GIFT OF 
 MICHAEL REE^E 
 

THE MANUFACTURE 
 
 OF 
 
 ELECTRIC LIGHT 
 CARBONS. 
 
 BY 
 
 O. G. PRITCHARD 
 
 Reprinted from " THE ELECTRICIAN." 
 
 WITH ADDITIONAL CHAPTERS. 
 
 PRICE ONE SHILLING AND SIXPENCE, 
 
 Post Fxee Is. 9d. 
 
 LONDON : 
 "THE ELECTRICIAN" PRINTING AND PUBLISHING COMPANY, LIMITED, 
 
 SALISBURY COURT, FLEET STREET, E.G. 
 
 COPYRIGHT. 
 
THE MANUFACTURE 
 
 OF 
 
 ELECTRIC 
 
 CARBONS. 
 
 . 
 
 BY 
 
 O. G. PRITCHARD 
 
 \\ 
 
 Reprinted from " THE ELECTRICIAN," 
 
 WITH ADDITIONAL CHAPTERS. 
 
 PRICE ONE SHILLING AND SIXPENCE, 
 
 Post Fvee Is. 9d. 
 
 LONDON : 
 
 : THE ELECTRICIAN" PRINTING AND PUBLISHING COMPANY, LIMITED, 
 SALISBURY COURT, FLEET STREET, E.C. 
 
 1 890. 
 
LONDON : 
 
 PRINTED BY GEORGE TUCKER, 1, 2 AND 3, SALISBURY COURT, 
 FLEET STREET, E.G. 
 
CONTENTS. 
 
 PAGE 
 
 INTRODUCTION ..... 5 
 
 PREPARATION OF MATERIAL 12 
 
 DRYING-ROOMS ..... 21 
 
 FIRING THE CARBONS 23 
 
 SATURATING ..... 26 
 
 CORING 27 
 
 DESCRIPTION OP PLANT REQUIRED . . . 33 
 
 COST OF PLANT .... 37 
 
 COST OF MATERIALS qo 
 
 *J */ 
 
 COST OF LABOUR .39 
 
 ESTIMATED PROFIT ... 40 
 
tJHIVBBSITI I 
 
 THE MANUFACTURE 
 
 OF 
 
 ELECTRIC LIGHT CARBONS. 
 
 BY 0. G. PRITCHAKD. 
 
 INTRODUCTION. 
 
 THAT hitherto we have been unable to cope with France, 
 Austria, Germany, or Bohemia in the production of electric-arc- 
 carbon candles is seen by the fact that our supplies continue to 
 be drawn from Continental manufacturers. Paris claims good 
 makers, such as Carre, Sautter-Lemonnier, Lacombe, Gaudoin, 
 and others, but their prices being high in great measure pre- 
 cludes their adoption on a large scale. Germany boasts Herren 
 Siemens and Schmeltzer, Austria Herr Hardtmuth, and the 
 Apostle carbon comes to us from Bohemia. These carbons 
 have of late years been much reduced in price, and leave little 
 or nothing to be desired. From America we obtain the Brush 
 carbon, solid and coppered, which eventually we may expect 
 to see superseded by the cored uncoppered carbon. Other 
 large makers might also be mentioned, to whose specialities I 
 need not now refer. Up to the present we stand committed 
 to Siemens, Hardtmuth, and Schmeltzer, who, between them, 
 
( 6 ) 
 
 and the Apostle carbons, divide the honours and the profits. 
 There are other first-rate makers, but they are but little known 
 to me, and have not as yet introduced their carbons on so large a 
 scale as the German makers. Strenuous efforts have been made 
 during the past ten years to introduce a home-made article, 
 efficient and good, but failure alone appears to have waited 
 upon endeavour. This has led to the idea that carbon-candle 
 making can never become an English industry. The writer 
 has, however, produced carbons on a scale of some magnitude 
 which were testified by experts to be quite equal to those of 
 the foreign makers ; and he now publishes the whole process 
 employed by him, and the results of eight years' experiments 
 and experience, in the hope that the arc-light carbon industry 
 may eventually be profitably introduced. 
 
 The difficulties in the way of producing a perfect carbon, 
 like all other difficulties, vanish with close application to the 
 subject. The various processes, so easily detailed and ex- 
 plained, however, have only been perfected after many years' 
 devotion to the subject and large expenditure of funds. How 
 not to do it is the antecedent to all discoveries ; and every 
 inventor, or would-be discoverer of hidden processes, has to 
 learn that only by the continuous registration of negative re- 
 sults can he hope to achieve positive success. I think it will 
 be instructive and also interesting to explain the many faults 
 to which carbons are prone, to diagnose the maladies, and to 
 point out the cause of failure, the necessary preliminary to a 
 cure. The ultima thule of carbon perfection, perhaps, has not 
 yet been achieved. Yet I am of opinion that, with the elimina- 
 tion of the faults hereinafter described, a carbon will be pro- 
 duced as near perfection as the severest critic can wish for. 
 
 I think we get the carbon and the light in their worst 
 form in the Jablochkoff system. The varying displays of 
 blue, red and yellow and green lights certainly afford a great 
 choice of colouring, but are undesirable, to say the least. 
 
( 7 ) 
 
 This system, however, being a thing of the past, need not 
 occupy our time. Hissing, also, may now be said to be like 
 the dodo dead. We come, however, to a serious defect in the 
 little gaseous flame that is sometimes seen to travel round the 
 point of the positive carbon whenever the arc lengthens beyond 
 the normal, and which throws a dark shadow upon the bowl. 
 This defect may be also enhanced by a faulty lamp not answer- 
 ing quickly to the change in the amount of current, and not 
 allowing the arc to become shortened. This has been attri- 
 buted to the lamp ; but seeing that by removing the carbon, 
 and placing a good carbon in circuit, the effect ceases, the fault 
 must be attributed to the carbon, and upon investigation it will 
 be found to arise from the presence of occluded gases in the 
 body of the carbon. This, again, arises from the improper 
 nature of the material employed, which renders the carbon 
 incapable of receiving or retaining the compression to which it 
 had to be submitted in the press ; or it may arise from want of 
 due saturation in an after process, whereby the want of com- 
 pression is got over, and the pores containing these occluded 
 gases are filled with carbonised matter. All or some of these 
 may be assigned as reasons. 
 
 But the original fault lies in the material employed. I make 
 this statement with boldness, and further assert that the purest 
 carbon is incapable of yielding a good carbon candle unless it 
 possesses qualities which have to be imparted to it mechani- 
 cally. The nearest approach which Nature affords for a proper 
 material is gas-retort carbon, inasmuch as the grain is hard, 
 good, and so well fitted to the purpose. But this material is 
 too impure for use, and is expensive to purify, which prevents 
 its adoption. When the preparation of material is touched 
 upon, this will be more clearly understood. Now, the grain 
 and character of the material, outside and beyond its purity, 
 are matters of moment. By way of more fully illustrating my 
 meaning, let me take the purest lampblack, perfectly suitable 
 
( 8 ) 
 
 per se for carbon-making as regards purity, yet, owing to the 
 absence of grain, it cannot be made to retain its compression ; 
 it occludes gases, and the perfection of the carbon is destroyed, 
 and its life considerably shortened. As far as my experience 
 reaches, I should say that a 10-millimetre carbon (cored posi- 
 tive), 42 .volts, 8-ampere current, T V n - arc, gives the highest 
 results. If the arc lengthen to Jin., red flames are produced in 
 the best carbons I have tested. 
 
