Author Topic: CW Multiplier Killing ZVS Driver  (Read 398 times)

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
CW Multiplier Killing ZVS Driver
« on: October 09, 2021, 04:51:17 AM »
I've been working on a HV ZVS driver and CW Multiplier for a long time now and I'm always running into snags it seems. My ZVS driver runs off rectified mains. It has a low voltage start up and then a push button for full power. It works fine while doing Jacobs ladders and drawing arcs. I estimate the open circuit voltage to be about 75-100kv. I connect it to my new multiplier and turn on low voltage. I get maybe 2 or 3 inch sparks. I go to hit the high voltage button and one spark shoots about 6 or 7 inches then the driver dies and both mosfets seem to be shorted. I connect a neon sign transformer to the multiplier to see if its been damages, but it works. I have an output resistor string in wax in the tube on top of the multiplier and it should limit the current to about 2mA. Diodes are rated to 5mA so I wanted to be extra safe because its all encased in epoxy. Does any one know why this is happening? My best guess is some high voltage spike going back into the driver some how, similar to what can happen to neon sig transformers in a tesla coil. Should I put a spark gap in parallel with the ZVS outputs? Should i be grounding one side of the multiplier?

In previous versions of this (30-40kv output on the driver) I was able to draw 14" arcs continuously so I'm not sure why this is happening.

Any help is appreciated. Thanks!



Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1325
  • Karma: +63/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #1 on: October 09, 2021, 05:41:50 AM »
Many possible issues.  Hard to guess which without circuit diagram and scope captures.  Here's at least a partial list:

HV transformers are dual-resonant due to winding capacitance.  ZVS drivers often lock to the upper pole.  Usually results in low output rather than destruction.  If it is switching modes back and forth occasionally, that can make excess voltage spikes as load current drops through ZVS input inductor.  I used several TVS diodes to clamp such spikes.  See this thread:
https://highvoltageforum.net/index.php?topic=831.msg5491#msg5491
Sudden load drops such as an arc striking or breaking cause similar transients.

You might have enough lead length from ZVS to resonant caps to get very-high-frequency oscillation based on FET capacitance and wire inductance.  I've had that happen.  HF is inefficient, so hot FETs.  ZVS works best with minimum stray inductance from FETs to caps.

If Q drops too low with arc loading, oscillation drops out and current ramps up to destruction.  If transformer coupling factor is below 0.85, this shouldn't be an issue.  Just hops to upper pole at high load.

Higher-voltage FETs may have too-high forward drop (high on-resistance), so not keep the opposite FET off.  I used IGBTs for lower forward drop.  I have another proposed solution, simulated but not built:
https://highvoltageforum.net/index.php?topic=1658.msg12681#msg12681

Good choice to have low-power start-up.  That solves one basic issue of most ZVS circuits.

Good luck with finding the real issue for your build!  Keep us posted.
« Last Edit: October 09, 2021, 05:44:58 AM by davekni »
David Knierim

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #2 on: October 11, 2021, 08:35:25 PM »
So I'm on my last set of IGBT's (my mistake saying mosfet before) I am running with TVS diodes to clamp and I put it all the in schematic. Only correction is the zener diode for the gate, its a 15/16 volt, I cant remember. And some of the capacitances may be off by a bit, but you'll get the idea. 

I'm still new to using my oscilloscope, but I think I might be having a problem with gate ringing? I took your advice about shortening the path between IGBT's and the tank caps, it did help a bit.

This picture is of the gate when the low voltage is turned on.



This next one is the gate while drawing an arc before shortening the leads from drain to caps.



This is the gate after shortening the leads from drain to caps.



This last one is of the drain, which was the same in the before and after.




I noticed some of those spikes in the gate went down a bit after shortening the leads, but the drain didn't seem to change much. Maybe a slight reduction, but nothing to write home about.

Here is the schematic.



Ill be doing more tests. My low voltage turn on is about 17.5 volts and I have gone above that to 35volts. No ill effects while drawing arcs from the multiplier at 35volts.


I did notice when drawing arcs the frequency dose change by a few kHz. When things get mess it might say its at 2 or 3 times the normal frequency. I figure this is a problem of the scope, not so much an actual change in frequency. I think this because the pulse width in the pictures doesn't actually change, just the reading does.

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #3 on: October 11, 2021, 11:36:09 PM »
I think I found one of the culprits. I dialed up the voltage to 50VDC and noticed a flash in the transformer container. Turns out as I crank up the voltage the HV secondary wire is drawn to the primary and gets close enough at 50VDC that it makes contact and arcs over. I took a video of it twitching. Ill need to put a ridged plastic tube around it to prevent this. I also need to see what effect grounding one side of the multiplier will have.


Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1325
  • Karma: +63/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #4 on: October 12, 2021, 06:12:08 AM »
Quote
Turns out as I crank up the voltage the HV secondary wire is drawn to the primary and gets close enough at 50VDC that it makes contact and arcs over.
Yes, that static-electric induced wire motion looks like a problem.  Good catch.

Quote
I'm still new to using my oscilloscope, but I think I might be having a problem with gate ringing?
Your scope traces are quite helpful.  Couple suggestions still.  First, change the probe (vertical) input coupling mode from AC to DC.  (Trigger is set to DC, but that is separate.)  Second, for analyzing glitches, add another image or two expanded horizontally (slower time per division) zoomed into a spike, one for rising edge spike and another for falling edge.

I believe the spikes are due to IGBT switching delay and IGBT internal diode reverse recovery time.  Those are the down-sides of using IGBTs for ZVS.  During turnoff time, one IGBT and the opposite diode conduct the full resonant current, not just the supply current.  Fast IGBTs with fast internal diodes help.  So will a low-inductance "ground" plane between the two IGBT emitters.  Spikes  may be OK as-is.  Not sure if they are a source of fried IGBTs or not.  If you end up needing any more IGBTs, I'd buy faster ones.

Do you have multiple 1.5KE300A TVS diodes in series?  As you see on the scope, Vce reaches 420V, and will go higher with 120VAC input.  Peak of 120Vac is 170V, making Vce peak of PI*170=534V.
David Knierim

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #5 on: October 18, 2021, 06:06:08 AM »
Thanks! I think it was for sure one of the problems. The problem was solved by grounding the lower end of the discharge column on the multiplier. Once I did that I noticed it worked better and the wire quit moving! I then took the input voltage to 50 volts and noticed that my reasoning behind the build was faulty. I thought because the capacitors share the voltage that didn't mean the top half of them was required to be covered. To a point I was right. and that point was 40VDC input. After that some hole poked through the capacitor insulation and I could feel the ion wind coming off of it. It was kind of fun because i could get charge up from it and sap people near by. The fix was hot glue. Lots of hot glue. I smothered the capacitors and it worked much better! no more leaks. However a new problem. I put the resistors in an external tube filled with wax. the goal is to limit the current to under 5mA. When I crank of the voltage to 60VDC the the arcs jump out my insulating wire and trace up the tube to the ball. Guess its easier to do that then go through the resistors! Its really cool and very intimidating. Ill take a video and share it later.

Do have a question about the output resistors. Do I actually need them? I'm looking around and see that power in has to  equal power out. So if I measure my input power at a max of 1500W on the high side and the output voltage is, on the low side, 75kv then the current should be about 20ma. then the multiplier is 6 times so if the voltage is multiplied 6 times the current is divided 6 times, meaning an output of about 3.3mA at most.  So I shouldn't really need output resistors. Is this making sense or am I totally out to lunch here?

I will for sure make the adjustments to the scope and see how it fairs. Im still learning to use the thing. thanks for the advice there!

I assume when you mean faster IGBT's you mean lower off and fall times? I notice alot of specs don't even mention on or rise time.

I have to of the 1.5KE300A TVS diodes in series.

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1325
  • Karma: +63/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #6 on: October 18, 2021, 06:49:11 AM »
Quote
Do have a question about the output resistors. Do I actually need them?
Yes.  Average current is fine as you have calculated.  Peak current as capacitors discharge into a spark is orders-of-magnitude higher.  Even if your caps can handle the peak current, diodes conduct half of the peak discharge current (one capacitor stack's worth).  Diodes won't survive.

Quote
I assume when you mean faster IGBT's you mean lower off and fall times?
Yes, and fast free-wheeling diode recovery time too (perhaps most important).
David Knierim

Offline MRMILSTAR

  • High Voltage Engineer
  • ****
  • Posts: 350
  • Karma: +11/-0
  • Cedar Rapids, Iowa
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #7 on: October 18, 2021, 03:14:21 PM »
It can be discouraging having built a nice multiplier and then have to spend almost as much to make an adequate load resistance. However if you want your diodes to survive that's what you need to do. As an example, have a look at the load resistance for my CW multiplier. I spent almost as much building this as the multiplier itself!  :)
Steve White
Cedar Rapids, Iowa
Retired electrical engineer

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #8 on: October 19, 2021, 04:51:16 AM »
It defiantly can be discouraging, but I honestly just love the process. And of course the massive sparks :D Ill post a video tomorrow and share my progress. But I was wondering, if we are trying to limit peak current form the discharging capacitors in the multiplier, could we use an inductor instead of a resistor? I feel like that should work, but something in me is thinking I might get some unwanted voltage spikes or wire resonant issues.

