Comparison of Methods for the Dissipation Factor Measurement at Practical Examples

Due to the changes at the liberalized electrical energy market, the cost optimization at utilities and industries is of great company interest. The infrastructure gets increasingly aged and therefore a change in the maintenance strategies could be observed. From the former time-based strategies nowadays condition-based management strategies are more and more used. As one knows the equipment condition, optimal maintenance strategies could be developed and used. For the evaluation of the equipment condition factors like the insulation integrity or thermal aging, various monitoring-, analysis- and diagnostic methods are available. Defects in the electrical insulation material lead to a change of the physical, chemical and electrical properties of insulation materials, which causes a change of dielectric parameters, like the dissipation factor (tan delta).
For condition estimation of electrical equipment the tan delta measurement plays a big role. With the knowledge of this value a good integral estimation of the insulation condition is feasible. As this value has already been measured in the past, major empiric experiences regarding the behavior of this value and the corresponding effect in the insulation are available.
The classical method for determining the tan delta is the measurement by usage of the Schering Bridge, which is an AC-balanced bridge circuit. This balancing could be done manual or fully automatic by using a micro controller. Nowadays more and more fully unbalanced digital dissipation factor measurement instruments are used. With these instruments the vectorial current of the reference and the test branch is measured and the dissipation factor will be calculated in different ways.
In this paper the basics of these two methods (Schering Bridge and unbalanced bridge) are shortly described and possible applications are shown. For a comparison of these systems different test measurements and results are presented and discussed.