US2076755A - Gaseous electric discharge device - Google Patents

Description

APril 3, 1937. w. R. WALKER 2,076,755

GASEQUS ELECTRIC DISCHARGE DEVICE Filed Dec. 13, 1932 uuonuuuu =F V I l I I r I! I I I 14 15 16 15 14 Patented Apr. 13, 1937 STATES PATENT FFICE to General Electric Vapor Lamp Company,

Hoboken, N. 5., a corporation of New Jersey Application December 13, 1932, Serial No. 647,016

1 Claim.

The present. invention relates to electric discharge devices, and particularly to apparatus for starting and operating thesame.

A particular object of the invention is to provide apparatus which will be extremely effective to start gas or vapor discharge devices into operation. Another object of the invention is to provide apparatus which will limit the potentials which are applied to initiate the discharge.

Another object of the invention is to provide ap-,

paratus to produce a high frequency high voltage discharge which is useful generally. Other ob jects and advantages of the invention will appear from the following detailed specification or from an inspection of the accompanying drawing.

The invention consists in the new and novel apparatus hereinafter set forth and claimed.

The starting of electric gaseous discharge devices, and particularly of those devices having 0 liquid cathodes, such as mercury vapor arcs, has long been an especially troublesome problem. High frequency alone is insufficient to initiate an arc discharge to a mercury cathode, it having been found that it is necessary to provide enough energy to produce a hot spot on the cathode before the arc can be started. For example, a voltage surge, produced by interrupting the flow of current through an inductance by means ofa mercury switch, has heretofore been used for the purpose of starting an arc in a mercury vapor arc lamp. I have now discovered, however, that by using a switch having solid contacts operating in a vacuum" to interrupt the current flow through this inductance a new type of discharge 30 is produced which is many more times more effective in starting the main arc than was the simple surge heretofore used. This discharge is not only of greater potential, for a given current and inductance, due to the more rapid interruption of the circuit in my novel switch, but also gives indications of having an extremely high frequency. This unique combination of high energy and high frequency is the ideal for starting a mercury vapor arc, since it not only produces the necessary hot spot, but also effectively initiates the main arc thereto. As a result instantaneous starting of the mercury arc, an ideal long sought, is now consistently attained for the first time. In some cases it is desirable to place 0 a limitation on the potential whichis produced by this apparatus, however, in orderto safeguard the insulation and the like. ,Such a limitation is readily attained by placing a spark gap across the terminals of the switch. I have also discovered a that in some cases improved results are obtained by connecting a high resistance in the lead to the so-called starting band.

For the purpose of illustrating my invention I have shown several embodiments thereof in the accompanying drawing, in which Fig. 1 is a schematic diagram showing my novel combination of a mercuryarc lamp with a vacuum switch,

Fig. 2 is an elevational view, in part section, of the novel switch which I prefer to use in my new combination,

Fig. 3 is a schematic diagram of a modification of the circuit of Fig. l, and

Fig. 4 is a schematic diagram of another modification of the circuit of Fig. 1.

In the drawing, with particular reference to i Fig. 1, there is shown a mercury vapor arc lamp l of the well known Cooper Hewitt type having a mercury cathode 2, starting band 3, and anodes 4. Said anodes i are each connected in a conventional manner through a ballast resistance 5 to opposite ends of an autotransformer 8. The midpoint of said autotransformer is connected through an inductance I to said cathode 2. From a point between said inductance and said cathode a connection is made through the vacuum switch 8, which is actuated by said inductance I, and a resistance 9 to one of the anode leads. The starting band 3 is connected to a point between said switch 8' and said resistance 9.

