Re: 3 phase to single phase, transformer wiring VTTC

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: FIFTYGUY@xxxxxxx 

In a message dated 8/26/06 10:30:38 AM Eastern Daylight Time, 
tesla@xxxxxxxxxx writes:

>Yup, 3x current, but limited to whatever the Y voltage is (otherwise
>the core will saturate). Phasing them all right will be the confusing
>part (if ones out of phase that's a short).  3 phase kva = V*A*1.73
>with V and A in relation to gound. But in single phase mode you'll
>end up with 3x original kva rating.

    I gave this some thought, but wouldn't you be limited to twice 
the current rating of an individual winding?
    Obviously, no single winding can handle more current than it's 
rated for. So that puts one limitation on the possibilities.
    But I think there would be a limitation due to flux density in 
the core. Since a one-piece 3-phase transformer designed for a 
balanced load shares the same core for all three windings, running 
each winding in single-phase would cause a core saturation problem somewhere.
    For instance, in normal 3-phase mode, each of the three "I" legs 
carries a flux that is shared by each of the other two legs. But in 
single-phase mode, with two windings in parallel, the "I" legs of 
each core support the full flux all the time. If you add the flux 
from the third leg, there's no "return path" for the flux from that leg.
    All the 3-phase transformers I've seen for balanced loads are of 
the "core" type construction, with the three "I" legs plus top and 
bottom core sections to link them. As I understand it, transformers 
for UN-balanced 3-phase loads have additional "I" legs to the 
outsides of the outside windings, and extended top and bottom cores 
to link the flux to them.
    Using a bank of 3 single-phase transformers (i.e., three pigs on 
a pole) is more versatile because each transformer is complete unto 
itself, and the flux in each phase is never magnetically linked to 
the other transformers. Such an arrangement would be readily 
convertible to single-phase operation. However, then you've got the 
problem of circulating current from imbalances, since it's not 
"sorted out" in a common core. Hence the issues with balancing chokes 
on big variacs. I've worked on high current, low voltage (12V 1000A) 
systems where circulating current problems with parallel transformers 
and parallel cable runs cause huge imbalances. Obviously not as much 
of a problem at "normal" current levels, but try to keep all 
connections and cables identical.

-Phil LaBudde