Royer without feedback winding?! / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Royer without feedback winding?!

Move Thread

LAN_403

Hazmatt_(The Underdog)

Fri Feb 09 2007, 03:09AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

My Power Electronics class was assigned to design an inverter that was self resonant D.C. to A.C. for another project the instructor has. So I looked around for a driver that would satisfy some requirements. It needed to work on a conventional center-tapped transformer and it needed to operate on 12VDC. The circuit appeared in Volume 3 of the Electronics Circuit Encyclopeida by Graf and first came from Popular Electronics.
The circuit had a simple problem, one of the transistors was flipped upside down, so the circuit would work, but only by half. I corrected the 'typo' and it works pretty well.

The first circuit is the one meeting the design requirements of the instructor. It operates on 12VDC and outputs about 40W of power. The graph of the output pulses is the one i'm showing here.
After that I got to playing around with the circuit to make one in a high-voltage flavor. It operates on 160VDC because that's what you would expect from rectified 110VAC. So this high voltage one would work from the rectified mains. I want to point out though, that you need to be careful how you divide down the voltage from the positive feedback. If you don't divide it down enough, the transistors will be pulling some serious current and die. This higher voltage one opeates at about 80W of power, so that's a bit under the 3055's maximum of 115W.

Have fun.

...

Fri Feb 09 2007, 03:16AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

That is almost the mazzilli driver, using bipolars instead of FETs (and the diodes to keep the gates from going negative and zeners to keep it from going >12v) But it is missing the series choke, and I am sorta curious as to how you managed to get yout circuit to work without it, unless it is relying on the core saturating... You might try adding a ~100uh choke in series with the centertap of the primary and see how it affects things.

Sulaiman

Fri Feb 09 2007, 08:31AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

What you have there is a 'blocking oscillator',
as mentioned above it works by saturating the core.
Very reliable BUT not as efficient as a 'Royer' (Current-Fed Parallel-Resonant) invertor
because energy is lost each half-cycle in saturating the core.

Marko

Fri Feb 09 2007, 12:51PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

That is almost the mazzilli driver, using bipolars instead of FETs (and the diodes to keep the gates from going negative and zeners to keep it from going >12v) But it is missing the series choke, and I am sorta curious as to how you managed to get yout circuit to work without it, unless it is relying on the core saturating... You might try adding a ~100uh choke in series with the centertap of the primary and see how it affects things.

It has no resonant capacitor so it's more of a some sort of blocking, or whatever, core-saturation oscillator as sulaiman said. It more resembles our '3055 flyback driver except it uses primaries as feedback.

Higher power inverters of this kind use the trick of a small current transformer wich gois in saturation and drives the transistors; main transformer never goes into saturation.

Are you allowed to use mosfets? At 12V, in any case, they would be much more efficient than 3055's. 3055's can have pretty high saturation voltage at just few amps, and will burn lot of power even in DC conditions.

If you used two IRFZ44 mosfets and built our popular 'mazzili' circuit you would have all your requirements and some pretty good efficiency, with already well-tried circuit.

Sulaiman

Fri Feb 09 2007, 07:17PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

If MOSFETs are used then diodes from Source-to-gate would be good
as the (fairly robust) 2N3055s can withstand the negative 'spikes' on their bases
whereas many MOSFETs will die due to gate-source overvoltage.

Using a separate small saturable transformer would definately be more efficient.

Hazmatt_(The Underdog)

Fri Feb 09 2007, 10:10PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

The parts of the driver are very specific, only BJT, the center tapped transformer which has to be an 'off the shelf' variety, and minimum amount of resistors and capacitors.

One of the requirements I failed to mention was 60Hz operation. So as-is, it does need to saturate the core and switch at that core resonant frequency. But you could use a ferrite if you wanted to drive a flyback.

A driver of a similar type with a second external saturable transformer is called a Jensen Inverter. I saw that one too, and I could have made the driver work that way but the instructor would probably start arguing with me about the driving transformer.

Some students submitted Monostables as their drivers, so I'm waiting to see what the instructor says about that.

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