COP 20.00 (2000%) Times, Reactive Power Energy Source Generator,

[G4RR3ττ]

Hello gents,

The site where I originally posted some of this info is down right now due to some complications involving a domain transfer, so I thought it couldn't hurt to post here in the mean time, as well as send a reply back to Wayne whom I was trying to write back to, and who looks to be on this thread.

Wayne,

Good eye! That circuit was drawn up late in the night and I didn't think much about the discharge path while putting it together. As for the fix, an NPN MOSFET can be used as a synchronous rectifier since MOSFETs can pass currents in either direction when turned on, this also helps reduce total conduction losses for the circuit under charging conditions as well.

For everyone else,

If you're interested in the circuit I've attached some of the diagrams I made recently, though the accompanying text that I had written on Garagehacker is a necessary compliment. For a basic summary: the circuit uses synchronous rectification and a simple DC series/parallel switching circuit along with PWM to force a 2xf sinusoidal current that would meet the requirements of Jim's switching circuit, particularly the Steinmetz reflection criteria. I feel this is a superior approach to both Babcock's convoluted circuit and Jim's original SERPS prototype built by "Eric," which appears to have used SSRs.

I think most people are caught up in the subterfuge of the circuit's description, particularly Jim's terseness and Babcock wanting to promote his "flyback" circuit. The only thing that really matters is the shape of the signal needed for creating the "reflection". If you know what the currents and potentials should be doing, then the circuit can be built however one would like as long as it satisfies the requirements of producing the correct wave shape and phase relations.

[TinselKoala]

If a circuit produces signals with higher harmonic frequencies that exceed the simulation's clock rate, is the sim likely to return correct math answers?

[MarkE]

If a circuit produces signals with higher harmonic frequencies that exceed the simulation's clock rate, is the sim likely to return correct math answers?

Nope.  The minimum simulation step needs to be carefully selected so as to capture all important information and still not make the output file unwieldy.

[hartiberlin]

@vasik041

LTSPICE simulation looks good !
So the last green scope traces show mostly negative input power
on average while the blue trace above it at the lamp resistor shows positive real active power ?


Well done !


Regards, Stefan.

[vasik041]

@vasik041

LTSPICE simulation looks good !
So the last green scope traces show mostly negative input power
on average while the blue trace above it at the lamp resistor shows positive real active power ?

Well done !

Regards, Stefan.

Hi Stefan,

Negative power in this context means power consumed from the power source.
The circuit is quite efficient, 7.1 W consumed and 6.8W dissipated on the load.
It is also creates interesting asymmetry in primary current. With "real" transformer it probably cause core saturation. May be this creates some "magic" ?

Regards,
V.