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

Re: THOR: First observations on streamer formation (try II)



Original poster: Bert Hickman <bert.hickman-at-aquila-dot-net> 

Marco,

Comments below...

Tesla list wrote:
>Original poster: Marco.Denicolai-at-tellabs-dot-com
>Hi Bert,
>  > Original poster: Bert Hickman <bert.hickman-at-aquila-dot-net>
><SNIP>
>  > Although it's not very surprising that you would see streamer
>  > growth during
>  > a single bang, especially if the initial corona breakout voltage is
>  > significantly below Vmax of your system, it is interesting
>  > that you saw
>  > forked and multiple streamers on a single bang.
>Yes, this was my point. Multiple streamers on a single bang.
>  >These are likely the
>  > effects of injected space charge from previous streamer(s)
>  > influencing the
>  > preferred direction taken during propagation of a new streamer.
>Yes for consecutive bangs, but not for one bang every (about) 5 seconds,
>which is my case.
>We just had 15 minutes time to observe this and then we had to get back
>to "family life". The setup has also been now temporarely dismantled as
>the university has to run some big HV transformer tests in between.
>A first-sight idea somebody of the university guys expressed was that
>the 15 cm streamer consumed so few energy that it left the toroid
>potential almost unchanged. So, stocastically speaking, the toroid
>"didn't notice" a streamer was born and the probability for any other
>area of it to start a new streamer was unchanged.

I agree with Antonio's comments. Suppose that initial corona and streamers 
initially occur at a terminal voltage that's significantly below the 
maximum voltage achieved during ringup. Under these conditions, there can 
be numerous opportunities for streamers and leaders to build off of their 
predecessors _during a single bang_.

Corona theory predicts that corona will initially form preferentially when 
the terminal is negatively polarized. However, streamer theory suggests 
that actual streamer growth will preferentially occur during increasingly 
positive portions of the waveform. Steamers and leaders grow in "jumps", 
each jump lasting 10's or hundreds of nanoseconds for small to medium-sized 
sparks. During each jump, an increment of charge is transferred from the 
toroid to the streamer. Since each jump causes a temporary reduction of 
toroid voltage at the time, the decrease voltage chokes off further 
streamer/leader growth. Because of this self-choking mechanism, the 
incremental charge loss is small relative to the total toroid charge, and 
larger (higher capacitance) toroids should create correspondingly longer 
jumps and hotter leaders. If the toroid voltage subsequently recovers and 
is then forced higher as part of ringup, further growth can occur from the 
tip of the existing leader.

Once the initial leader has formed, it should become the favored launching 
point for subsequent streamer and leader growth, since the leader "looks" 
much like a lossy wire tipped with a much smaller radius of curvature. And 
new streamers will head into a direction that is most opportune based upon 
the space charge left by their predecessor(s). One would expect the final 
path of the leader to be meandering and non-linear, and branched. Paper #12 
(in my earlier post) explains the forking mechanism.

The bottom line: a leader could grow, in multiple jumps, even within a 
single bang. And, we already know that they can grow from one bang to the 
another.

>  > Because there's significant charge transferred during each "step" of
>  > streamer growth, it should be possible to indirectly measure this via
>  > resonator base current.
>I once measured the secondary base current as rising up to 16A. During
>those 15 minutes we also found time to speculate about this. We thought
>that current measured though a grounded stick nearby could provide
>(maybe?) a better measurement as the sensibility could be tweaked more.
>The stick was not supposed to be reached by the streamer.

See paper #7 for a similar idea.

>  >Following are a dozen good references
>  > covering
>  > various aspects of corona and streamer formation, modeling, and
>  > theory/measurement. Paper #7 below discusses an alternative
>  >
>Thanks a lot. Next week I'll have to pay a library-day to dig for those.

>Best Regards

Great! Good luck and looking forward to further measurements.

Best regards,

-- Bert --
--