Computer particle simulation of high-voltage solar array arcing onset

Abstract

The operation of a high-voltage solar array in low Earth orbit may cause arcing on the negatively biased parts of a solar array. This sets a practical limit on the operational voltage of solar arrays.This paper is the extension of three earlier works 1 " 3 regarding high-voltage solar array arcing. The onset of arcing is reproduced by self-consistent computer simulations to verify the arcing onset model developed in the earlier work. It is shown that neutral gas is desorbed from the dielectric surface forming a localized neutral cloud over the surface, and the arcing onset occurs as the gas breakdown at a parameter pd (pressure times distance) much smaller than the Paschen minimum. Analytical expressions for the prebreakdown electron currents and the neutral densities are also derived and used to obtain a parametric formula of the breakdown condition. Arcing rates are calculated including the breakdown condition of the desorbed neutral gas. The theory is compared to the Japanese Space Flyer Unit High-Voltage Solar Array ground experiment and shown to give a reasonable explanation for data relating the arcing rate to the solar array temperature. A B.54 x 10-6 x 10 4 -52^1/2 /0 w , A/V 2 =6.53x 10 9 </>J; 5 , V/m -field enhancement factor induced by ion current = mean speed of gas, m/s = thickness of dielectric material d\ + d 2 , m = thickness of co verglass, m = thickness of adhesive, m = electric field, V/m = initial conductor (cathode) electric field, V/m -desorption energy, eV = electron incident energy, eV = electron incident energy for maximum secondary electron yield, eV = secondary electron emission energy, eV -electron incident energy to give secondary electron yield unity, eV = electric field parallel to dielectric side surface, V/m = electric field perpendicular to dielectric side surface, V/m = total current, A = current per unit length in the direction parallel to dielectric side surface, A/m -current per unit length in the direction perpendicular to dielectric side surface, A/m = incident electron current density to dielectric surface, A/m 2 = quasisteady incident electron current density to dielectric surface, A/m 2 = prebreakdown current density from conductor (cathode) before ionization begins, A/m 2 = enhanced field electron emission current density from conductor surface, A/m 2 = quasisteady current density due to negative space charge, A/m 2 = ion current density to conductor due to ionization of neutral, A/m 2 QEID SFNSreal S Ts V v a TSD y ee Tmax T chrg= desorption rate coefficient, s -1 = particle mass, kg = number of gas particles adsorbed per unit area, m~2 = adsorbed neutral surface density for monolayer coverage, m~2 = neutral density, m~3 = ambient neutral density, m~3 = critical neutral density for arcing onset, m~3 = pressure, Torr = particle charge, C = electron impact desorption cross section, m~2 = emission area at metal-dielectric interface, m 2 = emission area at dielectric-vacuum interface, m 2 = sticking probability of neutral particle to dielectric surface = temperature, K = ...

Ionization around a high-voltage body in magnetized nonflowing ionospheric plasma

“…The detail of this assumption is documented. 8 This treatment of q si is correct to simulate the steady state but incorrect to follow the unsteady development of ionizationdriven sheath expansion as studied in Refs. 4 and 5.…”

“…The details of the other collision cross-section data and their test against available swarm experiment data are discussed. 8 The MC-PIC code can reproduce the swarm experiment data for ionization coefficient and electron and ion drift velocities within the maximum error of 10% (Ref. 8).…”

“…8 The MC-PIC code can reproduce the swarm experiment data for ionization coefficient and electron and ion drift velocities within the maximum error of 10% (Ref. 8). …”

The role of water vapor and dissociative recombination processes in solar array arc initiation

“…The 1st model is commonly known as the Paschen Discharge Model (PDM). The second model takes into consideration the formation of a strong electric field between the metal, dielectric and plasma junctions [10,11]. The 2nd model has been coined the Triple Junction Model (TJM).…”

“…In this new model a field emission correction containing an artificially introduced field enhancement factor was set at approximately 1000 to 2000 [10][11][12][13][14]. The field enhancement factor seems unprovable.…”