In 1838 Gassiot made the discovery that the temperature of the positive electrode of an electric arc is much greater than that of the negative electrode. This is explained in electronic theory by the bombardment of the positive electrode by negative electrons or corpuscles of electricity. This temperature-difference was later taken into account in designing direct-current arc-lamps, for inasmuch as most of the light from an ordinary arc is emitted by the end of the positive electrode, this was placed above the negative electrode. In this manner most of the light from the arc is directed downward where desired. In the few instances in modern times where the ordinary direct-current arc has been used for indirect lighting, in which case the arc is above an inverted shade, the positive carbon is placed below the negative one. Gassiot first proved that the positive electrode is hotter than the negative one by striking an arc between the ends of two horizontal wires of the same substance and diameter. After the arc operated for some time, the positive wire was melted for such a distance that it bent downward, but the negative remained quite straight.
Charcoal was used for the electrodes in all the early experiments, but owing to the intense heat of the arc, it burned away rapidly. A progressive step was made in 1843 when electrodes were first made by Foucault from the carbon deposited in retorts in which coal was distilled in the production of coal-gas. However, charcoal, owing to its soft porous character, gives a longer arc and a larger flame. In 1877 the "cored" carbons were introduced. These consist of hard molded carbon rods in which there is a core of soft carbon. In these are combined the advantages of charcoal and hard carbon and the core in burning away more rapidly has a tendency to hold the arc in the center. Modern carbons for ordinary arc-lamps are generally made of a mixture of retort-carbon, soot, and coal-tar. This paste is forced through dies and the carbons are baked at a fairly high temperature. A variation in the hardness of the carbons may be obtained as the requirements demand by varying the proportions of soot and retort-carbon. Cored carbons are made by inserting a small rod in the center of the die and the carbons are formed with a hollow core. This may be filled with a softer carbon.
If two carbons connected to a source of electric current are brought together, the circuit is completed and a current flows. If the two carbons are now slightly separated, an arc will be formed. As the arc burns the carbons waste away and in the case of direct current, the positive decreases in length more rapidly than the negative one. This is due largely to the extremely high temperature of the positive tip, where the carbon fairly boils. A crater is formed at the positive tip and this is always characteristic of the positive carbon of the ordinary arc, although it becomes more shallow as the arc-length is increased. The negative tip has a bright spot to which one end of the arc is attached. By wasting away, the length of the arc increases and likewise its resistance, until finally insufficient current will pass to maintain the arc. It then goes out and to start it the carbons must be brought together and separated. The mechanisms of modern arc-lamps perform these functions automatically by the ingenious use of electromagnets.
The interior of the arc is of a violet color and the exterior is a greenish yellow. The white-hot spot on the negative tip is generally surrounded by a fringe of agitated globules which consist of tar and other ingredients of carbons. Often material is deposited from the positive crater upon the negative tip and these accretions may build up a rounded tip. This deposit sometimes interferes with the proper formation of the arc and also with the light from the arc. It is often responsible for the hissing noise, although this hissing occurs with any length of arc when the current is sufficiently increased. The hissing seems to be due to the crater enlarging under excessive current until it passes the confines of the cross-section of the carbon. It thus tends to run up the side, where it comes in contact with oxygen of the air. In this manner the carbon is directly burned instead of being vaporized, as it is when the hot crater is small and is protected from the air by the arc itself. The temperature of the positive crater is in the neighborhood of 6000° to 7000°F. The brightness of the arc under pressure is the greatest produced by artificial means and is very intense. By putting the arc under high pressure, the brightness of the sun may be attained. The temperature of the hottest spot on the negative tip is about a thousand degrees below that of the positive.
