[Trade Journal]
Publication: Proceedings of the National Electric Light Association 43rd Convention
New York, NY, United States
p. 664-667, col. 1-2
Insulator Maintenance
The investigation conducted by the Sub-Committee on Insulator Maintenance was designed to reveal:
Methods in use for maintaining line insulators. Difficulties of all kinds that have been encountered and their nature, particularly those pointing toward deterioration of insulators in service. If possible, the revelation of some of the causes of insulator deterioration.
The establishment of a continuing work on the part of the Overhead Systems Committee that will eventually lead to the accumulation of data on which the solution of the insulator deterioration problem can be based.
It is fully realized that many of the insulator troubles of the past have been of the same nature as those experienced in the development of any industry, viz.: those due to lack of knowledge of conditions, both in the manufacture of the product and in the service demanded of that product. The Committee further realizes that a vast amount of intelligent effort has been expended by manufacturers and users in investigation of these difficulties, which has resulted in marked improvement, both from the design and manufacturing standpoints. Thus the evidence appears to warrant more or less definitely the belief that service can be enhanced through the intelligent application of the designs now available to the specific needs. On the other hand, the Committee is not in a position to announce that a complete solution of the problem is imminent. It is felt that the solution has been materially delayed for lack of an agency through which all experience could be brought to a common point, permitting of impartial study and the development of the problems in all their phases, in such a way as to indicate the lines along which the work of solution should be conducted. This can be done only over a period of years by careful and consistent work, as insulator faults for any design properly applied to service do not develop fully until after several years' use.
As a basis from which to work, a questionnaire was sent out covering the basic data which it was thought must be secured to permit of an intelligent start. This questionnaire covered both suspension and pin type insulators, the lowest voltage on which data was requested being 11 kv. No attempt will be made in this report to take up each individual question, as some of the data were confidential, being requested only for the information of the Committee. Therefore, only the relevant questions will be taken up.
Suspension Insulators
Steel Towers.
Twenty-six companies submitted data on the following mileage of circuit.
On this mileage, approximately 694,000 suspension units were reported.
(b) Wood Poles.
Fifteen companies reported on the following circuit mileage.
Approximately 219,000 suspension units were reported on this mileage.
The question as to the number of units per string used for both suspension and tension units revealed a very decided lack of uniformity in practice. The information received is as follows:
The figures given are for the units in a single string. Many cases are reported in which multiple strings are used for reasons of mechanical strength. The figures give only the extremes reported. In the cases of 66 kv. and 110 kv. lines, the data are from a sufficient number of companies to indicate a general trend toward certain numbers of units per assembly if such a trend exists. Such a trend does not appear in the case of 66 kv. steel tower lines, though there is a tendency toward the use of the larger number of units per assembly on the theory that the provision of a larger initial factor of safety makes for a slower revelation of insulator depreciation. As examples, one very important 66 kv. steel tower line in the east uses eight units in suspension and nine in tension assemblies. A New England steel tower line uses six and seven units respectively, while a middle western steel tower line uses six units in both suspension and tension assemblies. For 66 kv. wood pole lines, the data as reported indicate a marked tendency toward the use of four units in suspension and five in tension assemblies. This tendency is particularly marked because the companies indicating the use of these assemblies are scattered all over the country. For 110 kv. steel tower lines, there is a marked tendency to use seven units for suspension and eight for tension assemblies. This is not consistent with the 66 kv. practice on important lines as revealed in the answers. This is due to the tendency toward the use of a larger number of units on 66 kv. lines appearing on the lines of recent construction, while all except two of the 110 kv. lines reported on are older developments. The latest 110 kv. steel tower line which was reported employs nine units in both the suspension and tension assemblies. One 110 kv. pole line recently constructed reports eight units in suspension and ten in tension assemblies. This line was constructed in 1918 and exhibits the tendency toward a larger number of units per assembly.
The question as to when deterioration was first observed produced thirty-three answers, of which five indicated that none had been observed, one that the line was not old enough to show deterioration and the remainder in periods ranging up to seven years. The majority indicated such observation from one to three years.
Methods of test brought thirty-three answers, of which seventeen indicated the use of the meg-ger, seven the buzz stick, one the oscillator, six no method at all, one visual inspection only and one a special device of its own.
The frequency of tests question was answered by twenty-five, of which eleven indicated yearly tests, two every six months, two every two years and the remainder either none or irregular.
The descriptions given as to how deterioration takes place indicated beliefs in the following causes, which however are not necessarily listed in the order of their present importance.
Porosity.
Minute cracks and other faults initially present which permit subsequent absorption of moisture.
Troubles from expansion resulting from dissimilarity in coefficients of temperature expansion of porcelain, cement, and hardware; and possibly aggravated by expansion of the cement in presence of moisture.
Of the three causes given, the first two are a problem in porcelain manufacture, while the third is a matter of finding a method whereby the conditions producing destructive mechanical strains can be avoided. Methods of mitigation of these mechanical strains have been widely discussed in papers before the various technical bodies and need not be referred to here. The evidence so far submitted does not indicate that the problem has been entirely solved. It has been felt that the difficulties due to porosity have occurred in insulators manufactured more than five years ago, and that the insulator manufacturers are now producing porcelain from which porosity need not ordinarily be greatly feared in the future. The questionnaire answers do not prove this conclusively, though a study of the data submitted, together with the comments seems to support such a view. The situation should be watched for some time, particularly on insulators of recent manufacture, before such a conclusion is reached.
The general feeling, expressed by the answers, as to the remedies for the difficulties encountered, is that the insulator manufacturers have made material progress in the design and manufacture of insulators in the past few years that will to some extent eliminate the extreme cases of difficulty occurring with the earlier designs. While these expressions are quite encouraging, it is felt by the Committee that there is still a lack of certainty for two reasons, the first being lack of sufficient time in service for the later designs to develop their weak points, and the second the lack of any method whereby the data gathered by companies all over the country can be brought together and intelligently analyzed.
The attempt in the past has been to analyze the troubles encountered by each individual company, and while important results have undoubtedly been produced, the analyses have not been sufficiently broad to lead engineers in touch with the situation to feel that the problem has been fully developed in all its phases. This situation is particularly emphasized in the question as to whether heavy mechanical strains influence insulator deterioration. Nine replies to the questionnaires answer the question very decidedly in the affirmative, while eleven return as decided a negative. Three state that the data available do not prove either conclusion. One western company states that tests on 75,000 units failed to show that dead end insulators were more liable to failure than suspension units. Another states that dead end insulators seem to be failing more rapidly than suspension insulators, but that no relation be-tween the mechanical loads and rates of failure can be traced. An eastern company operating approximately two hundred miles of 66 kv. circuit ascribes the noticeable increase in failures among tension insulators to more direct exposure of tension units, because of their position in service, to rapid temperature changes. Another suggestion is that the troubles supposed to be due to high loads may be caused by vibration and shock. The Committee has not been able to learn of any experiments with springs or cushioning devices at dead end points to determine whether the loss of insulators can be reduced. The suggestion offers a field of investigation that seems worthy of attention.
Under the heading of methods employed to keep insulators free from deposits in the vicinity of cement mills or other plants whose products tend to form a coating on surfaces, two suggestions were made.
Wash occasionally, one company suggesting the use of hydrochloric acid.
Wipe with an oily cloth.
The use of an oily cloth for wiping is not suggested as a cure, but as a means of making the deposit easier to remove when washing is undertaken. The answers in several cases suggest the use of a larger number of suspension units or higher rated pin type insulators where deposits are to be expected. A middle western company submits the following:
Insulators are removed from the line, brought into the storeroom and put to soak in a solution containing one pound of nitric acid to three gallons of water. Insulators are left in this solution for one-half hour. The cement and acid are then washed off with a ten per cent solution of ammonia water, using powdered pumice stone and a stiff brush to assist. We find that this method restores the glaze and we have put the insulators back in service without trouble.
In certain instances larger insulators have been used to permit of longer intervals between times of cleaning; likewise some success has attended the use of a special design with large umbrella-like upper petticoat which tends to shield the lower sections.
Under the heading of punctures due to lightning or surges, the answers received indicate that, while present, the cases are so comparatively few they have not left an impression of being of serious moment.
It is felt that the Committee's work for this year has been only a start in the study of this problem. Owing to the fact that the Association committee work does not start actively until September and must close in February with reports ready for printing, the time has been too short to permit of following this subject as it must be followed. The records of various companies are each kept in their own way and to meet their own needs. The questionnaire answers have furnished a splendid clue to the companies which have kept real records, and several of the questionnaires have furnished data that are carefully prepared and valuable. It has not, however, been possible to develop from these data any information of broad enough scope to include in this year's report. A form is shown which the Committee hopes will be made the basis of records that may be analyzed later so that more valuable data bearing on the problem may be developed. The lack of uniform records, or of records of any kind, is just now a bar to effective analysis of the data so far submitted.
The following information is given concerning the make-up of the form and, after having been in touch with a number of companies, the Committee believes that this record can be used by most companies operating high voltage lines. The data which would thus be made available are necessary if the general study of insulator performance is to be properly continued. It is hoped that the adoption of this record will be found practicable and that a great many companies will be ready to contribute such information to succeeding committees and possibly to next year's committee.
Side of Tower, If Tension. Designate by the name of the terminal of the line or section nearest which the insulators are located.
Conductor. All designations of conductors shall be determined by viewing the line facing in the direction in which the poles are numbered. Different arrangements shall be handled as follows:
(a) Vertical arrangement: "Upper," "Middle" or "Lower" as the case may be.
(b) Horizontal arrangement: Use the designation "Right," "Middle" or "Left" as the case may be.
Triangular arrangement : Use the words "Upper Right," "Upper Left," "Lower Right," "Lower Left," "Upper Middle" or "Lower Middle" as the position of the triangle may make necessary.
In every case indicate clearly the point at which the pole numbering begins and the point toward which it runs.
Unit Number in String. Designate the number of the unit by starting at the cross-arm and numbering toward the conductor in every case.
Visible Defects. Any defects in the unit which are apparent, such as small cracks in the porcelain or unusually large cracks in the cement around the stud, flaws due to manufacturing causes, variation in color of glaze from the normal and rusting of the stud where it enters the cement, should all be recorded.
Pin Type Insulators
Data were submitted on pin type insulators on 7597 circuit miles as follows:
It has been felt in the past that pin type insulators were not subject to the difficulties that suspension insulators have experienced. Random information coming to the attention of the Committee indicates that this may not be altogether true. Sufficient data to form an accurate estimate of the situation have not been available. From those available the following has been deduced:
Cracking is usually confined to the outer shell. Insulators not exposed to direct sunlight seldom crack.
Cracking is not entirely confined to the older types.
Instances have been noted on the same insulator where earlier lots have been almost entirely free from cracking, whereas later lots entirely identical in appearance have failed in large numbers. This would seem to indicate that factory methods, particularly cementing, play an important part.
Uncemented sections of insulators seldom crack, regardless of weather exposure.
As a bearing on the fourth condition noted above, one instance has been cited where rather severe difficulties were encountered with pin type insulators of an early make, in which part of the difficulty was attributed to the rather low grade of porcelain available at the time, and part to the fact that the shells were shipped separately and assembled into a complete insulator in the field. In this case the cracking was not entirely confined to any one shell, though it is usually noted under the conductor or tie wire and in the bottom petticoats. In the case of this particular property, no difficulty has been observed in the newer types of insulators or in more recent installations. One company reports failure due to cracking as developing after rains which have followed extremely hot weather, indicating troubles due to unequal contraction under quick temperature changes. Enough instances of trouble with cracking in pin type insulators have been cited to make the Committee feel that the situation needs more study, but not enough to warrant now the conclusion that serious trouble may develop.
