US 2444397 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
June 29, 194:; w; mm 4 4 2,444,397
' mcmtc mscmnax 1.4m 4
Filed Ka en 21, 1945 IN VEN 70R.
J S- Z. SrAnroRo Wmsnm can" av M W AnonusY an; .Iune 29.1948 v g f A 2,444,397 v UNITED STATES PATENT orncs ELECTRIC DISCHARGE LAMP Stanford Winston Cram, Flushing. N. Y., asslgnor to Sylvania Electric Products Inc., Salem, Masa, a corporation of Massachusetts Application March 21. 1945, Serial No. 585,033
1 2 Claims. (01. 176-122) My invention relates to electric gaseous dis- 4 the like in front of each electrode but not con-, charge tubes, and particularly to electric gasenected to either electrode. A further feature is ous discharge lamps and to an arrangement and a conductor for connecting the rings together. method for starting such lamps. Still another feature is a filling of inert gas at Heretofore lamps of this type have been started I comparatively high pressures such as 5 to 20 by connecting the filaments in series through millimeters, with the mercury pressure correan external circuit during starting. Thus. a' spondingly raised from its usual value of 8 miheating current is allowed to flow through the crons when 4 mm. of inert gas such asargon electrodes and after afew .seconds the starting is used. My, invention is not to be understood circuit is opened, thus causing a peak voltage 10 as limited to such high pressures; however, as to be pressed across the lamp to start it. In certain features of it, suchas the interconnected order to dispense with the time delay in startrings. are useful at low P ess es l ing lamps according to the above process, it has My invention will be more fully set forth in also been suggested to start the lamp with cold the following description referring to the accathodes by impressing a sufficient voltage across companying drawing, and the features of novelty the lamp tostart it practically instantly. Anwhich characterize my invention will be pointed other arrangement for providing substantially out in the claims annexed to and forming a part instantaneous starting without the use of a sufliof this specification. l
ciently high voltage to force an arc to strike di- R f ing t th awin 8- 1 I on i irectly between the electrodes is to employ probes nal sectional view of the device according to my and mechanisms between the electrodes to iniinventiom-Fig. 2 is a perspective view of one tiate a glow. It has also been suggested that e d 01 t e tube of 18- 8- 3 illustrates a modthe glow-producing devices be electrically conification of the auxiliary electrode construction; nected together externally of the lamp or within Fig. 4 is a curve of the voltage drop along the the envelope when the metallic connection is 95 lamp and will be used in a description of my otherwise insulated from the arc discharge. invention; and Fig. 5 illustrates the circuit for However, even with such glow-producing mechstarting my improved lamp. anism, it is still necessary toprovide sufficient Referring to the drawing, I have illustrated a voltage to cause the glow arc to transfer from tubel of glass or other light-transmitting mateone end to the other. rial sealed in each end by stems l8 and I1, one It is, therefore, an object of my invention to of which will ordinarily contain an exhaust tube. provide an arrangement for starting an elec-. ii. A coating of fluorescent material capable of tric gaseous discharge lamp relatively instantaneexcitation by the discharge in the tube. may be ously and without the necessity of providing sumprovided if desired to enhance the emission of cient voltage to force the glow arc transition to light or other radiation from the tube. At each be directly initiated between the two cathodes. end of the tube I. there are provided electrodes A further object of my invention is to pro- 3 and 4, preferably but not necessarily a coiledvide an electric discharge lamp which maybe coil tungsten wire coated with one or more of easily started without excess voltage under northe alkaline earth oxides, which is attached to mal conditions and also under high ambient hulead-in wires 5. 6, I, and l, which are in turn midlty conditions. sealed through the stems I and i1. 1 A further object of my invention is to pro- In order to facilitate starting of the lamp at vide a lamp which startseasily when filled with relatively low voltage, I provide auxiliary'elec inert gas with pressure of 5 to 20 millimeters of trodes, metallic members or rings 9 and lll'fwhichf mercury. are composed of any suitable conductive?mate- A further object is a lamp which utilizes a rial which are placed in front of orfl'adjacent vapor in addition to the'inert gas, and maintains to the electrodes 3 and I. The ringsmay be of its eillciency at such comparatively high gas presa diameter nearly as great as that of the tube} sures by operating with a mercury vapor pressure I, and they should be connected by a conducting above the usual value of about 8 microns used in wire H. I! or the like, for exam-pie, of metal of low pressure fluorescent and germicidal lamps. low resistance. This might appear toshort-cir- Still another object is a compact lamp of high cuit the portion of the discharge between the efllciency and higher brightness than the present two rings 8, ill, but I find that this is not the commercial lamps. case, and the discharge passes regardless of this A feature of the invention is a metal ring or II apparent short-circuit. One conductor, for ex,
ample, conductor Ii, is sufficient connection between the two rings additional conductor I! to provide a firmer support for the two rings 9, l0. One or both rings 9, l may be supported from a stem. such as I6, by' support wires l3, l4, which are sealed into the stem l6 from which they extend to attachment with ring 9. The other ring l0 may be similarly supported from stem I1, if desired, for additionalrigidity particularly if the tube l is long. If both rings 9, l0 are supported from their respective stems l6, H, the support wires extending from one of the stems may fit into sockets set in the corresponding ring, for ease of assembly. If both rings are-rigidly mounted on their respective stems, prior to assembly of the lamp, the wire I l connecting them may be made fine and flexible so that it will not interfere with the stems being pushed inward for assembling and sealing.