 Conductivity is of the greatest importance : the greater the 
 facility with which the current passes, the better the light and 
 the less the E.M.F. required. Thus I think I may state, without 
 experimental data to go upon, but by observation only, that 
 a candle of high conductivity will give as good a light with a 
 6-ampere current and 38 volts as one of lower conductivity 
 using 42 volts and 8 amperes. Indeed, when the conductivity 
 is very low, the current will not produce the arc, and the carbon 
 becomes intensely heated. This conductivity depends entirely 
 (given pure carbon) on the length of time to which it has 
 been submitted to the firing process. Graphite appears to owe 
 its peculiar condition of high conductivity to its lengthened 
 subjection to Nature's forces, unapproachable by human efforts. 
 Great pressure, continuous and intense heat protracted over 
 long periods, even if possible, would prove commercially im- 
 practicable ; yet I entertain no doubt that carbon submitted 
 to a white heat, such as would melt wrought iron, for a whole 
 week, would produce a candle more graphitic in character and 
 possessing higher conductivity than one that had been subject 
 to such treatment for a less period. As I deal with the subject 
 of firing, and my special economic arrangements for so doing, 
 this subject will be more fully dealt with. 
 
 I would like to mention (as it may not be thought sufficient 
 to take the raw graphite ignorant of the amount of pure 
 carbon it contains) a method of testing this raw material. 
 This, indeed, should be done repeatedly, and samples kept, 
 
( 9 ) 
 
 marked with the results of the analysis. By this means 
 experience will be gained, and the look will almost be suf- 
 ficient guide. Take and pulverise a small sample taken from 
 the bulk, and dry it well at a temperature of 380 F., so as to 
 make sure that all the moisture has been driven off. Take one 
 gramme of this powder and 20 grammes of oxide of lead, also 
 well dried ; mix them thoroughly together and pour them into 
 a hard glass test-tube, five or six inches long and half-an-inch 
 diameter. Weigh the tube and its contents carefully, and sub- 
 mit the tube to a white heat under a Fletcher's blowpipe until 
 all gases are driven off and the contents completely fused. Allow 
 the tube to cool, and weigh the residue in the tube. The weight 
 lost is carbonic acid, the oxygen of which has been taken from 
 the lead oxide, while the carbon is all that there was in the 
 graphite. For every 20 parts of loss there must have been 12 
 parts of carbon. 
 
 Again, the duration of the positive carbon is a matter of 
 moment, since it is de rigueur that the waste should not exceed 
 lin. per hour ; and as this is now an accomplished fact, it has 
 become a necessity. For some inscrutable reason, foreign 
 makers supply only lOin. positives unless otherwise ordered, so 
 that with lin. in the holder, and something to spare to save the 
 lamp from heating, little is left besides the Sin. required for 
 eight hours' light. I have met with no material in its native 
 state capable of reducing the consumption to this standard ; 
 the prepared material employed by the writer satisfies these 
 conditions. 
 
 Another defect, and one which makes the otherwise perfect 
 carbon valueless, is curvature. This takes place in drying and 
 firing. If not properly or sufficiently dried, carbons will not 
 only bend during the firing process, but break into several 
 pieces. The method of avoiding this will find its place further 
 on. With lampblack carbons, fully 50 per cent, will be 
 curved, but with the prepared material described at length, 
 
and with proper care taken in the drying, no anxiety need 
 arise. Another matter of grave importance, the consideration 
 of which, if neglected, will lead to most serious results, is the 
 following. A carbon may be selected, placed in the lamp, and 
 found to give perfect results with a normal f\in. arc. Another 
 candle from the same parcel, in the same lamp, gives only 
 a T 2 3-in. arc; whilst yet another finds its place of rest at Jin. 
 Now, to adjust these varying arcs to produce equal light, the 
 top in one case, and the bottom of the lamp in the other 
 case, will have to be weighted so as to adjust the arcs to a 
 normal length. But when lamps are running in series this 
 constant readjustment becomes impracticable ; yet without it 
 one lamp would be burning properly, another would be burn- 
 ing, but giving a minimum of light, whilst the other would 
 be flaming each carbon good in itself, but bad when burnt in 
 series. To what, then, is this irregularity to be ascribed, and how 
 is it to be obviated ? The answer is simple. The hydraulic press 
 must be fitted with an atmospheric gauge, capable of measuring 
 at every moment the amount of atmospheric pressure to 
 which the carbon matter is being subjected. The best amount 
 of pressure having been discovered, and care being taken that 
 all the carbons shall be made at that pressure, does it not stand 
 to reason that carbons pressed out at 50 atmospheres would give 
 different results to those pressed at 150 1 And when we hear, 
 as I have heard, that carbons are irregular, and not so good as 
 they used to be, can it not be well understood how a careless 
 workman in charge of the press, receiving a bucket of pugged 
 material, and passing it through the press without taking any 
 heed of the difference in pressure indicated by the gauge 
 amounting, perhaps, to 20 or 30 atmospheres, vitiates the per- 
 fection of the carbon, and does irreparable mischief 1 The 
 matter lies in the hands of the pressman, who has the gauge 
 for his guide. The man who pugs and prepares the material 
 requires an amount of experience that can only belong to an 
 
( n ) 
 
 expert to enable him to grasp a lump of material, and say 
 within 10 atmospheres what the pressure required to pass it 
 through the press would be. Several puggings, and at least 
 three days, are necessary in the preparation of the material. 
 From experience I can say that 120 atmospheres produce good 
 results nothing under. I have never been able to prepare 
 material requiring a pressure of more than 160; at 120 the 
 carbons leave the press quite warm. 
 
 The carbons made by good makers sometimes mushroom. 
 By this is meant the formation of a head resembling a mush- 
 room button upon the extreme point of the negative, which 
 appears to consist of particles of impure matter thrown down 
 from the positive carbon. If this mushroom top increases and 
 does not break off of itself, which it often does, it may touch 
 the centre of the positive cored carbon, in which case the cir- 
 cuit becomes closed and the lamp extinguished. This fault I 
 attribute to impurity of material and the presence of silica, 
 which fastens itself like a glass bead to the point of the negative, 
 and, being a powerful non-conductor, tends to arrest the current. 
 This bead is formed by the electric current searching out the 
 distributed particles of silica in the positive, but without power 
 to vaporise the globule which it deposits on the negative. 
 Pure prepared material is a certain cure. 
 
 The lengthening and shortening of the arc, often described 
 as jumping, and which causes the rapid fluctuations in the 
 amount of light, is the characteristic of a bad carbon, badly 
 made with bad material, and owes its existence to a complica- 
 tion of the faults previously mentioned in an exaggerated 
 form. 
 
 A word or two upon coppered carbons. Good wine needs nc 
 bush, and a good carbon needs no copper. This coating is 
 given to hide a defect ; perhaps more than one, since it may be 
 put on to hide a defective surface, whilst a perfect carbon can- 
 not have a bad one. But the chief reason is to get over want 
 
( 12 ) 
 
 of exterior hardness a very serious fault. Good carbons should 
 be extremely hard on the surface, gradually softening towards 
 the centre, which in the case of cored carbons consists of pure 
 graphite in powder. A carbon insufficiently hardened (hard- 
 ness being only acquired by repeated saturation and carbon- 
 ising) burns to a lengthened cone. This is due to the air 
 rushing in to supply the vacuum formed, and which, meeting 
 with little resistance, disintegrates the carbon, and works it to 
 a lengthened cone, presenting only a small point to support 
 the arc, the light in consequence becoming seriously diminished. 
 
