US1020323A - Electric lighting. - Google Patents

Description

C. P. STEINMETZ.

ELECTRIC LIGHTING.

APPLIOATION FILED AIR. 2s. 1900.

1,020,323. Patented Mar. 12, 1912.

2:4 e o E5 4 AN@ P 2 lg 9 Nlz Charles P Stenmet z.

Oqtt H.

UNITED sTATEs PATENT oEEioE. l

CHARLES P. STEINMETZ, OF SCHENECTADY, NEW YORK, .ASSIGNOR TO GENERAL ELECTRIC COMPANY, A CORPORATION OF NEW YORK.

ELECTRIC ALIGHTING.,

Specication of Letters Patent.

Patented Mar. 12,1912.

Application led April 25, 1900. Serial No. 14,224.

To all whom it may concern:

Beit known that I, CHARLES P. S'rEiN METZ, a citizen of the United States, residing at Schenectady, in the county of Schenectady and State of New York, have invented certain new and useful Improvements in Electric Lighting, of which the following is a specification. l

It is well known that there are in existence various devices for transforming the energy of electric current into light, which are superior in eiiciency to the ordinary incandescent lamp of commerce, but are 0bjectionable in that they require, for one reason or another, extremely high potentials in starting as compared with the potentials necessary to maintain them in operation. One lamp of this class is that in which the light-giving body is composed of what is sometimes called la conductor of the second cla-ss, or, as I prefer to call it,a pyro-electrolytic conductor, usually taking the form of a rod of oxid of magnesium or other metal oxid, or compound of metallic oxids. A lamp of this type may be designed to run successfully on a single-phase alternating circuit of 100 volts, but requires for its starting a potential which may rise to 500 or 1000 volts, or higher. By using as light-giving body a mixture of initially conducting and non-conducting substances, as a mixture of ferrite and chromite of iron, FeO4 and CrZFeCL, as set forth and claimed in my application Ser. No. 41,350, filed Dec. 28, 1900, this voltage may be reduced, but even with this improvement a starting device is advantageous.

Another lamp, possessing to a certain extent the same characteristics, is that described in my pending application, Ser. No. 7265, in which light is formed by the passage of an arc between electrodes vaporizable at low temperature, which electrodes frequently take the form of bodies of mercury. In some of the .forms shown in said application the mercury electrodes are inclosed in a glass vessel, from which the air has been exhausted to such an extent as to reduce the voltage necessary to strike across and start the arc, but even with the most favorable adjustment of the vacuum I find it difticult to start such lamps in multiple on ordinary constant-potential circuits without the addition of some extraneous starting device, although they are well adapted for use in connection with constant-current circuits.

I have, in the above-mentioned application, shown certain other lamps possessing the same general characteristics but arranged to start by relative movement between two electrodes, and such lamps are very suitable for certain classes of work. For certain other purposes, however, I find it better t0 dispense with the movement between the electrodes and to start the lamps by subjecting them to a very high potential. This is, as has been stated, extremely leasy in connection with series circuits, alternating or direct, but difficult in connection with commercial low-tension, multiple arc or constant potential circuits.

My present invention aims to provide a simple and convenient means for starting, on ordinary constant-potential alternating clrcuits, any lamps which require higher voltage at starting than during the running condition, whether of the type described or of other types.

In the drawings attached to this specification, which represent various modifications of my invention, Figure 1 is a view of a lamp of-the general character described in my application, Ser. No. 7265, provided with my improved starting device; Fig. 2 is a cross-section of a lamp based on the general principle of the lamp of Fig. l, but provided with certain improvements, and Fig. 3 is a view of an electrolytic lamp to which my present invention has been applied. Fig. 4 is an illustrative diagram.

