Faraday's law not obeyed by two switch flyback SMPS

[eem2am]

two switch flyback

Hello,

The two switch flyback, -detailed in this document, appears to disobey Faraday's Law:

V = N d(phi)/dt

http://www.powernucleus.com/application_no...ower_conver.pdf

...This is because the primary coil is clamped to the input supply voltage during secondary conduction.

However, during secondary conduction, the secondary will be referring a voltage of V(output) * Np/Ns onto the primary.

This must be so since both primary and secondary suffer the same d(phi)/dt, since they are wound on the same core.

So i cannot understand how the primary can be having these two different voltages imposed on it at the same time...

...the fact that during secondary conduction, the primary is clamped to the input voltage, means that Faraday's Law has been disobeyed, since the volts per turn reffered by the secondary cannot possibly be experienced by the primary, since it is clamped to the input voltage.

I wonder and am grateful if readers can shed light on how this two switch flyback can possibly work?

Added after 4 minutes:

Apologies for the above link not working...here is the two switch flyback...

**broken link removed**

 

[ghydda]

dual switch flyback circuit

Hi eem2am,

You are partly correct.
However if we assume that the secondary output voltage is sufficiently low the reflected voltage onto the primary side will not be higher than the input voltage and as such no law will be broken.

But I agree in that it is not common practice to design a flyback like this.


Cheers,
Ghydda

 

[eem2am]

flyback converter welder

Thankyou for replying,

..the problem for me is that even if the referred secondary voltage is lower than the input supply voltage.......which voltage wins out and ends up going across the primary?

.....from the schematic, there is no doubt that the primary has the input voltage impressed across it after mosfet switch-off......but this is not the referred secondary voltage......so i am wondering what has happened to the referred secondary voltage?.......by Faraday's Law it should be impressing across the primary and calling the primary voltage.

 

[ghydda]

the two-switch flyback converter

Hi again,


It would seem to be the perfect time to bring up the good old equivalent circuit of the transformer.

**broken link removed**

Sorry for the crude drawing, Bill's good old Paint was what were available

I have put a rectifying circuit and a dummy load on the output and also put on the switches and clamping diodes.



If we can agree on that the two resistors and the two inductors can move freely from primary to secondary if we transform them correct, then picture this:

When both switches turn of, the magnetizing current will continue to flow. This means that the ideal transformer is not locked down at both ends:

At the secondary the voltage is clamped to the capacitor-voltage (it cannot change instantly).

The primary is not clamped to supply rail because there is an inductor in series with it, which provides the necessary voltage to make ends meat.

Or if you choose: the primary is locked to supply and the secondary has too little voltage --> which the freely flying magnetizing inductance corrects by jumping into series with the secondary winding and "boosting" the output voltage of the ideal transformer.



So I guess that it is perfectly alright to use a dual switch this way. I never really analyzed this particular scenario before now.



Cheers,
/Ghydda

 

[FvM]

flyback leak

To my opinion, it's misleading to designate the circuit a flyback converter. It's usually understood as
push-pull converter and operated as such by keeping the stored energy low respectively the main
inductance high. It's very popular e. g. for kW arc welder converter.

I prefer to draw the primary as a one-quadrant full bridge converter. It's main advantage is to
limited the overvoltage from the transformer leak inductance to the supply voltage. This also
limits the possible flyback operation range and doesn't permit step-up action with meaningful energy
transfer. Thus it's not suited for a wide-range of-mains SMPS or similar applications.

P.S.: I think the leak inductance has a wrong place in the above drawing. In parallel to the
transformer, you find the main or magnetizing inductance, while the leak induktance is
series connected between primary and secondary, consult any transformer euivalent circuit in literature.

 

[ghydda]

two-switch flyback

You are correct.
I accidentally got the leakage and magnetizing inductance swapped.
My apologies. I was not thinking straight yesterday due to a mild flue. If you do not buy that excuse then I have others available

And as such my first conclusion is still vaild.
I however do see the sense in looking at it as one half of a full bridge. Makes it easier to identify the performance characteristica of the topology.



Cheers,
/Ghydda

 

[eem2am]

smps flyback advantage

Hi,

Thankyou for these replies, i believe Ghydda you are saying that the referred secondary voltage does get impressed across the primary, and then the leakage inductance gets the rest of the voltage to make it up to the input supply level, or am i wrong?

Also, i beg you FvM to call this a flyback though and not a push-pull......since its flyback nature is confirmed in that the energy is stored up in the core whilst the primary conducts and then, discharged when switch turns off.......in true flyback nature.....unlike the push pull, where current flows in primary and secondary simultaneously, but that the fields of these currents cancel each other out.