I have a Pansonic stereo system with separate high and low speaker
terminals. There is only one stereo power amp ... thus it is not a true
bi-phase arrangement. I can't find the answer to how the hybrids work, so I
am posting my guess here.

Here is what I think is going on. The right channel is phase inverted when
compared to the left. Thus, for a mono signal, the signal voltage across
the R-L connection is twice that from R to gnd or L to gnd. The clue is
that the low speakers are series connected across the R-L hot terminals and
so are the surround speakers.

Anybody know for sure? Thanks.

Sounds like you are prettty close in your assesment.

A BTL amp puts out 4 times the power of a single ended. When loaded
with a static impedance, you put twice the voltage you get twice the
current, which means quadruple.

Most likely they are feeding the woofers in paralell through a
crossover. Because 90% of the time bass is in phase and mixed equally
into both stereo channels this can work.

Used to be there was one way to design an amp, well, I take that back.
If you are interested in weirdo amp designs search this group for
"single transformer stereo amp", something like that. I got in it, but
the schematic is not supplied. Someone did email it to me though. Kinda
anteresting, and tubes no less.

If you like wierdo amp designs, also check out class G and class D
amps.

If you like all that, one of these days I am going to sit down and
design a class D power amp that uses only one output device. I have the
concept, and I am sure I can do it in the analog domain, but it is not
easy. Also, if I want it to cover the full range for audio, it is going
to be a very very fast four quadrant triac. I don't know if they make
them fast enough yet, but that doesn't mean the circuit can't be
designed, it just can't be built.

If I can get one that I can accurately PWN at like 60Khz or so, it is
feasible. Otherwise it might make a good low frequency amp. Of course
such a thing might already exist, I can't say that something does not
exist.

If you like all that, one of these days I am going to sit down and
design a class D power amp that uses only one output device. I have the
concept, and I am sure I can do it in the analog domain, but it is not
easy. Also, if I want it to cover the full range for audio, it is going
to be a very very fast four quadrant triac. I don't know if they make
them fast enough yet, but that doesn't mean the circuit can't be
designed, it just can't be built.



You can already get class D amps in a single chip needing only a couple
of passive components to work. Why bother making one from scratch?