US2363070A - Electric discharge device - Google Patents

Description

Nov. 21,1944. ELEMM RS 2,363,070

ELECTRIC DISCHARGE DEVICE Filed mg. 1, 1941 Inventor I Eugene Lemmers, fiis Aciror'ney.

Patented Nov. 211, 1944 ELECTRIC DISCHARGE DEVICE Eugene Lemmcrs, Cleveland Heights, Ohio, as-

signor to General Electric Company, a corpora- 1 tion of New York Application August 1, 1941, Serial No. 405,059

3 Claims. ('01. 176-122) This invention relates to electric discharge devices, and is useful in devices producing radiation for various purposes, such as germicidal tubes or lamps, and lamps used industrially for irradiati'ng or treating various substances and products, as well as for fluorescent lighting. The invention is useful for increasing the radiant output, and especially for increasing the resonant radiation as compared to the rest of the radiation from a working substance, such as the 2537 A. line of mercury. It affords means of increasing the electron temperature in an arc discharge, or, in general, of trapping positive ions of such a discharge.

The intensity of resonant radiation has been shown to depend on the electron temperature in the electric discharge producing it; or, in other words, on the electron energy, which depends onthe velocity of electrons in the discharge. More specifically, the resonance radiation in mercury vapor increases with the electron temperature, up to a peak value at between 40,000 and 50,000 K. The electron temperature also varies directly according to the ratio of the area of the enclosing envelope wall to the volume of the discharge. This is due to the fact that the envelope wall collects or traps" positive ions from the discharge, which results in an increase in the electron temperature, with an increased voltage gradient in the discharge to maintain it, and thus involves an increase in the wattage of the discharge. This is in harmony with the increase in resonance radiation resulting from increase in the electron temperature, as set forth above. In accordance with the relation of the electron temperature to the area of the enclosing envelope wall and the volume of the arc discharge as above stated, efforts have been made to increase resonance radiation by using envelope tubes of flattened cross-section, rather than circular, and thus of greatly reduced internal volume. For a similar purpose, baflles have been mounted in a discharge tube to divide it into parallel arc chambers having nearly the same aggregate volume, but a greatly increased aggregate perimeter. A centrally located filler has also been arranged to form an annular arc chamber, of increased perimeter and reduced volume.

The use of flattened tubes is successful as far as increasing resonance radiation is concerned,

to divide:' e. g., it occupies one chamber and refuses to enter the other; or it jumps from one chamber to another in a spasmodic manner. The use of a central filler in a tube results in a spiral canalizing of the arc, and a "precession of the armaround the filler in a manner suggestive of a rotating luminous barbers pole, a striking phenomenon, indeed, but objectionable from the standpoint of steady, uniform illumination. In addition, such measures defeat the purpose of increasing the electron temperature,

because the diffuse low-pressure discharge no longer fills the whole space in the tube (as it normally does) so as to come in contact with or wash all its wall surfaces, as well as those of the added structure; and so positive ions are not trapped out as effectively as when the tube bore is unobstructed.

I have discovered that the effective wall area enclosing the discharge space can be increased, and the space concurrently reduced more or less, if desired, by partition and filler means arranged in thedischarge space upon altogether different principles from those heretofore proposed. Instead of attempting to divide the discharge space in the general direction of its length, or to increase its perimeter from end to end without obstructing the discharge, I deal with the matter in a manner somewhat analogous to the modern military conception of defense in dept in a sort of labyrinthine fashion. For example, I may place across the tube a number of barriers all of which are readily penetrable by the electric discharge in many places, and none of which are of any great thickness, yet which in the aggregate present a very considerable surface but is subject to drawbacks in other respects,

such as non-uniform distribution of luminous.

output around the axis of the tube. The use of baflles to form parallel arc chambers has proved unsuccessful because of the refusal of the are that is effective in trapping positive ions out of the discharge, without anywhere materially interfering with its effective access to the envelope walls.

In the design and construction of such a barrier system in depth, which may preferably occupy the tube from end to end between its electrodes, certain points are particularly to be observed. The thickness of each barrier must be relatively small, as of the order of the length of the mean free pathof an electron in the discharge, so that the latter may spread. out or divide at all the barriers, rather than canalize itself in one single path of least resistance throughout the length of the tube, or jump about spasmodically over a variety of paths. The signiflcance of barrier thickness in thisconnection is that in a length comparable to the mean free path of an electron, the complete arc phenomenon as such is impossible. With a barrier whose' thickness is of this order, such as 1 to 2 mm., the maximum surface area can be obtained that is consistent with sufflcient width of passage past the barrier to avoid interference with normal discharge in a tube of given diameter.

The barriers should allow free passage of positive ions at points relatively close to the outer wall of the tube (i. e., at distances of an order comparable to the mean free path of a photon), so as not toshade or blanket appreciable areas or annular zones of this wall from receiving the desired photon radiation. In this connection, it is to be remarked that the 2537 A. photons do not obey optical laws in a discharge tube, be

touch the tube walls all around; although this statement is not without exception.

Various other features and advantages of my invention will become apparent from the following description of species and forms of embodiment of the invention, and from the drawing. In the drawing, Fig. 1 i a tilted sideview 'of an electric discharge tube embodying my invention in a form suitable for a fluorescent lampa mid-portion of the device being broken out and omitted to permit the drawing to be made on a larger scale; Fig. 2' is a cross-sectional view taken at the break in Fig. 1, illustrating the barrier means in Fig. 1; Fig. 3 is a View like Fig. 2 illustrating another form and arrangement of barrier means, also of sheet material.

As shown in Fig. 1, the electric discharge tube It is provided with the usual emissive coil cathodes H, H at its opposite ends, each of which has portions that act as anode during part of the alternating current cycle, when the device is operated on A. C. For a 15 watt fluorescent lamp, the tube Ill may be of 1 inch internal diameter and about 18 inches long, and may in general be similar in construction and opera-. tion to the positive column fluorescent tubes now in common use, including the usual coating of fluorescent material or phosphor I2 on the cylindrical tube wall. a mercury vapor pressure of substantially less than one millimeter Hg., usually a small fraction of a millimeter, e. g., about flve to twenty microns.

The barrier system for the'purposes of my invention comprises a number of transverse discs or diaphragms l3 uniformly spaced at intervals of about 1 inch throughout the length of the tube. are of the order of the tube bore. As here shown, the discs l3 are mounted on a longitudinal supporting rod I 5 whose ends are attached to the closed ends 16 of the tube It in any suitable manner, as generally indicated at H. The axial rod- 15 should preferably be of not over inch diameter, in order to avoid canalizing and spiralling of the arc discharge around it, as explained above. To an extent, this supporting rod l5 also serves to reduce the volume of the discharge,

Such lamps operate with? Thus the distances apart of the discs I3 without apertures saveat their peripheries; but

while increasing the wall surface. The discs I! and the rod l5 may be of metal (e. g., nickel), for strength, but may preferably have an insulative coating. I prefer, however, to make the discs l3 of white mica, and the rod l5 may be a suitably small glass tube. In general, any thickness of the discs, l3 consistent with desirable lightness will answer: e. g., with discs about mils thick, 9. tube of the dimensions above indicated operates satisfactorily on currents up to 0.5 ampere. For currents higher than this, the discs 13 should be made thinner.

As shown in Figs. 1 and 2, the discs I: are

the discharge in the tube It is allowed to pass by. means of apertures annularly distributed around the disc peripheries: e. g., the discs I! may be of only inch diameter, as compared the characteristics of its transverse barrier discs I30. and their relations to the discharge tube wall. Instead of being smaller than the bore -of the tube Illa, these discs I30, fit the tube bore peripherally with an easy fit; and instead of being imperforate, each'of them has a multitude of holes 18 which may be of about inch to inch diameter, arranged in annular rows around the central supporting rod 15a. The hole [8 may be so arranged that those in the outer row through the disc margin are about ,4; inch from the tube wall, and so that the shortest distance between holes in a row and between holes in adjacent rows is also about A; inch.

Situated and well spaced apart across the path of the discharge in the otherwise unobstructed tubes 10, Illa, the barrier discs l3, l3a are washed all over their surfaces by the discharge, which also spreads out around and-between them to fill the spaces between discs and wash the tube wall surfaces all over, and are effective in trapping out the relatively heavy positive ions,

thereby increasing the electron temperature in the discharge, the resonant radiation of the dis charge device, and its wattage. Of course positive charges build up on the barrier discs and supporting rod which tend .to repel positive ions, and must-be taken into account in the design of the barriers. Preferably, the barrier discs l3, l3a and the rod I5 are coated with fluorescent material or phosphor as indicated at 20 in Fig. 2, in addition to the phosphor coating l2 on the \wall of the tube Ill, so as to increase the luminous output. When this is done, the coating of phosphor l2 on the tube walls should not be too dense or opaque, in order to obviate excessive absorption of light from the phosphor 20 on the barrier system by that on the tube wall at l2, here shown as inside the tube.

lengthwise of the tube, and an lonizable operating atmosphere in the tube having in operation a pressure giving the discharge the abovestated diffuse character, of a series of discs of sheet material severally thinner than the mean free path of electrons in the discharge and mounted crosswise of the tube at distances apart which are of the order of the tube bore, but affording annularly arranged openings adjacent the tube wall for the passage of the diffuse discharge, which thus spreads around the tube at the discs through said annularly arranged openthe discharge tube during operation, as distinguished from high pressure and a discharge constrlctedthereby into av narrow cord, the combination with a discharge tube having its bore substantially unobstructed except as hereinafter set forth, discharge electrodes in said tube spaced apart to provide for a positive column discharge lengthwise of the tube, and an ionizable operating' atmosphere in the tube having in operation a pressure giving the discharge the above-stated difiuse character, of a series of discs of sheet material severally having a thickness of the order of thirty mils and a size of the order of half the tube bore, and mounted centrally in said bore crosswise of the tube at distances apart which are of the order of said bore, whereby the discharge spreads annularly through the openings around the discs, besides occupying the whole cross-section of the tube between adjacent discs, thus washing both the disc surfaces and the tube walls.

EUGENE IEMMERS.

Images