Timm, L. J.
Manager, Irrington Varnish & Insulator Division of Minnesota Mining and Manufacturing Co., Irvington, N. J.
Pages: 6 Published: Jan 1957
Since this symposium was initiated, I have felt the selection of the title “Significance of Minimum Property Values of Electrical Insulating Materials” was an unfortunate choice. The emphasis is on minimum property values and the significance of these values instead of stressing those values from which the quality level and dispersion within a lot or between lots may be determined. If we are interested in minimum property values, perhaps we should have a future symposium covering the theory of extreme numbers inviting papers by trained statisticians. My feeling at this time is that most of us are not aware of the full importance of this interesting problem which was touched upon by Mr. Endicott in his paper. However, it is recognized by the manufacturers of insulating materials that design and efficiency of electrical equipment is dependent upon their materials and that the highest quality economically feasible is desired by the consumer. The efficiency of equipment is determined by the design engineer with the assistance of the materials engineer who should be familiar with the quality level of materials. These men should be interested in the acceptable quality level and not in the minimum values except as derived from the quality level and dispersion within lots. It should be recognized that the quality level of any material is best determined by the manufacturer and only he is fully cognizant of the variations which are present in his manufacturing process. The manufacturer can sample his production over a longer eriod of time and by doing this obtain a truer mean or average value and the dispersion or standard deviation in his production. The purchaser can only sample his shipments and determine the quality level of shipments within specified confidence limits. The use of sampling plans will allow him at times to accept rejectable material and at other times to reject acceptable materials. These are risks inherent in any economical sampling plan. In my opinion, quality level can be ascertained by means of either of two techniques: a “go”-“no go” gage, or a test method which allows the user to accumulate data which subsequently can be analyzed using statistical methods and be of real use. The first technique produces limited data and the second produces an abundance of information. I would like to compare these techniques by the use of dielectric strength values. The continuous dielectric breakdown tester, a “go”-“no go” gage, is to be used for minimum values and the short time method with 1/4 in. electrodes is used for the accumulation of data.
Paper ID: STP46279S