Electric Illumination.

Progress, Volume II, Issue 7, 1 May 1907, Page 261

Electric Illumination.

Review of 1906. Never has greater interest been displayed in new methods of electric illumination than during the past year, 1906. If the promise held out by the inventors of metallic filament lamps are fulfilled we may soon witness the passing of the carbon filament bulb. Although the Nerst lamp, on which great hopes were based, because it only requires half as much current as the carbon filament, has proved too costly, and the osmium lamp has been found wanting for the same reason, and for the additional reason that its voltage of 47 is too low for ordinary circuits, the tantalum and tungsten lamps seem likely to be successors to the standard incandescent lamp. The tantalum consumes about as much energy as the osmium lamp, but its long filament renders its use possible on a 110-volt circuit, and on currents of even higher voltage. Its useful life of from 400 to 600 hours, and its maximum life of 1,000 hours and more, compare favourably with the best electric incandescent lamps in use. The filament is very delicate but able to stand greater variations than the carbon filament. When broken the ends readily fuse, so that the tantalum lamp's usefulness, although impaired, is not utterly destroyed. The present low cost of construction (50 cents) coupled with its high voltage, gives it a decided advantage over the osmium filament. Guelcher's irridium lamp is made only for low tensions (24 volts) : it consumes, it is claimed, only 1 to 1.5 watts per candle power, and costs about 87 cents. What its life may be it is impossible to state, inasmuch as no figures have been published. It is open to many of the objections levelled at the osmium lamp. More promising is the tungsten lamp, which is now made by four European firms using as many different processes. The normal tungsten lamp of Just and Hannaman seems to give about thirty to forty candles at 110 volts and consumes 1.1 watts per candle. Kuzel's tungsten lamp is said to show an efficiency of 1 to 1.25 watts for 19 to 32 candle lamps with a useful life of 1,000 hours, at the end of which the loss in candle power is said to be but 10 or 15 per cent. When broken the filament automatically welds together as in the tantalum lamp. The osmium-tungsten lamps have shown from 1.026 to 1.047 watts per candle at 110 volts. Whether these new lamps will fulfil the new hopes placed on them can of course only be determined by thorough tests under conditions approximating those ol actual service. At present the metallic filament lamp is m the experimental stage. The necessity of using the tungsten lamp m the inverted position may perhaps be regarded as a defect ; yet quite recently the inverted gas mantle has invaded the extensive field hitherto monopolised b\ the electric li^ht. — Scientific American.

The Tungsten Lamp. The Wolfram or Tungsten incandescent lamp was put on the market about the end of last year. It is the invention of Dr. A. Just and Mr. F. Hauaiuan. In appearance and design it closely follows the lines of the f ami]-

iar carbon filament lamp. These inventors were the first to make a tungsten filament, and their method of manufacture enables them, they claim, to make lamps commercially as low as 25 c.p. consuming one watt per candle,

and with only three filaments. The life of the lamps is put at no less than 1,500 hours. The European manufacturers are bringing on the market a standard tungsten lamp of 40 hefners c.p. consuming 40 watts. After tests Professor Reithoffer of Vienna says, in his report of last January, "A like efficiency has never been reached by any incandescent lamp, this efficiency, in fact, very closely approximates that obtained by arc lamps, and indeed, seems to have reached the limit of what possibly can be achieved m incandescent lighting." These lamps burn well on direct as well as alternating currents, differing in this respect from the tantalum lamp, which does not stand up so well on alternating current as on direct. The filament in ordinary sizes is 17 inches in length.

The Tantalum Lamp. Unlike the ordinary carbon spiral the filament of the tantalum is threaded up and down between two series of hooks arranged in circles, and connected with two terminals. This arrangement gives a large extent of glow and enables a more powerful distribution of light. Its chief merit, however, lies in the economy it effects. It consumes only half the current of a carbon arc of similiar voltage and candle power. When new its consumption is about 1.7 watts per candle power, as against 3.5 to 4 watts used by an ordinary lamp. It is claimed that its ordinary life is between 400 and 600 hours, but this is allowing considerable margin, as sometimes it endures for more than 1,000 hours. It is only where there are continuous circuits that full satisfaction can be obtained from the tantalum lamps, for, when on alternating circuits the results are somewhat irregular. Moreover, where local circuits run on 220 volts it is necessary to burn these in series, using two 110 volt lamps. The inventors claim that with these two minor restrictions, the tantalum lamp still has strong claims on the score of its efficiency and economy. The above lamp was discussed at the last meeting of the Institute of Electric engineers, on which occasion it seems to have come in for a very large share of attention. The discussion was upon a paper by Mr. Swinburne on "Metallic Filament Lamps." Its scope may be gathered from the following letter published in a contemporary, which besides throws some interesting light on the tantalum and the work it has done already, as well as its position in the market, which is, after all, the proper practical test. ' As there was not any opportunity to reply to several points raised by various members at the recent discussion at the institute on Mr. Swineburne's paper on "Metallic Filament Lamps," I should be grateful if you would allow me this opportunity of commenting on these points raised. "I note that nearly every one who spoke discouragingly of the tantalum lamps has used them on alternating currents in distinct opposition to the makers' recommendation It is, therefore, scarcely surprising that the results they obtained were not encouraging. As one having experience extending over two years with the tantalum lamp, I beg to state there is no doubt that the lraakers" claims are more than justified if the lamps are used under fair conditions. The " pairing " of the lamps is a matter of great importance, and when this is given proper attention, and the lamps are installed of voltage sufficiently high to cover the fluctuations of the supply (a fairly large order in some circumstances), the average useful life of the lamps on direct current circuits comes out well above 750 hours

with a reduction of not more than 25 per cent. to 30 per cent, in candle-power at the end of 1,000 hours. I was interested to hear from the carbonfilament lamp makers that there was nothing to prevent their placing a lamp (carbon filament) on the market of practically the same life and efficiency as tantalum. Why don't they do it ? There appeared to be an incubus of doubt as to the tantalum lamp being a commercial article. . I find, on enquiring at the makers that they are selling about 60,000 per month in this country, and could probably double this if they could obtain the lamps quicker.

The Osram Lamp. On the subject of this lamp the well-known contractor and electric lamp retailer, Mr. Cuningham, of London, gave the same contemporary, by way of illustration of the discussion, some very useful information. He says :— I installed at a double-fronted shop, in one window 24 16-c.r. carbon filament lamps, and the other window, 8 50-c.p. Osram lamps (metal f laments). The supply was 100 volts 60 periods. The Osrams now have been in use about 500 hours, during which time they have consumed about 250 units, which at the price charged, 5d., amounts to £5 4s. 2d. Two lamps gave out in the first 150 hours, therefore, allowing for 10 lamps at ss. each — £2 10s.— we have a total cost of £7 14s. 2d. for energy and lamps. The 24 Royal Ediswan lamps consumed 750 units, thus showing a cost, for current alone, of £15 12s. 6d. If I allow 245. for the cost of these lamps, the total cost of these and the energy is £16 16s. 6d. (as against £7 14s. 2d. in the case of the Osram), showing an actual saving of no less than £9 2s. 4d. Had the cost of the supply been 4d., 3d., or Id. per unit, there would have been effected a saving of £7 os. Bd., £4 19s. or 15s. Bd. respectively. The Osram also had the advantage as regards quality of light. In the matter of candle power the Osram gave 400 as against 384 of the Royal Ediswan. The latter, also, have all but ended their economical life. One of the Osrams requires renewing because it has blackened, but the rest appear as good as new. I have found these new Osram lamps satisfactory, and consider the small 2 and 4volt ones excellent. The Tantalum lamps I have found also very good, when burnt on a direct current. I believe we shall even yet see great improvements in incandescent electric lamps, and will give any new type a fair trial and accord the makers every encouragement."

The Mercury Vapour Lamp. This lamp is considered just as efficient, and is known as the " Cooper-Hewitt." Hitherto having been confined to indoor illumination, it is now being adopted for street lighting as well. The terminals are conducted to a small glass tube containing mercury, and a vacuum is formed in the tube, which is then sealed up. In operation the lamp has to be mounted in a lantern which must be tilted to obtain the light. This action causes the mercury to flow from one electrode to the other in a small stream in the enclosed tube. The efficiency of this lamp is claimed to be about twice that of the enclosed arc lamp, and six or seven times greater than that of the incandescent lamp. Further, it requires less attention to maintain, and its life is longer than any form of glow lamp. I^ike the tantalum, it is more adapted for direct circuits.