 PREPARATION OF MATERIAL. 
 
 Those who desire fuller information upon the nature of 
 graphite, the various parts of the world in which it is dis- 
 covered, and other details, may peruse " Spon's " with interest. 
 Two kinds are known, the amorphous and the foliated. The 
 amorphous is quite unsuited to our purpose, and contains often 
 only 45 to 50 per cent, carbon, the rest being argillaceous matter. 
 The foliated under careful test has shown as high a percentage 
 of carbon as 99'679. Crucibles and pencils are made of the 
 amorphous. The foliated graphite which comes to us from 
 Ceylon can be readily obtained in any desired quantity 
 from the Mincing-lane brokers, at prices varying from ,16 
 to ,22 per ton, and is well adapted to our purpose. It 
 comes to us in the form of nodules, which require careful 
 selection, and breaking under the hammer in small pieces to 
 extract any quartz discoverable. After careful hand-picking, the 
 graphite must be ground to a fine powder in a mortar or pug 
 mill. After this, the powder should be placed in cast-iron 
 mortars or crucibles, and mixed with chlorate of potash in the 
 proportion of 181b. of graphite to lib. of the chlorate, and 21b. 
 of sulphuric acid (sp. gr. 1 '8) to 1 Ib. of graphite. This mixture 
 must be moderately heated until the last fumes of the chlorous 
 
( 13 ) 
 
 gas are evolved, and then allowed to cool, when the excess of 
 sulphuric acid should be decanted. Upon this sulphated and 
 oxidised mass pour a small quantity of fluoride of sodium, 
 stir well together, and after the chlorous vapours have passed 
 off, the hydro-fluoric acid set free by the combination of 
 the sodium with the sulphuric acid will convert any silica pre- 
 sent into a gaseous fluoride of silica, which passes away in 
 vapour and leaves the mixture pure. This mass should after- 
 wards be thrown into water and well washed, placed again in 
 crucibles, and submitted to a red heat in a furnace. This will 
 cause the whole mass to swell and disintegrate, forming a light 
 flocculent powder floating on the surface, which must be collected 
 and dried. This impalpable powder is yet unsuited for carbon 
 making. A cutting sharp grain must still be imparted to it, 
 and to arrive at this the thin flat plates of gas-retort carbon, 
 taken from the top of the retorts, must be taken and ground 
 fine, in the proportion of one of gas carbon to three of graphite, 
 with which it must be well incorporated in the initial stage, so 
 that its impurities may be equally eliminated. You cannot obtain 
 these thin carbon plates without purchasing the heavier and 
 impurer shoulder-pieces. However, the refuse may be econo- 
 mically employed in furnace heating, as it gives great heat and 
 possesses great durability. 
 
 The material so far prepared must be thoroughly dried, and 
 pugged with sufficient caramel or carbonised sugar, mixed with 
 water, to make a moderately stiff paste, placed in ovens or 
 crucibles, and thoroughly carbonised. A material is thus pro- 
 duced possessing all the requisites desirable. It is perfectly 
 pure, and when ground fine possesses a sharp-cutting grain and 
 great hardness, allowing the molecules under pressure to com- 
 bine to the exclusion of occluded gases, which allows of a per- 
 fect homogeneity which could not be otherwise attained. It 
 will be necessary to crush these lumps so prepared in a stone 
 crusher, and after crushing grind them in a mortar mill, pass- 
 
ing the powder through a rotary sieve, working from the 
 counter-shaft, having 90 meshes to the lineal inch, or 8,100 to 
 the square superficial inch. 
 
 Having described the graphitic material or compound used 
 in carbon candle-making, the manipulation required in the 
 preparation of the stuff, also the way in which the granular 
 condition is acquired, and which experience shows is essen- 
 tial to the production of a good carbon, I now proceed to 
 explain the method of preparation of the saccharine matter em- 
 ployed in combining ' t the material to form the paste, or pugged 
 stuff possessing the special properties requisite. All materials 
 hitherto experimented upon and used to combine the powdered 
 graphite, such as tar, oils, gums, hydro-carbons, rosin oil, &c., 
 swell the carbons, after being subjected to a drying heat, in- 
 ducing porosity ; and no available after-process is satisfactory. 
 Saturation with carbonised sugar fills the pores, but leads to 
 much waste of time and enhanced expense, both of material 
 and labour. Wellnigh two of the first years which I devoted to 
 this subject were spent in efforts to overcome this difficulty 
 efforts not made in vain, since not only is the perfect drying 
 of the carbon without swelling now a fait accompli', but 
 also the valuable property of contracting under great heat has 
 been realised, so much so, indeed, that allowance has now to 
 be made in the gauge of the die, which, to allow for this con- 
 traction, requires to be increased -f^th of a millimetre upon a 
 10-millimetre carbon. Prolongation and intensity of firing 
 increase the contraction. The high conductivity possessed by 
 graphite is, no doubt, due to its peculiar dense condition ; but 
 to impart this quality of density we shall have to wait until 
 Nature invites us to work in her own laboratory. 
 
 The material I use to incorporate with the prepared graphite 
 powder is the best crystallised lump sugar. Crystallised 
 Demerara might do as well, but the difference in cost is so 
 slight that I prefer to adhere to the former ; and, therefore, 
 
( 15 ) 
 
 would cite " Tate's cubes " as the standard. I would urge that 
 no attempts be made to cheapen this part of the process. 
 Molasses and treacle are impure, containing, besides sugar, 
 acetate and sulphate of potassa, chloride of potassium, phos- 
 phate of lime and copper, mucilaginous and nitrogenous matters, 
 silica, water, and glucose. 
 
 Take, say, 201b. of Tate's cubes, put them into a boiler large 
 enough to hold four times the quantity, at least ; place the 
 boiler on a moderate fire, preferably made of small broken coke, 
 and allow the sugar to melt gradually, without burning, until 
 
 FIG. 1. 
 
 it partially carbonises. Careful stirring arid watching must be 
 exercised, and it must never be at any moment left, or the sugar 
 is certain to boil over, catch fire, and be wasted. To avoid 
 these contretemps, a low range of -firebricks may be built (see 
 Fig. 1), on the top of which bars, B, of iron are fixed, imme- 
 diately over the fire, and extending over a sunk pan, C. The 
 boiler A is free to slide upon the bars, so that in the event of 
 a sudden rise it can be pushed over the sunk pan, and any 
 overflow of the contents be recovered without loss. 
 
 I have signally failed in the use of beet sugar. When converted 
 into caramel, it decidedly loses the adhesive, sticky property 
 
( 16 ) 
 
 belonging to the cane sugar. Sugar heated to 400 F. loses 
 two equivalents of water, becomes brown, cannot be re- 
 crystallised, and is then known as caramel. At 410 F., the third 
 equivalent of water is set free, and complete carbonisation com- 
 mences. In this state the spec. grav. is 1'594. Sugar at this 
 stage ceases to rise in the boiler. By observation alone can the 
 point at which the caramel is in the exact state required be 
 determined. As a guide, however, it may be stated that the 
 process is well-nigh completed when the sugar ceases to rise in 
 the boiler \ but make sure that this is not owing to the lowering 
 of the fire. Small volcanic eruptions covering the surface, 
 bursting and emitting vapours which are caused by the third 
 equivalent of water becoming free, is another sign. If any 
 smoke accompanies the vapour, the operation is certainly com- 
 plete. The process had always better be carried a little too 
 far than not far enough. In the first case, the caramel is 
 only slightly less adhesive, leading to extra pugging ; but in the 
 second case the carbons will suffer, and this will not be ascer- 
 tained until the carbons have been made and placed in the 
 drying-room. To determine the right moment for closing the 
 operation above described needs practice, the appearances 
 shown whilst boiling depending much upon the state of the 
 fire, and being sometimes misleading and puzzling to any but 
 the expert. When the boiler has been duly removed from the 
 fire, add gradually small quantities of hot water. Adding cold 
 water in bulk is an experiment not likely to be repeated more 
 than once. Stir well while adding the water, and let this be 
 only sufficient to keep the caramel liquid when cold. When 
 required for use this stock must be diluted with cold water to 
 about the consistency of golden syrup one part, water two 
 parts, possessing a slight stickiness when rubbed between the 
 fingers. 
 
 The carbon powder may now be taken out of the storage 
 bin, in quantity sufficient for a pugging, with a pug-mill 
 
( 17 ) 
 
 6ft. Gin. over all, with rollers 18cwt. each. A 10 horse-power 
 engine is sufficient to drive the mill and the press. About 
 three tin buckets, holding some 201b. each of the powder, 
 may be taken as a charge, and this weight of material 
 will suffice to form 1,000 10-millimetre carbons. Pug this 
 stuff, adding sufficient of the liquid caramel to form a 
 paste ; when pugged, clear out the mill, and transfer the 
 paste to a covered bin, or keep it in the buckets, marking the 
 date when first pugged on the buckets. Continue making 
 these batches, repugging every day until ready for the press. 
 This pug-mill should be kept solely for the purpose of pug- 
 ging, and not grinding, as the pan could not be so thoroughly 
 cleaned as to avoid the risk of leaving some small fragments of 
 material, which would find their way into the carbon rods, tend 
 to divert the current when used, and give rise to complaints. 
 
 A portion of each day may be employed in pugging new 
 stuff, and previous puggings must also have the process repeated 
 daily, each day's batches being kept separate. Three or four 
 days, depending somewhat on the temperature, will bring the 
 stuff to a proper condition for forming into carbons. The paste 
 may be advanced, if required, by placing it in a warm room. 
 By squeezing the paste in the hand the expert can tell the atmo- 
 spheric pressure requisite to drive it through the press within 
 some 10 atmospheres. This pressure must not be under 120 
 atmospheres for a 10-millimetre carbon. This same paste, how- 
 ever, that requires 120 atmospheres for 10 millimetres, would in 
 the case of 14 millimetres pass through at 90 to 100 atmo- 
 spheres. The first few carbons issuing from the press will be no 
 guide, as the pressure will -increase until the carbon in the 
 cylinder becomes well compacted ; this will occur whenever 
 the cylinder has been emptied for the purpose of changing the 
 die or altering the coring needle. 
 
 If the cylinder has been charged, the pressure right, and 
 the carbons issuing from the die, the carbons will require 
 
 D 
 
cutting to normal lengths. For this purpose I use a small 
 instrument, which answers the purpose, and cuts off the 
 carbons to a standard length as they issue from the press. 
 It is made of tin, about three-quarters of an inch wide, 
 bent at right angles at D (Fig. 2). The carbon rod A, 
 
 FIG. 2. 
 
 shown in section, lies in the grooved top B, the thumb 
 slightly pressing the carbon, while the spring-back cutting part 
 is pressed forward with the fingers ; this closes over the groove 
 B, separating the carbon at the normal length. 
 
 f 
 
 t 
 
 N 
 
 U: 
 
 c 
 
 \ 
 
 D, 
 
 1 
 
 f\ 
 
 => G 
 
 21 
 
 B i 
 
 
 FIG. 3. 
 
 As the carbons are thus cut off they must be laid in grooved 
 plates, prepared by a machine (Fig. 3) constructed for the 
 purpose, consisting of a cast-iron frame, with grooved dies ; the 
 lower die B is fixed to the frame EE, whilst the upper die A slides 
 vertically in the frame. This upper die is actuated by a lever 
 handle C, working eccentric chuck-pieces, fastened to each end 
 
( 19 ) 
 
 of the bar, actuated by the lever, and turned by the handle 
 so as by its action to bring the dies in contact. The sheet tin, 
 being laid upon the guide-plate G, is pushed forward through 
 the open dies, and takes the form of the grooves upon lowering 
 the arm ; an iron rod at the back also guides the plate in its 
 passage, keeping the grooved plate straight. These dies may be 
 taken out of the frame to give place to others suitable for car- 
 bons of other gauge. These grooved plates (Fig. 4) upon which 
 the carbons are carefully laid as soon as they issue from 
 the press, are so constructed and figured that the carbon rods 
 are perfectly protected from the superimposed plates pressing 
 upon them, the plates resting upon each other, the carbons 
 being protected within the grooves ; thus a dozen plates filled 
 
 ^_ X^XTrvXiiX'.^ ^^X^^^xW^^^CX- ^^X^^SS'/C^X^^/?^ 
 
 SECTION. 
 
 FIG. 4. 
 
 with carbons may be safely piled one upon another before 
 leaving the pressman. These plates should be made two inches 
 longer than the carbons, of the best charcoal tin, and rather 
 stout. In these corrugated plates the carbons should remain 
 until removed for firing in the kiln, tin being a good conductor 
 of heat, accelerating the process of drying. When the press- 
 man has filled a dozen of these plates they may be placed 
 on a rack in the press-room. Here they may remain until 
 the evening, when they may be transferred to the first drying- 
 room. 
 
 I must here describe a fault which may arise by the time 
 this stage is reached. The carbon rods leave the press to all 
 appearance perfect ; half-an-hour afterwards every carbon is 
 lined with fine fissures at right angles to the length of the 
 
carbon rod, increasing in size and depth as they dry.' ^ [The 
 fault arises from the die. I had occasion at the time 
 I first noticed this defect to renew iny stock of dies, and 
 
 A 
 
 u. 
 
 A 
 
 "A 
 
 u. 
 
 Ul 
 
 2: 
 
 f 
 
 \ 
 
 
 21 
 
 
 / 
 
 
 
 cc 
 
 
 
 
 
 
 c 
 
 
 _ 
 
 
 FIG. 5. 
 
 gave a model to an experienced workman, who furnished me 
 with duplicates, apparently in every respect similar to model, 
 only slightly reduced in gauge. No eye-measurement "could 
 
 FULL SIZE DIE 
 
 MODEL 
 
 FIG. 6. 
 
 detect any difference ; but on slightly filing off the shoulder 
 at A (Fig. 5) the difficulty disappeared, and the carbons were 
 perfectly freed from this blemish. A model die for a 10- 
 millimetre carbon is shown in Fig. 6. 
 
( 21 ) 
 
 Prior to the discovery of the cause of the lateral fissures 
 referred to I conceived it possible that this fault might have 
 its origin in the speed of the hydraulic pump. This does not, 
 however, appear to affect the carbons, although a speed of 
 about 40 revolutions per minute would seem sufficient. 
 