No great demand arose for arc-lamps until the development of the Gramme dynamo in 1870, which provided a practicable source of electric current. In 1876 Jablochkov invented his famous "electric candle" consisting of two rods of carbon placed side by side but separated by insulating material. In this country Brush was the pioneer in the development of open arc-lamps. In 1877 he invented an arc-lamp and an efficient form of dynamo to supply the electrical energy. The first arc-lamps were ordinary direct-current open arcs and the carbons were made from high-grade coke, lampblack, and syrup. The upper positive carbon in these lamps is consumed at a rate of one to two inches per hour. Inasmuch as about 85 per cent. of the total light is emitted by the upper (positive) carbon and most of this from the crater, the lower carbon is made as small as possible in order not to obstruct any more light than necessary. The positive carbon of the open arc is often cored and the negative is a smaller one of solid carbon. This combination operates quite satisfactorily, but sometimes solid carbons are used outdoors. The voltage across the arc is about 50 volts.
In 1846 Staite discovered that the carbons of an arc enclosed in a glass vessel into which the air was not freely admitted were consumed less rapidly than when the arc operated in the open air. After the appearance of the dynamo, when increased attention was given to the development of arc-lamps, this principle of enclosing the arcs was again considered. The early attempts in about 1880 were unsuccessful because low voltages were used and it was not until the discovery was made that the negative tip builds up considerably for voltages under 65 volts, that higher voltages were employed. In 1893 marked improvements were consummated and Jandus brought out a successful enclosed arc operating at 80 volts. Marks contributed largely to the success of the enclosed arc by showing that a small current and a high voltage of 80 to 85 volts were the requisites for a satisfactory enclosed arc.
The principle of the enclosed arc is simple. A closely fitting glass globe surrounds the arc, the fit being as close as the feeding of the carbons will permit. When the arc is struck the oxygen is rapidly consumed and the heated gases and the enclosure check the supply of fresh air. The result is that the carbons are consumed about one tenth as rapidly as in the open arc. There is no crater formed on the positive tip and the arc wanders considerably. The efficiency of the enclosed arc as a light-producer is lower than that of the open arc, but it found favor because of its slow rate of consumption of the carbons and consequent decreased attention necessary. This arc operates a hundred hours or more without trimming, and will therefore operate a week or more in street-lighting without attention. When it is considered that open arcs for all-night burning were supplied with two pairs of carbons, the second set going into use automatically when the first were consumed, the value of the enclosed arc is apparent. However, the open arc has served well and has given way to greater improvements. It is rapidly disappearing from use.
The alternating-current arc-lamp was developed after the appearance of the direct-current open-arc and has been widely used. It has no positive or negative carbons, for the alternating current is reversing in direction usually at the rate of 120 times per second; that is, it passes through 60 complete cycles during each second. No marked craters form on the tips and the two carbons are consumed at about the same rate. The average temperature of the carbon tips is lower than that of the positive tip of a direct-current arc, with the result that the luminous efficiency is lower. These arcs have been made of both the open and enclosed type. They are characterized by a humming noise due to the effect of alternating current upon the mechanism and also upon the air near the arc. This humming sound is quite different from the occasional hissing of a direct-current arc. When soft carbons are used, the arc is larger and apparently this mass of vapor reduces the humming considerably. The humming is not very apparent for the enclosed alternating-current arc. The alternating arc can easily be detected by closely observing moving objects. If a pencil or coin be moved rapidly, a number of images appear which are due to the pulsating character of the light. At each reversal of the current, the current reaches zero value and the arc is virtually extinguished. Therefore, there is a maximum brightness midway between the reversals.
Various types of all these arcs have been developed to meet the different requirements of ordinary lighting and to adapt this method of light-production to the needs of projection, stage-equipment, lighthouses, search-lights, and other applications.
Up to this point the ordinary carbon arc has been considered and it has been seen that most of the light is emitted by the glowing end of the positive carbon. In fact, the light from the arc itself is negligible. A logical step in the development of the arc-lamp was to introduce salts in order to obtain a luminous flame. This possibility as applied to ordinary gas-flames had been known for years and it is surprising that it had not been early applied to carbons. Apparently Bremer in 1898 was the first to introduce fluorides of calcium, barium, and strontium. The salts deflagrate and a luminous flame envelops the ordinary feeble arc-flame. From these arcs most of the light is emitted by the arc itself, hence the name "flame-arcs."