Author Topic: Royer Refinements  (Read 154 times)

Offline flyrod

  • High Voltage Enthusiast
  • *
  • Posts: 3
    • View Profile
Royer Refinements
« on: November 28, 2017, 04:18:28 PM »
Hello all.  I found this forum after searching and reading about the royer induction circuits online.  I tried to register at 4hv.org but couldn't get in.  Anyway, I'm new to electronics but I've built several of these circuits and I was wondering if anyone with an oscilloscope has experimented with different components.  All I have is a volt meter so I can see changes in voltage/current draw but I don't know why I'm seeing the changes.  In my experiments I've noticed that there can be a wide range of outcomes using the same "recipe" with different components.

For example, I've noticed that increasing the tank capacitance increases the current draw.  My initial guess was that increasing capacitance would drop the frequency and also the current--but I observed the opposite effect.  Does increasing the tank capacitance increase the tank AC voltage?  i.e. the tank will swing further with a lager capacitance?

I've tried different toroid inductors and have not seen much difference.  Perhaps they are more critical at higher current draw?  I've wound the same cores with 30 turns and again with 40 turns and not seen a change.  Has anyone seen what happens if theses inductors are too big or too small?

I've seen some people put nuts or ferrite beads on the FET gate wires.  I tried some back-to-back tests with the only change being the beads, and I didn't see much difference.  How do you know what these parts do (without an oscilloscope)?

I've tried changing out tank capacitors with MMC or different types all with the same capacitance and seen wide ranges of current draw.  One circuit I have that's built using a heavy duty snubber capacitor pulls the most current (compared to other types or arrays of capacitors with the same capacitance).  Again, this was not intuitive for me because I thought a "better" capacitor would have lower internal losses and would thus use less power in the circuit.

I've tried a few different types of "fast" diodes and not seen much difference.

There are a lot of these circuits on ebay these days, and looking at the pictures they all seem to be about the same.  Except for this one:



it seems to have an over-current (or over temperature?) limiting circuit and it looks like they have some other type of gate control.  Note there are 4 small transistors hooked up to the gates of the power FETs.  Does anyone know what they're using for gate control on these boards?

At this point I think I need an oscilloscope to see what different components are doing, so I wanted to ask if other people have already done these sorts of tests.  Thanks!


Online Mads Barnkob

  • Administrator
  • High Voltage Engineer
  • *****
  • Posts: 396
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Royer Refinements
« Reply #1 on: December 03, 2017, 08:44:56 PM »
Hello and welcome to HVF :)

Hello all.  I found this forum after searching and reading about the royer induction circuits online.  I tried to register at 4hv.org but couldn't get in.  Anyway, I'm new to electronics but I've built several of these circuits and I was wondering if anyone with an oscilloscope has experimented with different components.  All I have is a volt meter so I can see changes in voltage/current draw but I don't know why I'm seeing the changes.  In my experiments I've noticed that there can be a wide range of outcomes using the same "recipe" with different components.

For example, I've noticed that increasing the tank capacitance increases the current draw.  My initial guess was that increasing capacitance would drop the frequency and also the current--but I observed the opposite effect.  Does increasing the tank capacitance increase the tank AC voltage?  i.e. the tank will swing further with a lager capacitance?

Increasing the capacitance will lower the primary circuit impedance and this is why you see a increase in current draw.

I've tried different toroid inductors and have not seen much difference.  Perhaps they are more critical at higher current draw?  I've wound the same cores with 30 turns and again with 40 turns and not seen a change.  Has anyone seen what happens if theses inductors are too big or too small?

Too big inductors will result in a too large voltage drop, as you also found out, they are not a critical component, but they do their job as voltage reserve.

I've seen some people put nuts or ferrite beads on the FET gate wires.  I tried some back-to-back tests with the only change being the beads, and I didn't see much difference.  How do you know what these parts do (without an oscilloscope)?

I would only think that is trying to suppress high frequency noise from interfering with the gate drive, as the Royer circuit can quickly get unstable and explode :)

I've tried changing out tank capacitors with MMC or different types all with the same capacitance and seen wide ranges of current draw.  One circuit I have that's built using a heavy duty snubber capacitor pulls the most current (compared to other types or arrays of capacitors with the same capacitance).  Again, this was not intuitive for me because I thought a "better" capacitor would have lower internal losses and would thus use less power in the circuit.

Better capacitors like a heavy duty snubber capacitor has a very low ESR, so while you have lower losses in it, it also lessens the overall resistance/impedance in the primary circuit. The dV/dt ratings are properly also much better on a good capacitor.

