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

Re: VTTC output

Original poster: "Dr. Duncan Cadd by way of Terry Fritz <twftesla-at-uswest-dot-net>" <dunckx-at-freeuk-dot-com>

Hi Ed, All!

>Original poster: "Ed Phillips by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <evp-at-pacbell-dot-net>
> Unfortunately, there is a bigger problem with VTTC's than all of the
>above, which primarily applies to an amplifier coupled to a resistive
>load.  The TC secondary is highly reactive, the load resistance and
>capacitance are both non-linear, and vary with the power level.  So,
>since in a TC both the primary and secondary are resonant, you have
>adjust either the primary or secondary frequency until both are the
>same.  While you're doing that the load impedance is changing as the
>streamer current varies, etc. etc.

All true and all fascinating and all _highly_ irritating ;-)

>Somewhere I have the manuscript of a
>paper from UCLA covering the theory and practice of coupling a power
>oscillator to a reactive load.  Work was done in connection with
>the D's of a cyclotron.  One thing I remember is that there is a
>possibility that, as you tune, your operation will jump
>from one unsatisfactory mode to another.

Magnetrons and klystrons do this too, but then there is considerable
similarity to a cyclotron.  And I suspect there are yet more examples
of this kind of thing.

> Bottom line is that educated cut and try is probably the only
>way to go, and that tuning/coupling/bias adjustments must be made at
>power level you want to use.  No comfort in this, but it's probably

Yes, 'fraid so.  I'd be interested in seeing this paper!

One upshot of my misbegotten attempts to pervert the NEC2 program to
the modelling of coils (it's all your fault, Skip, building a coil
with sufficiently few turns for my computer to have a go at!) is that
it would appear that when you take a perfectly well-behaved coil,
which at resonance over perfectly conducting ground exhibits a purely
resistive feedpoint impedance, and then stick it over a real ground
model, the point of resonance shifts in frequency as you might
expect - the resonant frequency goes up, according to NEC2. I'm not
sure I would have guessed that one correctly.  The resonant
frequencies predicted however are well wide of the mark :-( a lot
higher than measured.

The resonance over real ground no longer corresponds to a purely
resistive feedpoint impedance, and indeed the frequency at which the
reactive component is minimised is not the frequency of resonance as
indicated by maximum electric field and 90 degree phase shift . . .
Moreover, there appears to be no frequency at which the blessed thing
does have a purely resistive input impedance . . . Whether this means
there is some fascinating scientific tidbit awaiting discovery, or
simply that NEC2 should not be applied to TCs, the jury is still out.
I must say I have trouble totally discounting what NEC2 says because
it is built on some pretty fundamental stuff (method of moments).

However, NEC2 did confirm that, for the coil over perfect ground, the
maximum current is around 43% of the way up the coil.  (This was for a
model of a 17 turn coil of 8mm tubing BTW - I plan on using it in a
large VTTC)  So that at least is something.  On the downside, the
values of Q calculated seem to be low.  Plenty more milage in this I

Suffice to say that my computer is currently working very hard
overnight whilst I am in the land of Nod.  I think it will be doing
overtime for some weeks to come . . .