Author Topic: Royer converter, with MOSFETs for low power DC/DC  (Read 8175 times)

0 Members and 1 Guest are viewing this topic.

Offline BuriedcodeTopic starter

  • Super Contributor
  • ***
  • Posts: 1610
  • Country: gb
Royer converter, with MOSFETs for low power DC/DC
« on: February 22, 2019, 03:58:53 pm »
Hi,

For a while now I've been meaning to wind my own transformers for generating high voltage (for which I'm using flyback topology) and things like bipolar low-noise rails for opamp circuits, with the convenience of using a simple 12V power supply (12v -> 5V, +15V, -15V). 

Whilst there are many methods to generate a bipolar power supply, one I wanted to test was a resonant Royer converter.  What stopped me testing this was the requirement for a saturatable core, many of the cores I have for formers are generic N87, k2006 ferrites for <200kHz operation.  Am I correct in saying that a saturatable core is only required when using bipolar transistors? And When using bipolar transistors, one requires a feedback winding to drive these? 

The general requirements are <10W, >50kHz operation, with relatively constant load of 200mA max on each rail, and life would be easier if I didn't have to experiment with feedback windings, so would using MOSFETs mean one only needs a tapped primary?

Unfortunately the web is full of "ZVS mazilli" type circuits, specifically for high voltage and high power, and whilst there is a lot of detail info, there is also a lot of vague info posted by people who just want sparks. I think I understand the importance of the parallel capacitance (for the LC tank) and the choke in series with the centre tap, but I'm unsure if this kind of converter lends itself well to low power converters.  Was wondering if anyone had actually designed and used one for <10W ?
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2205
  • Country: mx
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #1 on: February 22, 2019, 04:23:25 pm »
Royer converters were simple, but I do recall the high current spikes during the switching transitions.

I would think if you employ a soft saturation core, the switching transitions will be slower, and efficiency will suffer.

But hey!, time to experiment a little.

 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #2 on: February 22, 2019, 04:27:27 pm »
Yes, sometimes I build them from stuff that's easily available.
Works with a center-tapped primary and a single choke, or with a single primary and two feed chokes, so I can use random transformers.
The current feeding choke(s) should be the same or one order higher inductance than the primary magnetizing inductance of the transformer, the use a capacitor to set the resonant frequency in the 10...100kHz range.
When you load these, the waveform gets heavily distorted, but still work. If you have an input voltage of say more than 12V, I'd recommend to use a small regulator to provide the gate voltage, otherwise one must clamp the gates, resulting in higher losses in the resistors.

For an example, look at the isolated power supply circuit of this otherwise pointless thing:
http://wunderkis.de/pwramp3/index.html
It's hidden at page 5 of the complete schematic

Another example: http://wunderkis.de/pulser/
(open the pulser.pdf file)

If you build them with bipolar transistors, you want a base drive feedback winding, but it's not required for MOSFETs.
And yes, sometimes I also build them to make some sparks:
http://wunderkis.de/spark/gallery/index.html
« Last Edit: February 22, 2019, 04:33:31 pm by capt bullshot »
Safety devices hinder evolution
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 5839
  • Country: de
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #3 on: February 22, 2019, 06:00:26 pm »
Please, don't go there.
Royer converters are from the days when transistors were really expensive, there is no need for them today, much better solutions and ICs are on the market.
The Royer converter is massively noisy, to the point where you kill stereo, radio and TV reception at your neighbours.

 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #4 on: February 22, 2019, 06:40:00 pm »
Royer oscillators are actually one of the cleanest switching converters due to having no sharp current or voltage transients. Jim Williams created a switching PSU with 200V output that has under 100uV of peak to peak noise using a Royer oscillator:
https://www.analog.com/media/en/technical-documentation/application-notes/an118fb.pdf

As for needing a saturatable core, luckily that is taken care of. Any core that is not an air core is saturatable. Every kind of ferrite saturates at some point, the question is only if the losses in the core at that magnetic flux density are acceptable. If not the efficiency at low load might be terrible or with high power converters the core might get hot.