And as far as the free wheeling diode, could I get a fast external one instead of worrying about the integrated one?

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1325
  • Karma: +63/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #9 on: October 19, 2021, 06:37:32 AM »
Quote
But I was wondering, if we are trying to limit peak current form the discharging capacitors in the multiplier, could we use an inductor instead of a resistor?
I've thought about the same thing, as I hate to waste so much power in resistors.  Try calculating the required inductance value :)

Quote
And as far as the free wheeling diode, could I get a fast external one instead of worrying about the integrated one?
Yes, but:  The IGBT part needs to have no internal diode (or have an internal diode with higher forward voltage drop), and the external diode needs low-inductance connection to IGBT.
David Knierim

Offline Andrew321

  • High Voltage Enthusiast
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CW Multiplier Killing ZVS Driver
« Reply #10 on: October 20, 2021, 02:40:34 AM »
The audio isn't so good, I need to talk louder. But I start at 17VDC and end at 62VDC. You can hear the corona. Ill need a bigger resistor chain and a larger top load.

/>
Good to know about the IGBT's. I'll keep an eye open because I will no doubt need to get more. My testing tends to be kind of destructive :D

I ran some numbers on the inductor and this is what I got so far just to get me in the ball park. I figure if the pulse is 1us (just a guess), that corresponds to 1 MHz. Given that frequency and an inductor that has a radius of 12cm, a length of 30cm and 1200 coils (about 30 AWG) I end up with a reactance of 1.7Mohms. If I estimate my input voltage at 75KV (will confirm this over the next day or so) and I have a 6x multiplier the output is about 450KV. Divide that by 0.005A (5mA) and I need an impedance of 90Mohms.

Not sure an inductor will do it for me. That thing would be massive lol. This is of course calculated as an air core coil. Maybe a large ferrite rod would work if well insulated?

High Voltage Forum

Re: CW Multiplier Killing ZVS Driver
« Reply #10 on: October 20, 2021, 02:40:34 AM »

 


