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

NST/cap sizes - was RE: Coil Woes



Original poster: "Lau, Gary" <gary.lau-at-hp-dot-com> 

I don't recall if I ever posted the estimated voltages from my
experiment.  If I assumed that my 15/60 NST puts out an open-circuit
peak voltage of 15kV * 1.414 with 120VAC in, the estimated voltages are:

C	V-Bang		120*.5*C*Vbang*Vbang
.02uF	24.4kV		714W
.03uF	21.9kV		863W
.04uF	19.6kV		922W
.05uF	17.1kV		877W

Note that with LTR caps, Vbang is considerably less than the peak cap
charging voltage.  For the .04uf cap, the peak voltage was about 22.2kV.

I'm not sure that our NST simulation models accurately predict what
happens when the mains voltage is raised above nominal and the magnetics
become non-linear.

Regards, Gary Lau
MA, USA



Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net>

Hi John,

That could very well be.  Unless I'm misinterperating the post, Gary's
4.0*Cres was the results of an experiment with a dummy load.  I'm just
trying to get closure on results in a running SRSG system and if the 4.0
Cres continued to hold up.  I'm most interested in the bang voltage of
the
15/60 system with 4.0 Cres.

Gerry R


  > Original poster: FutureT-at-aol-dot-com
  >
  > In a message dated 10/12/04 11:59:27 AM Eastern Daylight Time,
  > tesla-at-pupman-dot-com writes:
  >
  > >Hi Gary,
  > >
  > >I remember your post.  Since Cres for your 15/60 NST is 10.6 nf
(call it
  > >0.01uf), your experiment would correspond to ~ 1 to 5 * Cres.  Im
wondering
  > >if the dummy load is affecting the results.  My computer simulations
show
at
  > >values above 3*Cres, the inductive kick can not get things fully
charged.
  > >My actual measurements were between 1.6 and 2.5 * Cres where I
scoped the
  > >primary durning actual SRSG operation.  At 2.5 * Cres, I was drawing
1080
  > >watts from the power cord and charging to 24KV at the time of bang
(900
  > >watts thru the gap).  I haven't gone larger than 2.5 * Cres cause I
ran
out
  > >of cap.  So far the measured results agree with the simulation so Im
  > >wondering if 4 * Cres would hold up during real operation.
  > >
  > >Gerry R.
  >
  >
  > Gerry,
  >
  > It think possibly the secret is that Gary's NST is robust
  > and may deliver more current than expected.  This may be
  > partly due to the use of 140volts input voltage.   It seems
  > that all NST's are not the same regarding their current
  > capability.
  >
  > Cheers,
  > John
  >
  >