This paper summarizes the results of a functional life testing program conducted over a period of several years on low voltage resin encapsulated electronic transformers. The program was designed to establish how much environmental factors, such as thermal cycling, humidity, vibration, cold shock, overvoltage, and method of heating the units, affected the life of the units in thermally accelerated aging tests. It was concluded that thermal cycling was the most important factor influencing the life, which was usually terminated by a turn or layer-to-layer short circuit. It was further concluded that oven heating gave very similar results to internal circulating current heating of the windings. An important result of these tests was that this system, which incorporated a cellulose paper layer insulation, usually considered 110°C class layer insulation proved to have life values corresponding to a 130 to 155°C class system. This reemphasizes that temperature ratings of apparatus with composite insulation should be made on the basis of system performance rather than individual material performance. The life of the transformers followed the Arrhenius log life versus reciprocal Kelvin temperature relation.

Life Testing of Electronic Power Transformers-Part II

Life Testing of Electronic Power Transformers

"The Proposed IEEE Test Procedure for the Evaluation of Insulation Systems for Electronic Power Transformers" is based on the principle of accelerated thermal aging. In order to accelerate the thermal aging sufficiently to accomplish the evaluation in a reasonable time, the test temperature is often higher than the chemical stability temperature of some of the components of the system. This instability may render the system unsuitable for the use of this test procedure unless a way can be found to reduce the effect to a usable level. One of the more commonly used materials of this type is cellulose. With suitable resins and adequate embedment techniques, which in - cludes thorough impregnation, cellulose (papers, pressboard, etc) has been demonstrated to give good service at temperatures well above the previously accepted maximum temperature of 105°C. In this situation, one or more of the test temperatures will be sufficient to cause limited chemical decomposition of the cellulose. This does not necessarily render the material ineffective as a dielectric and incapable of evaluation by comparison, but a by-product of this decomposition is water which is chemically liberated. The water thus produced must be eliminated as it forms, otherwise failures will result which are not truly due to thermal aging as encountered at normal operating temperatures. This paper describes a series of tests which established the need for special precautions and procedures in the thermal evaluation of sealed insulation systems containing appreciable quantities of cellulose. A modification of the test procedure is suggested as a practical solution. This modification is considered to have a minor effect on the test results and permits comparison evaluations which otherwise are not possible.

E. Henry

Papers

A Problem Associated with the Evaluation of Insulation Systems for Electronic Power Transformers