Sparking potentials and ionisation coefficients in SF6

1981 annual report :

“…For highly nonuniform fields such as used in the measurements reported here, there are likely to be additional large uncertai nties in calculated avalanche size due to possible large uncertainties in the values of a-j at high E/N. This is demonstrated in figure 33 where the data of Mailer and Naidu [38] (also see Sec. II.…”

“…The expressions used for a-j and n a were obtained from polynomial fits to the data of Mailer and Naidu [38] and are given by .131 x IQ" 9 (E/P) 3 ]P, ( 29 ) for E/P > 117, 52 Figure 32.…”

“…Consistent with eq (36), we can also write dn_ dn _ o dt^o^dt (38) where the first term on the right gives the rate at which ions enter $j and the second term the rate at which they leave.…”

“…Thus 5^i n eq (39) is expressed as a function of position V . From eqs (38), (39), and j(a).da = dn_(a)/dt, we can write j £ (a).da where S 0 represents the surface of the initiation volume and £(a) is the maximum path length along a field line that an ion could travel through the initiation volume if it enters at a particular surface area element da. The above equation is approximate in the sense that we have neglected ion diffusion.…”

“…Mailer and Naidu [38] also extended the steady-state Townsend technique. In 1975 they reported values of a-j and n a for mixtures of SFg with other gases.…”