I admire your rube goldberg contraption! Great resourcefulness! Another
thought :
Besides the leakage effect, there is the capacitance effect - which is
easier to experiment about.
The simple theory presumes that the globe is isolated from all else, so
that the charge Q is uniformly distributed on the spherical surface and
its capacitance is (rel to a second infinite sphere at infinity) C =
4Pi*eps*R. Its potential (rel to infinity) is then V=Q/C.
Now nearby conductors (especially grounded) will invalidate these
assumptions of isolation, will disturb the spherical symmetry of the
charge, and will increase the capacitance of your globe (providing a
nearby second electrode), so that it's potential will be lowered for a
given charge, allowing it to accumulate more charge before getting to the
potential of the belt, etc. You might experiment with different
placement of all things involved and the effect on the spark length
obtainable. Consider the room walls and other objects as contributing
second electrodes for this capacitor. In your present situation it may
well be that humidity leakage is the controlling factor, swamping out all
of these effects. (Can you move to an air-conditioned room?)
-Bob
P.S. It confuses my software when you change the subject line with each
post! It wants stable categories!
(One per thread) Hi! 73's :)