Your browser doesn't support javascript.
loading
Electrochemical sensor for predicting transformer overload by phenol measurement.
Bosworth, Timothy; Setford, Steven; Heywood, Richard; Saini, Selwayan.
Afiliação
  • Bosworth T; Cranfield Centre for Analytical Science, Cranfield University, Silsoe, Beds, UK.
Talanta ; 59(4): 797-807, 2003 Mar 10.
Article em En | MEDLINE | ID: mdl-18968967
ABSTRACT
Transformer overload is a significant problem to the power transmission industry, with severe safety and cost implications. Overload may be predicted by measuring phenol levels in the transformer-insulating oil, arising from the thermolytic degradation of phenol-formaldehyde resins. The development of two polyphenol oxidase (PPO) sensors, based on monitoring the enzymatic consumption of oxygen using an oxygen electrode, or reduction of enzymatically generated o-quinone at a screen-printed electrode (SPE), for the measurement of phenol in transformer oil is reported. Ex-service oils were prepared either by extraction into aqueous electrolyte-buffer, or by direct dilution in propan-2-ol, the latter method being more amenable to simple at-line operation. The oxygen electrode, with a sensitivity of 2.87 nA microg(-1) ml(-1), RSD of 7.0-19.9% and accuracy of +/-8.3% versus the industry standard International Electrotechnical Commission (IEC) method, proved superior to the SPE (sensitivity 3.02 nA microg(-1) ml(-1); RSD 8.9-18.3%; accuracy +/-7.9%) and was considerably more accurate at low phenol concentrations. However, the SPE approach is more amenable to field-based usage for reasons of device simplicity. The method has potential as a rapid and simple screening tool for the at-site monitoring of phenol in transformer oils, thereby reducing incidences of transformer failure.

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Tipo de estudo: Prognostic_studies / Risk_factors_studies Idioma: En Revista: Talanta Ano de publicação: 2003 Tipo de documento: Article

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Tipo de estudo: Prognostic_studies / Risk_factors_studies Idioma: En Revista: Talanta Ano de publicação: 2003 Tipo de documento: Article