Referring more particularly to Fig. 1, E, E5, are terminals between which is maintained a substantially-constant alternating electromotive force, as by connection to the mains 2 2. From these terminals wires lead to the terminals E, ET, of the lamp B, which lamp consists of two glass tubes joined to a central chamber, as shown, with conducting points fused intoV the glass at E,E7. A third terminal, entering a depression or cup D4,'is provided as shown. The lower ends of the tubes, D, D7, and the cup D* are filled with mercury, or other suitable substance, as set forth in my application above referred to, and the surface of the mercury is in each case covered with a thin layer of a color-modifying substance such as a mixture of anhydrous halogen salts of earth metals, as iodides of lithium substance, as calcium fluorid.

and sodium, and some neutral or dilutant Two impedauces of different power factor, -a condenser 4 and a self-induction 5, are connected in series between the terminals. E4, E5, and a connection is made from a common point E8 between the condenser and gie self-induction to the central electrode With the parts arranged as follows, let

` us suppose that the switch is closed and that alternating currentat a potential, for example, of 100 volts be allowed to pass to the terminals, the vacuum in the lamp being supposed to be such as to produce the minimum dielectric strength, as for example, l mm. on the ordinary mercury gage. yUnder these circumstances the voltage 4, E1", will v form the base of a triangle, Fig. 4,-` of which the voltages E4, E8, and E8, E5, are the sides, and if the ohmic resistance of the self-induction coil 5, and the energy lost in the condenser, 4, are comparatively small the voltages E4, E8, and vE8, E5, may each of them be very many times the magnitude of the resultant voltages E4, E5. In the theoretical extreme case the voltage across E4, E8 would lag t-he voltage E4, E5 by 90 degrees, and the voltage E5, E8, would lead the voltage E4, E5, by 90 degrees, as shown in Fig. 4 in dotted lines, so that the phase triangle would be one of infinite altitude, and the voltages E4, E8, would be infinite. In practice this condition cannot occur, but. these voltages may be ten times the impressed voltage, or even higher. This being the case, it will be seen that that portion of the tube between the left-hand terminal D and the central terminal D4 will be exposed to the total voltage across the con denser 4, while the right-hand portion of the tube will be exposed to the tot-al voltage across the self-induction 5, so that although the total voltage across the tube may be only 100, yet the voltage across each half of the tube may exceed 1,000. This high voltage will cause the air, hydrogen, or other rarefied gas in one or bothvsides of the tube to break down, and an arc will be started from the left-hand terminal Dj to the central terminal D4, or from the central terminal D4 to the right-hand terminal D7, or both; If only one arc starts at first, the other will tend to be started, even on lower potential, since the whole tube is warmed i and lled with conducting vapors. This arc is formed of a column of mercury vapor evaporated from the terminals, and mixed with this mercury vapor are the halogen salts above mentioned, as, for example, lithium iodid. The mercury arc gives a bright light at high eiiiciency, and the lithium adds to the lightthe red line which is wanting. The lithium would attack the glass envelop and destroy it, but for the vmaintain the arc becomes very multiple with and quantity of the amount of color-modifying substances introduced, issues continuously from the tube.

The arc gradually warms the mercury until it reaches a temperature nearer the boiling point, when the voltage necessary to greatly reduced and the current ow; for a given voltage ,correspondingly increases, but as fast as current is taken on the left-hand side of the tube, for example, a resistance is introduced in multiple with the condenseig'and as fast as current is taken on the right-hand side of the lamp a resistance is interposed in the self-induction. The effect of resistance in multiple with a selfinduction or condenser is to decreaseits reactive effect, which causes the triangle of phases to collapse, as it were, as indicated in Fig. 4 by the triangle E4E5 EB. But if for any cause the arc should be extinguished on either side of the tube, as might happen, for example, if a very sudden-draft of air should cool one of the electrodes, the phase triangle would extend, the voltage would rise, and the lamp would automatically start itself again. The current is then forced to flow through the lamp under all conditions and automatically regulated.

It will be seen that my improved starting device is perfectly automatic and instantaneous in its action, and that it does not require 'any moving parts or hand regulation.` The relative values of the reactance of the condenser and self-induction should be adjusted in accordance with the wellknown formula to produce a resonant condition, though accuracy in this resp..:t is` not absolutely essential. No series resistance or other steadying device is necessary, and the efficiency is very high.