 The second fault we may have to contend with is the sudden 
 appearance of longitudinal fissures, increasing both in depth 
 and width, a few hours after the carbons have been placed in 
 the drying-room. This is due to the fact that the caramel has 
 not been sufficiently carbonised, and still retains some of the 
 water of crystallisation. 
 
 The third and last fault which has to be contended 
 with is the curving of the carbon rods as they issue from 
 the press; sometimes only a few carbons are affected this 
 way, sometimes many. It must be caused, I think, by a 
 difference of hardness in the pugged stuff, and, if so, the best 
 way to avoid it will be to have the stuff repugged just before 
 use. The curved carbons must be rejected, whether the cur- 
 vature is great or small. A curved carbon will aways come out 
 crooked in the firing. 
 
 I)RYIX< -ROOMS. 
 
 The most efficient and certainly the cheapest way of heating 
 the drying-rooms is by means of the slow-combustion stove. 
 The temperature can then be maintained, without sensible 
 variation, for quite six hours, if the fires are carefully built up 
 with small broken coke, and without renewal ought to burn 
 for 12 hours before extinction. The first drying-room should 
 be 20ft. long and 8ft. wide, and 8ft. to the springing of the 
 arched roof, with the stove in one corner nearest the door. By 
 this arrangement the furthest end of the room is only slightly 
 heated, whilst increased heat obtains near the stove. Fig. 7 is 
 a ground plan of the first drying-room. A is an iron sliding 
 door, BCD the shelving, framed and built up with angle iron, 
 
as already described. At the end E (Fig. 8), and on the lower 
 shelf the temperature will remain nearly constant at about 
 75 F., while the temperature at G will be about HOdeg. ; at 
 H, nearer the stove, the temperature will be 120deg., and at F, 
 
 
 D 
 
 { 
 
 
 
 
 
 Q "AZQ'Q 
 
 
 
 1 
 
 
 FIG. 7. First Drying-room (Ground Plan). 
 
 140 F. These temperatures will always bear a relative pro- 
 portion to each other, and never vary much. 
 
 The carbons are first placed on the shelves at E, removed to 
 G, and after 36 hours to F, where they remain for 12 hours ; 
 
 - G 
 
 -20-fO- 
 
 H 
 
 FIG. 8. First Drying-room (Section and Elevation). 
 
 afterwards removed to the second drying-room and kept for 
 two days at a temperature between ISOdeg. and 200deg. This 
 drying-room should be much smaller (about 10ft. by 8ft.), 
 after which the carbons should be ready for firing. They 
 ought not to be indiscriminately taken, but should be sepa- 
 
( 23 ) 
 
 rately handled, as some of the interior carbons may not be 
 sufficiently dry. Their condition is best ascertained by try- 
 ing to bend them ; if they feel the least inclined to bend they 
 are unfit, and must remain until they are fit. Firing before 
 fitness means breakage. 
 
 FIRING THE CARBONS. 
 
 I now describe my method of building the kilns. A cleanly 
 but costly system has been tried, but not generally adopted, I 
 believe, of firing carbons by means of gas and atmospheric air 
 combined. The system may be good, but the expense appears 
 to me to be too great. My system is to lay firebrick on the 
 flat, and well grouted with cement ; upon which build the two 
 long sides of the kiln B B (Fig. 9), with bricks on edge, well 
 laid in cement, composed of three parts of Stourbridge clay and 
 one part iron grindings. The sludge arising from the grind- 
 stone, well mixed, forms a valuable cement, and will not crack 
 under extreme heat. A crack in the kiln admits air, and dis- 
 integrates the carbons. The short ends of the kiln are built 
 with bricks on the flat, A A (Fig. 9) ; the thicker the ends are 
 the better, but 4Jin. are sufficient. The kiln must be 2in. 
 longer than the width of two of the plates, and 2in. wider 
 than the length of the plates, which should be 2in. longer 
 than the carbons. This allows for filling in with sieved ashes, 
 which should be kept perfectly dry ; any trace of moisture will 
 show itself in the partial disintegration of the carbons. 
 
 The lines at C C (Fig. 9) show the direction in which the car- 
 bons are to be packed. The ends A A are made thick to avoid 
 any lateral heat ; the fires being made at D D tend to keep the 
 carbon straight. Any heat in the direction of A A would in- 
 evitably twist them. After the kilns are packed and covered 
 over to a depth of 2in. with the ashes, the surface had better 
 be covered with a sheet of tin, and built in with firebricks, 
 well filled in between the joints with the prepared cement, 
 
( 34 ) 
 
 and pointed before using the kilns. After constructing the kiln 
 a fire should be made in and around, so that it may be per- 
 fectly dry when used. The trays in which the carbons are 
 fired must be kept solely for that purpose. The furnace con- 
 
 FIG. 9. Kiln (Ground Plan). 
 
 sists of a ring of firebricks, built as shown at E E E E (Fig. 9), 
 with interstices between them of about 2in., EE. (Fig. 10). 
 This circular wall should impinge upon the kiln at F F (Fig. 9), 
 to keep the heat from the sides A A. The bricks are laid loose 
 
 FIG. 10. Brick Wall of Furnace. 
 
 without cement, and are built to the height of the kiln. The 
 fires must be lighted at D D, gradually adding coke as the fire 
 advances, eventually increasing the height of the walls about a 
 foot higher than the kiln. If the fire was lighted the first thing 
 in the morning, it should not be piled over the top until the 
 
( 25 ) 
 
 evening, and this done only by covering with coke and banking 
 over with cinders and clay. Allow this fire to die out, when 
 the carbons will be found, on opening, when cold, sufficiently 
 carbonised for the after-process. 
 
 The second or final firing has for its object the contraction 
 of the carbon molecules, and increase in density and conduc- 
 tivity. The first firing, however, must be sufficiently pro- 
 longed to effect the carbonisation of the caramel, otherwise 
 when placed in the hot saturating bath the carbons would 
 become partially disintegrated. The second firing should be 
 conducted as follows : Bank up as before mentioned in the 
 evening; on the following morning rake out the lower holes 
 all round, and keep a strong fire all day all over the kiln. 
 Bank up partially in the evening, and more completely the 
 following morning. Three days after, on opening, the kiln 
 will be found at a white heat. When it is desired to with- 
 draw the carbons, pull down the retaining wall, and clear 
 away the fire and accumulated ashes, and expose the kiln to 
 the air. Do not open before you can bear your hand on the 
 outside. The top bricks may be taken off, and the kiln 
 emptied, if the carbons are not too hot to hold. I should 
 mention that straight carbons will twist if exposed to the air 
 hot from the kiln ; but if once allowed to cool, no after-firing 
 has the least effect upon them. This form of kiln gives out 
 an intensity of heat sufficient to melt wrought-iron, is very 
 easily regulated and modified by blocking up more or less the 
 interstices, and the firing can be prolonged for any length of 
 time by banking up with brick clay, which in itself partakes 
 of the nature of a slow-burning fuel. 
 