The connecting wires Ii, I! work well when fairly close to the tube wall I, but if the lamp has a fluorescent coating, they may be spaced somewhat from the coating l to prevent shadows thereon. In some cases, the rings 9, l0 may be formed directly on the glass tube l by condensing metal thereon.
The electrodes 3, 4 will retain their emittin material longer at higher inert gas pressures. and
I find that my starting rings also work somewhat better at high pressure. pressures even as high as 15 mm. are satisfactory. If a vapor such as mercury is used with the inert gas, the efliciency of radiation emission from the vapor, for example, the efficiency of radiation of the 2537 Angstrom Unit wave length of mercury is used, will not ordinarily be as great as at lower gas pressures, if the operating temperature of the lamp is not raised to increase the mercury vapor pressure correspondingly, so that the mercury atoms will be in about the sam .proportion to argon atoms as in the lower pressure lamps. Some increase in temperature will come from the increase in gas pressure. because of the consequent increase in lamp operating voltage, if the current is kept the same; but in some cases it may be necessary to increase the current somewhat, or even to change to a different inert gas to get a higher voltage drop across the tube. In this way, for example, we may have a morecompact lamp; for example, a fluorescent lamp of the geometrical size now used commercially for 9. 20-watt lamp, may be used at 40 watts, possibly at the same current and voltage as a 40-watt lamp.
My construction is particularly suitable for germicidal and sterilizing lamps, such as those used in domestic refrigerators, where compactness and low starting voltage is essential, even under the high humidity condition of some of the compartments in the newer refrigerators.
The electrodes 9, l0 may be considered for the purposes of this application as activated electrodes, for convenience in phraseology. The rings 9, l0 should not be activated, because if they are, the discharge will go directly to them from the main electrodes, and will not pass through the gas in the path between the rings.
The rings should be placed outside the region of the so-called -dark space" near'the electrodes.
With the pressures used, this will mean that the ring should clear the electrodeby about inch in most cases. I
.In the foregoing, I have generally referred to the auxiliary electrodes near the cathodes as rings for convenience, but other suitable forms .9, l0, but I have shown the of electrode may, of course, be used. For example, I have found the spiral wire electrode l8 of Figure 3 to be very satisfactory. The spiral may be in a fiat plane just parallel with the end of the cathode coil. With this electrode, and a gas filling comprising in addition to mercury vapor a filling of argon mixed with some neon, in amount up to 30% of the argon, I have been able to start a 48-inch long, 1% inch diameter, lamp at 210 volts with no preheating of the electrodes. Under such circumstances, my device provides a localized heating of the cathode bythe intense glow current, which makes the cathode capable of supporting an arc, and then the transition from glow to arc along down the tube is facilitated by the glow which occurs along the conductor connecting my spirals or rings. My device is thus most effective at gas pressures which permit the formation of such a glow along the conductor prior to the actual arcing of the lamp.
I have found that a lamp formed with spiral auxiliary electrodes such as illustrated in numeral ill of Fig. 3 provide a very improved and efficient arrangement for starting an electric gaseous discharge lamp. The steps in starting are as follows:
1. When a voltage is impressed across the lamp, which may be connected in a simple circuit as illustrated in Fig. 5, a glow will first show between the cathode and auxiliary electrode l8.
2. The glow then increases from the auxiliary cathode area to the wire along the column.
3. The glow then spreads from both ends of the tube cascading along the wire toward the center, and just before the glows on the wire meet, the glow shifts to the gaseous column.
4. The glow then shifts to an arc, and the lamp will then run in this final condition.
As has been mentioned above, the rings or auxiliary electrodes should be placed adjacent to the cathodes but outside the dark space. Referring to Fig. 4, I have illustrated a conventional lamp characteristic in which voltage is plotted on the ordinate axis and distance along the tube is plotted on the abscissa axis. It will be is reached.
As has been stated the auxiliary electrodes should not be placed closer to the cathode than that point corresponding to that point marked A shown on the curve, and it is also preferable that the auxiliary electrode not be placed further from. the cathode than a distance corresponding to that point B" on the voltage characteristic curve. If the auxiliar electrodeis placed further away from the cathode than the point correspondingly indicated by the point B, a correspondingly higher voltage will be required in order to initiate the glow between the cathode and the auxiliary electrode.
I have found that in order that the auxiliary electrode should not shield the electrical field existing between the two cathodes that the auxiliary electrodes should be placed radially from and in the same plane as the cathodes, such a construction being illustrated in Fig. 3 of the drawing. It will also be seen from an inspection of Fig. 3 that the inner coil of the auxiliary electrode is spaced radially from the cathode, and this radial space may be considered so far as the characteristic of Fig. 4 is concerned as a space along thetube.
I have further found that the area presented by the auxiliary cathode should be sufiiciently' large to cause a glow to be initiated by the cathode and the auxiliary electrode and still should not be too large so as to prevent the cascading of the glow along thewire. which connects the two spiral electrodes. For example, a 40-watt lamp which I tested using 7 millimeters of an argon-neon mixture and with about 8 microns of mercury operated satisfactorily with auxiliary electrodes having three convolutions as illustrated in Fig. '3, the auxiliary electrodes and connection being formed with the wire having approximately .03 inch diameter. In this particular device which was tested, the wire was formed of nickel and the resistance of the connection between the auxiliary electrodes was less than one ohm. It is desirable to have the value of the resistance of the connection between the auxiliary electrodes as low as possible so as to facilitate the cascading effect of the glow along the wire. I further found that a 40-watt fluorescent lamp having the characteristics described above started practically instantaneously when a voltage of between 210 and 260 volts was impressed across the cathodes.
Although I have shown and described particular embodiments of my invention, I do not desire to be limited to the embodiments described, and I intend in the appended claims to cover all modifications which do not depart from the spirit and scope of my invention.
A discharge tube according to my invention 85 may, of course, be used in various circuits, but I have found the circuit of Fig. 5 particularly useful. The autotransformer I9 is connected to the usual 110 volt A.C. line and through the inductance coil 20, which acts as a ballast to limit the rings 9 and I0, and their connecting conductor *current, to the lamp electrodes 3 and 4. The I H' are also shown, the entire diagram of connections being merely schematic. The coil 20 may be merely the leakage reactance of the transformer 49 itself, if the leakage reactance is large enough.
but outside the dark space for the What I claim is:
1. An electric discharge device comprising a tubular envelope, a filamentary electrode at each end thereof, a coating of at least one of the alkaline earth oxides on said electrodes, a plain metal ring electrode in front of each of said filamentary electrodes and of a diameter nearly as great as that of the tubular envelope, a filling of inert gas and mercury vapor in said envelope, and a pair of low resistance wires within said envelope electrically connecting said rings together and placed near and along but spaced from the wall of said envel pe, said wires being directly exposed to said gas inside said envelope and out of physical contact with said filamentary electrodes.
2. The combination of claim l'in which the inert gas pressure is above 5 millimeters and the mercury vapor pressure about 8 microns during operation, and, with the ratio of inert gas pressure to mercury vapor pressure being of the order of 500 to 1, and in which the ring electrodes are close to the respective filamentary electrodes gases and pressure used.
' STANFORD WINSTON CRAM.
REFERENCES CITED The following references are of record in the file of this patent:
UNITED STATES PATENTS Number Name Date 1,930,147 Pirani Oct. 10, 1933 1,961,618 Machlett June 5, 1934 1,968,823 Gaidies Aug. 7, 1934 1,980,534v Kirsten Nov. 13, 1934 2,042,147 Fairbrother May 26, 1936 2,085,561 Wiegand June 29, 1937 2,133,206 McCauley Oct. 11, 1938 2,182,732 Meyer Dec. 5, 1939 2,225,712 Price Dec. 24, 1940 2,301,670 Abadie Nov. 10, 1942 2,351,254 Elmendorf June 13, 1944