US 2244960 A
Description (OCR text may contain errors)
June 10, 1941. E. H. NELSON METHOD OF MAKING SEALS FOR DISCHARGE LAMPS Filed Aug. 16, 1940 Patented June 10, 1941 METHOD or MAKING SEALS FOR. DISCHARGE LAMPS Evan Herbert Nelson, Harrow Weald, England,
assignor to General Electric Company, a corporation of New York Application August 16, 1940, Serial No. 352,982 In Great Britain August 10, 1939 1 Claim.
The present invention relates to methods of manufacturing hermetic, electrically conducting seals for gaseous electric discharge lamps having an envelope of quartz or other vitreous material having a low coefiicient of thermal expansion.
Discharge lamps are now available commercially which comprise an elongated, tubular quartz envelope having an internal diameter of about 5 mm. and containing mercury in an amount suflicient to provide a vapor atmosphere having a pressure up to hundreds of atmospheres even when part of the mercury remains in the liquid state during operation of the lamp. The envelopes of such lamps are cooled by circulating fluid. The liquid mercury is located in the ends of the envelope and in contact with the inner surface of the envelope where it is cooled by the circulating fluid flowing over the outer surface of the envelope. A good contact between the mercury and the walls of the envelope is highly desirable for the purpose of dissipating heat from the liquid mercury which is in contact with current leads extending into the envelope.
There are two main methods by which current leads are introduced into the quartz envelopes of discharge lamps. One is by the use of intermediate sealing glasses; the other by the use of the strip seal. 0nd method, which is in some respects more convenient, has disadvantages when applied to lamps of the water-cooled type that are not present when it is applied to discharge lamps of other types.
Apparently during the collapse of the quartz tube onto the strip, which is an essential step in the making of a strip seal, material is distilled from the heated parts about the strip into the region between the electrodes which prevents the mercury from contacting properly with the walls, .or causes the quartz to devitrify, or both. The reasons why the same trouble does not occur in lamps of other types is probably (a) that the internal area of the discharge envelope is greater, so that any foreign material is less concentrated and (b) that no limit is imposed on the way in which; the liquid mercury, present when the lamps are cold, makes "contact with the wall. The object of this invention is to remove this disadvantage and to make strip-seals as convenient and useful in lamps of the water-cooled type as they are in lamps of other types by providing a method for manufacturing such seals. Still further objects and advantages of the invention will be apparent to. those skilled in the art from' We have found that the secseparating two wider portions of which one has the bore of the final dischargetubular envelope and the other of which is adapted to be collapsed onto the strip, (2) introducing into the said second part of the tube a thin strip of molybdenum or the like attached at one end to a wire of tungsten or other refractory metal, the diameter of the wire being such that it passes through the constriction with a small clearance and arranging the wire so that its free end protrudes through the constriction-for the required distance, (3) heating and then collapsing the constriction onto the said wire, (4) heating and then collapsing the quartz tube onto the strip. The part of the tube collapsed onto the strip is preferably of the same bore as the discharge envelope; but this is not essential, so long as it is wider than the constriction. The end of the strip not connected to the said wire is preferably, as usual, attached to another wire that forms the external current lead. The free end of the wire protruding from the constriction preferably serves as one electrode of the lamp and protrudes slightly from the liquid mercury in full operation; but it may in known manner be always covered by mercury, which serves as the electrode, so that the wire is a current lead thereto. Preferably two constrictions are formed in the tube separated by a distance rather greater than the length of the final discharge envelope and a seal is made at each end in accordance with the invention.
The success of the method according to the invention apparently depends on the fact that,
since the wire just fits the constriction, the quartz collapses on to it so rapidly that noxious material has no time to distil into the envelope, and that the collapsed quartz fit the wire so closely (although it does not form an air-tight seal with it) that noxious material evaporated when the tube is collapsed onto the strip cannot pass through the gap and reach the discharge envelope.
The single figure of the accompanying drawing illustrates the quartz tube I after the constrictions 2 and 3 have been formed therein. The constriction 3 is shown collapsed onto the currentinlead wire 5, such as a tungsten wire, which extends into the space between the constrictions 1 2 and3. The molybdenum strip 4 is attached at one end to the wire 5 and at the other end to the V made at the other end of the tube 1 by the method described above so that no noxious material is distilled into the part of the tube I between the two seals during the manufacture of the seals.
What I claim as new and desire to secure by Letters Patent 0! the United States is:
A method of manufacturing a hermetic, electrically conducting seal for a gaseous electric discharge device having a-container of vitreous material of a low coeflicient of expansion which comprises the steps of first forming a constriction in a tube of said material, introducing into saidtube an electrically conducting linear element comprising a thin, metal strip and a wire, said strip being attached at one end to said wire, said wire having a diameter closely approaching the inner diameter of said constriction, said strip being in one of the portions on either side of said constriction and a portion of said wire being within said constriction, heating said tube at the constricted portion thereof and collapsing it into contact with said wire and then heating the portion of said tube surrounding said strip and collapsing it into contact with said strip to form a hermetically tight juncture therewith.
EVAN HERBERT NELSON.