3 phase ac SEPIC converter

[Johansen]

**broken link removed**

the tentative concept is 480 vac in and +/- 200 vdc out.

thoughts?
is this insane?
topologically the switches could be arranged in delta but then you lose the rail splitting effect.
(that and its difficult for me to imagine the switching waveforms in my mind, when the inductors and switches are connected delta)

its an improvement over other 3 phase buck/boost capable power factor converters, at the expense of twice the magnetic volume as compared to boost only PFC (I think, depending on the input voltage ratio) and you need three relatively large capacitors, as well as the WYE and or delta connected input capacitors.

other advantages are intrinsic inrush current limiting, and voltage isolation subject to the 60hz line impedance of the capacitors..

 

[goldsmith]

Hi Johansen

the tentative concept is 480 vac in and +/- 200 vdc out.

Why this way ?

thoughts?
is this insane?

ha ha ! somehow it is ;-)

Why you are going through this way ? why not usual ways for this purpose ? are you familiar with ordinary and popular ways for such an aim ?

Best Wishes
Goldsmith

 

[Johansen]

are you familiar with ordinary and popular ways for such an aim ?

I am not aware of any buck boost 3 phase PFC circuits.

there are several topologies of boost only, i've read lots of documents on those.

i've run across buck only pfc, but it can't get 100% (hardly ever needed anyway)

and you could make a buck boost from a fully controlled rectifier, however this requires 12 switches.

the 3 phase ac sepic i propose requires 3 ac switches and 6 fast diodes, at the expense of:
Double the inductor volume compared to boost only (i think this is true, but it depends on the input voltage to output voltage ratio, as does the boost)
3 600vac rated capacitors to push the "kilowatts" of power through.

 

[goldsmith]

i've run across buck only pfc, but it can't get 100% (hardly ever needed anyway)

Hi Johansen

What you mean by buck only PFC ? are you meaning a buck converter as PFC ? but it is not a good idea because buck converter is dealing with reactive power factor which must be improved via another circuit too . did you mean that really ?
What you mean by 100 % ? are you mean 100 Cos phi= 1 ? did you know that cos phi of one can be considered as a good thing at fir5st vision but it won't be good for power line and generator lines and also dispatching systems ?
If you look from the top of these things you'll see in power plant sections they intentionally won't let cos phi be exactly about 1 . did you know this and do you know why ? ( all of these things are applied if you are speaking about cos phi and i could catch your meaning correctly )

the 3 phase ac sepic i propose requires 3 ac switches and 6 fast diodes, at the expense of:
Double the inductor volume compared to boost only (i think this is true, but it depends on the input voltage to output voltage ratio, as does the boost)
3 600vac rated capacitors to push the "kilowatts" of power through.

I'll be happy if you describe your design more . would you please ? because it sounds interesting .

Best Regards
Goldsmith

 

[BradtheRad]

For a simple conception of how a 3-phase AC-fed buck-boost converter might work, here is my simulation:



It does not have all the windings which a SEPIC type has.

It shows one polarity, or one-half of the layout.

Capacitors are installed across each supply, to provide power factor correction.

The SCR's are switched on soon after the waveform goes positive. Notice with an AC supply you cannot obtain greater than 50 percent switch-On time. This limits the possible range of output volt levels.

By experimentation, the coil values are sufficiently high to result in about 200 V to the load (1.24 kW) at the 60 Hz operating frequency.

 

[Johansen]

well after googling "buck boost 3 phase power factor correction"
i've run across a number of circuits..

none of which are as elegant as my ac sepic converter (lmao)

Rad, did you see the waveforms from my 4 phase boost converter? the damn thing works surprisingly.
i'm working on a 400 volt version for my induction heaters.

 

[mtwieg]

After a bit of browsing on IEEE I've found numerous papers on buckboost PFC, some of which are based on SEPIC. This seems to be the simplest configuration: