Re: THOR Bang energy vs. streamer length measured

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

Original poster: Terry Fritz <teslalist-at-twfpowerelectronics-dot-com>

Hi Marco,

At 04:05 AM 7/26/2004, you wrote:
>Hello all,
>
>I mailed already once this on July 5th. I had almost no feedback. Was it
>because the most of you was on vacation or because this material is
>simply not for you (don't care, too difficult, useless, just crap,
>etc.)?
>To get an answer to this question I'm posting it again here below.

Vacations, summer, and it takes a long time to absorb all the data...


>*************************************
>
>I have completed the first set of measurements on Thor and I was
>thinking to share with you my findings.
>
>I have changed the bang energy (primary capacitor voltage), the RSG
>rotating rate and the grounded target distance. For each position I have
>measured "how well" the target was reached. In particular I was able to:
>
>- document how the streamer grows length from bang to bang
>- model the hit probability with a Weibull distribution
>- show that a change in the RSG rotating rate DOESN'T influence the
>streamer length
>
>Read the whole story (includes diagrams and data) at:
>
>http://www.iki.fi/dncmrc/meas/performance.htm
>
>My "performance" measurement method offers a very good repeatibility and
>can be easily used with any SSTC. It is very easy to test a supposed
>performance improvements by using it.
>I hope to receive some feedback also on it, considering also the recent
>debate about the "energy and power" stuff.
>
>Suggestions, questions, comments and corrections are all welcome.
>
>Best Regards

Your data looks extremely good and is probably the best look we have into 
these issues by far!!

The first set of results regarding strike distance Vs. number of bangs, is 
comfortably what we expect to finally see someday.  Short distances are 
overwhelmed by the high voltages in comparison to the distance and few 
bangs are needed.

However, as the distance increases, streamer growth becomes extremely 
dependant on previous bangs and ionized path formation.  One thing that is 
interesting from the Weibull curve fit is that distance becomes fairly 
proportional to the number of bangs.  There are definite good and bad 
places to be on those charts.  Even though the number of available bangs is 
limitless, the statistical chance of the streamer failing increases over 
time until it becomes improbable that the streamer will be able to connect.

Although we do not have the specific numbers and equations to predict such 
things with arbitrary coils now, your data definitely shows such relations 
do exist and they are not terribly complex.

Your BPS data also suggest that there is a "right" BPS.  The 10.7J data 
seems to suggest that ~325 BPS was optimal.  It also shows that the optimal 
BPS is not far from what we use now, although it might be ~~2X what we in 
the 120BPS world are used to.  I suppose we could start thinking about 
optimal capacitor/charging/gap systems for 240 and maybe 360 BPS sync 
operation...  A daunting task indeed!!

>P.S: Coming next -> New measurements with RSG gap amount reduced from 4
>to 2. Any improvements? We'll see...

I think the Hit% Vs. BPS data is the most intriguing and something we can 
address fairly quickly.  Your data suggest that 120 BPS is too low and 
could greatly hinder streamer distance.  If that is true, we can fix that...

Cheers,

         Terry



>**************************************
>
>Regards