 The condition of the carbons as they leave the kiln may, in 
 some measure, be determined by the sound emitted when 
 struck one upon another. This sound should be distinctly 
 metallic, the more so the better. This metallic ring increases 
 in the finished carbon to the clearness of a small silver bell ; it 
 
( 26 ) 
 
 is only heard in good carbons. A carbon, however, is not neces- 
 sarily perfect, although the ring may be perfect. It is very 
 desirable to know in what condition the carbon leaves the 
 kiln, to be able to determine what it lacks, and what it requires 
 to perfect it. The feebleness of its metallic ring will indicate 
 that it is soft. If you insert the end of one of these carbons 
 in a clear hot fire until it is red hot, and on taking it out 
 observe it closely, you will find it smoulder for quite five 
 minutes before it dies out. If you scrape this end with a knife, 
 much carbon powder will detach itself. Observe also the 
 colour of the powder ; if reddish, the carbon is impure. The 
 colour of the burnt part of a good carbon is a bluish steel-grey. 
 This holds good for all the best foreign carbons. If the carbon 
 just taken from the kiln had been put in circuit, the current 
 would have made it red hot from end to end, but no arc would 
 have formed for want of conductivity. By imparting increased 
 density to the carbons, these symptoms of an unfinished carbon 
 are altered, and give place to easily recognised signs, which 
 are always the characteristics of a good finished carbon, viz., 
 rapid extinction of the heat after firing the end, and hard- 
 ness of surface, which becomes as hard as a rod of iron, and 
 does not allow of the surface being disturbed or injured by 
 scraping. 
 
 SATURATING. 
 
 Now, to effect this change, the carbons must be saturated, 
 and the operation is carried out as follows : An iron tank, A 
 (Fig. 11), is provided, partitioned off at one end to contain hot 
 water, and the tank is placed upon a range, with two fire- 
 places, B B. Fires being lighted, the tank must be filled to 
 the depth of 1 2in. with water and a sufficient quantity of sugar 
 to form a syrup. The boxes, D, constructed of tin or zinc, 
 with perforated bottoms, are filled with the carbons packed 
 vertically, and placed in the tank. After boiling half an hour 
 
the boxes are removed, and placed upon iron bars, E E, to drain. 
 When cold replace them in the tank, and repeat the process 
 three or four times. Drain them well, and wash the carbons 
 freely in the hot-water end. Remove them when cold (by 
 which the liquor is drawn into the centre) into the third dry- 
 ing-room, and keep them there for 24 hours at a temperature 
 of between 180 and 200 F., after which the whole process 
 may be repeated until the carbons show the characteristics 
 described above of a finished carbon. 
 
 000000060660000000000000000000000 
 
 A 
 
 FIG. 11. Saturating Tank. 
 
 Some time ago, in hunting over patents at the Patent Office, 
 I came across a patent (provisional) for saturating carbons. I 
 have never tried it, but I give it here for what it is worth, and 
 I hope not to the injury of the patentee. Steep the carbons in 
 a dilute solution of calcic chloride, afterwards into a saturated 
 concentrated bath of potassa chlorate (at a boiling temperature), 
 dry afterwards by heat, and whilst hot steep in another bath 
 of fused sodic nitrate, and wipe clean whilst hot. What 
 effect this Turkish bath may have upon the carbons I am not 
 prepared to say, but I think it might be worth a trial. 
 
 CORING. 
 
 Numerous experiments on the different modes of coring 
 enable me to say that the method of coring is of less im- 
 
( 28 ) 
 
 portance than the condition of the material. A hard core 
 is useless. The best possible form is that of a compressed 
 powder, breaking up easily with the point of a knife ; but if 
 too softly packed the current is apt to throw it out, and leave 
 a vacant place which degenerates into a hole, and the light 
 becomes diminished. 
 
 Every manufacturer of carbons has, doubtless, a way of his 
 own of coring carbons. I have experimented on three very 
 different methods, and I now explain them. The most certain, 
 the best, but I must also add, the slowest way, is as follows : 
 Take the finished carbon, and fit a small funnel with a nozzle 
 large enough to fit close exteriorly, partially filling the funnel 
 with graphite powder, purified, and fired at a white heat in a 
 graphite crucible, and then tamp the powder. With a little 
 practice this is quickly done, but scarcely quick enough for 
 commercial purposes. The core, however, by this method can 
 be made hard or soft, as desired. The second method is per- 
 formed much more rapidly by using a specially-constructed 
 coring machine. Before, however, describing it more particu- 
 larly, I must mention certain difficulties. If the powdered 
 graphite is mixed with water, on applying pressure with the 
 screw the water separates, rising above the piston, whilst the 
 powder hardens with the pressure and becomes like stone. 
 Powdered graphite mixed with caramel fills the carbons easily, 
 but after firing the carbon becomes harder than the ordi- 
 nary lead pencil, and is spoilt. If oils are used, or 
 liquid gums, the core is partially driven out in drying, the 
 medium having a tendency to swell, so that the carbon is but 
 partially filled, and the paste becomes attenuated and honey- 
 combed when it should be close and solid, and the carbon may 
 flicker. The only medium that answers, as far as my experience 
 goes, is heavy mineral oil, with a specific gravity of l'S2. Mixed 
 with this, I have not found the carbons swell nor get too hard 
 when fired. The third and last method which I wish to men- 
 
( 29 ) 
 
 tion is the gravitation method. A cask partially filled with the 
 powder is placed under a tap of running water. This cask com- 
 municates by means of pipes with boxes packed with carbons, 
 and as the cask runs over it fills the boxes, and the fine powder 
 floating in the water gradually subsides, filling the core-holes. 
 The cask requires stirring up occasionally. Where many thou- 
 sands have to be cored this process will be employed, the system 
 being automatic. The surface water should be decanted when the 
 carbons are completely covered, and the boxes removed to the 
 
 FIG. 12. Coring Machine. 
 
 third drying-room, and left to dry, when the carbons may be 
 withdrawn, wiped clean, and dried upon the shelves. 
 
 The coring machine (Fig. 12) explains itself. A is the piston 
 actuated by a screw revolving upon the wheel E. C is a tap 
 fixed to the nozzle B, which, tapering, enters the core-hole of the 
 carbon. The finger is placed at the end of the carbon. Two 
 or three turns of the screw will put sufficient pressure upon the 
 material to fill several carbons, by simply turning the tap at C, 
 which must be turned on and off as required. This method of 
 coring must be adopted after the carbons are saturated and 
 have been washed for the last time, the core hole being wet and 
 allowing of the free passage of the paste. The pointing of the 
 
( 30 ) 
 
 carbons is most easily effected on an emery wheel, revolving, as 
 all emery wheels should to be of any use, at a high speed, 
 1,200 to 1,500 revolutions per minute. 
 
 I now come to an important matter, the construction and 
 adjustment of the dies and coring needle. The pressure to which 
 this needle is subjected when embedded in the pugged stuff is 
 enormous ; 120 atmospheres have to be resisted. On more than 
 one occasion I have had the wings of my coring needle crumple up 
 like paper ; and to prevent this a proper form must be employed. 
 A A (Fig. 13) represent the sides and base of the cylinder, E E 
 a loosely-fitting die resting upon this base. C C is the die proper. 
 
 FIG. 13. Die and Needle. 
 
