Direct-voltage Tests Research Articles

Surface arcing of insulators due to bentonite contamination

Abstract Impulse, alternating and direct voltage tests together with optical observations have been done under clean and polluted surface conditions with respect to bentonite, which is treated as a pollutant. The insulator materials tested were polymethyl methacrylate and polythene. Bentonite pollution may affect several surface flashover characteristics such as voltage at breakdown and 50% breakdown voltage under negative lightning impulse (1.2/50 μs) and the time to breakdown under wet condition for both negative and positive impulse voltage. There are four types of paths taken by the discharge channels, based on which the degree of degradation of the material surface may vary.

Modeling of the Dielectric Response of a Stator Winding Insulation from a DC Ramp Test

The aging due to mechanical, thermal, electrical and environmental stresses of a stator insulation system inherently involves alterations of the material properties which are detrimental to its service operation. When these properties are deteriorated to the point where the material can no longer operate safely under normal stress conditions, it implies that it has reached the end of its useful life. To prevent such forced outages and even to improve the useful life, condition-based maintenance and diagnostic tests are periodically conducted on stator insulation system. Among the tests performed to assess the condition of winding insulation, various methods are commonly used like partial discharge measurements, hi-pot, step voltage and ramped voltage tests to name just a few. The I-V curve measured during a ramped direct voltage test, as the one obtained in a polarization/depolarization test, includes the contribution of various current components, the capacitive, the absorption, the surface leakage and the bulk conduction currents. These contributions, especially the absorption and the conduction (surface or volume) currents, are not as easy to separate in a ramped voltage test as in a polarization/depolarization test (step voltage test). A modeling of the I-V curve obtained with the ramped voltage test, allowing an easy separation of the leakage current from the absorption and the capacitive currents, is presented in this paper. This modeling is compared with a field measurement on complete stator winding and with a lab measurement on spare coil. The calculated values of the absorption and leakage currents obtained during a ramped direct voltage test are compared with the same components deduced from a polarization/depolarization test.

Assessing the condition of hydrogenerator stator winding insulation using the ramped high direct-voltage test method

Ramped high direct-voltage tests have proven to be an effective and nondestructive means for reducing in-service stator winding failures. The measured current versus applied voltage response obtained during these off-line tests can be used to detect cracks and fissures, moisture absorption, tape separations, conductive surface contamination, lack of epoxy cure, and internal delamination. Ramped voltage test results have been used to determine when corrective actions are needed and what the appropriate actions are. When done properly, the test can identify stator windings that are approaching failure without accelerating the deterioration process.

Effect of duct configuration on oil activity at liquid/solid dielectric interfaces

A test cell has been designed to allow simultaneous determination of the currents flowing at the surfaces of two spacers in a 5mm oil gap. The spacers were asymmetrically disposed so their surfaces were at different distances from the bulk oil. Electrical measurements and Schlieren observations showed that the currents and activity in the oil at the spacer surface are influenced to a considerable degree by the length of duct: the current increasing by an order of magnitude in the longer duct. The technique tends to reduce the difficulties often experienced when attempting to compare results from direct voltage tests carried out in different cells. During the application of 50Hz alternating voltages to the system, different phenomena were observed at the spacers. These are considered to be due to the effect of dielectric losses in the spacer material. At higher stresses (of the order of 4 × 10 6V/m) EHD effects appeared to be present in certain regions resulting in movement of the oil: an increase in partial discharge activity was recorded for this condition. Higher frequency tests, up to 1000Hz, and similar stresses indicated that no EHD effects were present: it is postulated that the changes observed were due to losses in the oil itself. The work is continuing in an attempt to explain in more detail the various phenomena and to determine their possible significance in technology.

Effect of magnetic field and magnetic particles on conduction current in transformer oil

The paper presents the results of an experimental investigation into the effect of a perpendicular magnetic field on the conduction current in transformer oil. Direct voltage tests on samples of Voltesso 35 oil show that the presence of a relatively weak magnetic field can cause a reduction of more than 65% in the conduction current under typical voltage stresses. The effect of impurities in the form of magnetite particles is also investigated by measuring the conduction current with and without an applied magnetic field. Three theoretical models are examined for possible explanation of the observed results in plain-oil samples.

Puncture risks when testing large synchronous machines with direct voltage

The electrostatic field distribution is calculated for the insulation of the coil ends of a winding undergoing a direct-voltage test. It is found that a relatively high voltage can occur across the insulation of the coil ends if these are covered with a layer of dust or moisture, and a puncture may occur. By considering the effect of such a puncture upon the current in the test circuit, it is demonstrated that a puncture at some distance from the stator core may be left unnoticed. Such punctures have been discovered in coils subjected to field tests with direct voltage.

Behavior of Dielectrics A Study of the Anomalous Charging Current and the Variation of Dielectric Energy Loss and Capacitance with Frequency in Solid Dielectrics

The paper describes a research undertaken with the object of throwing light on the relations between the electrical properties of dielectrics or insulating materials, especially on the relation between the ``anomalous charging current'' under a direct voltage test and the dielectric energy loss for an impressed alternating voltage. After a review of the theory of this relationship, the methods developed in this study for observing the anomalous charging current in condensers in the time range from 0.0007 to 0.100 seconds after impressing a steady potential are described. The methods used in observing the ``normal leakage current'' and the alternating energy loss and capacitance in the frequency range from 200 to 4000 cycles per second are also described. The results of measurements by these methods on mica, glass, varnished cloth, and paraffin paper condensers are given. These measurements include a series giving the effect of varying temperatures on the properties of the glass and varnished cloth condensers. Finally, the comparisons between the dielectric energy loss and capacitance variation with frequency predicted on the basis of the d-c. anomalous charging current tests and the variations which were actually observed are given. These comparisons show that a fair agreement exists between the predicted and observed curves.