Moisture and aging strongly influence the dielectric properties of oiycellulose insulation systems. Dielectric measurements can therefore be used for assessing the condition of oUpaper insulation. In this paper, we report on measurements and modeling of recovery voltages and polarization and depolarization current measurements on composite oiYpaper insulation. A numerical model allows us to model the response of the composite system from knowledge of the dielectric properties of the constituents (03 and paper). In particular we demonstrate the consistency between recovery voltages and polarization and depolarization current measurements and show the influence of material properties on dielectric response. I. Basic Theory of Dielectric ResponseThe basic theory of dielectric response as it pertains to the analysis described in this paper have been developed in references [ l,2] and reproduced here for convenience. Assuming a homogeneous electric field E(t) is applied to a singZe dielectric material, the current density through the surface of the material can be written as:The current density J(Q is the sum of the conduction current and the displacement current, where (T is the DC conductivity and D(t) is the electric displacement given below. o(f) = E O E , E ( f ) + @ ( t )(2) so and E, are respectively, the permittivitty of vacuum and of the dielectric material. P(t) is the dielectric polarization and is related to the response function of the material by the relationship:The response function can be derived from measurements of polarization (charging) and depolarization currents. If we consider the case where an insulation system with geometrical capacitance CO is exposed to a step voltage, U, , the charging current is given by, (4) and the depolarization current is given by.where fChqinz is the length of thc charging period. Another method of obtaining the response of a dielectric is by measuring the recovery voltage of the dielectric system. A DC voltage, U,, is applied to the composite insulation for a period of time tl. The insulation is then grounded for a subsequent time period t2-tl. After this grounding period, the ground potential is removed from the insulation and for time period t~2 , a voltage recovers across the insulation as the test object is recharged by its own depolarization current.The current density that applies during a recovery voltage measurement can be obtained by substituting Equations 2 and 3 into Equation 1.4k@?i&l=-c$?a G-(9-f(t+bg,>>During the recovery voltage period, the current density is zero (0), so Equation 6 is solved with J(t) set to zero.

No abstract

No abstract

The continuous, non-intermitted service of electrical grids relies on the reliability of their assets, e.g., power transformers. Local insulation defects can result in serve failures such as breakdowns with severe subsequent costs. The prevention of such events is crucial. Hence, partial discharge (PD) activity at power transformers is evaluated directly in the factory before shipment. Additionally, PD activity can be monitored during service using the ultra-high frequency (UHF) method. In this contribution, a calibration procedure is proposed for the UHF method. The calibration process is required to ensure both, reproducibility and comparability of UHF measurements: Only a calibrated UHF measurement procedure can be introduced supplementary to IEC 60270 in acceptance tests of power transformers. The proposed calibration method considers two factors: The influence of the UHF-antenna's sensitivity and the PD recorder characteristics including accessories such as cable damping, pre-amplifier, etc. The former is addressed by a characterization of UHF sensors using the standard antenna factor (AF) in a gigahertz transverse electromagnetic (GTEM) cell. The PD recorder's influence is corrected by using a defined, invariable test signal as reference for all recording devices. A practical evaluation of the proposed calibration procedure is performed in a laboratory setup using different UHF recording devices and UHF sensors using artificial PD signals and real voltage-driven PD sources.

No abstract

For reliable verification of the quality of both, new and aged insulating systems, not only the correct measurement of significant internal partial discharges (PD) but also an exact localization and physical interpretation of the detected PD-signals is needed. A step forward towards the interpretation of conventionally measured PD-signals is realized by a digital data processing. This paper describes PD-measurements performed on different highvoltage apparatus using a computer controlled phase resolving partial discharge analyzer (PRPDA). The PRPDA-system records partial discharge (PD) activity of a test object during a given number of cycles of the applied ac voltage, or during a fixed time. Some typical results obtained in SFg-insulated systems (GIs), power transformers and power generators in off-line and in on-line condition are presented and discussed.

The temporal variation of the spark resistance RFS of a tip-plane and a rod-plane electrode gap immersed in n-hexane, n-pentane, n-decane, and 2,2 dimethylbutane (neohexane) was measured by employing fast high-voltage and current measurement techniques. Three distinct regions were found: (1) rapid decrease of the resistance to 103 Ω, (2) development of the spark channel with RFS decreasing to 102 Ω, and (3) an exponential variation of RFS in time.

A fast coaxial discharge line for the study of the electric breakdown of liquid hydrocarbons is described. It allows the simultaneous measurement of discharge current (up to 5 kA) and voltage (up to 50 kV) with nanosecond time resolution. The spark gap is of tip-plane geometry. A detailed frequency analysis of the discharge and measurement circuits has been performed. Examples for the spark breakdown of n-hexane under a variety of conditions are presented.