US1573593A - Electric bulb - Google Patents

Description

A. A. WILKINS Feb. 16 1926.

ELECTRIC BULB Filed April 19, 1922 Patented Feb. 16, 1926.

UNITED STATES ALBERT A. WILKINS, OF MIDDLETON, MASSACHUSETTS.

ELECTRIC BULB.

Application filed April 19, 1922. Serial No. 555,609.

To all whom it may concern:

Be it known that I, ALBERT A. IVILKINS, a citizen of the United States, and a resident of Middleton, in the county of Essex and Commonwealth of Massachusetts, have invented a new and useful Improvement in Electric Bulbs, of which the following is a specification.

The present invention relates to bulbs, particularly filament-containing bulbs, and to methods of making the same, and the object of the invention is to improve upon present-day bulbs and their manufacture.

The accompanying drawings show diagrammatically, for illustrative purposes, several types of bulbs that may be improved in accordance with the present invention. Fig. 1 is a diagrammatic view of one of several types of X-ray tubes: Fig. 2 is a similar view of an audion; and Fig. 3 representsan incandescent tungsten-filament lamp that is used for lighting purposes.

The glass or other casing of the bulb is indicated in all the views by the numeral 2. The illustrated bulbs all contain a fila ment 4 that is adapted to be heated to incandescence by the passage therethrough of an electric current, but it will be understood that some important features of the invention are not restricted to filament-containing bulbs. In addition to the filament, the X-ray tube of Fig. 1 is provided with an anode 6, and the audion of Fig. 2 with a grid 8 and a plate 10. The X-ray tube and the audion are shown completely manufactured. The lamp of Fig. 3 is shown partly completed.

The best method now known of carrying out the invention will first be described, and a discussion of theoretical principles will then follow. The invention may be described in connection with the manufacture of an incandescent lamp. such as is illustrated in Fig. 3. In this figure, the filament 1 is shown already enclosed in the glass casing 2. The bulb is as yet unsealed, being provided with the usual tubulature 12.

According to the best method now known of practicing my invention, the filament 4, before it has been enclosed in the glass, in the position shown in Fig. 3 has been coated, or partly coated, with sodium nitrite. This may be done by dipping the filament in a solution of the salt and drying the filament before it is mounted in place. Though preferred, it is not necessary that the sodium.-

nitrite be applied directly to the filament, for very good results have been obtained when the nitrite was introduced in other ways, as when th filament supports 14.- have been coated instead, and even when a little of the salt was applied to the interior surface of the glass casing 2 itself. It is believed to be essential that the sodium nitrite be reasonably pure and free, uncombined with some other substance that may prevent it from functioning roperly. In Fig. 3, a very small portion 0 the filament is shown coated with the nitrite, as indicated by the heavy line numbered 16.

The bulb is then exhausted in the ordinary way. by means of a pump that is connected to the tubulature 12. The degree of exhaustion need not be high; indeedrvery satisfactory results are obtainable with an ordinary (iaede mercury pump. The use of highly refined pumps, such as condensation pumps, and involved present-day processes, like refrigeration, is wholly unnecessary and, it is believed, contributes nothing to the efliciency of the lamp. During the later stages of the exhaust, when the highfrequency-current test yields only a slight fluorescence, a current is preferably passed through the filament to heat it. After a suflicient stage of exhaustion, a few seconds later, the filament is preferably cooled and the tube is sealed in the customary manner. The filament is then again heated, preferably at normal temperature, for two or three minutes. Little flashes are observable in the bulb at this time, which become larger and larger, if the filament heating is increased, until they finally disappear. Raising the filament to so high a temperature would have caused the filament to rupture in the absence of the sodium nitrate. The resulting lamp burns very brightly at ordinary lamp temperatures. with very little heating of the glass wall of the bulb, at very high efficiency, for a much longer period than is possible with present-day commercial bulbs. and without any preceptible degree of blackening of the walls of the bulb.

As has been explained above, the invention is not restricted to lamps alone. It is applicable also to other bulbs, as X-ray tubes of the. type shown in Fig. 1, and audions like the audion shown in Fig. 2, when prepared according to the method of the presappear to be endowed with Nor is the inent invention, most unusual properties.

vention restricted to filament-containing Some theories have attributed this result bulbs alone, for an X-ray tube without a filament, of a type hitherto unknown, has been produced by the present method, the electrodes of which are very close together in the center of the tube. The term bulb is therefore employed in the specification and the claims in a generic sense.

The above description covers the best method now known of practicing the invention, but it is obvious that changes may be made within the skill of the artisan without departing from the spirit and scope of the invention. The use of sodium nitrite is recommended, because it has been found, in practice, to work very well, but other substances may advantageously be employed instead. It is not necessary that the steps inthe method above described be followed exactly as described, and some of the steps may be omitted or others added. And the bulb of the present invention may conceivably be produced in some other manner than by the method herein described.

The above description sets forth the actual steps of the best known method of practicing the invention, without any attempt at explaining why these steps lead to the desired result. In the nature of things, an y theory of the operation must be purely conjectural, and it is therefore desired not to limit the invention to a theory that later maybe found not to be correct. The invention resides not in a theory. but in actual accomplishments and results. Several explanations may, however, for completeness of description, be advanced for what they are worth. 7

It has been stated above that the lamp of the present invention may be burned for months at ordinary filament temperatures without perceptible blackening of the walls of the bulb. Such blackening is usually caused by disintegration of the. filament. Bit by bit, the filament disintegrates during the use of the ordinary lamp, and becomes deposited upon the inner surface of the glass of the bulb, This disintegration is one of the most important among the factors that are chiefly responsible for the short life of electric lamps and other filamentcontaining bulbs. In the first place, the fila- -ment becomes thereby weakened, until it ultimately bursts;

and secondly, the disintegrated portions of the filament that become deposited upon the glass form an opaque layer that shuts out much of the light coming from the filament, necessitating discarding the lamp even though the filament has not yet become broken, and though it is otherwise still in good condition. Disintegration is particularly serious in the case of metal filaments, like tungsten.

Various theories have been advanced to account for this filament disintegration.

to the effects caused by residual gases remaining in the bulb after it hasbeen exhausted and sealed. Just how the gases came to remain in the bulb is not clear among other explanations, they may have originated in the atmosphere, and could not fully be swept out by the pump; or they became stuck, film-like, upon the Interior walls of the glass; or they remained hidden away, for the time bein between the molecules of the filament itsel However theygot there, their action, according, to the theorists above mentioned, is something like this: The residual gas particles in the exhausted bulb first combine chemically with particles of the filament. This chemical union is caused by the heat of the lighted filament. The particles so chemically combined then break up again chemically when they come into contact with the comparatively cold glass of the bulb. In so breaking up, they deposit the filament particles upon the glass. The gas particles thus freed soon again approach the heated filament, through convection currents or otherwise, when they again'attack the filament, combining with it chemically, and they later again deposit the filament particles upon the glass. A very small amount of residual gas is thus enabled, after the lapse of a comparatively small time interval, to do much damage.

The present filament-attacking gases remaining in the bulb after it hasbeen exhausted and sealed are usually oxygen and water vapor. The harmful effects of the oxygen are considered to be inconsiderable. The chief source of trouble is, therefore, the Water vapor; at any rate, if there are any additional destructive gases present, they all act like the water vapor. The water vapor is believed to give up its constituent oxygen and hydrogen under the influence of the heat emitted by the heated filament. Dry hydrogen seems to have no effect upon the filament, but the oxygen, in its nascent state, immediately attacks the filament, forming an oxide of the latter. This oxide ultimately gives up the metal of the filament to the glass of the bulb, when coming into contact with the latter, as about described; and the oxygen thus freed combines with hydrogen contained within the bulb to reform water vapor. This cycle is repeated over and again, until the lamp has lostits usefulness.

