Fixing that damn ZVS / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Fixing that damn ZVS

Move Thread

LAN_403

Adam Munich

Wed Jan 18 2012, 03:01AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Falstad seems to have fixed his sim, and the ZVS renders correctly now.

Changing the +36V to something higher like +170V breaks the circuit. Just like in real life. My reasoning for this is the gate-pulling-down diodes. As you can see in the sim, they conduct in very short bursts, probably around 30 or more amps.

As you turn up the power, these bursts get bigger and bigger, and at a certain point you have something like 200A flowing through the diode! Because of this, a voltage drop appears across the junction and the gate can't get pulled below its threshold voltage.

Thus, circuit go boom.

-Edit-

Forgot the dang filtering inductor. Derp.
But my point still stands; we need a better feedback method to make this circuit not explode.

So it looks like the problem with the ZVS is the direct diode feedback. It sucks nips, to be frank about it. I'm going to see if I can't figure out a new feedback method to fix this issue...

New falstad link.

You might have to disconnect a mosfet for a second to get an oscillation going, since the sim assumes both components are identical.

m4ge123

Wed Jan 18 2012, 03:15AM

Registered Member #4118 Joined: Mon Oct 03 2011, 04:50PM
Location: MD
Posts: 140

for LTSpice
If you insist on using falstad, at least go to options->other options and turn the timestep down to 100n or less so you can use high freqencies.

Edit: for that file I posted, you'll want to turn down R1 to about 1k... I'm still working on this circuit and testing stuff with it.

Patrick

Wed Jan 18 2012, 03:18AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Grenadier wrote ...

So it looks like the problem with the ZVS is the direct diode feedback. It sucks nips, to be frank about it. I'm going to see if I can't figure out a new feedback method to fix this issue...

do you mean better diodes, or a better circuit?

m4ge123

Wed Jan 18 2012, 03:21AM

Registered Member #4118 Joined: Mon Oct 03 2011, 04:50PM
Location: MD
Posts: 140

A better circuit. The issue with the diodes and feedback has to do with the mosfet's internal capacitances, specifically Cds and Cdg.

Patrick

Wed Jan 18 2012, 04:35AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Im looking through this PDF at the moment. ->
]mosfet_switching.pdf[/file]

im trying to figure out how MULTIsim10 works with the mosfet spice models... i have no idea how to do so.

like this... I dont know the standards of spice 3 though.
AAV
]2n7002lt1.txt[/file]

will you guys get up to date libraries, for your simulators? because ive always found new better mosfets before the spice models are released for update... in a "better" circuit where the stray capacitences matter, you and i may need up to date simulation of known mosfets.

I may have found a spice model of the STW13NK100Z ,
]stw13nk100z.txt[/file]

If you guys could figure out a better circuit for flybacks, and it output higher V. it would be a great gain.

Adam Munich

Wed Jan 18 2012, 06:14AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

I had a flaw with my model. The new one works just fine, but the high current diode issue still stands.

Falstad did one hell of a job with this thing

Problem though, it assumes both mosfets are identical, so you need to disconnect one for a second or so in the beginning to get that 'lopsidedness' to start the oscillation.

Antonio

Fri Jan 20 2012, 03:40PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

The problem with the "ZVS" oscillator is that if it ceases to oscillate it turns both transistors on, and if the power supply allows, destroys them. Oscillation stops if the series inductor is too small, if the load is too small, and if the series inductor saturates. The addition of something to impede excessive current drain from the power supply would turn the driver safer.

Adam Munich

Fri Jan 20 2012, 03:54PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

I've noticed that adding a 0.5 ohm resistor greatly increases reliability, but if you are pushing 20A then it'd dissipate 200W!

Maybe an 'active resistor' could help. After a certain current the resistor would be shorted to reduce power waste. Thermistors are no good, since once they are hot you have to wait for them to cool down.

But this only bandages the problem, doesn't fix it :-/

Dr. ISOTOP

Fri Jan 20 2012, 04:05PM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

Grenadier wrote ...

I've noticed that adding a 0.5 ohm resistor greatly increases reliability, but if you are pushing 20A then it'd dissipate 200W!

Maybe an 'active resistor' could help. After a certain current the resistor would be shorted to reduce power waste. Thermistors are no good, since once they are hot you have to wait for them to cool down.

But this only bandages the problem, doesn't fix it :-/

That doesn't work...once you hit an overcurrent condition the resistor is bypassed and the overcurrent fault continues to destroy transistors.

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