US1092410A - Means for rating incandescent lamps. - Google Patents

Description

A. J. SWEET.

MEANS FOR RATING INGANDESGENT LAMPS.

I APPLICATION FILED MAY 16, 1910.

Patented Apr. 7, 1914.

2 SHEETSSHEET 1.

.Per Ccm Candle Big.

WITNESSES:

vzggw MLWM ATTO'RNEY COLUMBIA PLANOURAPH CO, WASHINGTON, D. c.

A. J. SWEET.

MEANS FOR RATING INGANDESOENT LAMPS.

APPLIOATION FILED MAY16, 1910.

Patented Apr. '7, 1914. 2 sums-sum 2.

6 RP. 56 WATT CARBON 6 5.256 WAT S-CARBON ATTdRNEY COLUMBIA PLANDGRAPH C0,.WASIHNGTON, n. c.

'tlNlTTED STATES PATENT OFETQE.

ARTHUR J. SWEET, OF NEWARK, OHIO, ASSIGNOR TO WESTINGHOUSE LAMP COMPANY, A CORPORATION OF PENNSYLVANIA.

MEANS FOR RATING INCANDESCENT LAMPS.

Specification of Letters Patent.

Patented Apr. a, 1914.

Application filed May 16, 1910. Serial No. 561,621.

To all whom it may concern:

Be it known that I, ARTHUR J. SWEET, a citizen of the United States, and a resident of Newark, in the county of Licking and State of Ohio, have invented a new and useful Improvement in Means for Rating In candescent Lamps, of which the following is a specification.

My invention relates to photometers that are utilized by incandescent lamp manufacturers and others for determining the ratings of lamps, and it has for its object to provide a method of and means for eX- pedit-iously and conveniently determining the rating of an incandescent lamp in terms of the watts consumed per candle power at a given voltage or amperage of the lamp.

In order that purchasers of incandescent lamps may be informed of the voltage or current at which each lamp is best adapted to be used, it is essential that one elementof the rating, that is marked upon a lamp by the manufacturer, should be such voltage or current value. Also, in order that the character of service for which the lamps are adapted may be known, another element of the rating should be the candle power of light that the lamp will give at the rated voltage or current, or the watts consumed by the lamp at the rated voltage or current, or the watts consumed per candle power of light at the rated voltage or current.

It has heretofore been customary to rate lamps in terms of candle power at a given voltage or current, but such a rating is not entirely satisfactory because it does not give definite information regarding the efficiency of the lamps. Since the watts consumed per candle power of light given by a lamp is a measure of its efficiency, a rating in terms of watts per candle power, at a given voltage or current, possesses many advantages over the common rating in terms of candle power at a given voltage or current.

The life of a lamp depends upon the temperature at which the filament is operated. The temperature of the filament is not a function of either the candle power or the watts consumed alone but is a function of the watts per candle. In order to obtain the most uniform life, therefore, from a lot of lamps, the filaments should all be operated at the same temperature, which means that the lamps should all be rated in terms of the same watts per candle. It follows that if all lamps are rated by the manufacturer in terms of watts per candle, at a given voltage or current, it is possible for the purchaser to obtain greater uniformity of life from a lot of lamps of the same rating than is possible under the present system of rating. Moreover, it has been found, in practice, that the average life of a lot of lamps having the same rating in terms of watts per candle is greater than the average life of a lot having different, or widely varying, ratings in the same terms, but the average of the ratings of which is the same as the uniform rating of the first lot. For instance, the average life of a lot of ten lamps rated at 8.5 watts per candle is greater than the average life of a lot of ten lamps in which five are rated at 3.4: watts per candle and five at 8.6 watts per candle. A greater average life results in a lessened cost of renewal lamps to the purchaser. has been found further, that, with a given regulation of the operating circuit voltage, there is a certain lamp efficiency wthich means lowest cost of operation, so that, if lamps are rated in watts per candle, which is a measure of a lamps efliciency, those that are adapted to give the most economical operating conditions may be selected.

lVhile the advantages of rating lamps by the voltage and the watts per candle power, rather than by the voltage and the candle power, have heretofore been appreciated, such a rating has not been utilized commercially because of the difficulties and expense of determining it. The heretofore known methods of rating a lamp in terms of the voltage and the watts per candle power have usually been, in a practical, commercial way, cut and try methods, and require the taking of several sets of instrument readings, or the making of computations, before even an. approximation of the watts per candle power can be arrived at. According to the present invention, however, with a certain assumed watts per candle power, the voltage necessary to give that efficiency may be determined quickly and accurately with one set of instrument readings and without the exercise of an unusual or a high degree of skill on the part of the operator. Thus the desired rating in terms of the watts per candle power and the corresponding voltage for any particular lamp may be obtained in such a simple and effective manner as to render the method capable of commercial utilization.

The curves of Figure l of the accompanying drawings are characteristic curves of lamps which are useful in explaining the present method and apparatus. Fig. 2 is a diagrammatic view of apparatus embodying, and for practising, the present invention. Fig. 3 is an enlarged face view of a portion of the said apparatus. Fig. 4 is a sectional view along the line IVIV of Fig. 3 of the same and Fig. 5 is a face View of a modification of one of the parts of the apparatus.

The present invention is based upon the fact that the characteristics of all lamps of the same type, or having the same structural and physical features and properties, are identical or substantially so, and also upon the fact that a lamp may be caused to operate at any desired watts per candle, within limits, by simply adjusting the voltage or current applied to it. The characteristic curves of lamps demonstrate these facts and also permit the determination of the exact amount of voltage or current adjustment that is necessary to effect a desired change in the watts per candle of a lamp. The curves of Fig. 1 are certain of such characteristic curves and represent the average of the results of tests of a considerable number of lamps that had previously been accurately rated at watts per candle, which is the nominal efliciency of most lamps having carbon filaments. The tests consisted in obtaining measurements from the lamps at diiferent voltages, ranging, by approximately half volt steps, from about 12% below the voltage at 3.5 watts per candle to about 12% above the voltage at 3.5 watts per candle. From these curves, it isseen that a certain percentage increase or decrease of voltage will cause a certain percentage incrase or decrease of candle power and will also cause a certain decrease or increase of watts per candle or efiiciency. The voltage of a lamp at any given watts per candle may, therefore, be readily determined from the curves. For instance, assume that a lamp consumes 56 watts when its candle power is 17.5 and its voltage 108.6. The efficiency of the lamp is then 3.2 watts per candle. If it is desired that the lamp should operate at an efliciency of 3.5 watts per candle instead of 3.2, the voltage for such an eificiency is ascertained by referring to the curves of Fig. 1 and locating the point upon the watts-per-candle curve corresponding to the value 3.2, which point is at A. The voltage value corresponding to this watts-pen candle value is at B upon.the voltage curve directly above the point A and upon the same ordinate. This voltage value is found to be 102.4%; that is, 108.6 volts is 102.4% of the voltage at which the lamp will give an efliciency of 3.5 watts per candle power, which voltage is 108.6 divided by r 1.024 or 106 volts. Then, if the lamp is operated at 106 volts, it will give an efficiency of 3.5 watts per candle power. This is only an example from which it is seen how the curves may be used in other specific cases. When applied to shunt lamps, therefore, the method consists broadly in first measuring the candle power and voltage ofv each lamp when it consumes a certain number of watts, which is the same for all lamps of the same kind, and which is preferably the nominal wattage of the lamps. Then, in order that all of the lamps may be rated at the same efficiency, a correction is algebraically added to the measured voltage, which correction represents the difference between the measured voltage and the voltage at which the lamp gives the desired or chosen watts per candle elliciency, the correction being deter mined from a previously prepared chart, diagram or scale similar to Fig. 2. lVith series lamps, the current is measured, of course, instead of the voltage, but the method is the same, in other respects, as for shunt lamps. V

In practising the invention in connection with shunt lamps, the candle power of the lamp 1 to berated is compared with that of a standard or pilot lamp 2 by. the usual photometric means, here indicated as a screen 3 that is movable between the two lamps. The lamp 1 is caused, by means of a rheostat 4, to consume a predetermined number of watts, or its nominal wattage,

which is measured by a wattmeter 5 and which, in the case of ordinary carbon filament lamps 1s usually 56, and the voltage of the lamp 18 measured by means of a voltmeter 6 when the lamp consumes the given number of watts. At the same time, the photometric screen is operated until a balance is secured, the position of the screen being indicated by means of a pointer or indicator 7 that is connected therewith and is carried by a slide 8 operating in a grooved stationary member 9. Another slide 10, that is provided along the edges of its outer face with scales l1 and 12, is mounted in the stationary member 9 and is adapted to be operated manually. The scales 11 and 12 are calibrated in volts, each division being proportional, in length, to the change that is produced in the candle power of lamps of the same kind as the lamp being rated by a change of a volt in the voltage applied to the lamps, the calibration being determined from the curves of Fig. 1 and from the photometer in the manner above indicated.