You can also use mosfets in a Royer oscillator if you want, but they won't really provide any benefit as this kind of circuit does not drive the transistors into saturation. This is the reason why this converter type is so incredibly low noise, but due to the transistors operating in the linear regime it means there is a lot of power burned on the transistors. So the efficiency is not that great and scaling it up to high power means heatsinks on the transistors are a must.

The ZVS oscillator design is just a bit worse for noise (but not by much, currents are still fairly smooth, just a bit discontinous) but has a big advantage that it does drive the transistors into saturation so you get lots of benefit from low RdsOn MOSFETS and from the name ZVS (Zero Voltage Switching) says the transistors switch at zero crossing giving them practically no switching losses. This means there is very little power wasted in the transistors, so little that they stay cold at reasonabe powers. The voltage applied to the transformer is directly proportional to the input voltage so a correctly designed transformer will never saturate. Regulating the input voltage is also the only way you can reasonably regulate the output voltage tho.

Recently i was building a 20W supply that takes 12V in and outputs +/- 200V and i needed it to be low noise. So i ended up using a ZVS driving a standard transformer from Coilcraft with a switching regulator in front to drop the input voltage to the ZVS cirucit in order to regulate the output voltage. The output of the transformer is then a nice sine wave you can rectify. Worked really well and was pretty efficient at high load, tho at low load the efficiency is not great due to the resonant LC currents constantly flowing trough the transformer even at no load, causing resistive heating of the primary winding.

There is no reason a ZVS can't be built for even 1W output power, just have to choose the components accordingly.
« Last Edit: February 22, 2019, 06:43:17 pm by Berni »
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #5 on: February 22, 2019, 06:53:40 pm »
Please, don't go there.
Royer converters are from the days when transistors were really expensive, there is no need for them today, much better solutions and ICs are on the market.
The Royer converter is massively noisy, to the point where you kill stereo, radio and TV reception at your neighbours.

That is true for the original Royer, but the OP is describing a circuit that only superficially resembles the original - the addition of current feed chokes and a resonating capacitor in parallel with the primary inductance (magnetizing + leakage) turns the seriously crappy Royer into the moderately crappy current-fed parallel-resonant push-pull. The advantages that make it less crappy are: sine wave operation and lossless transitions; easily implemented with MOSFETs, as long as the supply voltage is above 10V, anyway; transformer is very tolerant of poor winding technique - leakage inductance is incorporated into the resonant circuit; a modest amount of voltage multiplication as long as loaded Q is above, say, 3.

Disadvantages which make it not-awesome: lossless switching is lost at heavy loads while MOSFETs are lost if output is shorted; supply voltage must be rapidly applied and above the minimum necessary to fully enhance the MOSFETs - slowly ramping up the voltage leads to one MOSFET latching on and usually destroying itself if the supply isn't current (and energy) limited. Circulating current in the tank is surprisingly high and remains that way regardless of loading; transformer has to be sized for at least 3x the power rating vs. a hard-switched push-pull; that same voltage multiplication also applies to the MOSFET's, which have to withstand at least Pi*Vsupply; startup current is massive and output voltage invariably overshoots - particularly if unloaded - so it is good practice to hang a zener or 3-terminal regulator on the output.

But, this circuit can make a very quiet isolated supply. Attached is an LTSpice simulation file of one that I tweaked to supply +/-15V from a 12V nominal input. Component selection was a bit "how ya doin'" to use Dave's favorite phrase, but it works both in simulation and in real life with an EFD25 core gapped to an Al value of 160nH/t² (IIRC - I built this years ago to supply gate the gate drivers in something else; I no longer use this circuit for that purpose - or any purpose, really - mainly because of its sensitivity to the supply voltage ramping up too slowly).
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #6 on: February 22, 2019, 07:09:23 pm »
You can fix the ramping voltage problem by adding a undervoltage protection cirucit to the base supply of a Royer oscillator. You can ramp up the supply voltage as slowly as you want when the base driving supply is disconnected as the transistors stay off, once it has ramped up high enugh you suddenly switch on the base drive supply and it will kick into life. If your main supply browns out the base supply also suddenly disappears and the circuit stops.