Various getters have been employed to counteract this harmful effect of the water vapor. As soon as it is released from the water vapor, under the influence of the heated filament, the nascent oxygen is caused to enter into permanent chemical union with the getter. The getter must, of course, have a sulficiently strong affinity for the harmful gas, as the nascent oxygen from the water vapor, such that the harmful gas will combine with it in reference to attacking the filament, and Will remain in combination. Phosphorus, for example, either alone, or in combination with other substances, was used as a getter for many years, and is still used to-day. Not the only trouble with these getters is that they do notprevent blackening in lamps; and in the case of X-ray tubes, audions, and the like, they are not effective. Either the theory is wrong, or the getters introduced other complications.

As possible further light upon this theory, if it is correct, it is suggested here that these getters may not wholly eliminate a .l the harmful gases, or that, possibly, the getters, either alone or combined with the harmful gases, become, in their turn, harmful to the filament Thus, it has been found that sodium sulphite, potassium nitrite, and certain oxides each has something of the effect produced by sodium nitrite when used as described above; but each, unlike sodium nitrite, causes blackening at ordinary filament temperatures; and each, again unlike sodium nitrite, causes the bulb to become very hot during the lighting of the filament. It may he, therefore, that sodium sulphite and potamium nitrite each causes the presence, in the bulb, of a heatconducting gas that facilitates the blackening process. But neither of these other two substances is nearly so bad in its effects as, for example, arsenic; for the latter, though it acts much more slowly, acts much the same way, except that the heating of the glass is much more intense and the blackening ever so much more rapid. \Vhen it is borne in mind that even a sodium-nitrite treated bulb will blacken, and the glass become heated, at extremely high filament temperatures, the theory is suggested that possibly each substance employed will prevent blackening if the filament is kept at a low enough temperature; but that at higher temperatures, the degree of which is dependent upon the particular substance employed, gases are produced which, in some way, facilitate blackening. Either no gases are produced at ordnary temperatures by the sodium ni-e trite, or such gases as are produced are of the harmless variety. Viewed in this light, the present invention has produced, in sodium nitrite, a more eflicient getter than has ever been discovered before, and the invention comprises within its scope all such getters that are capable of yielding a lamp that will not perceptibly blacken at ordinary filament temperatures.

It has been disputed, however, that the presence of residual gases is the true cause of blackening, at least in well exhausted bulbs, and the theory has been advanced, instead, that the real cause of blackening, in such bulbs, is that the filament has a tendency to evaporate. The evaporated filament particles fly off to the interior walls of the glass, according to this theory, and being cooled by contact therewith, become deposited thereon. Inert gases, like nitrogen and argon, were found to decrease the rate of evaporation. If this is the correct theory, then obviously the sodium nitrite acts, in some way, to prevent evaporation at ordinary filament temperatures, and the filament disintegration at higher temperatures is caused not by the presence of some gas, but by expected filament evaporation. A better efficiency is obtained according to the present invention than in the nitrogenfilled lamps, and without perce tible disintegration, while the filaments o nitrogenfilled lamps do disintegrate. Viewed from this angle, the sodium nitrite counteracts the tendency of the filament to evaporate at ordinary temperatures, and the invention consists in treating a filament so that it shall not evaporate at ordinary filament temperatures.

It should be noted, in this connection, that the bulbs of the present invention may also be tilled with an inert gas, like nitrogen, before the sealing step. In a lamp so filled with nitrogen, upon the filament being heated, distinct flashes are observable, as above described, showing that the gas is becoming further purified. If it may be assumed that the disintegration of nitrogenfilled lamps is caused by impurities in the nitrogen or other inert gas, then the present invention includes an improved lamp of the inert-gas-filled type.

Many other theories of lamp behavior have been proposed, but they will not all be considered here.

