[Trade Journal]
Publication: Journal of the American Institute Of Electrical Engineers
New York, NY, United States
p. 1257-1258, col. 1-2
Test Results on the Performance of Suspension Insulators in Service
BY C. F. BENHAM
Associate, A. I. E. E. Great Western Power Co., San Francisco
Review of the Subject.—Megger test results on suspension insulators on the lines of the Great Western Power Company covering records since 1908 giving the percentage of depreciation by districts, viz., mountain, valley and coast conditions; also results covering the different types.
THE suspension insulator has been the subject of much discussion, particularly during the last decade, when the faults of most of the products of earlier manufacture were brought to light and the great increase in their use, due to the general tendency toward higher voltages, considerably extended the interest in their behavior.
Extensive investigation has been carried on as to the cause of their deterioration, which was so pronounced in the older units and from which the later output is not entirely free. Many valuable data have been collected, from observation of insulators in actual service,—and experimentally as well,—and many conclusions have been drawn. At the present time the most generally accepted theories attribute failure to improperly fired porcelain,—either porous from under-firing or brittle from overfiring,—and even more to the mechanical stresses set up by the unequal expansion of the various elements of the usual cap and pin type of unit.
Most of the manufacturers have profited from the light which has been thrown on the subject; by a combination of better porcelain through more scientific mixing and heat control of the firing and the elimination of a large part of the temperature stresses through better design and methods of assembly, they have produced an insulator which is fairly long lived, as compared to the article supplied fifteen years ago.
Records are available covering the performance of a fairly large number of suspension units in use in the high-tension lines of one company,—a majority of which have been in service for about fifteen years,—and most of the balance from eight to twelve years.
After about six years of operation, the frequency of insulator failures was increasing at an alarming rate, and interruptions to service became so numerous that it was necessary to adopt a system of field testing to eliminate the faulty units.
From the indications of the first testing it was deemed advisable to replace one type of insulator completely—as well as the defective units of all types—using a new insulator of improved design and material. Subsequently other types were replaced as a whole and all insulators in certain sections where the rapid depreciation coupled with the inaccessibility made testing very costly. As a result failures have been reduced to only one or two sporadic cases in a year, at the most, with one period of twenty months in which none occurred at all. Testing intervals have also been increased to a fairly reasonable period, varying from one year in a few sections to from two to three years for the majority. This is governed partly by the type of insulator preponderating, but more especially by the climatic conditions prevailing, which vary considerably over the entire system.
Periodic testing has been carried on for the past nine years, and an analysis of the test records, with reference to the effect of different climatic conditions and also comparing the behavior of various types, is presented graphically in Figs. 1 and 2.
Fig. 1 brings out the decided effect of temperature changes on the life of an insulator, and is rather conclusive proof that expansion stresses are a more serious problem than porosity. The four geographic divisions selected are quite sharply defined with distinct characteristics as to climate; while they vary greatly in extent, each is large enough to include a representative number of each of the various insulators and hence affords a fair comparison. The mountain and coast hill sections are subject to almost daily ranges of temperature of considerable magnitude, while in the upper and lower valley sections the only variations of any consequence are seasonal, though the daily range in the lower valley is perhaps a trifle greater on account of the winds. Again, the mountain and upper valley districts are comparatively dry, while the lower valley and coast sections are quite humid, the one being in a river delta and the other subject to fogs. Thus we have four distinct combinations of temperature cycles and humidity.
Fig. 2 covers the general performance of several different types of insulator, and is particularly interesting in the manner in which it brings out a comparison between different designs and stages of development of the same manufacture. The slight tendency of Curve B-1 to show a decrease in the deterioration can be accounted for in the fact that it includes insulators of two different ages. It has not been possible to segregate them generally in the test records, as they are alike in appearance, but a few specific cases indicate that the newer insulator has a much greater life due to improvements in porcelain and details of assembly. The proportion of older insulators is gradually being reduced through replacement, with a resulting decrease in the rate of depreciation of the combined group.
The most striking feature is the very slight deterioration of the insulator shown in Curve B-2, which represents a fairly recent output of a design which is standard at the present time. Actual figures are perhaps more conclusive. Of some 10,000 units subjected to test, no failures developed in six years, and only nine in the seventh year and 10 in the eighth year or a total of only 19 units out of 10,000 after eight years of service. This would rather indicate that the insulator question is no longer a very serious problem, at least under the conditions which prevail among those in the foregoing study.
Presented at the Pacific Coast Convention of the A. I. E. E., Del Monte, Cal., October 2-5, 1923.
