1920.] Birks and Ferguson.—High-tension Insulators. 179 
Year. 
Breakdowns. 
Insulators replaced. 
Time lost. 
1915 .. 
12 
34 
24 minutes. 
1916 .. 
15 
50 
53 
1917 .. 
25 
85 
72 
1918 .. 
26 
96 
lot „ 
Totals .. 
78 
265 
159J minutes. 
The transmission-lines are each approximately sixty-three miles in 
length, and a total of about 5,200 insulators of the pin type shown in 
fig. 1 are in use on both lines, so that during the four years of operation a 
little over 5 per cent, of the insulators broke down in service. A further 
1*5 per cent, have been removed under suspicion following night observations, 
when they were observed to glow or “ crackle ” unduly. 
The bulk of the replaced insulators were taken from the substation 
“ third ” of the lines. This section is most subject to mists and fogs, and 
it is possible that some of the latter are slightly laden with salt swept 
inland from Lake Ellesmere—a large salt-water lagoon about ten miles 
from the lines. It suggests itself that these fogs are a possible factor in 
the breakdown of the insulators, and no doubt they contribute to some 
extent to the failure at least of partially-broken-down units by lowering 
the flash-over voltage. 
It was obvious soon after regular operation commenced that break¬ 
downs were in certain circumstances directly associated with atmospheric 
conditions. If a line that had been dead for some hours were subjected 
to pressure even raised from zero, under certain conditions of mist, it would 
almost invariably break down, although under the same conditions the 
other line kept continuously alive showed no trouble. Evidently the 
humidity of the atmosphere is an important factor, and experiments were 
made to determine the effect of misty conditions. 
An insulator was placed in a wooden box into which water-vapour was 
introduced by means of a pipe leading from a small boiler, thus raising 
the humidity of the air inside the box. The insulator used in these tests 
was first thoroughly tested at 50 cycles and at high frequency, and its ohmic 
resistance tested “ infinity ” with a 1,000-volt megger. Curves No. 2, 
fig. 2, show the result of a test made by connecting the megger across the 
top shell, neck groove to joint, and increasing the humidity slowly. The 
humidity was measured by means of wet- and dry-bulb thermometers inside 
the box. Curve No. 1 is a repetition of the above test but with the megger 
leads connected from neck groove to the first cement joint. These tests 
were repeated with the insulator-surfaces dusted with road-dust, and the 
surface resistance was about the same as for the moisture coating alone, 
as shown in fig. 2. The curves are interesting as showing the decrease 
in surface resistance when the surfaces are exposed to a humid atmosphere. 
The flash-over voltage is very much reduced under these conditions, and 
it thus appears probable that an insulator which is already broken down 
in one or two shells may arc-over during a fog with a moderate rise in 
voltage on the line, although it is capable of standing up to the pressure 
without trouble under better atmospheric conditions. 
