3 phase current fed forward converter for PFC

(It appears my font is messed up again so I will use words instead of ascii art)

A somewhat clever idea:

From the input rectifier we have a series inductor feeding the common point of a Y connected 3 phase transformer.

Each of the phases connects to a rectifier that connects to the output capacitor Each phase also has a connection to a MOSFET to ground.

The 3 phase transformer is made as 3 windings on the 3 legs of an E shaped core.

The output capacitor voltage is tuned to be only about 10% higher than the peak of the input voltage.

Near the peak, the transistors switch on one at a time in a non overlapping manner, With significant dead time between.

When just one transistor is on, the common point of the transformer pulls down to

2/3 the input voltage. The other two diodes conduct. The sum of the current in the diodes is twice the current in the transistor. This means that we are storing some energy in the inductors core but transferring more to the output capacitor.

When the transistor switches off, the energy in the inductor is transfered to the load side.

When the input voltage is at 2/3 the output capacitor voltage, the transistors are working such that one is always on and the inductor is not doing any real work. Below this, the on times start to overlap.

When the on times over lap, the common point goes down to 1/3 the capacitor voltage and the transistors conduct twice the current between them as is going through the one diode which is conducting. The inductor is again doing work.

When we get down to 1/3 the input voltage, the conduction times will start to meet. When all three transistors are on, the common point drops to zero and at this point all the energy is getting stored in the inductor.

The advantage is that the inductor never is doing anything like processing all of the power that needs to go to the load side. The magnetics should be fairly small for the amount of power.

The disadvantage is the complexity.

Just an idea for those who are interested :)

Reply to
MooseFET
Loading thread data ...

Hmm... I think I get it, but a diagram would be handy.

Sounds like something I've seen, then again none of them used wye.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
 Click to see the full signature
Reply to
Tim Williams

mmmmm .....

***** lllll They line up

ASCI art: T1 o ----)))))------+-------->!-- ! ! ! L1 ! o ! ! Bridge ----)))))-----------+----)))))---------+----->!--+---- Capacitor ! ! ! ! ! o ! ! ! ----)))))------------+-->!-- ! ! ! ! ! ! o o o MOSFETS / / / ! ! ! --+--+--+---GND

he

e
Reply to
MooseFET

I think that the 'commom point' voltage will not be as tame as you anticipate. It could swing between the moon and Mongolia without a current path coupled to a lower impedance point. As it is, there are only stray capacitance there.

RL

Reply to
legg

..

f

-------->!--

=A0 =A0 =A0 ! =A0 =A0 =A0 =A0 =A0 =A0!

=A0 =A0 ! =A0 =A0 =A0 =A0 =A0 =A0!

=A0 =A0 =A0 ! =A0! =A0 =A0 =A0 =A0 !

=A0 =A0 ! =A0! =A0 =A0 =A0 =A0 !

-----+-->!--

=A0 =A0 =A0 ! =A0! =A0!

=A0 =A0 =A0 ! =A0! =A0!

=A0 =A0 =A0 o =A0o =A0o MOSFETS

=A0 =A0 =A0 / =A0/ =A0/

=A0 =A0 =A0 ! =A0! =A0!

=A0 =A0 --+--+--+---GND

n

ut

an

e

Perhaps I shouldn't have put dot on the coils because they are misleading.

Because of the way the 3 phase transformer is constructed, the common point to the diodes has a low impedance. I will attempt a little ASCII art to show this:

CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC CCC CCC CCC CCC + CCC + CCC + WW CCC WW WW CCC WW WW CCC WW WW CCC WW WW CCC WW WW CCC WW WW CCC WW WW CCC WW WW CCC WW - CCC - CCC - CCC CCC CCC CCC CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC

Only the leakage inductance will make the common point have a high impedance.

If the left most coil is driven, the two others will share the dB/dt between them and have half the voltage of the opposite polarity on them. In this condition, the transformer looks like a step down.

If the left and center coils are drive, the right one gets the combined dB/dt and it steps up.

Reply to
MooseFET

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.