John touched on a key point here.When the gap quenches at 1st notch, all the cap energy was transferred to the secondary in the first half cycle and was dissipated in the secondary and in sparks. If the secondary and top terminal are not capable of dissipating all the energy in the secondary, the remaining energy will transfer back to the primary "while the gap is still conducting" during the second half cycle. It will again transfer back to the secondary and hopefully be dissipated. This will go on and on until the energy has finally dissipated allowing the gap to stop conducting or quench.
If the energy is quenched by the 1st primary notch, there is only a single ring up "time" for the spark channel itself. If we quench on the 2nd primary notch, the "time" of the secondary sparks are controlled by the ring up time between each spark and the energy bussed to the developed spark channels. It's not difficult to see that faster quench times may not always be preferred.
In second notch quenching, the secondary has rung up twice and produced a spark in each ring up and likely in the same channel as it is hot and ionized by the first ring up event. I expect the channel to be more conductive and possibly allow a little longer protrusion on the second ring up event.
But there is a limit. The second ring up has less energy than the first. I think the overall control mechanism is the air around a terminal (the medium the sparks are burning through) and the allowed energy in the spark channel controlled by the top terminal size and the energy transferred to the secondary.
When we increase pulse impedance by means of higher primary inductances, we are limiting the energy transfer time affecting the secondary's energy envelope over it's ring up, and in effect, helping to balance the secondary's ability to dissipate the energy.
Take care, Bart
The key to understanding this is that *all* the energy is always transferredby the first notch no matter what. But some of it comes back to the primary if quench does not occur. The spark streamer meanwhile utilizes the portion that it can during the first transfer. Thus a tight coupling gives longer sparks, regardless of how good the quench is (assuming no power arcing).
_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla