Discussion on “electrical measurements with special reference to lamp testing” (Edwards), and “incandescent lamps as resistances” (Amrine), Boston, Mass., June 28, 1912. (see proceedings for June, 1912)

Clayton H. Sharp: Speaking of the paper by Mr. Edwards, the author gives a curve of corrections for the alternating-current voltmeter as used in checking the voltages on the life testing rack. He emphasizes the importance of that correction curve. I do not think he emphasizes it quite enough and I doubt if a laboratory standard voltmeter alone, even though carefully checked, is quite sufficient to maintain the very high degree of accuracy which is called for in the measurement of life testing voltages. One-tenth of one per cent on the life testing voltage will make several per cent difference in the life of the lamp, and may have very serious commercial results under certain circumstances; so that in all lamp testing, the most important thing is the accurate determination and checking of these voltages. No single method is sufficient, but rather, various methods must be used and checked against each other. For instance, in our own laboratory, we use a multicellular electrostatic voltmeter with a mirror and scale, a method introduced by Dr. Kennelly some years ago. This is used as a transfer instrument to check directly against the potentiometer´, so that the chances of errors of the instruments themselves are as nearly as possible eliminated. Other instruments are used in the same way.