Re: [TCML] 3-500Z tube coil project

[Jason Johnson <jasonmsusolar@xxxxxxxxx>]

On Tue, Jul 13, 2010 at 11:28 PM, <bturner@xxxxxxxx> wrote:

> From what I heard, the grid winding can start out with the same number of
> turns as the primary - main thing is enough grid voltage to bring the
> triode to cutoff. Approximate the volts/turn on the primary, assume that
> for the grid coil, knowing the maximum grid current for the tube will then
> tell you what sort of grid resistance range you will need.
>

Amazingly simple and concise. I hadn't gotten it boiled down to this point
yet in my mind...thanks for the reasoning here.


>
> My understanding (and how I built my VTTC - using a pair of 805's) is to
> use a fixed number of turns on the primary and add or remove tank
> capacitance to bring it into resonance. Varying the primary turns can
> alter the plate load too much.
>
>
This is what I was thinking to begin with. But, it seems to me that my
(somewhat) arbitrary choice of ~22 turns may not really be the best for the
tube anyways.

I have worked through the math a bit and calculated C_pri by setting Q equal
to between 10 and 20 and figuring the load impedance of the tube to be some
2500 ohms, then I get between 1.4 and 2.8 nF by following John Freau's math
on Steve Ward's webpage (based on the proper care and feeding of power tubes
manual by Eimac, I believe...). Of course, I was assuming F_res of 457 kHz
as predicted by JavaTC; If I use the 386 kHz predicted by WinTesla, I get
other values....

So as always with TCs it looks like a tradeoff between L and C. I am used to
spark gap coils where I would try to err on the side of higher primary surge
impedance by keeping L on the high side and C on the low side of acceptable
values. I have enough caps to make C_pri as high as 8 nF, but I would like
to not beat up on the tube _too_ much.


> Be sure to decouple the plate power transformer from RF - use a mica
> transmitter cap across the transformer winding - 0.01uf should be enough
> and also have a means to adjust the cathode voltage. The thoriated
> tungsten cathode can be damaged if the voltage (thus current) is not
> within spec. Too low and electron emission suffers ('sucking' electrons
> from the tungsten), too high, and it can physically weaken the cathode.
> Use a balanced-current method - cathode transformers normally have a
> center-tap to use for the remainder of the cathode circuit.
>

Yes, good advice. I have a _real_ filament transformer (from the 1950's, so
you know its heavy duty). I was planning on bypassing each side to the
grounded center tap with a 2.2 nF 800 V polypropylene cap (though I might
add some more C, since you recommended 10 nF). Also, I'm throwing in a
variac on the filament transformer and a voltmeter across the filament pins
for good measure.


>
>  - b
>
>
>
Thanks for the food for thought.

- Jason
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla