[Trade Journal]
Publication: Proceedings of the National Electric Light Association 43rd Convention
New York, NY, United States
p. 672-673, col. 1-2
Testing of High Tension Insulators
Either of the two following methods of testing is usually selected by operating companies which follow a general policy of having insulators tested before they are installed, these tests being made at the manufacturer's plant or on the company's premises.
High Frequency Tests
Insulators are tested with high frequency, the potential being as a rule just below the flash-over value.
Normal Frequency Tests
Insulators are tested under high potential, and at normal operating frequency, the voltage being raised to the flash-over value, although some companies prefer to keep the potential just below the flash-over point.
There is a wide difference of opinion regarding the relative merits of these two methods of testing, some of the advantages and disadvantages claimed being as follows:
There is more likelihood of damaging a good insulator with the high frequency test be-cause of the steeper wave-front and the greater dielectric loss. The high frequency test set is more portable than the low frequency set.
The high frequency test gives a more even distribution of corona, and for that reason picks out any faults more readily than the low frequency test. The low frequency test is the one commonly used by manufacturers.
In addition to the tests made before installation, quite a number of companies periodically test insulators in service. This practice is largely confined to companies operating lines of 60,000 volts and above. Of the companies answering the questionnaire, sixty operate transmission lines, and twenty of these test insulators after installation. The methods employed by these companies are as follows:
Megger Method
Eight companies report the use of the megger for testing insulators. The resistance of the insulators is measured with a megger, which is applicable to either pin type or suspension type insulators. It is necessary to take the line out of service in performing which is not the same for the various companies, each one having its own standard. These values vary from 500 to 5000 megohms. The results secured seem to be generally satisfactory, although there are certain types of defects in insulators which this test will not reveal.
"Buzz Stick" Method
Six companies report the use of the "buzz stick" method of testing insulators. This test has been developed for both pin type and suspension type insulators and is made with the lines alive. It is more especially adapted for tests on the latter type, and the description given applies to suspension insulators.
This method depends upon the electrical capacity of the insulator unit. The only tool used in testing is the "buzz stick." This is a long handled tool with a two pronged metallic fork on the end, with which the insulator units may be short-circuited. The test in general consists of two steps: the first is to indicate whether there are a sufficient number of good insulators in the string to make it safe to apply the second step, which involves short-circuiting each individual unit. The first test consists of touching one tine of the fork to the cap of each insulator, beginning with the one supporting the conductor. As contact is made, a small spark is produced which represents the charge taken by the prongs of the testing stick to bring them to the potential of the cap. Since the cap potential decreases as the test progresses from the conductor to the point of support, the spark produced diminishes. If an insulator is defective, its cap will have the same potential as the cap of the one preceding it, and the spark will be the same. The tester judges by the noise made by the spark whether there are any defective insulators in the string.
If this test indicates that there are enough good units in the string to permit the short-circuiting test to be applied, each unit is then short-circuited, beginning with the one supporting the conductor. In this test, the spark produced is caused by discharging the insulator which is, of course, acting as a condenser. Each succeeding insulator is charged to the voltage of the cap of the preceding insulator, and since the capacity of an insulator would be decreased if it were defective, a bad insulator is indicated by the absence of a spark when short-circuited, or by an abnormally small spark. This test is much more this test. The instruments employed vary in capacity up to 5000 megohms-1000 volts. Two men are needed to make this test, one stationed on the pole or tower holding the leads on the insulator, and the other operating the instrument. Insulators are considered defective if the resistance is less than a pre-determined figure, accurately whether or not an insulator is defective.
Since this method depends upon the electrostatic capacity of the insulator for its operation, it will naturally indicate only the faults existing in the porcelain between the cap and pin, and will not indicate faults in the disc which would lower the flash-over voltage. Furthermore, before this test will indicate defective porcelain between the cap and pin, the defect must have decreased the electrostatic capacity of the insulator.
The companies using this test state that it gives very satisfactory results, the lines being kept remarkably free from insulator failures.
Telephone Receiver Method
Five companies report the use of telephone receivers with exploring rods for testing insulators. The tester uses a high resistance telephone receiver, one terminal of which is mounted on the end of an exploring rod and the other terminal grounded. The person testing walks along the line holding the ungrounded terminal of the receiver high in the air, and bringing it close to each pole. A defective insulator is indicated by a distinctive grating sound in the receiver, which is easily distinguished from the hum caused by the small leakage current flowing over the surface of the insulators. This kind of test is not so well adapted for suspension insulators, as it does not readily indicate defective units in the string. Similarly, on multipart pin insulators it does not give an indication when only one part is defective. It is, of course, necessary for the line to be alive to employ this test.
"Keokuk" Method
One company reports the use of the "Keokuk" method of testing insulators. The apparatus used is a self-contained testing fork, the tines of which are connected across the high tension winding of an automobile induction coil mounted on the handle. The low tension winding of the induction coil is energized from three dry cells carried in a container at the end of the fork handle, a snap switch being provided in the battery leads. In series with one of the high tension leads, connected to the tines of the testing fork, is an ordinary spark plug which forms an air gap. The tester places the fork across the insulator to be tested, and if the insulator is defective, sparking occurs across the spark plug. In making this test it is necessary to take the line out of service.
