Megger Tests

The insulation resistance meter test method for determining the condition of electrical insulation has been widely used for many years as a general nondestructive test method.

A serious limitation of this test is that its operating voltage of 500 to 1,000 volts will not always detect insulation punctures, whereas the higher voltages used by the high-voltage, DC testers will detect these punctures.

The insulation resistance meter test will show following parameters:

(a) Relative amount of moisture in the insulation,
(b) Leakage current over dirty or moist surfaces of the insulation, and
(c) Winding deterioration or faults by means of insulation resistance versus time curves.

Description of Test

A dc voltage of 500 or 1,000 volts is applied to the insulation and readings are taken to the insulation resistance versus time. Data should be recorded at the 1-and 10-minute intervals and at several other intermediate times.

Test Equipment

The hand-cranked insulation resistance meter has been the standard instrument for many years for checking insulation resistance. The hand- cranked instrument is satisfactory for “spot checks” but is not recommended for routine dielectric absorption tests, because very few men can continue cranking for 10 minutes without tiring and slowing up the cranking speed toward the end of the period.

Motor driven or electronic insulation resistance testers operating from a 115-volt, ac source or a self-contained battery are available and should be used for this purpose.

Because the value of insulation resistance varies with applied voltage, it is important that the test instrument have sufficient capacity to maintain its rated output voltage for the largest winding being tested, and the output voltage be constant over the 10-minute test period.

For this reason, some of the smaller test instruments may not be suitable for tests on large generators or transformers which draw a large dielectric absorption current.

For occasional checks on the calibration and proper function of insulation test instruments, it is recommended that a resistor in the 100-megohm range be attached to the inside of the instrument cover for use as checking standard. It is recommended that the same test instrument be used for each periodic test on a certain piece of equipment, as differences in instrument output characteristics may affect the shape of the dielectric absorption curves, especially at the lower end.

Dielectric Absorption Curve

Insulation resistance is not a definite measure of the voltage an insulation will withstand, but when properly interpreted affords a useful indication of the suitability of the winding for continued service. It should be remembered that values of insulation resistance, even on identical machines and for identical conditions, may vary over a wide range.

Changes occurring in insulation resistance are more significant than certain absolute magnitudes. This curve is called the curve of dielectric absorption.

The test voltage should be applied for a standard period of 10 minutes, with readings taken at intervals of 1 minute or less.

Any such curve which reaches a constant and lower than normal value in about 3 minutes or less, indicates high leakage current (due to the leakage current being large in proportion to the absorption current), and the winding should be thoroughly cleaned and retested or further investigated. Such cleaning should preferably precede all insulation resistance tests.

In case of very damp insulation, the dielectric absorption curve may start upward and then droop to a value lower than at the start of the test.

Minimum Values of Machine Insulation Resistance

“Recommended Practice for Testing Insulation Resistance of Rotating Machinery,” IEEE Standard No. 43, November 1974, indicates the recommended minimum insulation resistance Rm for armature and field windings of ac and dc machines can be determined by the equation:

where:

Rm = recommended minimum insulation resistance in megohms at 40 °C of the entire machine winding
Vt = rated machine terminal to terminal potential, in rms kilovolts

The winding insulation resistance obtained by applying direct potential to the entire winding for 1 minute must be corrected to 40 °C to be used for comparison with the recommended minimum value Rm. The insulation resistance of one phase of a three-phase armature winding with the other two phases grounded is approximately twice that of the entire winding.

Therefore, the resistance of each phase, when the phases are tested separately, should be divided by two to obtain a value which, after correction for temperature, may be compared with Rm.

If guard circuits are used on the two phases not under test when each phase is tested separately, the observed resistance of each phase should be divided by three to obtain a value which, after correction for temperature, may be compared with Rm. For insulation in good condition, insulation resistance readings of 10 to 100 times the value of Rm are not uncommon.

It should be remembered, however, that decreasing values of insulation resistance obtained from periodic tests are more indicative of deterioration of the insulation than low values. Machines rated at 10,000 kV-A or less should have either the polarization index or the insulation resistance (at 40 °C) at least as large as the minimum recommended values to be considered in suitable condition for operating or for overpotential tests. Machines rated above 10,000 kVCA should have both the polarization index and the insulation resis-tance above the minimum recommended values.

When the end turns of a machine are treated with a semiconducting material for corona elimination purposes, the insulation resistance may be somewhat lower than without such treatment.

Transformer Insulation Resistance

Although the foregoing paragraphs apply more specifically to generator and motor windings, they also apply, in general, to transformers, except that no insulation values have been established for transformers. Also, the technique of measuring transformer insulation resistance is not well known or standardized. If the transformer windings are not immersed in oil, the insulation resistance will behave much like generator insulation resistance.

The insulation resistance will be less after adding the oil, because the insulation resistance of the oil is in parallel with part of the solid insulation. Therefore, insulation resistance readings alone cannot be used to indicate the progress of dry out of the winding because the winding and the oil resistances cannot be separated.

The change of insulation resistance with temperature when the transformer windings are oil-immersed is similar to that in generators, and curves similar to those of figure 3 are useful for temperature standardizing.

Whether the slope of these temperature correction curves is affected by moisture content in the oil is not fully known. At the present state of the art, it is believed that the power factor test gives a better indication of transformer insulation condition than the insulation resistance test. Tests should be made between each winding, between each winding and ground with the other windings grounded, and between each winding and ground with the guard circuit connected to the other windings but not grounded.

Cable Insulation Resistance

The most frequently used test on high-voltage cables is insulation resistance measured by means of an insulation resistance meter. The most informative test for high-voltage cables is the dc, high-voltage test modified to combine a modest voltage withstand with insulation current/voltage measurement.

Insulation resistance testing of cable differs from the testing of apparatus windings mainly because of the high capacitance, if the cable is long, which takes a longer time to charge, and in the difficulty of obtaining a satisfactory temperature measurement, insulation resistance measurements are of value for comparison rather than for conformance to stated minimums.

The temperature of the cable is important and should be recorded with the insulation resistance. This will be difficult if the cable is partly indoors and partly outdoors, partly underground, partly above ground, partly exposed, and partly in conduits.

It may be necessary to estimate the temperature of the various lengths, and a weighted average computed. Tests should be made between each conductor, between each conductor and ground with other conductors grounded, and between each conductor and ground with other conductors connected to the guard circuit but not grounded.

SOURCE: TESTING SOLID INSULATION OF ELECTRICAL EQUIPMENT VOL.3-1