I'd expect quite the contrary: the gap voltage dropping to the point where the arc extinguishes. Not to zero. (Wire) inductance will result in lower voltages than defined by the gap width.

Thus the gap width will primarily define the remaining voltage. If you have a flexible wire geometry this will influence the remaining voltage - depending on current wire inductance - resulting in voltage variations.

AND: the gap voltage is also influenced by tool wear!

Hi, look for literature on a effect called "dialectric absorption", this could cause a non-zero voltage even after a cap gets shorted (after removing the short, the voltage will grow up a bit like from 0V to 1% of original voltage).

Before anything discussed in the result in; I would first check my measuring circuit. What Probs was measured, is the DOS calibrated properly, I have a fixed or unstable ground point, you experience additional parasitic capacitances and inductances.

Since I can never be measured directly on the electrode surfaces, needs are regularly checked, how big are the parasitic elements to the contact. The right choice of parameters resolution can be important.

Best regards

Hans-Peter Schulze