The Effect of Surface Charges in Vacuum Discharge Tubes

Abstract

Current-voltage curves obtained with a four electrode tube with two rather widely separated gauzes, showed discontinuities at voltages which were evidently not critical potentials since they varied with the gas pressure. Since charges on the glass surface were thought to be responsible, a simple case was studied. Electrons were accelerated from a filament through two gauzes, 3.3 cm apart and held at the same potential, to a plate, the space between the gauzes being surrounded by a metal cylinder $C$ insulated from both. This cylinder became negatively charged until the loss or leak just equaled the gain due to the faster electrons. As the accelerating voltage $A_1$ was increased, a value $b$ was reached at which both the voltage of $C$ and the current suddenly increased greatly, and these larger values were maintained until $A_1$ was decreased to a limit $a$, considerably lower than $b$. A study of the equipotential surfaces gives an explanation of these effects. The discontinuities may be eliminated by controlling the voltage of the cylinder $C$. In experiments in which electron velocities must be known, it is evidently necessary that insulated "floating" surfaces be avoided and that the potential of all parts of the wall be controlled.