Randy said:
Chris wrote:
[email protected]
I was asked this question by a curious owner. I don't know but, to
reduce temptation, I think a solid answer is needed.
rg
There are parts of the core that bridge between the sections (legs)
that have coils on them that carry flux from more than 1 coil. When
the coils are excited with 3 phase, the coils, in effect, take turns
using that bridging part of the core.
If two coils are energized at the same time, there are two
possibilities. Either their fields around the loop through those two
legs adds, so that that loop has twice the rated peak flux, or they
buck each other, putting twice the normal flux over through the third
leg. In either case, at least you get the same volts per turn and
(with voltage derating for the saturation taking place at least
somewhere at half the rated volts per turn) it still acts like a
transformer. This assumes that you can get to all the coil ends and
wire then either in series or parallel, not delta or Y like they were
connected for 3 phase.
With 3 coils energized, simultaneously, either they all create field
in the same direction so that there is no return path for all that
flux, or two fields are in one direction and the third is in the other
direction, so that the odd leg still has twice the flux of the other
two. Since volts per turn are proportional to the peak to peak flux
(for a given frequency) you either get twice the volts per turn on the
odd leg, the same volts per turn on each coil, but no closed path, so
very low inductance and very high magnetizing current and lots of
fringing fields. I'll leave the first case as an exercise how you
series parallel connect the coils to get some kind of transformer.
The second case is just a low impedance reactor across the line.
Is this what you were trying to picture?