* Recent Topics and Posts

post Re: Gap distance between main bus bar
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
October 26, 2021, 11:08:03 PM
post Re: Syntherrupter - A Feature-rich, Polyphonic Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
SimonNwardUK
October 26, 2021, 10:37:39 PM
post Re: Ultra High Voltage Lab (UHVLab) | website | youtube
[Hobby Sites]
AstRii
October 26, 2021, 09:33:18 PM
post Re: Gap distance between main bus bar
[Dual Resonant Solid State Tesla coils (DRSSTC)]
SAMGO
October 26, 2021, 08:29:08 PM
post Re: My Switches Don’t Switch
[Voltage Multipliers]
abstruse1
October 26, 2021, 07:51:08 PM
post Re: Gap distance between main bus bar
[Dual Resonant Solid State Tesla coils (DRSSTC)]
TMaxElectronics
October 26, 2021, 06:38:24 PM
post Gap distance between main bus bar
[Dual Resonant Solid State Tesla coils (DRSSTC)]
SAMGO
October 26, 2021, 01:13:09 PM
post Re: Ultra High Voltage Lab (UHVLab) | website | youtube
[Hobby Sites]
Mads Barnkob
October 26, 2021, 11:20:58 AM
post Re: Problems with Franzoli electronics' easy flyback driver
[Electronic Circuits]
futurist
October 25, 2021, 07:14:25 PM
post Problems with Franzoli electronics' easy flyback driver
[Electronic Circuits]
Anon_v3
October 25, 2021, 07:01:28 PM
post Re: My Switches Don’t Switch
[Voltage Multipliers]
Vtroxi
October 25, 2021, 04:38:37 PM
post Re: Mains Staccato / T-200 VTTC
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 25, 2021, 01:58:47 AM
post Re: Mains Staccato / SGTC
[Spark Gap Tesla Coils (SGTC)]
304er
October 24, 2021, 10:42:48 PM
post Re: My Bestest HV ZVS Driver and Jacobs Ladder
[Transformer (Ferrite Core)]
Duane B
October 24, 2021, 08:21:45 PM
post Re: Mains Staccato / T-200 VTTC
[Vacuum Tube Tesla Coils (VTTC)]
Duane B
October 24, 2021, 08:04:02 PM
post Re: Mains Staccato / SGTC
[Spark Gap Tesla Coils (SGTC)]
Duane B
October 24, 2021, 07:56:31 PM
post My Switches Don’t Switch
[Voltage Multipliers]
abstruse1
October 24, 2021, 06:50:05 PM
post Mains Staccato / T-200 VTTC
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 24, 2021, 11:18:22 AM
post Mains Staccato / SGTC
[Spark Gap Tesla Coils (SGTC)]
304er
October 24, 2021, 09:44:10 AM
post Some more VTTC controllers and SSR's being applied
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 24, 2021, 06:21:17 AM
post Re: My Bestest HV ZVS Driver and Jacobs Ladder
[Transformer (Ferrite Core)]
davekni
October 24, 2021, 05:55:33 AM
post VTTC Staccato/Interrupter Controller modified so SSR is useable for LShPS
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 24, 2021, 05:47:46 AM
post Re: My Bestest HV ZVS Driver and Jacobs Ladder
[Transformer (Ferrite Core)]
MRMILSTAR
October 24, 2021, 05:03:08 AM
post My Bestest HV ZVS Driver and Jacobs Ladder
[Transformer (Ferrite Core)]
Andrew321
October 24, 2021, 02:21:37 AM
post Re: Finally finalizing my 304er VTTC
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 23, 2021, 09:14:00 PM
post Re: Finally finalizing my 304er VTTC
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 23, 2021, 08:48:22 PM
post Re: Finally finalizing my 304er VTTC
[Vacuum Tube Tesla Coils (VTTC)]
Mads Barnkob
October 23, 2021, 08:33:22 PM
post Re: Finally finalizing my 304er VTTC
[Vacuum Tube Tesla Coils (VTTC)]
MRMILSTAR
October 23, 2021, 03:40:46 PM
post Re: Finally finalizing my 304er VTTC
[Vacuum Tube Tesla Coils (VTTC)]
304er
October 23, 2021, 10:29:37 AM
post Re: Battery powered DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
TMaxElectronics
October 23, 2021, 01:07:13 AM
post Re: DRSSTC tuning
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
October 23, 2021, 01:05:04 AM
post Re: Hi! 811 A tesa coil
[Vacuum Tube Tesla Coils (VTTC)]
Duane B
October 23, 2021, 12:48:40 AM
post Re: DRSSTC tuning
[Dual Resonant Solid State Tesla coils (DRSSTC)]
kubajed
October 23, 2021, 12:01:14 AM
post Re: Hi! 811 A tesa coil
[Vacuum Tube Tesla Coils (VTTC)]
jpvvv123
October 22, 2021, 11:52:23 PM
post UD3 CPL file for smt assembly by JLC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Franksmather
October 22, 2021, 08:56:59 PM
post Tesla Coil Show Controller - Analog Audio Interrupter and Test (Part 6 of 8)
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Mads Barnkob
October 22, 2021, 11:58:51 AM
post Re: Next Gen DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Netzpfuscher
October 22, 2021, 07:03:26 AM
post Re: Where to buy DIY or home made capacitor book(s)?
[Capacitor Banks]
Twospoons
October 22, 2021, 05:04:51 AM
post Re: Idea for increased electromagnetic coin shrinkage
[Capacitor Banks]
davekni
October 22, 2021, 04:43:23 AM
post Re: Where to buy DIY or home made capacitor book(s)?
[Capacitor Banks]
davekni
October 22, 2021, 04:27:13 AM
post Re: Where to buy DIY or home made capacitor book(s)?
[Capacitor Banks]
HVuser
October 21, 2021, 11:15:28 PM
post Re: Tesla Coil Music
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Intra
October 21, 2021, 09:45:55 PM
post Re: Coin shrinking video
[Capacitor Banks]
MRMILSTAR
October 21, 2021, 08:32:37 PM
post Re: Coin shrinking video
[Capacitor Banks]
klugesmith
October 21, 2021, 06:10:06 PM
post Re: 3kJ coin shrinking
[Capacitor Banks]
klugesmith
October 21, 2021, 06:09:19 PM
post Re: Idea for increased electromagnetic coin shrinkage
[Capacitor Banks]
klugesmith
October 21, 2021, 06:07:54 PM
post Re: Next Gen DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Intra
October 21, 2021, 01:27:29 PM
post Re: Interrupter frequency?
[Dual Resonant Solid State Tesla coils (DRSSTC)]
GrantV
October 21, 2021, 12:37:00 PM
post Re: QCW problem with ZVS and phase lead
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
October 21, 2021, 06:34:26 AM
post Re: Idea for increased electromagnetic coin shrinkage
[Capacitor Banks]
davekni
October 21, 2021, 06:06:50 AM

Sitemap 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 
SimplePortal 2.3.6 © 2008-2014, SimplePortal