Re: srsg + mots - happy couple?

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "Dmitry (father dest)" <dest@xxxxxxxxxxx>


> Original poster: Steve Conner <steve@xxxxxxxxxxxx>

>>here are the model of my psu:

> It looks great to me. My only question is where will you get the 20
> henry inductor that can withstand 10kV?

hey, Steve - i think you`re able to do more difficult things than this
one :-) and what`s the problem -  it`s 100hz hv, not hf. look - the
insulation of an ordinary enameled wire can withstand more than
1000v, so for 10kv we need only 10 layers, but there would be also an
insulation between layers,  that`s able to keep 1000v (i`ve got a
whole bunch of the transformer paper, that`s impregnated  with
something like paraffin). the only problem - insulation from the
core, but i guess it wouldn`t spoil the deal - 5mm of the teflon
would help us. or something is wrong with me again? :-D

as for the 20H - everything depends from the mots Ls  - maybe we`ll
need a 40H inductor, maybe  10H, or maybe we can keep even without it
at all. i have to measure the coupling coefficient between the
primary and secondary of each of my particular mot. i`m planning to do
this in such way:

1. feed the primary, say, with 200v, then measure the voltage on the secondary,
x1=Usec/Upri

2. feed the secondary with, for example, 1900v, measure the primary voltage,
x2=Upri/Usec

coupling coeff. K=sqrt(x1*x2)

is it OK? i`ve thinked it out in 5 min, could you give me the advice
how to make it more precisely?

and one more question - i need Ls as much as possible - are the shunts
influence this? they limit the flux between the pri & sec, but don`t
accumulate the energy - is it the same as decreasing  Ls at large flux?

> Also, wouldn't it behave the same if you put the inductor in the
> primary side of the MOTs, reducing its inductance by the square of
> the turns ratio? (40:1 since it's a 4 MOT stack) The inductor current
> is discontinuous, so it shouldn't make much difference whether it's
> rectified or not.

it wouldn't behave the same - because in such case the inductor adds
the voltage not only to the capacitor, but on the primaries of mots,
so in my case they`ll see 411 amplitude voltage instead of 268! and in
case of full mains voltage there would be 480v, so - no, thanx :-D

> That might make it easier to find an inductor, but
> then again, having the inductor on the HV side protects the MOT stack
> from spark gap transients so it might be better.

the inductor can protect us only from the _current_ transients, not
voltage - isn`t it? anyway - do the mots need such protection? a
friend of mine has killed an nst, in the list  archives people are
killing nst very often, not a long before there was a topic "not again
- i killed  this puppy :-(" - i hate nst based psu and i want to make
my own one - more better.

"When running an NST through a Variac, the optimal phase setting of the
SRSG will vary depending on the Variac setting."

it`s crap.

"Additionally, NST's are not the well-behaved linear devices that
intuition and simulation models would have us think."

even greater crap. why people must fight with such crap? they deserve
better lot - MOT :-D