Paper Authors

Abstract

NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract

Session 1559

A QUADRUPOLE MASS SPECTROMETER BASED GAS ANALYZER FOR POWER TRANSFORMER FAULT ASSESSMENT

Saleh M. Sbenaty Department of Engineering Technology and Industrial Studies Middle Tennessee State University

ABSTRACT

This paper describes the feasibility of using a quadrupole mass spectrometer-based gas analyzer to predict malfunctions in high-voltage oil-filled power transformers using the analysis of dissolved gases in oil technique. A vacuum system for the quadrupole analyzer and the gas introduction system is designed and built for this purpose. A method for the extraction of the dissolved gases in oil is developed and the gas introduction system is described. The spectra of the thirteen gas species of interest are obtained and the sensitivity factors for each gas are calculated. The spectra of a standard gas mixture and the dissolved gases in oil from a faulty transformer oil specimen are obtained. The data reduction procedure is described and is used to reduce these spectra to the partial pressures of the 13 gas species. Finally, the concentration of each gas is calculated and methods for fault assessments are presented.

INTRODUCTION

Gas analysis is one of the best available techniques for transformer fault prediction and diagnosis. Scientists have found several decades ago that a loaded power transformer tends to evolve gases1, 2. These gases, the result of normal aging processes of the insulating materials, dissolve in the transformer’s oil. The concentrations of these gases are usually small in normal operation and depend upon the solubility of each gas in oil and the transformer’s type. When a fault or a combination of faults exists in a power transformer, however, the concentrations of the evolved gases are comparatively large3. It was also realized that the amount of evolved gases and gas species depend on the nature of the fault. For example, hydrogen (H2) and acetylene (C2H2) are the main gaseous constituents when arcing in oil occurs. Hot spots mostly evolve carbon oxides (CO2 + CO) and light hydrocarbon gases. On the other hand, partial discharges (corona) produce hydrogen and other light hydrocarbon gases. In addition, slowly developing faults were found to produce decomposition gases4. By analyzing these dissolved gases, therefore, one can detect an incipient fault and reveal the operating condition of a power transformer before a costly and/or catastrophic accident occurs such as interruption of service and explosion.

FAULT DIAGNOSTIC TECHNIQUES

Several diagnostic methods using gas analysis can be used to predict faults in a power transformer5, 6. x The Dornenburg Ratio Method uses the following four gas ratios: methane/hydrogen (CH4/H2), acetylene/ethylene (C2H2/C2H4), ethane/acetylene (C2H6/C2H2), and acetylene/methane (C2H2/CH4). These ratios are calculated from the measured concentrations

• Citation

• Format
  • APA
  • APA - LaTeX bibitem
  • MLA
  • MLA - LaTeX bibitem
  • Bibtex
  • EndNote - RIS

Sbenaty, S. M. (1997, June), A Quadrupole Mass Spectrometer Based Gas Analyzer For Power Transformer Fault Assessment Paper presented at 1997 Annual Conference, Milwaukee, Wisconsin. 10.18260/1-2--6752