 The coring needle F must be screwed into the wings D D, 
 which are four in number. The size of the core-hole will be 
 determined by the size of the needle at H, no matter what the 
 thickness of the shoulder. The die proper C C is sunk below 
 the outer die E E, so as to allow the wings D D to find a 
 bearing against the outer die E E. The wings are made sharp- 
 edged above, gradually thickening to 4 mm. at the base.. 
 The coring needle, when properly adjusted, will always be 
 central, unless the wings get twisted by the pressure. The 
 motive for making the die E E fit loosely is to facilitate the 
 removal of the needle and the inner die. When necessary this 
 is most effectually done by taking a copper tube, placing it at 
 M M, and striking it with a heavy hammer ; this will bring up 
 
the pugged stuff and the dies, instead of picking out the 
 pugged stuff; the proper die will also be preserved from 
 damage. 
 
 The hydraulic press (Fig. 14) which I had constructed was 
 made of steel, but this is not necessary, as cast iron will do 
 perfectly, since it will not be required to stand a greater pres- 
 
 rl1 F 
 
 FIG. 14. Hydraulic Press. 
 
 sure than 200 atmospheres ; the ram should be 1 2in. diameter, 
 and the plunger quite Sin. diameter, if it is required to turn out 
 20,000 10mm. carbons per week. With a cylinder holding the 
 pugged material lOin. deep, the plunger is keyed into the ram 
 at E, so as to be removable, if necessary. The weights F F 
 consist of iron slabs, Icwt. each, suspended from the chains 
 
( 32 ) 
 
 N N. Five slabs on either side are quite sufficient to raise the 
 piston. The chains are fastened to the lugs B' B', which are 
 part of the casting. The cylinder L with collar C drops into 
 the framing MM, fitting tight, to prevent rocking. These 
 parts must be made with greatest nicety. When the 
 piston A has descended to its greatest depth it is sure to jam, 
 unless the cylinder is strongly fixed. This may be done in 
 many ways, and need not be explained, but must be attended 
 to. I have employed a leverage of at least 50 tons to separate 
 the piston from the cylinder. Not only must the cylinder be 
 immovable, but the piston must be treated in the following 
 manner. The piston must be lengthened, say 1 Jin., by a piece 
 screwed on to the piston end by the screw C, the head well let 
 into a counter-sink. The addition must be a shade larger in 
 diameter than the piston proper, with a small groove, D D, at 
 junction. Before the piston enters the cylinder a little heavy 
 lubricating mineral oil should be rubbed round the piston. 
 After I had adopted this very simple plan the pugged stuff 
 never clogged the piston higher than the groove D D, and the 
 piston always rose unaided. 
 
 The manner of setting up the press needs no description ; 
 but being some five tons in weight, requires a good foundation. 
 A word about the pumps. Sometimes the high-pressure pump 
 gets out of order, a fact known by the arrested motion of the 
 descending carbon-rod. Either the suction valve, or the cut-off 
 valve, is out of order, or leaks. To dismount the pump is 
 needless. Empty the tank, and wash it out with hot (not boil- 
 ing) water. Throw washing soda, or soft soap, into the water, 
 making it pass through the pump to cleanse the valves, which 
 in time invariably become clogged with grease. The ram has 
 constantly to be greased to keep the collar soft and pliable. 
 Sometimes the piston of the pump requires repacking. Besides 
 these annoyances easily remedied the pumps should work 
 indefinitely. 
 
DESCRIPTION OF PLANT REQUIRED. 
 
 Having fully described the various methods employed in the 
 manufacture of carbons, I am led to the belief that the de- 
 scription of the plant required will be helpful, also a ground plan 
 
 -320-- 
 
 FIG. 15. Carbon Manufactory. 
 
 of premises (Fig. 15) suitable to an output of 20,000 carbons 
 per week, showing the various rooms required and their general 
 arrangement. In the ground plan the rooms, as will be ob- 
 served, are large enough to contain two presses and two pug- 
 
( 34 ) 
 
 mills, so that a larger number than that stated could be manu- 
 factured without increased accommodation. The yard, how- 
 ever, would not be large enough for over 20,000. Of course, it 
 will be seen that the plan is drawn rather for the purpose of 
 showing the size of the rooms required and the most conve- 
 nient method of arrangement, which may be modified as cir- 
 cumstances require. I place my engine-room facing the open 
 for the convenience of installing both the engine and the 
 general machinery. The pug-mill is best situated near the 
 engine-room, whence the motive power can be directly supplied 
 without the intervention of the counter-shafting, by means of 
 which all the lighter machinery will be driven. The store 
 room N receives all the material required for the processes. 
 From this store-room the material passes at once to the 
 crusher I, from there into the grinding mill E, and thence 
 it is removed into the yard to undergo the purifying process, 
 and to the carbonising pans V. After being pugged and 
 recarbonised it is ready for crushing, grinding, and pugging, 
 and it is placed in the bins H until required. These bins 
 are situated for convenience close to the pugging mill. When 
 ready the material is taken to the table L, close to the press J , 
 whence it issues in the form of carbon rods, which are placed in 
 their separate trays and removed to the shelves H. From here 
 the carbons are removed to the first drying -room Q. After drying 
 the carbons in the first drying-room they are to be removed into 
 the second, T ; and it will be observed that by the adopted 
 arrangement every process, beginning from the engine-room, 
 passes gradually through the rooms. From the second drying- 
 room they are taken to any of the vacant kilns U U to be sub- 
 mitted to the first firing. After this they are at once removed 
 to the saturating pan or tank, W, whence they are again re- 
 moved to the third drying-room. After the saturating process is 
 complete and the carbons are dried, pointed, and cored, they are 
 removed to the store-room A. The writer has erected premises 
 
( 35 ) 
 
 upon this model, and his study has been to avoid unnecessary 
 carriage of material from place to place, especially in the case 
 of the unfinished carbons, which in the early stages are tender 
 and easily broken. 
 
 Having said so much upon the subject, it will be desirable to 
 state the various speeds which I have proved most suitable for 
 carrying on the work. Given the speeds, it will be easy to 
 apportion the diameters of the pulleys, with reference to the 
 diameter of the main driving-wheel, to arrive at any speed 
 required. 
 
 Tt will be well to set down at first a 16 horse-power engine 
 capable of driving two pug-mills and two presses ; but this 
 will, of course, depend upon the scale upon which it is pro- 
 posed to establish the works. The pug-mills may be allowed 
 to make twelve revolutions per minute, not more. Even at 
 that rate of speed the material in course of pugging clinging 
 to the rollers will occasionally be ejected from the pan. To 
 avoid this, special provision must be made, such as a shield of 
 stout sheet-iron, fastened with bolts and nuts to the periphery 
 of the pan, and rising 18 inches. By the adoption of such a 
 shield the material falls upon it and returns into the pan. A 
 sliding door should be fitted to allow of access to withdraw the 
 pugged material. It will be also necessary to provide a collar 
 either of tin or (preferably) leather to surround the central axis 
 of the mill, and prevent the admission of the graphite powder, 
 which otherwise grinds away the axis, and makes it heavier to 
 work. The scrapers must also be specially designed, and 
 require to be made strong enough to resist the strain, which is 
 enormous. The pug-mill should be fixed upon a very solid 
 foundation, built of concrete, with stone pillars, to support the 
 iron legs of the mill, which must be bolted down to the stone- 
 work and run with lead. The grinding mill must be similarly 
 erected, running at a speed of 12 revolutions per minute ; but 
 to avoid the waste of powder and the dust arising from the 
 
