[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Three phase TC



Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-qwest-dot-net>" <jimlux-at-earthlink-dot-net>



> Original poster: "Basura, Brian by way of Terry Fritz
<twftesla-at-qwest-dot-net>" <brian.basura-at-unistudios-dot-com>
>
> Yes. I'd like to get the RF between the three secondaries 120-degrees out
of
> phase with each other (3-phase as it's being called here on the list).
This
>
> Your idea of using a RSG firing at 120-degrees of the RF cycle seems
> impracticle to implement. If we have a 200khz coil then each cycle would
> only be 5uS (1/200,000). If we then take 1/3 of that it gives us 1.67uS.
It
> seems like that timing would be impossible to obtain using a RSG. Let's
> assume an 8-electrode rotor with a 11" diameter (through centerline of
> electrodes). Rotational speed would need to be around 5million RPM (Yikes)
> and tip speed would need to be close to warp speed (154k MPH).

Err.. Actually, I didn't advocate a RSG for this:

> Another way is with a precision fired triggered gap....fire the second
> coil's gap 120 degrees worth of RF time after the first, and the third 120
> after that.  That's the approach I'm thinking of trying...

that's a set of triggered gaps.. with microsecond timing.


Now that I've contemplated it since you proposed the idea.. I think that
some combination of L's and C's should give you the right phasing.  Tuning
might be a challenge (hah, it WILL be a challenge).  But, I've build antenna
tuning networks to generate precise phase and amplitude variations on
separate radiators.  The problem will be the detuning effect of the
streamers.. fortunately, the secondary is loosely coupled.  The trick will
be coming up with a suitable network topology. It's fairly easy if you use
resistors to get the phase shift.

http://home.earthlink-dot-net/radio/ has some links to equations for calculating
pi and T network values. You can probably start with an arbitrary network,
calculate all the values, and then start combining reactances.  Alternately,
set up some equations in Excel, and let the "solver" grind to find
solutions.