The Tantalum. There are immensely valuable deposits of tantalum ore at Wodgina, West Australia. Tantalum is worth about 2s. per ounce, and its properties are numerous and valuable. It makes an admirable filament for electric lamps, and tantalum lamps are now on the market. It makes good nibs for fountain pens, and is far less costly than gold or radium. It is not acted on by acids or air, and thus has many of the qualities of platinum, which now costs £7 per ounce. It is a hard, grey metal, heavier than iron, and can easily be worked. When hammered it becomes exceedingly hard, so that its use has been proposed on drills, replacing the diamond. Its properties in alloy with steel render it of great metallurgical importance. It gives a strong, hard, and tough metal, specially suitable for high-speed lathe work. The Times correspondent adds that " there is little room for doubt that tantalum will be used for armour piercing shell points and for armour plate," and urges that its export should be forbidden, " pending investigations as to its military value by the British Admiralty." Krupp's firm are known to have experimented with it. It is quite possible that by the use of tantalum alloyed with steel, the resistance of armour plates might be increased 25 per cent., or even more. It is curious to note how attempts are being made to secure a monopoly in the rare metals. Radium is virtually an Austrian monopoly, yet no great conimercial importance as yet attaches to it. The sources of supply of thorium found in Brazil with a salt, of which gas mantles are prepared, are in German hands. Osmium, of value for the filaments of electric lamps, is more or less controlled by German finance. American financiers have tried, not unsuccessfully, to obtain control of the Canadian deposits of cobalt and nickel.

Tungsten. Tungsten is a very hard and brittle metal which is sold in the form of a black powder, or as ferro-tungsten. It was for a long time considered unfusible, but the electric furnace showed of course that it could be melted. The powder is difficult to squirt, even mixed with a good deal of tragacanth. Kazel has invented what seems to be an admirable way of getting out of the difficulty. He gets the tungsten in the form of an exceedingly fine powder by employing a method that was used by Brederg for getting what was known as colloidal platinum. An arc is made to play under water between tungsten electrodes, and this is said to produce a very finely divided form of metal. This is collected and worked up into a stiff enough paste and squirted. Tungsten is not an expensive metal, so the only cost is in making the filaments. Whether the f laments of this paste can be squirted so as to be fine enough for 200 volts will be a matter for the future to decide.

Osmium. Osmium is a crystalline metal which cannot be drawn into wire. It is very hard, scratching quartz. The Welsbach osmium lamp is said to be made by making a paste of finely divided osmium and an organic binding material, and squirting it. The filaments are then baked and heated electrically to a very high temperature to eFminate the carbon. The osmium lamp so far produced is for low pressures as might be expected, but it has a very high efficiency. There is some doubt whether the lamps known as osmium are made of osmium or an alloy of osmium

and tungsten. Tungsten is a curious metal, and it is not very easy to get into alloys, but it may alloy with osmium perfectly, though there is no record to that effect.

Probable Effect of the New Lamps. In the first place, they will increase the output of stations, just as machinery increases labour. But there is more difficulty in foreseeing the result of high efficiency hampered with low pressure. A probable solution is that people will gradually take to using large lamps taking the same pressure, and about the same power as carbon lamps, but giving, say, four times the light. As to the lamp-making industry, one might prophesy without much danger that the present makers will merely alter their manufacture and make metal lamps. This will pay inventors better, because the existing makers have their commercial organisations and their facilities for distribution. Besides, all the works, except the parts devoted to the making of filaments, will be available. It is possible that new works will be set up to make filaments, and that the lamp-makers will buy the filaments and make them up into lamps. There are so many possible ways of making metal filaments, that it is doubtful whether large monopolies can be secured by patents ; and it is much more likely that most of the present carbon lamp makers will work out particular processes of their own, and will put their own metal lamps on the market. — Electric Engineering.