I believe the reason humidity affects van de Graaff generators is not
because of air discharge, but because of surface discharge along the
insulating column. Supposedly the dielectric strength of moist air differs
little from dry air. But high humidity environments can increase the level
of "surface water" on the insulator, which in conjunction with other dirt,
fingerprints, etc. can increase the conductivity of the insulator.
We have experienced the same problems Tom McCarthy mentioned during our
high-humidity season (spring), and we have had some success in the following
manner. (Note, our demo van de Graaff has a Plexiglas tube as its
insulating column.)
(1) Wipe off the tube (inside and out) with clean paper towels or cotton
rags using a bit of isopropyl alcohol. This will clean the tube of
fingerprints and other grime, and remove some of the water.
(2) Dry the tube with a heat gun (inside and out) with low enough heat
setting that you don't do any damage to the plastic parts. A hair dryer
works. The goal is to get the whole tube 20-30 degrees hotter than room
temperature. At the elevated temperature the water should evaporate and it
should not re-condense until the tube cools back down.
(3) The rubber belt should be heated also (and will be if you aim the heat
gun down the tube). We typically do not have to clean the belt with
alcohol, although I suppose you could. Too much cleaning with alcohol could
tend to remove any plasticizers present and could cause the belt to get
harder.
(4) Once the tube is clean, try not to touch it with hands or other things
that will get it dirty again. If it stays clean, the next time you use it
you can just use the heat gun and skip the cleaning step.
Michael D. Edmiston, Ph.D. Phone/voice-mail: 419-358-3270
Professor of Chemistry & Physics FAX: 419-358-3323
Chairman, Science Department E-Mail edmiston@bluffton.edu
Bluffton College
280 West College Avenue
Bluffton, OH 45817