The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	The role of electrical noise in screening transformers prone to failure
Author:	Shirodkar, Rajeev D.
View Online:	njit-etd1983-016 (v, 99 pages ~ 3.6 MB pdf)
Department:	Department of Electrical Engineering
Degree:	Master of Science
Program:	Electrical Engineering
Document Type:	Thesis
Advisory Committee:	Misra, Raj Pratap (Committee chair) Zambuto, Mauro (Committee member) Ball, W. H. Warren (Committee member)
Date:	1983
Keywords:	Electronic noise. Electric transformers.
Availability:	Unrestricted
Abstract:	Insulation used as a barrier between the high voltage and low voltage windings of a transformer has sites of in- homogeneities like pinholes, cracks or voids, developed during the manufacturing process. Thus, these minute air pockets are stressed much more than the solid dielectric, on the application of a voltage stress, because the dielectric constant of the solid insulation is many times that of the air dielectric. As a result, excitation or ionization takes place in the air (gases). The amount of excited or ionized molecules is proportional to the volume of the inhomogenei- ties within the insulation. Therefore, experiments were conducted to monitor the electrical noise generated due to partial ionization of the tiny air gaps in the insulation. This was done by placing the transformer in a plexiglas tubing, around which an antenna was wound. The antenna was connected to a highly sensitive Hammerlund receiver, whose carrier level meter was calibrated in S-units. For 5S units of noise to be generated, the voltage stress required to be applied to the insulation was measured for a sample of three hundred transformers. The voltage readings were analyzed statistically and were found to have a normal distribution. Seven units which gave high voltage readings and seven units which gave low voltage. readings were subjected to accelerated life tests. This data has been analyzed statistically and has been found to be significant. The coefficient of correlation has been found and a linear regression equation has been formulated. The F-test of significance has been applied to the regression equation and has been found to be statistically significant.