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

Re: Streamer length vs. power delivery time...



Original poster: Steve Ward <steve.ward@xxxxxxxxx>

Hi Terry,

I see you have been busy!

On 6/19/05, Tesla list <tesla@xxxxxxxxxx> wrote:
> Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>
>
> Hi again Steve ;-)
>
> I made this file of what I think your coil is doing:
>
> http://drsstc.com/~terrell/modeling/StevesCoil85uS.pdf

Yes, that is finally the correct current profile!  It just rings past
zero maybe once and then its dead.  The rise looks about right too, it
should be slightly flatter at the top... heck, i will just have to
take a picture of it next time (why didnt i think of that last
night!).

I found i had to set Rprimary to .15 ohms in pspice for it to match my
primary current at 700A.  using .5 ohms lowers Ip to 500A, though the
primary current rise profile is correct (it just rings out for a long
time afterward).  That long ring out is very important because it
means more time for the secondary to pull out energy (in the model).
That means that (bang energy constant) the streamer *power* is much
lower in my simulation where the primary continues to ring for a long
time.  In real life the ringing (at high power only) is gone in 1
cycle or so, indicating that the streamer has used it all up.  At low
power, the coil rings down for about as long as it rings up (110uS
both ways).

>
> 700 Primary amps peak. Fast Ipri shutdown and all....
>
>  From page four...
>
> 701 amps peak
> ICPriRMS = 3.4 * sqrt (120) = 37.25 amps RMS
> VCpriPeak = 3247V
> 8.09 joules/bang x 120 = 971 watts
> Load Energy = 2.02 * 120 = 242watts
> Rise time = 70.5uS
>
> Does this all seem about right?

I think you need to increase my toroid to about 42pF to get the right
operating conditions.  This requires Lp to be 15uH.  My operating
frequency is right at 56khz or so on the primary side so that should
be nearly 110uS burst length.  I think that will make the bang energy
larger.

>
> The program says your Lp = 10.27uH.  If it were 12.1uH, the program says
> (last page):
>
> 401 amps peak
> ICPriRMS = 2.033* sqrt (120) = 22.27 amps RMS
> VCpriPeak = 2049V
> 6.15 joules/bang x 120 = 738 watts
> Load Energy = 3.54 * 120 = 425 watts
> Rise time = 86.7uS
>
> So going from Lp = 10.27 to Lp = 12.1uH would increase streamer power 75%
> (spark length 32% further) while reducing all the input levels....  Rise
> time would increase 23%....

I do notice that tuning is pretty picky.  Maybe i still dont have it
tuned for best sparks yet ;-).


I will have to work on my simulation some more to see if i can get the current to head to zero faster. I altered my "gate driver" in pspice so that all switches would be OFF at the end and the diodes would have to take care of the rest, but it didnt change anything (somewhat expected). I think i will be running the coil again tonight, if so i will snap some scope shots at various power levels.

Steve

>
> Cheers,
>
>          Terry