I've tried a few different types of "fast" diodes and not seen much difference.

You would properly only see bad results if you used diodes too slow, either it would not work at all or quickly you would see exploding mosfets

There are a lot of these circuits on ebay these days, and looking at the pictures they all seem to be about the same.  Except for this one:



it seems to have an over-current (or over temperature?) limiting circuit and it looks like they have some other type of gate control.  Note there are 4 small transistors hooked up to the gates of the power FETs.  Does anyone know what they're using for gate control on these boards?

It could be that it uses logic gates and gate drive ICs for active gate drive control and phase shifting, instead of relying in the self oscillating nature of the Royer oscillator.

At this point I think I need an oscilloscope to see what different components are doing, so I wanted to ask if other people have already done these sorts of tests.  Thanks!

It is always a must to have a oscilloscope for electronics, but be aware that you need to take care of grounding issues when measuring on inverter and oscillator circuits. If you do things wrong, like not using a isolation transformer or differential probe, you risk burning out your oscilloscope. This video tells you all you need to know

/>
I have not done tests like this on a Royer induction heater, but I did a lot of measurements on a quasi-resonant induction heater topology.: https://highvoltageforum.net/index.php?topic=104.0
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline petespaco

  • High Voltage Enthusiast
  • *
  • Posts: 28
    • View Profile
    • French Creek Valley Activities
Re: Royer Refinements
« Reply #2 on: December 04, 2017, 05:23:27 PM »
You can get an inexpensive  single channel digital 'scope for as little as about $USD30, in kit form, or about USD$50 assembled.  I bought one a while ago just out of curiousity and in can work help you to "see" into the circuits.
Here's one example"
https://www.amazon.com/JYETech-Battery-Powered-Oscilloscope-Display-DSO112A/dp/B01BMYS10S/ref=sr_1_1?ie=UTF8&qid=1512404141&sr=8-1&keywords=dso112

All of my experience is limited to the 1000 watt ZVS heaters that you can see for sale around the world right now, so if your Royer circuits are
a lot differenct, pay no attention to me at all.
  You have talked a lot about current draw, but you don't say whether your measurements are taken with or without a workpiece in the work coil.
To me, the change in "idle" current vs loaded current tells the story.  I hope you are using either steel workpieces or graphite crucibles to load down the system.  Other materials like copper and aluminum don't soak up much power from systems running at 40Khz to about 120Khz, as far as I can tell.
You also don't talk about the capability of your power supply.  If it's voltage sags to different levels with changes in the tank, you may not be comparing apples and apples.
  I concentrate on gate waveforms, so, if you are asking about what others are measuring, go to my page on the subject:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm
About 2/3 of the way down the page you will see some scope waveforms that may help.

Pete Stanaitis
---------------

Offline flyrod

  • High Voltage Enthusiast
  • *
  • Posts: 3
    • View Profile
Re: Royer Refinements
« Reply #3 on: December 05, 2017, 07:56:12 PM »
Thanks for the replies, guys!  Yes, my circuits are the basic version seen in a lot of places.  I'm just seeing a lot of variety in performance using slightly different parts and I'm curious about why.

I've tried a few different types of "fast" diodes and not seen much difference.

You would properly only see bad results if you used diodes too slow, either it would not work at all or quickly you would see exploding mosfets

I recently got some nice diodes to try (MUR140RLG 35ns 0.88Vf) and was surprised to find that the FETs quickly got very hot.  I suspected that they must be ringing at a much higher frequency than the tank resonance, so I tried adding back in the ferrite beads on the gates.  Now the beads had an effect and the FETs did not get nearly as hot.  However, the plain FS105 diodes (250ns) pulled from a junk board still seem to work better.

It could be that it uses logic gates and gate drive ICs for active gate drive control and phase shifting, instead of relying in the self oscillating nature of the Royer oscillator.

That board pictured above seems to be an evolution of this one:



After the over-current protection parts, it seems to have all the exact same components as the other generic ZVS/Royer boards on ebay, except that this new version has those extra 4 small transistors going to the gates.  So it seems like they are trying to better control the gate signal while maintaining the simplicity of the self-resonant circuit.  They have also rotated the capacitors 90 degrees, I suppose for better cooling.  The ebay pictures are not clear, but I also suspect that the new screw is holding a little DC-DC converter board that they are using to drive the 12v fan/pump and the gates as well.  This would give a constant 12v supply for a range of input voltages.

I'm using a nice 30v 30A CC/CV power supply, so I have not blown up any parts so far.  I have seen the "lockup" condition, but with my setup the power supply just current limits and nothing explodes.  With batteries I can see that a fuse would be needed.  So I'm reading the current draw off of the power supply, and I've been doing back-to-back tests with open coil, loaded coil, etc. to test the effects of changing out different components in the circuit.  One recent change I've made is using a 7812 regulator to supply 12v for the gate pull-up, along with lower value resistors (150ohm instead of 470ohm), and a low power switch that lets me turn the circuit off/on just with the gate supply.

So I guess it boils down to getting an oscilloscope.  But I am curious if there are simple ways to clean up the gate signal without going full PLL or solid state tesla coil driver.  Suppose I hook up a scope and see the messy gate signal that I suspect.  What kinds of things can I try to improve it? 



Offline petespaco

  • High Voltage Enthusiast
  • *
  • Posts: 28
    • View Profile
    • French Creek Valley Activities
Re: Royer Refinements
« Reply #4 on: December 05, 2017, 10:47:57 PM »
Quote
So I guess it boils down to getting an oscilloscope.  But I am curious if there are simple ways to clean up the gate signal without going full PLL or solid state tesla coil driver.  Suppose I hook up a scope and see the messy gate signal that I suspect.  What kinds of things can I try to improve it? 

I don't see a "dirty" gate signal that need "clean up" my testing.  It simply doesn't get up to "full on" very quickly when the gate voltage rises too slowly.
I suggest that you carefully watch your power supply's voltage AND current as you experiment.  Sure, you have current limiting, but that is a double-edged sword, so to speak.  If the supply goes into current limiting, it does so by lowering its output voltage, right?
  And this lowering of power supply voltage is seen by the gate circuit.  This, in turn slows down the rise rate of the gate signal and causes an increase in Mosfet heating.  Maybe you have or will solve this by creating a separate gate supply, but that will come with its own complexities, won't it?

If you saw the gate waveforms on my webpage, you can see that even 30 volts applied (see the 24 volt and 36 volt traces) to the gate circuit doesn't give you really fast Mosfet turn on times.  And, for instance, my 24 volt and 36 volt waveforms were made using car batteries for power.

Pete Stanaitis
---------------   

Offline flyrod

  • High Voltage Enthusiast
  • *
  • Posts: 3
    • View Profile
Re: Royer Refinements
« Reply #5 on: December 06, 2017, 03:28:36 AM »
If you saw the gate waveforms on my webpage, you can see that even 30 volts applied (see the 24 volt and 36 volt traces) to the gate circuit doesn't give you really fast Mosfet turn on times.

I did see your web page--thanks for sharing your results!  The gate signal does look much better at 48v.  The "shoulders" on the lower voltage signals look like they could be high frequency ringing.  How high of a frequency can your scope see?  I think this ringing at many times the resonant frequency was causing the FET heating I saw with extra fast diodes.  I would be curious to see if those "shoulders" change shape or size with some sort of noise filtering on the gates.

The 470ohm resistors limit the charge rate of the gates and I think the 470 value is just to cover a range of possible input voltages.  I've had good luck with using lower value resistors with the lower voltages I'm using.  I  would be curious to see if you can get a nice looking gate signal (like your 48v input signal) using a separate 12v source and lower value resistors (maybe between 100 and 150ohms?).  Charging the gates with 12v through a lower resistance should be equivalent to charging them with 48v through a higher resistance.


High Voltage Forum

Re: Royer Refinements
« Reply #5 on: December 06, 2017, 03:28:36 AM »

 


* Recent Topics and Posts

post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
matt0852
Today at 03:20:44 PM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
matt0852
Today at 03:06:07 PM
post Re: My first drsstc
[Dual Resonant Solid State Tesla coils]
oneKone
Today at 01:04:04 PM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
Today at 09:10:16 AM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
chjarup
Today at 06:53:35 AM
post Re: new problem
[Dual Resonant Solid State Tesla coils]
Garyf
Today at 06:23:00 AM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
matt0852
December 13, 2017, 11:42:27 PM
post Re: new problem
[Dual Resonant Solid State Tesla coils]
Max
December 13, 2017, 04:16:45 PM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 13, 2017, 08:41:17 AM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
matt0852
December 13, 2017, 03:35:30 AM
post Re: Highway to DRSSTC
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 12, 2017, 09:36:30 PM
post Re: Highway to DRSSTC
[Dual Resonant Solid State Tesla coils]
Laci
December 12, 2017, 08:49:16 PM
post Re: Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 12, 2017, 07:39:45 PM
post Re: DRSSTC: IGBT selection
[Dual Resonant Solid State Tesla coils]
matt0852
December 12, 2017, 04:09:13 PM
post Steve Ward's Universal Driver 1.3: using IFD95 fibre optic receiver
[Dual Resonant Solid State Tesla coils]
matt0852
December 12, 2017, 04:07:25 PM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 12, 2017, 11:21:12 AM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
matt0852
December 12, 2017, 10:11:22 AM
post Re: DRSSTC: IGBT selection
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 12, 2017, 09:35:46 AM
post DRSSTC: IGBT selection
[Dual Resonant Solid State Tesla coils]
matt0852
December 12, 2017, 03:43:27 AM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
matt0852
December 11, 2017, 11:52:40 PM
post Re: new problem
[Dual Resonant Solid State Tesla coils]
Garyf
December 11, 2017, 11:27:15 PM
post Re: new problem
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 11, 2017, 09:05:14 PM
post new problem
[Dual Resonant Solid State Tesla coils]
Garyf
December 11, 2017, 08:51:12 PM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 11, 2017, 07:14:11 PM
post induction heating/Blacksmithing
[Electronic circuits]
petespaco
December 11, 2017, 04:30:06 PM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
matt0852
December 11, 2017, 02:49:10 PM
post Re: First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 11, 2017, 02:10:54 PM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
Mads Barnkob
December 11, 2017, 01:08:29 PM
post Re: FPS1000HD how-to, information and work flow collection
[High speed filming]
Mads Barnkob
December 10, 2017, 08:32:12 PM
post FPS1000HD how-to, information and work flow collection
[High speed filming]
Mads Barnkob
December 10, 2017, 08:31:47 PM
post First DRSSTC: PCB Critique needed!
[Dual Resonant Solid State Tesla coils]
matt0852
December 10, 2017, 03:43:19 PM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
chjarup
December 10, 2017, 10:49:22 AM
post Re: Highway to DRSSTC
[Dual Resonant Solid State Tesla coils]
Laci
December 09, 2017, 06:37:09 PM
post Re: What kind of oscilloscope to buy?
[Electronic circuits]
Hydron
December 09, 2017, 03:58:31 PM
post Re: What kind of oscilloscope to buy?
[Electronic circuits]
futurist
December 09, 2017, 03:35:22 PM
post Re: What kind of oscilloscope to buy?
[Electronic circuits]
petespaco
December 09, 2017, 05:08:54 AM
post FPS1000HD unboxing
[High speed filming]
Mads Barnkob
December 08, 2017, 11:15:54 PM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
Mads Barnkob
December 08, 2017, 07:54:06 PM
post Re: What kind of oscilloscope to buy?
[Electronic circuits]
Mads Barnkob
December 08, 2017, 07:29:29 PM
post Re: What kind of oscilloscope to buy?
[Electronic circuits]
futurist
December 08, 2017, 05:57:07 PM
post What kind of oscilloscope to buy?
[Electronic circuits]
petespaco
December 08, 2017, 04:39:48 PM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
petespaco
December 08, 2017, 04:22:31 PM
post Re: Hacking the IKEA 2000 Watt induction stove (5 parts)
[Electronic circuits]
Mads Barnkob
December 07, 2017, 10:03:07 PM
post Re: Tardief's DRSSTC
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
December 07, 2017, 07:27:45 PM
post Re: Audio modulation by BUCK power supply
[Solid state Tesla coils]
Acid Byte
December 07, 2017, 06:33:43 AM
post Re: Tardief's DRSSTC
[Dual Resonant Solid State Tesla coils]
Tardief
December 06, 2017, 03:27:42 PM
post Re: WTS [DK]: NWL W00238 Capacitor 815 KVAR 1200 VAC 680 Ampere @ 10 kHz
[Sell / Buy / Trade]
Max
December 06, 2017, 01:56:20 PM
post Re: WTS [DK]: NWL W00238 Capacitor 815 KVAR 1200 VAC 680 Ampere @ 10 kHz
[Sell / Buy / Trade]
Mads Barnkob
December 06, 2017, 12:30:21 PM
post Re: WTS [DK]: NWL W00238 Capacitor 815 KVAR 1200 VAC 680 Ampere @ 10 kHz
[Sell / Buy / Trade]
Max
December 06, 2017, 11:04:26 AM
post Re: Royer Refinements
[Electronic circuits]
flyrod
December 06, 2017, 03:28:36 AM