Said switch 8 is shown in detail in Fig. 2. This novel switch, which is claimed in a separate application filed concurrently herewith, consists of a sealed vitreous envelope H which is highly evacuated. A pair of inleads I! extend through opposite ends of said envelope, a horizontal contact plate l3 which extends from wall to wall of said envelope being welded to the inner end of each of said inleads. A hubbed disc ll of porcelain or the like is mounted on each inlead between said contact plate and the seal-and is freely rotatable on said inlead. Said discs serve as stops for the endwise movement of a bridging conductor l5 which is supported on the contactplates l8. Said bridging conductor 55 is tubular, and may conveniently be formed of tungsten wire wound. into a tight helix. Within said bridging memberthere is retained in any suitable manner a rod I6 of magnetizable material, such as iron. Said rod has a diameter which is appreciably less than the inner diameter of said tubular bridging mem-. ber and hence has considerable freedom of movement therein. 7

In the use and operation of theforegoing apparatus, upon energization of the autotransformer 6 potential is applied between the anodes 4 and the cathode 2, but this potential is insufflcient to initiate a discharge. Current flows, however, from one end of said autotransformer through the resistance 9, switch 8, and inductance I back to the midpoint of said autotransformer. As soon as said inductance l is thus energized the iron rod [8 is raised and strikes a hammer blow on the surrounding bridging member l5. This hammer blow, plus the turning moment exerted by the pullon the free end of said bridging member, is suflicient to-free said bridging member even though it is frozen or partially welded fast at one end, as frequently occurs,

and hence this bridging member is immediately raised to the top of the envelope ii, and the circuit through said switch 8 is opened. Said inductance I thereupon generates a voltage surge or discharge, as a result of which said cathode 2 is momentarily made intensely negative with ree spect to the anodes I and the starting band 3. A disruptive discharge thereupon occurs at said cathode which produces a hot spot on the mercury cathode and suflicient ionization in the are tube to permit the main arc to start from the anodes I. The nature of the surge or discharge generated by this novel combination of the inductance l and the switch 8 is unique, in that it has a high frequency component which remarkably assists in the initiation of the desired arc,

it having been found that with this novel com-.

bination of elements a mercury vapor arc lamp can now be started with the same certainty and despatch as an incandescent lamp, an ideal which has been unattainable heretofore.

In some cases, however, it becomes desirable to limit the maximum potential to some given value, in order to safeguard the insulation and to, prevent undesirable arcing .between leads. This may be accomplished, as shown in Fig. 3, by-placing a spark gap l1 across the terminals of the switch 8. This spark gap may be of any desired type, but as here shown theelectrodes are sealed into a vitreous envelope which may eitherbe evacuated, or contain a suitable gaseous atmosphere. Such devices are well known and hence need not be further described here. The breakdown of this gap is so chosen as to be the maximum desired potential. As a result this circuit operates in exactly the same manner as that of Fig. 1, with the exception that when occasional surges tend to exceed the desired maximum the spark gap l1 breaks down and drains of! the surplus energy.

In some cases, where there is only a relatively short gap between the starting band. I and the inlead to the cathode 2, there may be a tendency for arcing to occur therebetween. I have found that this tendency is readily overcome by placing a ,high resistance It in the lead'to said starting band 3, as shown in Fig. 4. Said resistance It isof the order of 15,000 to 30,000 ohms and is conveniently made of carbon impregnated paper. since the current carried thereby, is relatively small. The circuit shown in this figure is otherwise the same as that of Fig. 1 and operates in the same manner, the lamp invariably starting with great certainty.

My novel combination of a vacuum switch with an inductance is not limited, of course, to use in starting a mercury vapor arc, since it may be used in various other applications where the unique voltage of the discharge ll'oduced thereby is of advantage, such as in starting a discharge in any type of discharge device, including those with thermionic electrodes, or in operating a vice, which permits thestarting to be produced with great certainty and celerity, even at very low temperatures at which the apparatus heretofore employed has been ineii'ective.

While I have described my invention by reference to particular embodiments thereof, it is to be understood that various changes, omissions.

'and substitutions, within the scope of the appended claim, may be made therein without departing from the spirit thereof.

I claimas my invention:

In combination, a mercury vapor arc device having a mercury cathode, at least one anode, and a conducting sheath in electrostatic relation to said cathode, an inductance in the lead to said cathode, a connection from said cathode to the lead to said anode, said connection including an evacuated metal to metal contact switch, and a high resistance connection to said sheath from the side of said switch which is connected to said anode lead, said last mentioned connection being distinct from said connection between the cathode and the anodelead.

WARREN R. WALKER.

Images