( 36 ) 
 
 process it will be imperative to cover in the pan with an iron 
 (sheet) cover, fitting round the top, clearing the rollers, and sus- 
 pended by a cord and pulley, to be elevated, when the ground 
 powder requires to be taken out for sieving. The rotatory 
 sieve should work off the counter-shaft with a speed of about 
 60 revolutions per minute. The meshes of the sieve should be 
 of brass, and must be 90 to the lateral inch, or 8,100 meshes 
 to the superficial square inch. It will be inadvisable to pass 
 the ground powder through the machine direct ; it should be 
 passed first through a hand sieve, with 40 meshes to the lineal 
 inch, to extract any small fragments which may have escaped 
 being ground. Unless this is done, the feeding rollers of the 
 sieving machine will be arrested, and require constant atten- 
 tion. The delicate wire sieving will also be torn, allowing the 
 passage of these coarser particles to mingle with the finer 
 powder. In order to be certain that this sieve is sound, it will 
 be necessary at all times, before clearing the middle box, to 
 carefully pass some of the powder between the finger and 
 thumb so as to detect the presence of grosser particles. As 
 the machine is closely covered in, and cannot be examined 
 without some trouble, this will be seen to be clearly necessary; 
 and if grit is met with the machine must be dismounted 
 and the sieve renewed or otherwise repaired. The hydraulic 
 press and pumps have been described. The crushing machine, 
 so long as it effectually does its work, may be of any type, but 
 a disintegrator is not necessary nor advisable. 
 
 It may be as well to make some mention of the arrangements 
 of the yard, a portion of which must be covered in ; and this had 
 best be done with sheets of corrugated iron, 7ft. in length. The 
 plan shows the proposed method of arrangement sloping inwardly 
 from the outer walls. By means of this roofing the kilns will 
 be protected from the rain, whilst the centre of the yard is 
 open, affording a suitable place for treating the crude material 
 during the process of purification with the acid, and allowing the 
 
( 37 ) 
 
 escape of noxious fumes and generated chlorine gas. Storage 
 room also is required for coal and coke and clay. 
 
 The carbonising pans and range arrangements (Fig. 16) consist 
 of a series of cast-iron vessels, A A A, with close-fitting lids, built 
 into a range with suitable fireplaces, C C C. In these vessels 
 the purified material can be readily carbonised, the first process 
 of purification being carried out in the open yard, as well as 
 
 FIG. 16. Carbonising Pans. 
 
 the washing. It would be convenient and inexpensive to illu- 
 minate the premises with incandescent lamps instead of gas, 
 with the dynamo placed in the office to avoid the carbon dust. 
 
 COST OP PLANT. 
 
 I now propose to deal with the cost of the plant required for 
 an output of 20,000 10 mm. carbons per week, leaving it to 
 those immediately interested to make their calculations for any 
 larger number. 
 
 The practicability or otherwise of establishing factories in 
 this country for the manufacture of carbon candles must 
 depend upon the relative values of material and labour here 
 and abroad. A close investigation of the probable cost seems 
 absolutely necessary, nor would the subject be complete with- 
 out some knowledge of the kind. It would be unfair to take ex- 
 treme prices either way. The maximum would obtain in esta- 
 blishing works near London, where price of fuel alone would 
 
 ( UNIVERSITY 
 
( 38 ) 
 
 militate against success, and the rent of premises even more so. 
 I will not, therefore, base my calculations upon extreme prices, 
 but place the works somewhere in the vicinity of gas-works, 
 where coal and coke are cheap. If my prices are too low, 
 perhaps the margin of profit may, even with the addition, prove 
 satisfactory. It appears to me a matter of small question 
 where the works are established the cost of transit of the 
 material employed need scarcely enter into calculation ; but the 
 price of coal and coke is an important matter, overriding 
 every other consideration. In the neighbourhood of coal-pits 
 breeze can be had for next to nothing. This, mixed with clay, 
 a fuel much used in Belgium and the frontiers of Prussia, 
 makes an excellent fuel suitable to every purpose save the 
 engine, which requires coal. 
 
 The calculations I have made are based upon an output of 
 20,000 carbons cored per week, 10 mm. diameter and lOin. 
 long, or 15,000 as above and 15,000 negative solid carbons 
 6in. long and 6 mm. diameter. 
 
 The cost of the machinery delivered from the works will be 
 
 as follows : 
 
 s. el. s. d. 
 
 1 hydraulic press 100 
 
 1 pug-mill 60 
 
 1 grinding mill 35 
 
 J sieving machine 25 
 
 1 crushing machine 12 10 
 
 3 coring machines 25 
 
 1 corrugating machine 8 10 
 
 16 H.-P. engine 180 
 
 446 
 
 Iron-work, tank, cast-iron crucibles, furnace 
 
 tank 60 
 
 Counter- shafting, belting, &c 50 
 
 Adapting premises, &c 120 
 
 Kilns, furnaces, iron roofs 45 
 
 275 
 
 Rent, rates, and taxes , 50 
 
 771 
 
COST OP MATERIALS. 
 
 These calculations are based upon an output of 20,000 
 positive cored carbons per week, as before mentioned, upon 
 which a week's labour has been charged. 
 
 s. d. s. d. 
 
 900 Ibs. graphite, at 18 per ton 8 2 
 
 300 Ibs. gas carbon, at 25s. per ton 3 6 
 
 2,400 Ibs. sulphuric acid, at Id. per Ib. 
 
 (half of this is saved i 5 
 
 60 Ibs. chlorate potash, at 7d. per Ib ... 115 
 
 240 Ibs. Tate's cubes, at 3d 300 
 
 3 tons coal (engine), at 12s 1 16 
 
 8 tons coke (tiring), at 8s 340 
 
 23 6 
 
 COST OF LABOUR. 
 
 1 superintendent, per week 5 
 
 1 engine-driver, at 25s 150 
 
 1 yard man. at 20s 100 
 
 1 man (pugg), at 18s 18 
 
 Iboy at 12s 12 
 
 1 boy (yard), at 12s 12 
 
 1 boy (general), at 10s 10 
 
 3 girls for coring, at 8s 1 4 
 
 11 1 
 
 34 1 6 
 
 The price of foreign carbons is difficult to arrive at. Large 
 orders naturally command large discounts, hence the published 
 pricey are no guide to the consumer. In some cases I have 
 known 50 per cent, discount. Even more, I have no doubt, 
 has been allowed. The reader must therefore find out for him- 
 self what the best foreign carbons can be supplied at. 
 
 Here I take the liberty to fix the wholesale price, net, at 
 .5. 10s. per thousand cored 10 mm. carbons, as the market 
 price, below which I doubt much if they will be ever sold, the 
 competition being very small, and the demand daily increasing. 
 
( 40 ) 
 
 I have not taken notice in these calculations of interest on 
 capital or deterioration of plant. 
 
 ESTIMATED PROFIT. 
 
 *. .1. 
 Cost of 20,000 foreign cored carbons, 10 mm. diameter, 
 
 10 inches long, at 5. 10s. per thousand 110 
 
 Cost of material for 20,000 home-made carbons 23 6 
 ., labour 11 1 
 
 34 1 6 
 
 Profit on home-made carbons 75 18 6 
 
 Thus showing a gross weekly profit of no less than 75 18s. 6d. 
 on 20,000 carbons, the profit increasing largely with the 
 diameter of the carbon. 
 
 PRINTED BY GKOKGK TUCKER, AT 1, 2, & 3, SALISBURY COURT, FLEET STREKT, E.C. 
 
UNIVERSITY OF CALIFORNIA LIBRARY