I use the same trick with my ZVS power supply and have never managed to get it to latch, even when messing with it during testing. Infact when the transistor input drive supply comes from a separate source you can ramp the main supply all the way down to 0V and the circuit keeps running.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #7 on: February 22, 2019, 07:11:48 pm »
I am not sure why there or here is so much fuss about the royer circuit? It is particularly craptacular circuit, using core saturation to dictate the commutation.

Royer is NOT resonant.

On the other hand, the resonant variant is very quiet in terms of switching noise and most of the disadvantages listed above can be relatively easily mitigated by using a few extra components.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #8 on: February 22, 2019, 07:13:07 pm »
You can fix the ramping voltage problem by adding a undervoltage protection cirucit to the base supply of a Royer oscillator. You can ramp up the supply voltage as slowly as you want when the base driving supply is disconnected as the transistors stay off, once it has ramped up high enugh you suddenly switch on the base drive supply and it will kick into life. If your main supply browns out the base supply also suddenly disappears and the circuit stops.

I use the same trick with my ZVS power supply and have never managed to get it to latch, even when messing with it during testing. Infact when the transistor input drive supply comes from a separate source you can ramp the main supply all the way down to 0V and the circuit keeps running.

That does not fix it really, but is a good addition.

What fixes it instead, is a R-S flip-flop added before the mosfet gates, to prevent both from turning on.
 

Offline BuriedcodeTopic starter

  • Super Contributor
  • ***
  • Posts: 1610
  • Country: gb
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #9 on: February 22, 2019, 07:15:59 pm »
This is the sort of confusion I'm talking about, some see them as noisy (square wave push-pull), others clean (resonant, with a Cap between collectors/drains).  I was thinking more of the latter, and with reasonable filtering on the output, the PSRR of opamps, and perhaps even linear regulators should make it fairly low noise.

As mentioned, my main concern is high current on startup, or perhaps inability to get the thing started.  Again as mentioned, there are better ideas for this application but I'm doing this as a challenge and investigation.  My main reason for going with MOSFET's isn't really for efficiency, but to negate the need for a feedback winding.  I'm winding my own transformers anyway, so an extra winding really isn't much of a hassle, but the ZVS-royer-like-circuits seem much simpler.

Capt Bullshot: thanks for the links. I see the design in pulser.pdf just has one primary, but with two feed chokes.  Thats probably the simplest design I've seen!  I noticed the chokes are pretty high value.  In my LTspice simulations, the higher the inductance, the lower the current spike at the start (makes sense).  Most sources indicate a minimum of 4x the primary winding value, I used 10x.  The isolated 12V supplies also look interesting - I have been meaning to make a fully isolated set of 9V 100mA isolated supplies for guitar pedals and that looks ideal!

Berni, thats what I'm cautious of, component selection.  Generally I use off-the-shelf solutions where component selection isn't that critical (as long as the inductor has the right inductance, DCR and saturation current..) but with self-oscillating ones, the transformer specs can make or break it.   I also have that Linear Tech app note on my desktop.  It is partly what started this.

High efficiency isn't really required, although of course it would be nice to keep it above 70% purely to reduce the need for heatsinks.  I'll get cracking with LTspice before I knock up something real. Cheers folks!
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #10 on: February 22, 2019, 07:57:39 pm »
Well the magnetic parts determine most of the behavior in ZVS, as well as you want some good low loss capacitors in the tank circuit or they could get warm. The transistors simply just need to be big enough for the power rating and for a ZVS that's not very big due to how little power they burn across them.

