A kickstart is needed to get the Ammann device started.

I forgot to mention that it may be necessary to raise the amplitude of the snap charge of voltage delivered to the Leyden jar (or, capacitor) to be the minimum amount necessary to get the noble gases within the tubes and spheres to become a plasma and ignite, such as: 60 to 90 volts for neon gas. Previously, I was only considering the tubes and spheres as filled with air and so I just assumed that a small charge delivered to the Leyden jar was all that was necessary to get things going. But if they are filled with a noble gas instead of air, then this is an indication that ionization short of arcing may not be enough to run the device. It may be necessary for the noble gases inside of the tubes and spheres to be in a state of plasma and arcing throughout their interior?

An air-based spark gap has a voltage breakdown of around 1kV so the gap or space between the spheres will still continue to be a non-plasma and in a state of a pre-arcing ionization even though the interior of the tubes and spheres are a plasma and are busy arcing across the helium or neon noble gases which have a voltage breakdown of approximately 10 to 20 times less than the atmosphere (45V for helium).

This more recent development of conceptualization is significant because it creates two types of spark gaps within the circuit. One type is arcing and the other is not. And the type that is arcing is giving us negative resistance from inside of the tubes and spheres while the other type is not arcing and not in a state of a plasma but is merely ionizing in a state of pre-arcing in the space between the spheres.

This space between the spheres, in some respects, is the most significant of both types of spark gap activity due to its ability to boost the generation or amplification of reactive power within circuit components outside of itself without being powerful itself.