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

Re: The Freau number...



Original poster: Jimmy Hynes <jphynes@xxxxxxxxx>

A few things I forgot to mention...

On the topic of marx generators, I remember reading that the huge russian one that did 100m sparks actually had a rise time of a couple milliseconds, for best spark length. Their problem was forming multiple streamers. Unless that is out our time range, it could make things more difficult than a simple linear function =\

I am also willing to do some tests like the one I suggested once I get my little DRSSTC going. It should be done really soon, but you never know..

On 6/18/05, Jimmy Hynes <<mailto:jphynes@xxxxxxxxx>jphynes@xxxxxxxxx> wrote:
Hey all,

If you guys are going through all the work to factor in streamer formation time (which is interesting, and very important), it seems like you would like to refine the equation in other ways as well.

The main problem I have with it is BPS. Everyone knows that a coil running 500bps isn't going to live up to the expectations. It might be tough to get an accurate prediction of how the BPS would affect spark length, especially since the 'corner frequency' when the sparks start growing varies from coil to coil. The 'spark ceiling' (if it exists, and I believe it does) would also have to be predicted for each coil. Maybe you could do this based on the Fres and predicted single shot length of the coil. Doesn't sound too simple, but neither is predicting the streamer growth time's effect.

For finding the effect of streamer growth time, I'd do it single shot. It will be easier to compare coil to coil, since 100bs on one coil might mean something different to 100bs on another. For a single coil, the test is easy (Steve has done very similar tests on his smaller coil). You just have to vary the on time, and voltage to keep same bang energy with different times, and measure the max length. This test could be done single shot, or at some fixed BPS, since it's the same coil.

Steve's test was fixing the spark length, and finding the voltage needed to reach the target with different on times. He basically found that shorter is better, but check his site for details.

<http://www.stevehv.4hv.org/DRSSTC3.htm>http://www.stevehv.4hv.org/DRSSTC3.htm under UPDATE: 1/22/05

Steve and I have both done tests to find the spark lenth versus input voltage curve. My coil showed a fairly linear response, and Steve's fit a line really well. This would suggest that the 0.5 power is close enough, just varying as the load varies.



On 6/18/05, Tesla list <<mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx> wrote:
Original poster: Terry Fritz <<mailto:teslalist@xxxxxxxxxxxxxxxxxxxxxxx> teslalist@xxxxxxxxxxxxxxxxxxxxxxx >


Hi Steve,

I think it will always have to have a dimension.  It will always be some
length/some energy term...

We could easily write it as:

Spark length (meters) = F x (bang energy x BPS)^0.5

That way we can change the "0.5" if needed.  So,:

F = (Spark length (meters)) / ((bang energy x BPS)^0.5)

The dimension really gets odd if we us 0.6...

I liked Freau number since I never can spell "coefficient" ;-)  There is
also "Freau Factor"....

Cheers,

        Terry




At 05:03 AM 6/18/2005, you wrote:

>>        Spark length (meters) = F x SQRT(bang energy x BPS)
>>
>>"F" would stand for "Freau" so John will be eternally famous!!  I would
>>suggest going to meters just to keep things "scientific".
>
>
>I agree that "F" is a handy way of measuring coil performance, a bit like
>the specific impulse for a rocket or such like. But I wonder if there is a
>way to make it into a proper dimensionless number. Otherwise you have to
>call it the "Freau coefficient" with its rather unwieldy dimensions of
>"metres per root watt".
>
>Also I remember Nick Field mentioning that the law may not be quite a
>square root. Square rooting is raising to the power 0.5, and he suggested
>the exponent might actually be nearer 0.6 or something.
>
>I suggest the "Watts number" (streamer length divided by resonator length)
>as another handy parameter. Some day I ought to try making a graph of F
>vs. Wa for a coil and see if it throws up any interesting results.
>
>Steve Conner