A large percentage of electrical apparatus failures are due to the deteriorated condition of the insulation. Many of these failures can be anticipated by conducting routine testing and timely maintenance. Changes in the normal capacitance of insulation indicate abnormal conditions such as the presence of a moisture layer, short circuits, or open circuits in the capacitance network. Tan delta (or dissipation factor) / power factor measurements indicate the following conditions in the insulation of a wide range of electrical apparatus:
- Chemical deterioration due to time and temperature, including certain cases of acute deterioration caused by local overheating
- Contamination of water, carbon deposits, bad oil, dirt and other chemicals
- Severe leakage through cracks and over surfaces
The interpretation of measurements is usually based on experience, recommendations of the manufacturer of the equipment being tested, and by observing these differences:
- Between measurements on the same unit after successive intervals of time.
- Between measurements on duplicate units or a similar part of one unit, tested under the same conditions around the same time, e.g., several identical transformers or on one winding of a three-phase transformer tested separately.
- Between measurements made at different test levels on one part of a unit; an increase in slop (tip-up) of a dissipation/power factor versus voltage curve at a given voltage is an indication of ionization commencing at that voltage.
An increase of dissipation/power factor above a typical value may indicate conditions such as those indicated above. If the dissipation/power factor varies significantly with voltage down to some voltage below which it is substantially constant, then ionization is indicated. If this extinction is below the operating level, then ionization may progress in operation with consequent deterioration. Some increase of capacitance (increase in charging current) may also be observed above the extinction voltage because of the short-circuiting of numerous voids by the ionization process.
An increase of dissipation/power factor accompanied by a marked increase in the capacitance usually indicates excessive moisture in the insulation. Increase of dissipation/power factor alone may be caused by thermal deterioration or by contamination other than water.
Unless bushing and pothead surfaces, terminal boards, etc. are clean and dry, measured values do not necessarily apply to the insulation under test. Any leakage over terminal surfaces may add to the losses of the insulation itself and may give a false indication of its condition.
FACTORS AFFECTING TAN DELTA / POWER FACTOR TESTS
Influence of Temperature
Most insulation measurements have to be interpreted based on the temperature of the specimen, as the dielectric losses of most insulation increase with temperature. It is important to determine the dissipation/power factor temperature characteristics of the insulation under test, at least in a typical unit of each design of apparatus. Otherwise, all tests of the same spec should be made, as nearly as practicable, at the same temperature. To compare the dissipation/power factor value of tests made on the same or similar type of equipment at different temperatures, it is necessary to correct the value to reference temperature base, 20° (68°F). The MIDAS 2881, MIDAS micro 2883 and IDAS 2823 do this automatically, when the DUT is defined correctly.
The capacitance of dry insulation is not affected by temperature; however, in the case of wet insulation, there is a tendency for the capacitance to increase with temperature.
Influence of Humidity
Exposed surfaces under adverse humidity conditions, may acquire a deposit surface moisture which can have a significant effect on surface losses and consequently on the results of a dissipation/power factor test. Surface leakage errors can be minimized if dissipation/power factor measurements are made under conditions where the weather is clear and sunny and where the relative humidity does not exceed 80%, or in an environmentally controlled laboratory or test lab.
Influence of Leakage Current
Any leakage over the insulation surfaces of the specimen will be added to the losses in the volume insulation and may give a false impression as to the condition of the specimen. Even a bushing with voltage rating much greater than the test voltage may be contaminated enough to cause a significant error. Surfaces of potheads, bushings, and insulators should be clean and dry when making measurements.
Electrostatic Interference
When tests are conducted in energized substations, the readings may be influenced by electrostatic interference currents resulting from the capacitive coupling between energized lines and bus work to the test specimen. The lower the specimen capacitance (and its dissipation/power factor), the more difficult it is to make an accurate measurement. It is also possible that a negative dissipation/power factor reading may be obtained. The MIDAS 2881 and MIDAS micro 2883 both contain an interference suppression feature, which minimizes the influences. The influences may be minimized further by:
- Using the maximum voltage of the test set, if possible
- Disconnecting and grounding as much bus work as possible from the specimen terminals
- Making measurements on a day when the weather is sunny and clear, the relative humidity is less than 80%, the wind velocity is low, and the surface temperature of exposed insulation is above the ambient temperature.