That design with dual chokes does save you a winding on the transformer but requires a bigger choke for the same ripple current and two of them. So if there is space on the transformer core for a dual primary that tends to be the better choice in my opinion. And yes you want the choke to be at least 2x the transformer inductance when using a center tapped transformer. Bigger is better but the advantage falls off exponentially so going over 10x probably makes no sense while it makes the input choke very large physicaly as it has to handle the entire input current. Efficiency is mostly determined by how low loss you can make your transformer and choke.

With my ZVS design that separately powers the transistors gates and the transformer i get no inrush current at all. Enough power leaks over from the gate drive to start the circuit oscillating before the main supply is even connected. Or the other way around you can apply the main supply before and it will draw zero current before you give it the gate supply too. Connecting the two supplies together and ramping it slowly is what kills MOSFETs, when ramped up separately with a time delay between them it always starts up cleanly.

One thing about the ZVS design tho is that the "output impedance" out of the transformer is really low for common transformer designs without intentionally built in leakage. This means if you have large DC smoothing capacitors on the other side of the transformer the ZVS circuit will draw a huge current from the input to try and charge those capacitors as quickly as the transformer will allow it to. Tho this is also good because the output doesn't sag very much when loaded down heavily, it mostly stays a fixed ratio to the input voltage. Soft starting a ZVS can be difficult tho. My PSU design uses a opamp in the feedback loop to limit the output current because my load is very violent at times.
« Last Edit: February 22, 2019, 07:59:56 pm by Berni »
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2205
  • Country: mx
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #11 on: February 22, 2019, 09:24:21 pm »
Royer oscillators are actually one of the cleanest switching converters due to having no sharp current or voltage transients. Jim Williams created a switching PSU with 200V output that has under 100uV of peak to peak noise using a Royer oscillator:
https://www.analog.com/media/en/technical-documentation/application-notes/an118fb.pdf


I positively love all of the app notes that Jim Williams wrote.
Not only was he capable to explain in exquisite (and clearly understandable) detail all the minutiae required for successful, high performance circuits without being pedantic....... his last page cartoons were always a bonus!

He is sorely missed. One of the greatest electronics designers ever.
 

Offline BuriedcodeTopic starter

  • Super Contributor
  • ***
  • Posts: 1610
  • Country: gb
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #12 on: February 22, 2019, 09:54:16 pm »
Well the magnetic parts determine most of the behavior in ZVS, as well as you want some good low loss capacitors in the tank circuit or they could get warm. The transistors simply just need to be big enough for the power rating and for a ZVS that's not very big due to how little power they burn across them.

That design with dual chokes does save you a winding on the transformer but requires a bigger choke for the same ripple current and two of them. So if there is space on the transformer core for a dual primary that tends to be the better choice in my opinion. And yes you want the choke to be at least 2x the transformer inductance when using a center tapped transformer. Bigger is better but the advantage falls off exponentially so going over 10x probably makes no sense while it makes the input choke very large physicaly as it has to handle the entire input current. Efficiency is mostly determined by how low loss you can make your transformer and choke.

With my ZVS design that separately powers the transistors gates and the transformer i get no inrush current at all. Enough power leaks over from the gate drive to start the circuit oscillating before the main supply is even connected. Or the other way around you can apply the main supply before and it will draw zero current before you give it the gate supply too. Connecting the two supplies together and ramping it slowly is what kills MOSFETs, when ramped up separately with a time delay between them it always starts up cleanly.

One thing about the ZVS design tho is that the "output impedance" out of the transformer is really low for common transformer designs without intentionally built in leakage. This means if you have large DC smoothing capacitors on the other side of the transformer the ZVS circuit will draw a huge current from the input to try and charge those capacitors as quickly as the transformer will allow it to. Tho this is also good because the output doesn't sag very much when loaded down heavily, it mostly stays a fixed ratio to the input voltage. Soft starting a ZVS can be difficult tho. My PSU design uses a opamp in the feedback loop to limit the output current because my load is very violent at times.

