Vacuum Switching Experiments at California Institute of Technology

Abstract

Synopsis.-Successful experiments in switching or breaking a circuit in a high vacuum have been made at the California Institute of Technology. This paper is a report on three sets of these experi ments which extended over a period of three years.The conclusions drawn from the experiments may be summed up in the statement that vacuum breakers of laboratory type have been successful in breaking circuits and offer a possible solution of the circuit-breaker problem.The results show that switching in vacuum affords the advantages of no pitting of contacts, quick break, the arc always going out on the first half cycle, small voltage rise across the switch, and small dis tance of travel necessary for the switch blades.Making the vacuum switch practical calls for a solution of the problem of making commercial apparatus with vacuum-tight joints,, and the elimination of the use of liquid air with the vacuum pump. E XPERIMENTS on breaking an electrical circuit in a high vacuum have been made during the last three years at the California Institute of Technol ogy in connection with the study of switching highvoltage, high-power circuits. These experiments were undertaken as a result of the well-known limitations of oil circuit breakers. A large number of tests was made on high-vacuum breakers of laboratory type. Some very promising results were obtained in interrupting large currents.When these experiments were suggested, the question immediately presenting itself was: Will the vacuum maintain itself at the time the arc is formed between the separating metallic parts of an opening switch?This doubt was quite generally substantiated by the commonly recognized theory of the electric arc, 1 viz., that the maintenance of an arc is dependent upon the giving out of thermions from hot spots on the electrodes between.which the arc is formed, with the attendant vaporization of the metal. If this were true, a large current could not be interrupted in a vacuum because the formation of even a small amount of gas would reduce the vacuum and cause it to become a conducting vacuum rather than an insulating vacuum.The fact of the matter, however, is that if the vacuum is sufficiently high and all adsorbed gases have been removed from the metal electrodes, very large currents can be broken without formation of enough vapor to maintain an arc.Dr. R. A. Millikan has shown 2 that, with cold elec trodes suitably prepared, millions of volts of potential gradient are required to obtain discharges of any kind between metal surfaces. He has also worked out with much care the conditions necessary for denuding metal surfaces of gases and preventing the impairment of the vacuum through the evolution of gases. A. Janitzky 3 also has reported experiments showing that currents will not flow across the space between cold electrodes in a vacuum provided the electrodes have been completely outgassed.

Multiply stripped ion generation in the metal vapor vacuum arc

“…One of the earliest publications in the field is that of Sorensen and Mendenhall in 1926 ; early work was severely impeded by the rudimentary vacuum techniques of the era. An historical survey of the field, pre-1960s, has been given by 2 Cobine .…”

The power Application vacuum switch

“…His report also shows some success with fine outgassed copper wire fuses in highly evacuated sealed-off glass tubes. T h e progress report, a 1926 paper given at Salt Lake City, U t a h , 1 and a thesis by M e n d e n h a l l , 2 together describe experiments with the three switches of Fig. 1.…”

Multiply stripped ion generation in the metal vapor vacuum arc

“…The fundamental phenomenon which drives the vacuum arc is that of cathode spot formation -minute regions of intense current concentration which reside on the surface of the cathode and at which the solid cathode material is vaporized, ionized, and The study of the metal vapor vacuum arc discharge, also called the vacuum arc or metal vapor arc, had its origin in the high power switching field. One of the earliest publications in the field is that of Sorensen and Mendenhall in 1926 ; early work was severely impeded by the rudimentary vacuum techniques of the era. An historical survey of the field, pre-1960s, has been given by 2 Cobine .…”