John Byrns wrote:

OK, I see we are going in opposite directions on the coupling effect.
If the coupling were perfect when the core is not saturated, then the
light would be as bright as it could get due to the bucking effect of
the two windings.

Right - up to the "as it could get" with the core NOT saturated - then
the coupling is good - the windings buck - bulb dim.


When the transformer becomes saturated decreasing the
coupling, the brightness would have nowhere to go but down due the
residual inductance remaining in the two coils.

This is where you go off track - when the core saturates two things
happen - the inductance of the coils drops through the floor - less
inductance - higher current. The coupling also starts to fail as well -
so the bucking effect also decreases - bulb brighter...


News Flash, I am really liking the theory I advanced a few messages ago
that the "transformer" is not wound like an ordinary transformer, but
instead has one of the secondaries wound on each outside leg of the
E-core.

Not news - this would be common for a saturable reactor.

The reason I am really liking this theory now is that I looked up the
service data for the General Electric E-155 which has the similar
Colorama tuning system, and the "transformer" is constructed exactly as
I speculated in my earlier message. If the S-W "transformer" is built
the same way then it is a whole new ball game.

Why? If in fact (as I've postulated most recently) that reactance is
indeed the greater control factor - then whether one coil of the two is
reversed or not would only effect the linearity of the control, not that
it works. I looked up a few site - and found a couple interesting ones...

http://www.allaboutcircuits.com/vol_2/chpt_9/7.html

Down towards the bottom of the page - just above the Scott-T. Notice
that in the final example - they reverse one of the two control windings
- the same thing would be accomplished by leaving the control windings
in phase and reversing the the load winding... Here they use two
reactors to accomplish symmetry - here is a way to do it with one:

http://www.tpub.com/content/neets/14.../14180_137.htm
note that this is the first of several pages - you'll have to page
through them to get to the main point. Notice that by the time they get
to the three legged configuration - (and torrids) - the load windings
are opposed to provide symmetry.

Again not knowing the construction of the S/W transformer leaves some
guessing - but as I noted before - if all three windings are wound on a
C form - then bucking may be a significant part of the control
equation - if three legged on an E core (I doubt torrids were in use
back then) - then bucking would not be significant (as that coupling
would be minor compared to the reactance change).

best regards...
--
randy guttery

A Tender Tale - a page dedicated to those Ships and Crews
so vital to the United States Silent Service:
http://tendertale.com