This is exactly the sort of info I was hoping would be posted, thank you :)  Lots of stuff to try, without being too vague.  Plenty of ways to change the gate-drive turn-on, a delay between main power (transformer) and gate drive sounds interesting.  This way one can "add on" things so there's a nice scale of complexity of things to try - ranging from the very simple original resonant royer, to a full on brown-out, current fed (buck) royer with regulation. I'll start off simple, and tinker.

The output impedance thing is the reason I never used a royer for applications that require charging a large cap at high voltage, such as xenon flash/strobes, or CDI.  It is often used for these applications, but I always thought flyback was better as the output is essentially a current source, and current-mode control is relatively easy (one can control charge time/input current with peak current or off-time). The only time I used a royer-like converter for cap charging was a couple of decades ago attempting to make a stroboscope from scratch. It worked, but must have been very inefficient, got pretty damn hot.
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #13 on: February 22, 2019, 10:58:13 pm »
Yes flyback topology is the switcher of choice for a simple isolated DC/DC that is pretty well behaved. Its easily controlled by the PWM signal and is reasonably efficient. But they are also one of the electrically noisiest typologies out there and become pretty unpractical for large power levels due to the transistors and diodes having a lot of switching losses.

The ZVS on the other hand is not as well behaved but is electrical quieter than all PWM based switchmode designs and very efficient at the same time. You can reduce increase the output impedance of it by having a special transformer with a magnetic shunt (Like microwave oven transformers) or just add a choke to the output. This increases the transformers leakage inductance and that will limit the current, but also make for worse output regulation under heavy load as the output is more "compliant".

Still if the goal is to charge a big capacitor flyback is the way to go in most cases. While ZVS is well suited for providing clean floating analog supplies.

EDIT: Fixed mistake
« Last Edit: March 04, 2019, 06:29:00 pm by Berni »
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #14 on: March 04, 2019, 05:54:22 pm »
How a magnetic shunt decreases output impedance? Please explain. (It does exactly the opposite - increases leakage inductance, decreases coupling!).

Flybacks can be used for some pretty awesome power densities too, so not really true they are just for small power.
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #15 on: March 04, 2019, 06:28:22 pm »
Sorry i did make a mistake there, i meant increase the impedance. If one wanted to decrease it a way to do it would be to get feedback from the output and form a regulation loop.

I'm not saying flybacks can't be made more powerful by simply making them bigger. Its just that as they are scaled up that diminishes there advantages of being very simple and cheep and exposes there disadvantages such as being hard on the switching elements (Big expensive semiconductors with lots of heatsinking) and creating a ton of switching noise. Since the PSU is now getting big and expensive it makes sense to go with some of the more complex designs that offer grater efficiency and power density. This could make the supply cheaper, smaller, more powerful and more efficient. There is a reason why so many different switchmode topologies are in use, each one fills its own niche.
« Last Edit: March 04, 2019, 06:37:41 pm by Berni »
 

Offline flyrod

  • Contributor
  • Posts: 11
  • Country: af
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #16 on: March 04, 2019, 09:12:53 pm »
You can add a separate inductor to the tank circuit so you can't completely short out the resonance with a load on the transformer:

https://digitalcommons.calpoly.edu/theses/112/

 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #17 on: March 05, 2019, 06:18:33 am »
Ah nice i could have used that research paper when making mine.

I ended up designing the same thing except that my buck PWM driver is simply a switchmode converter IC (I only needed 20W) with integrated switches and i didn't use any fancy gate drive on my MOSFETs, instead just used the classical ZVS Mazilli circuit (With the separate gate supply modification to fix startup problems). In my opinion there is not that much benefit in the fancy fast gate drive circuitry since the mosfets experience very small switching losses due to doing it at zero crossing.

