RE: THOR Bang energy vs. streamer length measured

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Denicolai, Marco" <Marco.Denicolai-at-tellabs-dot-com> 

Hi Gerry,

 > Now I'm confused,  if the bang energy is constant and BPS is
 > varied, then the power draw will vary

yes

 > and spark length will vary.

Not necessarely. Look, this is just what we want to measure below 100 Hz
and what I have already measured from 200 to 400 Hz.

If we were warming a water pot with those bangs, your argument would
hold right. But we aren't. IMHO there are a number of mechanism (e.g.
static charge formation) that are NOT linearly dependant on the BPS.
Doubling the power fed you won't double their effect. Therefore, you
don't have to halve anything to compensate for the BPS doubling (or the
other way round, as you wish).

 >I'm not sure what this proves other than there is an
 > optimum BPS per bang energy.

I believe there is a MINIMUM BPS to achieve the top streamer length (for
the SAME primary capacitor voltage, i.e. bang energy). The rise in
length is very steep vs. the BPS. This minimum BPS ought to be below 200
Hz (IMHO). Above this BPS value the streamers just grow intensity (more
"fat") or happen slightly more often.
Now you can call that minimum BPS value an optimum. It is an optimum as
it maximizes the length for the minimal average power fed.

 > If one has a fixed power source, it seems that one want to
 > find the optimum BPS for that source which is a different experiment?

No, it is a subproblem.
The fact is simply that if you want to detect the influence of factors A
and B on result C, it doesn't help to variate both A and B and measure
C. You can't discern which of A or B affected C's change. There is a
whole discipline called "experimental design" devoted to this problem.
In a nutshell, following the KISS principle, change A keeping B
constant, measure C. Then change B keeping A constant, measure C.

After collecting this family of measurement you can combine them as C =
f(A,B). From that you can transform into your subproblem of maximizing C
(the length) vs. A*B (the product of energy and BPS, the average power).

Best
---------------------------------