Upon the stationary member 9 is a mark or index 13, opposite which the division of the movable scale 12 corresponding to the voltage, as measured by the voltmeter 6, is placed during the operation of rating the lamp. Since the position of the pointer or indicator 7 is dependent upon the candle power of the lamp being rated, the distance between the index 13 and the pointer or indicator 7 represents the number of volts that should be added to or subtracted from the measured voltage of the lamp in order to produce such a change in the candle power and wattage of the lamp that it will operate at the given efficiency. The final step in the method, therefore consists in reading upon the scale 11 the voltage opposite the pointer 7 this voltage being that at which the lamp will give the desired watts per candle eliiciency. Thus, specifically and briefly, the method consists in measuring the voltage and candle power of the lamp when it consumes a given number of watt-s, then so adjusting the slide 10 that the division of the scale 12 which corresponds to the measured voltage is opposite the stationary index 13, and finally reading upon the scale 11 opposite the pointer 7 the voltage at which the lamp will give the desired elliciency, as measured in watts per candle power.

The scales upon the slide may be calibrated for any given voltage at which the lamp consumes the predetermined number of watts, and they will, of course, be entirely or exactly correct only when used for lamps that consume the predetermined number of watts at the calibrated or given voltage, though they are sufliciently accurate for practical use in the rating of lamps that consume the predetermined number of watts at voltages differing by two or three per cent. from the given or calibrated voltage. In practical operation, therefore, two or three slides bearing scales are suflicient, and, in order to minimize the number oi slides, both faces thereof may be calibrated. As illustrated, the slide of Fig. 3 is calibrated for lamps that consume 56 watts at 110 volts, but it is suitable for use in the rating of lamps varying in voltage from 107 to 113. These limits are conveniently designated by means of adjustable pointers or indicators 14: and 15 that are secured to one edge of the stationary member 9.

In Fig. 5, a slide is shown that is calibrated for a lamp that consumes 56 watts at 116 volts, but it is suitable for use with lamps varying in voltage from 113 to 119. Two slides, or a single slide with scales on each face, therefore, provide for a range of rating between 107 and 119 volts, and

others may be provided if a greater or different range is desired.

The method has been explained specifically as applied to the rating of shunt lamps, but, as before indicated, it is applicable also to the rating of series lamps, in which case the current will be measured, instead of the voltage, when the lamps consume a given number of watts, and the scales upon the slide 10 will be calibrated in amperes, rather than in volts, curves similar to those of Fig. 1, with current measurements substituted for voltage measurements, being provided as a basis for such calibration.

I claim as my invention:

1. Means for determining the rating of an incandescent lamp comprising a movable photometric member, an indicator operated thereby, and a movable scale to cooperate with the indicator, the divisio-ns of which are proportional to the changes in the candle power of the lamp that are produced by the changes in the voltage or current, as represented by each division thereof.

2. Means for determining the rating of an incandescent lamp comprising a movable photometric member, an indicator operated thereby, and a movable scale to cooperate with the indicator, the divisions of which are proportional to the changes in the candle power of the lamp that are produced by the changes in the voltage or current, as represented by each division, and a stationary index with which the movable scale also cooperates.

3. Means for determining the rating of an incandescent lamp comprising a wattmeter and a voltmeter or ammeter for determining the volt-age or current of the lamp when it consumes a given number of watts, a movable photometric member, an indicator operated thereby, a movable scale to cooperate with the indicator, and a stationary index opposite which the division of the movable scale that corresponds to the measured voltage or current is placed, the rated voltage or current of the lamp being then designated by the indicator upon the movable scale.

In testimony whereof, I have hereunto subscribed my name this 26th day of April,

ARTHUR J. SWVEET. Witnesses:

S. B. PEACOCK, B. D. GRAHAM.

Copies of this patent may be obtained for five cents each, by addressing the Commissioner of Patents, Washington, D. 0.

Images