The major efficiency hog at low load is the transformer because high resonant currents continue to flow trough it and at high load the buck switchmode in front of it, its pretty much the only thing that gets warm in the whole circuit. Well apart from the magnetics getting warm due to just simple resistance losses.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #18 on: March 05, 2019, 07:31:20 am »
Note that the constant-current (or inductor) fed, resonant version was developed by Baxandall.  Royer developed the saturation-commutated circuit.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #19 on: March 05, 2019, 10:44:39 am »
Like others said, replacing the bipolar transistors with MOSFETs in a Royer converter still requires the separate control winding.

Another Royer implementation Tektronix had in the early 22xx oscilloscopes used a separate saturating and output transformers.  That way the current spikes during saturation can be much smaller for a given output power and the same small saturating transformer can be used in multiple designs.

What I would consider for a low noise universal design though is one of the push-pull controllers from Linear Technology like the LT1533 or LT1683 which support adjustable current and voltage slew rate limiting.  If configured for open loop operation, they can also be used to drive a low noise inverter with primary side regulation like Tektronix was fond of doing.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #20 on: March 06, 2019, 06:02:11 am »
Oh hey, so especially that: if you use a separate feedback transformer (which saturates at less flux), which is slightly loosely coupled (say with resistors) to the output, then the drive saturates rather than the output, and peak currents go away.

That's basically half of what I did here,
https://www.seventransistorlabs.com/Images/LED_Light2.png
A half wave (monostable rather than astable) with forced hFE = 5 and a Schmitt trigger to drive it (controlled by a single transistor as error amp).  The #43 core saturates in, whatever it was, something like 5us, heavily shorting out the base charge (giving turn-off about as fast as possible) while the stored current swings it to negative base voltage quickly, after commutation is complete (transformer current is able to go to zero after the catch diode turns on).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #21 on: March 06, 2019, 02:52:50 pm »
They are unrelated but that reminds me of the blocking oscillator inverter Tektronix used in their military 465 style oscilloscope.  It was their last iteration of this type of inverter and has refinements over their earlier designs.  It might be considered a singled ended Royer oscillator and it could be adapted to low power regulated DC to DC conversion which is essentially what it was doing as shown below.

L554 and Q556 operating as a synchronous rectifier function to scavenge energy which would otherwise be lost improving efficiency.  This may have been due to a requirement that the oscilloscope operating with passive cooling only.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #22 on: March 09, 2019, 03:14:32 pm »
Oh hey, so especially that: if you use a separate feedback transformer (which saturates at less flux), which is slightly loosely coupled (say with resistors) to the output, then the drive saturates rather than the output, and peak currents go away.

That's basically half of what I did here,
https://www.seventransistorlabs.com/Images/LED_Light2.png
A half wave (monostable rather than astable) with forced hFE = 5 and a Schmitt trigger to drive it (controlled by a single transistor as error amp).  The #43 core saturates in, whatever it was, something like 5us, heavily shorting out the base charge (giving turn-off about as fast as possible) while the stored current swings it to negative base voltage quickly, after commutation is complete (transformer current is able to go to zero after the catch diode turns on).

Tim

Shouldn't the diode be rather connected to the emitter directly? This way it is an incorrect design, with additional inductance (the 3 turn winding) presenting inductive spiking to the power NPN.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21606
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #23 on: March 09, 2019, 05:46:18 pm »
No, that would force continuous load current through the transformer, and it wouldn't come out of saturation, or at least very quickly.

Peak voltage is constrained by the primary side circuitry, which actually isn't noted in this schematic, there should be a load resistor or zener in parallel for this. :)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline adammunich

  • Contributor
  • Posts: 26
  • Country: us
Re: Royer converter, with MOSFETs for low power DC/DC
« Reply #24 on: December 01, 2020, 10:51:57 am »
What fixes it instead, is a R-S flip-flop added before the mosfet gates, to prevent both from turning on.

So I tried it and as it turns out, it does not fix the problem. Instead, one mosfet latches on and then blows up.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf