posts: 35

Hi Dan,

When I said 'transient spikes' I meant:

That is 'transient spikes' caused by the parasitic L and C from the gate. Maybe you thought I meant spikes from noise or fast D-S dv/dt. Anyway just symantics, your description gets the point across better. Thanks. I didn't understand what you meant by:

...but the hard limit on how fast the gate is charged is not due to the size of the resistor, but rather characteristics of the parallel diode

As to Tonic's question, maybe this is a better explanation.
The diode effectively bypasses the resistor. As current is pulled out of the gate it flows through the diode instead of the resistor. It's like you put in a resistor for the turn-on time and replaced it with a short during turn off.

Just to make sure - with turn-off time you mean turn-off delay time plus fall time?

Yes, just in general how long it is taking to turn-off the FET or IGBT.

Summarizing this, I would say that adding those diodes to MOSFETs would give bigger benefit, even if there are transients.

The oscillations on the gate hurt the FET/IGBT by either exceeding the voltage the gate is capable of, maybe -20 to +20 volts or by turning the device on when it shouldn't be on and causing shoot-through or some other horrible thing to happen.

I don't have much experience with sizing bridge decoupling caps. I think more would be better. Although you have to watch out because you could be causing a resonance issue between the decoupling caps and the parasitic L in your PCB or bus bar layout making oscillations on the DC bus even worse.

Ken M.

Thu Jun 05 2008, 09:49PM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

Its taken awhile but I finaly got my CDE Caps, however fit appears they are smaller then the ones most coilers use for DRSSTCS, Heres a pic with the test data under it and a AA battery for comparison. Just need to figure out how to do the test wiring for the H-bridge the conventional Gator clamps probbly wont work here

Straight uploading of photoes didn't work so I just linked from the bucket.

Tom540

Mon Jun 09 2008, 03:52AM

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

Ken M. wrote ...

Its taken awhile but I finaly got my CDE Caps, however fit appears they are smaller then the ones most coilers use for DRSSTCS, Heres a pic with the test data under it and a AA battery for comparison. Just need to figure out how to do the test wiring for the H-bridge the conventional Gator clamps probbly wont work here

Straight uploading of photoes didn't work so I just linked from the bucket.

They're smaller because the capacitance value is smaller. Most of us use the the .1uF caps. Ive used the .047uF (which are the same size as the ones you have) and .1uF caps myself. They all work the same it's just a matter of combing them in a way you get the voltage and value you need.

previous 1 2 3 4

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