It has already been explained that fiashes of light take place during the heating of the filament after the bulb has been scaled Such flashes have ordinarily been regarded as indicating that gases within the bulb are disappearing. It may he, therefore, that a higher vacuum is obtained in the bulb by the use of sodium nitrite than has been ob tainable before, and that the improved lamp is simply the result of an improved vacuum of a degree never before attainable. This theory gains force from the fact that the glass of the sodium-nitrite-treated lamp remains comparatively cool, but this phenomenon is equally consistent with the presence of a gas that is not a good heat conductor. The operation of the X-ray tube shown in Fig. 1, however, though the tube is exhausted with an ordinary Gaede pump, is most remarkable, and it has been regarded as settled that the operation of such a tube depends upon a very high vacuum; so much so that even the presence of gases released from the walls of the bulb may be suflicient to permit conduction by ionization between the cathode and the anode, and thus to destroy the tube. It 1s to be noted, however, that after this X-ray tube has been out of use for a time, the

flashes above described recur for a very brief interval when the current is turned on again If the high-frequency tests shows sible.

It is well known, again, that audions,

like the audion illustrated in Fig. 2, willnot operate properly except in a high vacuum. This is especially true when the audion is used as an amplifier. After an audion has been in use for a long time, it seems that gases become liberated therein, which destroy' its usefulness. This has been found to be the case, for example, in audions prepared with phosphorus. as a getter. Either the hosphorus does not wholly destroy the Earmful gases, or' it produces other harmful gases in the course of time. "Whatever the explanation, the condition produced by the phosphorus does not remain fixed, the phosphorus apparently losing its effect after considerable use. Phosphorus, furthermore, seems to affect the life of the filament, and especially a filament made of tungsten. Audions made according to the present invention, and using only a Gaede pump, have given wonderful service. Both in thecase of the audion and in the case of the X-ray tube, the longer that the bulb is used, the better condition it seems to be in; instead of the volume of undesirable gas within the bulb increasing, it decreases with the lapse of time; which is just the reverse of what takes place with bulbs of present-day construction, which become gassy through continued use} In fact, the bulb may be removed from the Gaede pump while still in a very imperfect condition, and the use of the tube,

will alone perfect it. It has hitherto not been possible to produce this character of X-ray tube and audions, using only a Gaede pump.

The question as to whether the sodium nitrite produces a higher. vacuum than is obtainable with present-day methods-eondensation pumps and liquid-air refrigeration included-is therefore not answered cal condition.

here. It is preferred to suppose that an improved 0 crating condition of some (at present) in efinite nature is created in the bulb. The reason for the supposition is that tests seem to indicate that extremely high vacuum is not so essential as the proper character or state of the vacuum. This 1mproved condition, therefore, may, or may not, be a higher vacuum; or it may be something else, such as a more favorable electri- This improved electrical condition may be of such a nature as, accepting, for the moment, the evaporation theory, to prevent the evaporation of the filament particles or to repel. them back to the filament; or to increase the electrical resistance between the filament and the glass wall of the bulb. In the case of X-ray tubes, audions, and other bulbs having a plurality of electrodes, the improved electrical condition seems to cause a higher electrical resistance between the electrodes. Theresistance increases during operation. From this viewpoint, a feature of the invention resides in the use of a substance that will act primarily to increase the electrical resistance of the bulb,other substances, like phosphorus, for example, having some other primary function. As above pointed out, the energization of the device may be expected whether or not the bulb contains a filament.

As above pointed out, the not reside in a theory, but in actual accomplishments and results. The theories discussed may or may not be true to fact; the action may be physical or chemical; the

secret may lie in the production of a superior 95 filament; planation. invention pler and duction.

\Vhat is claimed as new is:

1. A bulb into which sodium nitrite has been introduced before the bulb was sealed;

2. An electrical lamp comprising a paror there may be some other ex- The fact remains that the present yields a superior bulb by a simmore economic method of protia-lly evacuated bulb having a filament to which sodium nitrite has been applied.

3. A partly evacuated chamber containing a filament part at least of coated with sodium nitrite.

' 4. A bulb containing an inert gas into which sodium nitrite was introduced before the bulb was sealed.

In testimony whereof, I have hereunto subscribed my name this 17th day of April,

ALBERT A.. IVILKINS.

invention does 90 which has been

Images