Old Single stage PFC design never made it...why?

[Faringdon]

Hi,

Please can you ratify this old offline, single stage, PFC'd SMPS design, which never gained popularity despite being around some 10 years ago.........its a single stage, isolated , PFC’d SMPS and can do 150W from a PQ26/25 and no other magnetic components. Its one that I actually saw many years ago, but since its so dis-used, I thought to start using it, since it only has one small magnetic component. I can not draw it, but can explain it to you so you can draw it yourself if you wish?.....

Firstly imagine a Boost PFC, followed by a Forward converter…and then “roll them both into one”…as follows….

1…First of all, draw a Boost converter PFC with the usual inrush diode from mains bridge output to the output capacitor.

2…Make the Boost PFC’s output diode a synchronous rectifier FET.

3…Now add a secondary coil to the Boost’s inductor.

4….Add a secondary side two-diode rectifier, but make it capacitor coupled from the secondary coil, so that it can transmit power to the output on both power “strokes”. Make the secondary side rectifier diodes synchronous if you wish. The secondary rectifier is as in the following...
....Fig 5, page 9 of the LT1248 datasheet.....
https://www.analog.com/media/en/technical-documentation/data-sheets/1248fd.pdf
...but leave out the C3 and R1 and short D3, and make the diodes synchronous (if you wish).

5…Be sure to have a large leakage inductance in this transformer, since that leakage inductance will now act as the Boost PFC’s inductor. This leakage inductance will also act as the forward converter’s output inductor.

6…Switch the two primary FETs as you would a synchronous Buck’s. That is, as anti-phase pulse trains, with 500ns or so of dead time.

7…. Be sure that volt.seconds(ON) = volt.seconds(OFF) for the transformer primary. (you need software control so you can assure that this always holds) You can then vary the power throughput by varying the frequency.

For example, for say a 37V output, with 100-265VAC in, you would have Lp=480uH, Ls = 7.5uH and Leakage inductance with sec shorted measures as 70uH.
__---___

Years ago the company where I worked was trying to do a patent for this. Why did it never get popular?
BTW, it definitely worked, I ran it at 150W output (37v) from 75VAC to 265VAC. Obviously it had a few volts of 100Hz ripple on its 37V output. The transformer was just a PQ26/25.
___---____

BTW, the output rectifier is also seen here...(fig 1 , page 60 & Fig 3, page 61)
https://www.newark.com/pdfs/techarticles/onSemi/PowerFactorCorrectionHandbook.pdf
....in fact, this is pretty much the entire circuit!, except that the Boost PFC diode has been made synchronous, and the transformer has an increased leakage inductance.

[magic]

BTW, it definitely worked, I ran it at 150W output (37v) from 75VAC to 265VAC. Obviously it had a few volts of 100Hz ripple on its 37V output.

Could this be the answer to your question? Conventional two stage SMPS have virtually no 100Hz ripple.

Maybe it's a viable replacement for heavy line frequency transformers in (somewhat) ripple tolerant devices like audio power amplifiers, but then it still needs large secondary capacitors to smoothen the line frequency ripple to a manageable level. Chances are that a two stage SMPS may be cheaper overall AND have better output regulation.

[Faringdon]

Thanks, though we dont mind a few volts of 100Hz ripple.
Its the ripple current in the 400V capacitor that is the crux of the matter.....however, its taking time to rig up a calculation/simulation to investigate this.
""If"" the 400V capacitor is low in ripple current, then the fact that it only uses a single PQ2625 for 150w from 100-265VAC,  and no other magnetic component, will mean that this is the smallest solution out there.