Referring now more particularly to Fig. 2, it will be seen that the tube B appears in a U-shape, with a cent-ral terminal D4 at the lower part and two condensing chambers G1, G2, containing the electrodes D", D7, which electrodes are annular in Shape, well shown in the drawing. The tube B is supported on a baseplate N 8,' which in turn is carried by the external casing N12, urovided with terminals'E4, E5. The tube B is held in place by a protecting plate N43 with an asbestos washer N14 interposed. A condenser 4 and a self-induction coil 5 are suspended from the cap N6, and are connected according to the scheme of connections shown in Fig. 1. The operation of this lamp will be readily understood from what has. preceded, and further description is, therefore, unnecessary.

Having thus described the operation of my invention-as applied to the lamp shown in my application Ser. No. 7265, I will nowv proceed to describe its applicationI to the pyro-electrol tic lamp shown in Fig. 3. In this figure is a rod lformed of metallic oxids, either pure or mixed, as is well understood in the art. This rod is carried by a central support 6, and by side supports 7 and 8. The side support 7 is connected to the terminal E, and the side support 8 is connected to the terminal E". The condenser 4 and the self-induction coil 5 are connected in series, and from the common point is taken a lead to the central post 6. Obviously, the lamp may be suspended in an inverted position, and supplied with a glass globe, as shown in Fig. 2, if so desired. In starting, the rod does not take any substantial amountof current, owing toits very lhigh resistance, but the voltages across the condenser and self-induction rise to a considerable value, as above explained, until finally current is This action may be aided by external heat,

if it is so desired. Asfsoon as current begins f in'the other above described, it is 'for one side to start first, and to re luce the to flow through the rod the phase triangle collapses and the voltages are reduced, as shown in Fig. 4. In this form of lamp, as ossible voltage'on the other side. If the rod be long and thick one side only will burn, and the single lamp thus formed is within my present invention, but by making the rod fairly short and thick the active side will be caused to heat the inactive side, whereby the whole lamp will be started.

What Il claim as new and des'n'e to secure by Letters Patent ofw the United States is 1. The combination with a vapor electric device having an exhausted tube and at least three electrodes therein affording a plurality of paths for current flow, of a source of alternating current, means for deriving from saidf source a plurality of voltages out vmy hand this 21st day of April, 1900.

of phase with each other, and means whereby said out-of-phasevoltages are impressed upon said electrodes whereby coperating y currents are caused to flow in different paths in said device. i

2. The combination with a vapor electric forced through the rod.'

devicer having an exhausted tube and at least three e ectrodes therein, a source of alternating current, means for deriving from said source high voltages for starting the arc-in said tube, and means for automatically reducing the voltage as the current in the tube increases.

3. The combination with an exhausted tube provided with whereby a pluralit of paths for current in the tube is aor ed, of impedances, and meansfor subjectinfr the tube at starting' to high alternatingvoijtages created by the interaction of said impe ances.

4. The combination with an electric lamp of two impedances of different power factor connected in series across an alternating circuit, and means for supplying separate portions of the lamp with the voltages created by the resonant rise between said impedances. l v

5. The combinationdwith an electric lamp which requires in starting a voltage greater than that required in normal operation, of a condenser and self-induction in series with each other across the circuit, connections from terminals of the lamp to the sides of the circuit, and a connection from an intermediate portion of the lamp to a point between the self-induction and the condenser.

6. The combination vwith lan electric lamp 'which requires in starting a voltage higher than that required in normal operation, of a condenser and self-induction in series with each other across' the circuit, connections from terminals of the lamp to the sidesl of the circuit,l and a connection from an intermediate portion of the lam to a point between the self-induction an the condenser.

7. The combination with a hermetically sealed envelop provided with at least three electrodes one at least able material, a source of alternating current connected directly to two of the electrodes, and impedances in shunt to the source having a point therein connected to the third electrode.

In witness whereof I have hereunto set CHARLES P. STEINMETZ.

Witnesses:

BENJAMIN B. HULL, MABEL E. JAooBsoN.

at least three electrodes of which is of vapomz-

Images