Re: [TCML] tank capacitor help

[Gary Lau <glau1024@xxxxxxxxx>]

Hi James,

When you say that a 15/60 NST "needs" a 10.6nF cap, let's just be clear that
this is not really a requirement for anything.  The 10.6nF value is the cap
that would resonate with the NST secondary winding at 60Hz, but this is not
a necessary condition for a TC to work, in the sense that the primary tank
circuit resonant frequency MUST match the secondary resonant frequency.  It
was once believed that the most efficient transfer of power through the NST
(i.e. you'll pull and process the most power from the wall) occurs when the
cap is selected to be mains-resonant with the NST, but the normal practice
now is to use a cap that is Larger Than the mains-Resonant value, or LTR.
One benefit of deliberately avoiding a mains-resonant (MR) value is that the
likelihood of having NST damage from using a too-wide static gap is
minimized.

Last year I performed a novel experiment to determine how much power was
processed using a wide range of cap values, for several different 15/30 and
15/60 NST's.  I was surprised to learn that the cap value that resulted in
the most power being processed, was actually closer to the MR value than the
LTR value.  But it's important to understand that the experiment measured
only how much power would be processed by a static gap into an incandescent
lamp dummy load.  How that correlates to spark length may be another
matter.  It is known that bang size (i.e. cap size) correlates with spark
length, but of course, one can't arbitrarily increase cap size beyond the
LTR value and ignore the fact that less power is being processed.  The
complete report is here:
https://docs.google.com/viewer?a=v&pid=explorer&chrome=true&srcid=0B0afsm3K5nFYNGQ3MmJlZGEtYjQ5My00OTJhLTkwNDMtOWEwMjRkYWJkZjAy&hl=en

The only indisputable finding is that if you can, run the NST with a Variac
that is capable of boosting the voltage up to 140V.  The boost in processed
power is very significant.

So in summary - the "old" advice was to use the MR cap value.  The accepted
practice for about the last decade has been to use the LTR (about 1.5X the
MR value).  My study suggested that for maximizing processed power, a value
closer to the MR value _might_ be best, but this ignores the safety benefit
of using LTR, and doesn't address whether spark performance might be
different.

And whatever you use, don't sweat the value too much.  There's not a narrow
sweet-spot that you'll get greatly diminished performance if you deviate
from it.  As long as it's in tune.

The cap you asked about from Allied - the CDE940 - is similar to the CDE942
series that is most commonly used (but mure expensive and harder to find)
in MMC's, but it may fail if pushed hard with a lot of current.  It's
difficult to generalize whether it will or won't fail.  But I would ask you
to consider why you chose the .047uF value, when these caps are also
available with values of 0.1 and 0.15uF.  Much more bang for the buck and
fewer devices to assemble with the higher values.

Regards, Gary Lau
MA, USA


On Thu, May 19, 2011 at 4:57 PM, James Hutton
<b-u-r-t-o-n-boy@xxxxxxxxxxx>wrote:

>
> hello.I have heard that raising the tank capacitor value actually works
> better then the capacitor match.That if i have a 15kv 60ma transformer,
> (needing 10.6 nf) i should multiply this value by 1.5,giving me 15.9nf. can
> someone please confirm this?also, if someone could confirm that these caps
> will be suitable, that would be great :
> http://www.alliedelec.com/search/productdetail.aspx?SKU=8620508
> thanks!
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla