• iifc'.-'V 3; 126 005 THE ART OF PAPER-MAKING: % §mk to \\}t Cjjeorg anir |ractke OF THE MANUFACTUKE OF PAPER. BEING A COMPILATION FROM THE BEST KNOWN FRENCH, GERMAN, AND AMERICAN WRITERS. EDITOR OF ''THE PAPER MILLS DIRECTORY." SOLD BY KENT & CO., PATERNOSTER ROW. 1874. Price 25*. UNWIN BEOTHHES, PBINIEBS. PKEFACE. It is very remarkable that since the Art of Paper- making first struck root in this country, now more than three hundred years since, no work of any kind has appeared on the subject, of any authority or otherwise, of a technical character relating to the manufacturing process strictly ; and adapted to the wants of the Paper- maker. The French Paper-making trade and the German have — although only within the last ten years — produced some writers on the subject, whose writings and treatises have obtained authority to some extent among them ; and quite recently a German writer in America has written a treatise of considerable practical merit. And the simple matter of making selections from these for the benefit of the English makers, is all we lay claim to in these pages. In the early part of the work the former have been drawn upon, and for the later portion the latter. Hence it is that some parts of this work are, we are aware, either crude, obsolete, or useless, in comparison with our most modern modes of working. But it will at least serve the purpose in laying a foundation for future efforts, of showing what our foreign neighbours, whom we are so frequently threatened are to be our future formidable competitors, have adopted, and do still use, as text-books of their modes of operation. London, 1874. CONTENTS. CHAPTER I. A Sketch of the History of Paper and Paper-making Invention 1 CHAPTER n. Raw Materials Employed in Paper-making 5 CHAPTER m. Manufacturing Process Machine-made Paper 11 CHAPTER IV. Manufacturing Process Machine-made Paper 66 CHAPTER V. Manufacturing Process Hand-made Paper 72 CHAPTER VI. Chemical Analysis of Paper-making Materials 94 CHAPTER VII. Washing Water, Water Power, Steam Boilers, etc... 122 INDEX. NO. Air Roll 106 Alkali Test 35 Alums 38 Ancient Period 1 Anticlilorine (p^g^) 31 37 Apron 74 . „ H4 Bar Screens 76 Beating 13 „ . 17 Bleaching 15 Boiler Tests 62 „ Explosions 64 Boilers 57 „ 62 Boiling Rags 12 Brass Rolls 99 Calenders 113 Chlorine 15 Chlorometric tests 36 Chemical Impurities 47 Chilled Rolls 114 Chimney Draught 60 Clutch 107 Colouring 19 20 Pulp (page) 47 „ Matters 41 Combustion and Fuel 59 Condensation 66 Construction of Boilers 62 Couch Rolls 86 Cutters 118 Dandy Roll 89 Deckles 94 Doctors 100 Draining 14 Draught Chimney 60 Drying 31 Ill Drying Cylinders 110 Dusting 10 NO. Dyes 19 Engine Power 68 ,, 18 Engines, Steam 65 „ 67 Esparto ... 5 „ Paper 23 Examination of Papers • 43 Expansion, Steam 65 Explosions, Boiler 64 Fan Pump 73 Feed Water 63 Felts 101 Felt Rolls 102 „ Washer '. 105 ,, Management 108 Finishing 22 32 Firing 61 Foreign Rags (P^g^) 13 Fourdrinier Machine 71 Fuel 42 ,, Combustion 59 ,, for Drying 112 Gates 95 Gelatine 29 Glazing 22 Grate Surface 61 Half Stuff 13 Hand-made Papers 26 Heating Surface 58 Housing 98 Ibotson's Strainer 78 Kaolin 39 Laboratory 44 Limes 36 Machine 21 Wire 85 Manganese 36 Middle Ages Period 2 Mill Dams 53 Mixing Box 73 Modern Period 3 8 INDEX. NO. Motive Power 120 Overshot Wheels 56 Paper, Examination of 43 Paper in liolls 119 ,, Carrying Rolls 102 ,, Machine 21 „ 71 ,, Shavings 8 Plate Screens 77 Power 66 „ Water 52 ,, Engine 68 „ Motive 120 ,, Loss of 69 Press Felt 103 „ Rolls 98 Pulp 16 „ Diluting 72 ,, Colouring 20 „ M (page) 47 ,, Dresser 74 ,, Regulator 72 Pumps 51 Rag Papers (page) 1 1 „ Boiling 12 ,, Dusting 10 ,, Washing 11 Rags 4 ,, Foreign (page) 13 Reels 116 Refining 17 Reversed Screens 81 Revolving ,, 80 Rolls, Air 106 ,, Brass 99 „ Chilled 114 „ Felt 102 ,, Paper Carrying 102 ,, Rubber Cased 99 ,, Spread 104 „ Stretch 104 Rubber Cased Rolls 99 Sand Tables 74 Save-all 90 Screen Vat 83 Screens, see 13ar Screens. ,, „ Plate do. ,, „ Revolving do. ,, „ Reversed do. Shaking motion 93 NO. Size and Speed 121 Sizing 18 „ 27 „ Room 28 Sorting and Cutting 9 Spread Roll 104 Starch 40 Steam Boilers 57 ,, Engines 65 ,, , 70 ,, Utilisation 67 Steaming Paper 115 Slitters 117 Strainers 75 ,. 78 „ : 82 ,, Suction 79 Straw 6 Straw Papers 24 Stretch Rolls 91 104 Stuflf 16 „ Catchers 92 Suction Boxes 88 Tinting 19 Trimming and Cutting 117 Tube Rolls 87 Turbines 55 „ 66 Ultramarine (page) 45 Utilisation of Steam 67 Water Distribution 49 W^ater-marking 33 W^aters 34 Washing Rags 11 „ Water 45 50 W^aste Paper 8 Water Pipes 90 ,, Power 52 ,, Quantity 60 ,, Wheels 54 Wood 7 „ Paper 25 Webs and Paper 119 Wet Felt 108 Wire Cloth 85 96 Wire Guides 97 ^o.J. Price THE AET PAPER-MAKING: % iufee k tj)e Cljeonr an!b ^rattia MANUFACTURE OF PAPER. COMPILED BY THE EDITOR OF THE ''PAPER MILLS DIRECTORY." SOLD BY UNWIN BROTHERS, 24, BUCKLERSBURT. B. DONKIN & CO.'S SPHERICAL ROTARY RAG BOILER. To CONTAIN FROM 20 TO 25 cwT. Or Rags. This Boiler being spherical, is twice as strong as a Cylindrical one of the same diameter and thickness. The Plates used are ' notwithstanding of the usual substance, thus rendering it perfectly safe, durable, and suitable for high pressure steam. The Spherical shape has also another important advantage, viz. allowing all the rags to fall out by themselves, when the boiler is revolving with its cover off. Inside the Boiler are Strainers to take oflP the dirt, and Lifters to agitate the rags during the process of either boiling or washing. To avoid Cement or run lead Joints, the Gudgeons and the Boiler are turned true in a lathe to fit each other, the joints being simply made with red lead. B. DONKIN & CO., ENGINEERS, MILLWRIOnTS, MACniNISTS, AND IRON FOUNDER>, Blue Anchor Eoad, Bermondsey, London, 8.E. Manufacturers of Steam Engines, Turbines, Water Wheels, Flour Mills, (fcc, Paper i\Iaking Machines, and everything connected with I'aper I^Iilis, Rag Engines, Rag Cutters, Cutting Machines, Glazing and Hydraulic Presses, Pumps, Hoists, &c., Steel Bars and Plates, Wires, Felts, and Deckle Straps. Machines for Printing in two Colours. Gas and Water Valves. 512G005 MANUFACTURING PROCESS. 65 are crushed, and appear as transparent yellow spots when the leaf is held up to the light. The work of a satining a^^paratus requires a number of hands ; three or four gangs besides one or two work- men to hand the plates. One will be sufficient, if the rollers work in both directions so as to return when reversed. A boy or girl may place and remove the sheets of paper, and a workman lays on the zinc plate. Some- times this work is confided to a woman, but a man is indispensable for the larger sizes. Calenders, or three cylindrical presses, are sometimes used, the middle one being of paper and the other two of metal. This apparatus is absolutely necessary to glaze paper in rolls. The paper is wound upon rollers placed at the head of the calender, and passes first between the uj)per and middle cylinder, and then between the middle and lower ones. This finishing in rolls, especially when the paper is very wide, presents some difficulties. It is necessaiy that there should be a uniform pressure upon every point of contact, otherwise the paper will be creased and a con- siderable loss occur. This accident is less to be apprehended when the paper is rolled in separate sheets. Three hands are necessary to attend to the work of a calender. The first engages the leaf between the two upper cylinders ; the second returns it between the middle and the lower ; and the third receives the sheet. The calendered paper preserves its whiteness, which is somewhat impaired by zinc. All things being equal, the plates are better for thin papers. F 66 MANUFACTUICE. The glazing done, there remains the operation of counting and putting up into reams. The sheets are counted by special hands, who acquire great skill. The quires, after being arranged in reams, with the backs placed alternately for folded papers, are put under the press at night, and the next day are ready to be packed. Bundles are composed of two reams. It is well to weigh before packing ; and to label each ream, indicating the quantity, quality, &c. A careful manufacturer should see that the packing is always done with care, and even with elegance, in the case of fine papers. Chapter IV. MANUFACTUKE. 23. Esparto Paper. — There are found in Esparto two or three qualities as imported for Paper-making, all of which should be sorted and cleaned by hand before they are suitable for putting into the engine. This may be done on the ordinary rag-cutting tables. Weeds, roots, or flowers should be carefally cut off" and thrown out. It should, like straw, be washed and boiled with caustic alkali, washed and bleached with a chlorine solution. Before subjecting this material to the boiling and bleaching processes, it is essential to crush it in the direction of its length, by means of fluted rollers ; as the MANUFACTURE. 67 fibres are more permeable to chemical agents when separated. It may be put into the boiler in the dry state and in its full length. Esparto, after being boiled with lye, retains enough tenacity to be separated into long shreds. It contains a red colouring matter which becomes soluble under the combined influence of chlorine and caustics. The waters used in washing assume a blood -red tinge. The waste of material may be thus estimated : — Yellow colouring matters .... 12 Red ,, ,, .... 6 Resin and gum 7 Salts constituting the ash .... 1-500 26-500 Paper fibres 73-500 100-000 The yellow colouring matter being soluble in alkalies, is eliminated after the first boiling. By mixing Esparto with rags of poorer quality, very good printing paper may be produced. Stationary boilers are considered to be best adapted for this material. In the preparation of the lye it is essential that the soda should be rendered caustic. The yield of stuff from Esparto is considered to be not less than 40 or more than 50 per cent.; the variation being caused by the varying purity and quality of the grass itself. F 2 ^H MANUFACTURE. 24. Straw Papers. — The first attempts to make paper from straw go back to the beginning of this century. The process consisted in subjecting the material to the action of a \\e made of a mixture of lime and soda, or potash, until the substance was softened enough to be crushed between the fingers. After washing followed by trituration, the stuff was converted into paper. The numerous patents taken out with the same object are based upon analogous principles. Packing paper of considerable strength has been made by mixing 50 per cent, of coarse rags with straw pulp, prepared in the following manner : — 220 lbs. of wheat straw, finely chopped, and 176 lbs. of quick-lime, were placed in a boiler with a sufficient amount of water to form a kind of pulp. The mixture was stirred, and poured into a second boiler every day for a fortnight. . ■ , . This material was then reduced to pulp, and mixed T/ith the rag pulp in the beating engine. The product obtained was half-sized, of a j'ellowish tinge and great strength. Straw contains a yellow colouring matter, which is more or less communicated to the paper, unless subjected to a succession of boilings and bleachings, with gaseous or liquid chlorine. In this case the waste of material is very much increased. The majority of manufacturers who employ straw are satisfied with subjecting it to maceration with lime, and succeed in making common wrapping paper. If, however, it is wished to make common printing paper, such as newspaper, &c., of straw pulp, it is indispensable to bleach MANUFACTURE. 69 with repeated chlorine and acid baths. The mixture is then made in the proportion of — 25 to 40 per cent, of straw pulp. 75 ,, 60 ,, ,, rag pulp. The manufacture is thus only profitable in localities where chemicals are at a low price. Very fair papers have been made with 80 to 90 per cent, of straw, but it remains to be seen whether the profits will render the process practicable. The nature of the strawy and the composition of the soils in which it is grown, are not unimportant mat- ters to the paper-maker. The hardness of straws pro- ceeds from the quantity of silica interspersed throughout their substance, forming an obstacle to their conyersion into paper, by binding together the fibrous parts of the stalk. The straw to be chosen is that of wheat, as being the most tender. This fact is in accordance with chemical analysis, which gives the following proportions of silica contained in the three principal cereals of our soil : — 1st. Wheat straw, 4-3 per cent, of silica. 2nd. Rye ,, 6-3 ,, 3rd. Barley ,, C9 ,, ,, The knots of graminious plants, in general, are in- jected with a much greater quantity of silica than the intermediate parts, and should therefore be carefully removed when the straw is to be converted into white paper pulp. Among other straws which have been tried, we may mention maize, which gives a naturally sized paper of 70 MANUFACTURE. great strength, anti which at one time engaged some attention. Maize stalks unbleached, only boiled with lye, and added to rag pulj), are suitable for making packing-paper. The product possesses a certain tenacity not offered by that of other straws. This substance may then answer very well in countries where it grows in sufficient quan- tities to allow of its employment. 25. Wood Paper. — For twenty years wood has been the object of numerous, and at first fruitless, experiments. At the present day, owing to the persevering efforts of a few inventors, processes are known which admit of the use of this material, when mixed with rags, for making the commoner kinds of printing paper. In France and Belgium, white and resinous woods, such as pine, fir, &c., are emploj-ed. It has been shov;n by experiments that wood is not made up of cells solidified by encrusting materials, as was long supposed, but of two superimposed layers ; one exterior, short-grained, and brittle, the other internal, supple, and fibrous. In order to make wood paper, all that is necessary is to dissolve the former and employ the remaining substance. This first layer is acted upon by alkalies, hydrochloric acid, and the h3'pochlorites. The machiucs employed to separate the fibres of wood are of difierent kinds. That invented by M. Voelter consists in a grindstone, which consumes several logs of wood placed against its circumference and pressed in such a manner as to keep them always in contact with it during the entire duration of the work. MANUFACTURE. 71 Another machine crushes the chips between cast iron rollers, and the material is triturated by means of a cylinder engine. The pulp is strained through a sieve, placed in the cap of the engine, and may then be used immediately for the manufacture of common papers. Wood alone gives too brittle a paper, and the short thick fibres do not yield that suppleness which is characteristic of rag paper. The motive force necessary to separate the fibres of wood is very great. It is calculated that a 25 horse power is required for 24 hours, in order to produce 6 cwt. of wood pulp. We give as an example of the mixture the proportions admitted by M. Voelter, of Heidenheim. Per cent. Per cent. Writing paper . . . . 33 of poplar wood, 67 of rags. Fine printing paper . . 20 ,, ,, 80 Common printing paper . 50 of pine wood 50 Brown wrajiping ,, .40 ,, ,, 60 Grey ,, ,, . 50 „ ,, 50 Blue paper 33 ,, ,, 67 In Belgium, paper for newspapers is made of — 60 parts birch wood, 20 ,, kaolin, 20 ,, coarse grey rags. The wood pulp is added a quarter of an hour before the end of the beating process. The wood being neither boiled nor bleached, the proportion of bleaching should be diminished, as a part of the colouring matter is not destroyed by acids. The use of chemical agents to bleach wood pulp requires much skill, otherwise there is great risk of 72 MANUFACTURE. having a pulp without consistency, giving rise to con- siilcrable loss of material. The introduction of unboiled and unbleached pulp is considered so great an economy, that several manufac- turers of Germany, Belgium, and France have decided to make additions to their paper-mills, in order specially to prepare wood pulp. Chapter V. MANUFACTURE. 26. Hand Made Paper. — Paper-making from the vat, or by hand, requires the labour of three workmen for each vat : 1st. The vatman ; 2nd. The coucher ; 3rd. The layman. The first, by dipping the mould into the vat, makes the sheet ; the second stretches or couches it upon the felt ; the third, when the post has been pressed, succes- sively detaches each damp sheet intei'posed between the felts. The vrorkman, holding the mould in both hands by the two short sides, dips it into the pulp at an inclination of about 60 or 70 degrees, and raises it horizontally, after having taken up enough of the pulp to obtain the re- quired thickness of the paper he intends to make. By means of a double oscillating motion, called balanc- ing, he distributes the pulp, as uniformly as possible, over the entire surface of the mould. Gradually the water drains through the wires or the meshes of the wire gauze ; the pulp solidifies, and assumes a peculiar shiny MANUFACTURE. 73 look, which indicates to the workman that the sheet is formed. The workman then lays the mould upon the plank ; takes off the deckle, which he places at his right upon the hridge, and then hands the mould to the coucher. This workman raises it, and places it upon a small, curved, wooden stay in such a position that its inclina- tion will favour the drainage of the water. The vatman then applies his deckle to the second mould, and makes another sheet. In the meantime, the coucher seizes the mould with his left hand by the short side nearest him, and grasping it again by the upper long side — which, by turning the mould, is on his right — he applies it to the felt, making it describe the quarter of a circle. When this is completed, the coucher rises, and slides the empty mould along the bridge. It is after this operation that the vatman pushes for- ward the second mould deprived of its deckle, which the coucher places upon the curved stay to drain. This workman then takes a felt from the bench on his right, and applies it accurately to the leaf which he has just couched. The operations just described continue until the post is completed, that is, until the coucher has exhausted all the felts of which it is composed. The number of the felts in a post is variable ; in the older mills it was composed of eight quires, that is to say, 209 felts containing 26 x 8 = 208 sheets of paper. . The two workmen should regulate their work so that the manipulations of each may coincide. 74 MANUFACTURE. The rapidity of operation, however, depends upon the nature of the pulp and of the wire cloth of the mould. There are two methods of couching : the French and the Swiss. In the first the coucher applies the mould vertically upon the felt, and then reverses it. In the second, the mould has been rotated 180 degrees, so that it occupies an almost reverse horizontal position ; the coucher first leans it on the long side, and then making it describe an arc of 90 degrees, he couches every part of the leaf upon the felt successively. This last method is more rapid than the first, but re- quires more skill on the part of the workman to couch the leaves evenly one upon the other. It should, however, be adopted exclusively in working with pulp not readily drained, as the weight of the pulp causes it to gravitate towards the lower edge, and more or less destroy the homogeneous character given to the sheet by the vatman. It is important that the coucher should not allow any drops of water to fall upon the pages in raising the mould, which would produce so many spots impossible afterwards to remove. On beginning a post, it is prudent to place several felts one upon the other, so as to make a softer bed for couching the first sheets. Without this precaution, shreds of pulp remain attached to the mould, and pro- duce breaks in the paper. Couching by the long side of the mould ofiers the double advantage of accelerating the work and faciHtating the extraction of the pulp from between the laid wires, as each interval will develop the outline of the surface generated by the motion of the coucher. MANUFACTURE. 75 Another cause of waste proceeding from couching are bubbles, that is to say, blemishes upon the sheet, produced by the interposition of a volume of air between the felt and the sheet, which, in escaping when the next sheet is couched, leaves a depression in the sheet of paper bearing more or less resemblance to a bottle. New felts are soft, and taking less hold of the surface of the sheet are liable to this kind of accident, which also occurs when the felt is badly stretched. When the post is completed, the three workmen unite to carry and place it under the press ; the central portion being generally more elevated than the edges, the differ- ence in thickness is made up by bits of wood in the shape of an elongated triangular prism. The workman then inserts a strong piece of oak, from 4 to 7 inches thick, upon which the force of the press is exercised. It is preferable to employ, instead of the hand-press of the older mills, an hydraulic press, the force-pump of which should be of sufficient diameter to accelerate the operation of pressing. In this case it would be well to place a strong oak plank under the platten of the press, as the elasticity of the wood renders the pressure more uniform. The requisite degree of pressure attained, the work- man brushes off the drops of water that ooze out along the felts, and which would be absorbed by them, and thence by the sheets of paper, when the pressure is taken off. When the post is carried away, the work of the third hand begins^ while the other workmen are filling the vat 7G MANUFACTURE. with a quantity of pulp proportioned to tlie weight of the previous post. The hiymau successively detaches the moist sheets, placing them one upon the other, and throws back the felts upon the bench to the right of the coucher. The operation of raising the sheets is accomplished in two ways : either upon an inclined or horizontal plane. In the first method, generally employed in France, benches are used with an inclination of about 50 to 60 degrees. The layman is generally assisted by an apprentice whose duty it is to remove the felt, while the w^orkman lays his sheet upon the preceding one. This assistant may be dispensed with when the work is done upon an inclined plane, but is necessary when the direction is horizontal. He holds a sort of flat rule, over which the layman throws the sheet as soon as de- tached from the felt, still holding it with both hands by the two corners of the short side nearest him. The sheet thus supported by the four corners is placed without difliculty upon the preceding one. Without the aid of the assistant, the sheet, still very moist, vrould adhere too readily, and it would be difficult to avoid the presence of wrinkles. When the sheet is adjusted, the assistant withdraws the rule, removes the felt, and the operation is repeated. If the post has been but slightly pressed, it is difficult to raise the sheets, the pulp has little consistency, and breaks, the comers tear off", and the result is the loss of a great number of sheets. If, on the other hand, the pres- sure has been too great, the sheets adhere too firmly to the felts and carry away some fibres of wool. It is the MANUFACTURE. 77 duty of the layman, therefore, to see that the pressure is sufficient and neither too light or too heavy. When all the sheets are detached, -the layman takes several felts to cover this first post of paper, and pats it forcibly with both hands to make it into a sort of com- pact cake, less liable to be torn in handling. After several days' work, the felts become greasy, and on being detached from the sheets, give rise to a peculiar creak. It is then necessary to wash them with brown soap in order to recover their softness and absorbing qualities, so as to facilitate the operations of couching and separating the sheets. There also accumulates around the sides of the vat after a certain length of time, a peculiar kind of grease which, if mixed with the pulp, would produce spots upon the paper. It is advisable to clean the vats at least once a month, or even every fortnight. This cleaning should take place on Saturday, at the close of the day's work. The pulp is used up as far as possible by making a last post of paper somewhat thinner than the others. The pulp is then removed with great care, without touching the sub- stances at the bottom of the vat, which are most im- pregnated with this grease. By means of a brush the internal surface of the vat is perfectly cleansed, and, after repeated rinsing, the pulp which had been removed is replaced. The greasy pulp may serve for making common kinds of paper. In some mills the posts of paper, after having been subjected to a light pressure, are carried to the drying room. But if it is desired to obtain a superior 78 MANUFACTUKE. quality of paper, it is essential to lift and change the sheets. Lifting consists in detaching one, two, or several sheets together, and forming new posts before hanging them up to dry. In the change, the sheets are raised one by one, and replaced one above the other in a different order from that which they at first occupied. The workman charged with this work takes alternately a sheet from two posts placed beside him and makes of them a new one, in such a way that the middle sheets are at the outside, and vice versa. The grain of the paper is softened, and it acquires a greater degree of firmness, and assumes that velvety feel which characterises the Dutch papers where this method is universally practised. After four or five such operations, alternating with pressure applied stronger as the paper gains consistency, the sheets are carried to the drying-room. These posts of paper being thicker in the middle than at the edges, their irregularity may be compensated by bands of felt inserted when subjecting them to the action of the press. The paper is hung in spurs of four or five sheets, which are placed upon the ropes or tribbles by means of poles. Paper, when dried in warm weather, shrinks, and ac- quires great hardness. The material when too soft yields to its own weight, expands out, and forms a wrinkle on the back of the sheet — a very objectionable defect. This imperfection is less evident when the paper is delivered folded. MANUFACTURE. 79 In France, the drying-rooms are placed upon the top of the house. In Holland they are less elevated, and cooler stories are preferred, in order to prolong the drying. Some manufacturers hang the paper in spurs of seven or eight sheets, in order to economise space in the drying- room. This method is bad in every respect, as the sheets on the outside dry more quickly than the under ones. The air does not circulate, and the drying is not uniform. The nature of the tribbles very much affects the clean- liness of the paper. Hempen tribbles ought not to be used, as they stain the back of the sheet yellow. Those of aloes and of the fibres of several kinds of cane are advantageously employed : those of horse-hair being the cleanest and best. With thick tribbles, the air circulates better, and the back of the sheet, being more rounded, is not so apt to crease. As soon as the paper is dry, which may be known by the peculiar rustling produced among the sheets when they are moved by the hand, the operation of sizing follows. The duration of the drying depends upon the season, the situation of the drying-room, the diameter of the tribbles, the number of leaves in a spur, the operation of exchange, the thickness of the paper, and the pulp used in making it. To be kept continuously at work, a well -organised mill ought to be furnished with drying-rooms warmed by hot-air stoves, so as to be able to dry and size the paper at all seasons. 80 .AIANUFACTUKE. 27, Sizing. — To render liaud-made paper impermeable, or fitted to receive writing, it is dipped in a warm solution of size or gelatine raised to a temperature of from 68*^ to 95° Fahr. This constitutes animal sizing. This size is prepared by dissolving by long boiling the refuse of hides, cartilages, ears, hoofs, tendons, &c., bought from tanners or dealers in them. This refuse animal matter is dipped in milk of lime, which preserves it from putrefaction, and after drying in the open air is sold under the name of scrolls, pates, hide pieces, dc. Notwithstanding the use of lime, if the desiccation has not been well attended to, this size will undergo a certain kind of fermentation which deprives it of a part of its adhesive qualities. This defect may be recognised by the ammoniacal odour generated while the material is boiling. Size obtained from young animals, such as lambs, calves, &c., is easily prepared, and of a white colour ; whereas that produced by the hides of oxen, cows, &c., is darker, but stronger and tougher, and gives more firmness and resonance to the paper. To obtain the solution of the gelatine contained in the scrolls, these last are placed in a copper boiler, sur- rounded by brickwork, and usually heated by a naked fire. It is well first to wash the scrolls in order to remove the various impurities which they contain. The duration of the boiling varies according to the nature of the materials. For those obtained from oxen, from twelve to sixteen hours at least are required. The scrolls should not touch the bottom of the vessel. MANUFACTUEE. 81 as the parts in contact with the metal become too greatly heated, and produce a brown discoloration of the solu- tion. To remedy this objection, in some mills a bed of straw is laid at the bottom of the copper. But though the nature of the evil is changed, the effect remains much the same, as the straw gives out its own yellow colouring matter. During the first hours of boiling, fatty matters which appear upon the surface must be carefully skimmed off. To facilitate this operation, they are sprinkled with lime, so as to form a calcareous soap. When it is thought that the solution is sufficiently concentrated, the liquid is run off into a lower vessel, and the copper again filled with water. By again boiling a second weaker solution is obtained, which is generally mixed with the first. Another method is to let into the boiler a quantity of water proportioned to the volume of the solution drawn off by the stopcock, so that the level of the liquid will remain the same. The boiling is continued, and the fire is removed when it is judged by the amount drawn off that the size-pieces have been deprived of all their soluble principle. The solution thus obtained is never clear ; it contains suspended in it different substances of which it is to be freed. This constitutes the operation of clarifying the size. A small quantity of powdered lime should be added, and, after being stirred, the liquid is allowed to rest. If the impurities have not completely settled to the bottom, a half a hundredth part of sulphuric acid is added. An insoluble sulphate of lime is formed at once, and its precipitation clarifies the solution. 82 MANUFACTURE. The liquid is clecanted and filtered through several folds of felt, which retain any remaining impurities. This solution contains a mixture of gelatine and chon- driue, two substances differing very greatly in their pro- perties. The latter being an obstacle in sizing, it is indispensable to eliminate it. Fortunately, this material may be completely precipitated by a concentrated solution of alum. Sulpliate of alumina and sulphate of iron pro- duce the same eflect. The size, when again filtered, is ready for use ; if it is too concentrated it may be diluted with warm water. The temperature and the strength of the size both vary according to the nature of the paper and the condition of the atmosphere. Papers manufactured from hard rags require a thin size with a high temperature. The reverse is the case with those made from soft rags. All things equal, it is important that the strength should be greater in summer than in winter. The papers are dipped in a copper of dimensions vary- ing with the size of the paper. To keep up an even temperature, the copper should be heated by a small furnace : preferably by the heat of a water-bath or steam, by means of a worm or a copper with double sides. The workman employed in the operation of sizing, takes a handful containing from 100 to 150 sheets at a time and dips them into the size, separating them like a fan, and manipulates them so that every part of the sheets shall be uniformly saturated. This requires a certain amount of skill. He then carries the package to a press near the sizing tub. MANUFACTURE. 83 A series of packages are piled one above the other, and when they have attained a height of about 2 feet, he places a plank upon, and presses them in order to drive out the excess of size, which flows back into the tub placed below. The degree of pressure should vary according to the nature of the paper ; so that, in all cases, the sheets may be readily removed singly and without tearing. The moist sized paper must then be carried to the dr3'ing-room, and hung upon the tribbles in spurs of two, three, or four sheets. Careful makers exchange the paper when sized before carrying it to the drying-rooms. The drying is benefited by this, and the sheets acquke more strength and firmness. M. Payen, in his Chimie Industrielle, gives the follow- ing theory of gelatine sizing : — *' In order that paper may be well sized, it must be properly dried. This should be done gradually and slowly, without, however, being carried so far as to allow the spontaneous decomposition of the gelatine to take place. " This accident sometimes occurs in summer, especially in damp and stormy weather ; the size then Hquefies, loses its adhesive qualities, and the operation miscarries. If the drying is too rapid, the size remains disseminated through the entire substance of the paper ; but if the process is carried on with proper slowness, the moisture contained in the paper gradually finds its way to the sur- face, and carries with it the gelatine, which forms a superficial and impermeable coating. " The drying is moderated by means of ordinary win- G 2 84 MANUFACTURE. dow blinds, the openings of which may be regulated at Avill. It will be understood that by drying slowly the jjelatinons solution is allowed to come to the surface as fast as the evaporation of the water takes place, and that therefore the greater part of the gelatine is concentrated at the surface, and renders the paper impermeable, whereas if dried too quickly this material would remain disseminated throughout its entire substance. *' It is easily ascertained whether the paper is sized on the surface only, by scratching it, and then drawing an ink-mark over the denuded part. The paper sized with gelatine will absorb the liquid, whereas the ink will re- main unaffected by machine-made paper sized with resin through the entire thickness." This explanation is entirely confirmed by the practice of damping the paper when too quickly dried, and ap- pearing to be imperfectly sized. Such paper placed between wet sheets becomes moist, and by again passing thi'ough the drying-room, the size disseminated through- out the substance of the sheets comes out on the surface as the water evaporates, and the paper becomes imper- meable without any additional gelatine. This process gives the paper firmness, and some manu- facturers were formerly in the habit of always employing it for certain kinds of paper, finding in the excellence of the result a sufficient compensation for the cost of the double manipulation. The character of the water has a marked effect upon sizing. It is therefore essential that the water which is to be used in the different manipulations of the pulp should be clarified with much care. If the water is very highly MANUFACTURE. 85 charged with calcareous matters, it is well to precipitate these substances with carbonate of soda, or even alum. The proportion of these reagents ought, of course, to vary according to the composition of the water. Certain well waters containing a large quantity of sul- phate of lime, absorb 28 grs. of pure dry carbonate of soda, before being enabled to dissolve soap. River waters require from 150 to 400 grs. of alum to 22 gallons of liquid to effect this precipitate. This fact explains the superiority of the papers produced by some mills situated upon streams with granite beds yielding no calcareous elements. 28. The Sizing-Room. — In a paper-mill, the sizing- room is an apartment set aside for the operation of sizing alone. It contains : — A furnace furnished with a copper boiler about five feet in diameter by three in depth. It is in this boiler that the size is made, or the gelatine extracted from the sub- stances which contain it. Also a copper vessel called the sizing tub, in which the operation of sizing is effected. This vessel may be about 3 feet in diameter, and about 2 feet deep. It is mounted upon an iron stand, under which is placed a small stove to keep the size at a proper temperature during the pro- cess of sizing. The tub is placed near the press, so that the size in running off from the saturated paper may return to it, and not be wasted in the passage. 29. Preparation of Gelatine. — G-elatine enters into the composition of the soft and solid parts of animals. In 80 MANUFACTURE. this condition it is found in the muscular fihre, skin, ligaments, cartilages, and tendons; the membranes contain a large proportion of gelatine, and it constitutes about half the weight of bones. Gelatine is heavier than water, without taste or smell, colourless, and without reaction upon litmus paper ; it is, therefore, neither acid nor alkaline. Decomposed by fire, gelatine again offers us the same phenomena as these substances, but is easily distinguished from them by certain of its properties. This material is very soluble in boiling water, but very sparingly so in cold. When two parts and a half are dissolved in a hundred parts of water, the liquid con- geals on cooling. The jelly sours in a few daj^s, especially in summer ; it then liquefies, and before long takes on all the phenomena of putrid fermentation. Gelatine or glue is generally prepared for the uses of commerce from parings of hides, parchment and gloves, and from the hoofs and ears of oxen, horses, sheep and calves. All size is more or less transparent. Some sorts are of a blackish and some of a reddish-brown, while others are of a slightly yellow-white. The most transparent and least coloured are the purest, and these kinds are cmj^loyed by the paper-maker. There are several methods of preparing size in our paper-mills. In some of them it is considered sufficient to put the raw materials into a boiler, containing a suit- able proportion of boiling water, and to boil them until all the gelatine is extracted. The boiling is kept up from twelve to fifteen hours, for three hundredweight of hide cuttings. MANUFACTURK. 87 It is evident that the strength of the size is propor- tional to the quantity of raw material employed. In other mills a smaller amount of water is, at first, added, and when the decoction is concentrated enough, the liquid is drawn off and replaced by another smaller amount of water. In this manner, three or four solutions are obtained, which are either mixed together or preserved separately. This method is preferable ; because, the pieces of hide being- unequally soluble, the first portion of gelatine extracted deteriorates by remaining in the boiler until the rest has been melted. Besides this, the quality of the materials may be better ascertained from the amount of size they have furnished. Whatever method may be adopted, it is important that the size pieces should not touch the bottom of the boiler, as in that case they would be burnt, and give the size a darker colour. In the Batch paper-mills the raw materials are held in a wicker basket, which is let down into the boiler and withdrawn by means of a pulley, when the size is to be removed. This apparatus is very simple, and allows of ascertaining whether there still remains any gelatine un- dissolved. With whatever care size is extracted, the decoction is never quite clear ; it holds in suspension a great quan- tity of undissolved gelatinous matter, which would not be precipitated, even after long rest, with the fluidity of the size constantly maintained ; but if by any means an abun- dant precipitate can be obtained, the particles of suspended matter will be carried with it, and the size will then become perfectly clear. 88 MANUFACTURE. 30. Operation of Sizing. — A tub in which the operation of sizing takes place is situated near the press, so that the size which runs off from the handful of sheets just dipped by the workman in the tub, and laid on the bed of the press, may not be lost, but fall back into the same vessel. The upper surface of the press-stand, which is con- structed with a view to great firmness, is raised some thirty inches above the floor. This stand is surmounted by a frame two inches broad all round, and about an inch in thickness. The frame is strongly fixed upon the stand by screw bolts, the heads of which are buried in the frame above, and fastened by a nut under the table. In place of these bolts, strong wood screws may be used, having their conical heads sunk in the frame, which should fit the upper surface of the table so accurately that water may not be able to pass between them. It will be seen that by this arrangement the press-stand presents a regular hollow. In the left-hand corner of the press- stand, and below the lower surface of the frame, a hole is pierced, slanting upwards, and opening in front of the inner border of the frame, into which is fitted a copper pipe, projecting far enough beyond to pour the excess of fluid continually into the tub. This pipe should be well cemented to the edge of the hole, so that no liquid may be able to run off by any other outlet. The workman generally stands at the tub, with the press at his right hand, and on his left a bench support- ing the sheets, just as they have been carried from the drying-room after the stiffness has been taken out. The tub is then filled with tepid size, and the heat applied MANUFACTURE. 89 under it. A quantity of alum is thrown in, varying according to time and circumstances. The alum is dis- solved in hot water, and well stirred to mix it with the gelatine. The alum prevents the size from decomposing, and preserves it for a considerable time. Some manu- facturers add white vitriol (sulphate of zinc). The workman places by his side three or four wooden pallets, by means of wliich he manages the sheets of paper in the manipulations of sizing. These pallets are pieces of wood, flat upon one face, rounded on the other, and shghtly conical at the two extremities. The shape of this implement may be imagined by considering it as formed of a cylinder, three or four inches in diameter by twenty-two in length, terminated at each end by a blunt elongated cone, then cut in two, and by this division forming two pallets. Standing in front of the tub, the workman takes a handful of paper in his left hand, tind supports it from beneath by one of these pallets ; he seizes the sheets on the opposite side with his right hand, and takes care to separate them w^ith the fingers of that hand, in order that the size may the more readily penetrate between them ; he submerges that entire end of the paper by dipping his hand into the size ; he then lifts the bunch with his left hand, and holds it above the tub to drip, which brings the sheets tosjether. The end held in the ricrht hand is allowed to rest upon a second pallet, wdiich generally floats upon the size, while with a third he seizes it above so as to catch it between the two, and lets go the other end, which he held in his left hand. He again separates the leaves with his fingers as before, and plunges his left hand into the size with this end. The bunch of sheets is 00 .MANUFACTURE. held suspended for some time to allow the size to run off and the pages to adhere ; and the workman, after having raised the lower end with his left hand, carries the lot, with both hands to the hollow surface of the press- stand. This operation is continued until ten or twelve liand- fuls have been sized. A turn is then given to the press, which causes the size to penetrate the substance of the paper, and the excess of the liquid flows back into the tub by the pipe already mentioned. This operation requires considerable care ; for if the paper is pressed too hard, an excessive amount of gelatine is thereby ex- pelled, and the interior of the handfuls of sheets v/ould remain unsized. From experiments made it has been estimated that a quantity of unsized paper weighing eighty -six pounds, absorbs in sizing six pounds of dry gelatine. 31. Drying after Sizing. — ^When the process of sizing is completed, the workman presses the entire mass of paper gently and slowly, not carrying the operation too far, so that the size may have time to become fixed in the sub- stance of the leaves. The paper is only left long enough under the press to ensure a uniform penetration of the size, without allowing it to dry, as in that case there would be danger of the leaves becoming so tightly glued together as to be no longer separable. When the moment of taking off the pressure has ar- rived, the paper, still wet, is delivered to women who separate it sheet by sheet, and thus hang it upon the tribblcs, in the drying-room, by means of T-shaped lifters, beginning from above. These lifters, with handles vary- MANUFACTURE. 91 ing in length with the height of the tribbles, are employed to avoid the necessity of mounting upon trestles. The entire length of one line is covered with sheets before beginning upon the next. The Dutch employ a peculiar mode for the preparation of paper after sizing. They introduce the operation known as the exchanoe, which is done as it leaves the sizino;- room, as follows : — " When the paper has remained long enough under the press, the workman carries it away, in portions of one or two handfuls, and distributes it along the table ; he then begins with the nearest lot and exchanges the leaves one by one, lifting them by the corner, so as to form a new pile, wiiich differs from the first only in that the surfaces, which before touched and had been pressed against each other, are made to correspond with the surfaces of other sheets. By thus mixing the sheets in a new dis- tribution each surface is detached from those of the con- tiguous leaves, to which it adhered, and applied to others, against which it is again pressed. It is of little conse- quence whether the paper is still warm or not, so long as it is wet. Care must be taken, however, to see that the leaves are not replaced under the press, after the exchange, until the paper is cool ; for if still warm, the size would be fluid and liable to be expelled from the leaves by the action of the press, or to exude unequally upon the sur- face, thus producing irregularities and destroying the advantage of the exchange. It will be seen that the process of sizing is one of the most difficult and uncertain operations of paper-making. We are frequently obliged to begin anew, either by the season, the unfavourable condition of the atmosphere, or 0'2 MANUFACTURE. the temperature of the size itself. If the size is too hot it injures the paper ; if too cold it will not penetrate it ; if too thick it attaches itself only to the surfaces and runs off when the paper is dry ; and lastly, if too thin the sizing will be insufficient. The atmosphere also acts a most important part in the process. 32. The Process of Finishing. — Hand- made paper, is the same as that described in the case of paper made by machinery. On leaving tlie drying-rooms, the paper is piled up and subjected to repeated pressure in order to remove in part the trace of the fold on the back. A series of packages is made up, between each of which packages is placed a wooden board of the same size as the paper, in order that the action of the press may be more uniform. Such papers as are not to be rolled require a more heavy pressure. It is also well not to pile them up when they are too dry, as a slight trace of moisture greatly facilitates the pressing. The nature of the wood from which the boards are made is not a matter of indifference : wood without knots is required ; and very dry walnut answers the purpose well. Sometimes these boards are covered over with glazed board, or the latter is used alone. Whatever may be the nature of the hard substances placed between the pressings (200 to 500 sheets), it is essential that the pressure .should be applied gradually in the beginnmg, otherwise the sheets crease, or the grain is crushed, and never gives that uniform surface which is characteristic of the products of the best mills. MANUFACTUllE. 93 Before being put up into reams and packed, the pressure of an hydraulic press wouki much increase the histre of the paper. The principal defects which are found in paper made by hand, are knots, water spots, bubbles, folds, torn edges, ragged edges, size stains, holes, and tears. When the paper is put in reams, it should be per- fectly dry, and be laid in a dry and well-ventilated place, otherwise, by the action of moisture, it would soon sour. 33. Water-marked Paper in general. — The paper of bank- notes, when held to the light, generally presents inscrip- tions or designs which, in the trade, are termed water- marks. There are several kinds of watermarks. The most simple are obtained by sewing to the laid or wove wires of the mould fine brass wire twisted according to the outlines of the design, or, still > etter, a thin copper leaf cut out with a puncher or graver. The increase of thickness in this place, and the raised wire preventing the pulp from draining, produces a reverse effect upon the paper, the design of which is seen to be lighter than the rest of the sheet. First, another method is the use of a dark watermark with light letters obtained by depressing the wire cloth itself, in such a manner as to produce a sort of rectangular cavity in which a greater quantity of pulp may be deposited, and then sewing the letters to the bottom of this depression. Next, shaded watermarks, the richest of all, allowing of obtaining by a sort of moulding process, every variety of relief, of whatever nature. 94 CHEMICAL ANALYSIS OF MATERIALS. The principle of depressing the face of wire cloth or the preparation of watermarks, is that every part of the surface should be readily stripped from the paper in order to faeiUtate tlie operation of couching. Moulding the ware cloth, for shaded watermarks, re- quires the skill both of a moulder and an engraver. The manufacture of watermarked paper requires espe- cial care. The workman should manage to produce a leaf of exactly uniform thickness throughout its wdiole extent. This is necessary to secure a clear impression. The coucher should see that this impression is sufficiently distinct, and notify the foreman at once, when a letter or any other part has become detached or unsewed. Chapter VI. CHEMICAL ANALYSIS OF MATERIALS EMPLOYED IN PAPEE-MAKING. The nature of cotton, linen, and hempen rags may be determined by chemical tests, and by microscopic examination "when the material is comparatively new. The fibres of cotton and hemp seen through the microscope, present the appearance of rigid cylindrical tubes, with intercepted intervals like straws, reeds, &c. They are elongated cells, glued together by a material of denser texture, which is susceptible to the action of acids and alkalies. The diameter of hemp fibres may be estimated to be ToV'txi to ToVoo of an inch. Those of linen are still finer, and have a silkier appearance. CHEMICAL ANALYSIS OF MATERL\LS. 95 The fibrillse of cotton, on the contrary, are transparent tubes, flattened in the middle throughout their entire length, presenting the appearance of two parallel rolls, united by a very thin partition. They cannot be better compared than to a T rail, twisted several times upon itself. The diameters of the fibres vary from yo^oo to tuotj of an inch. When these fibres have been brought into the con- dition of textures, more or less worn, these microscopic characteristics are insufficient; the straight and rigid conformation of the hemp and flax fibres having disap- peared. It is then necessary to resort to chemical reagents. Linen and cotton fibres may be distinguished in several w^ays. 1st. A boiling solution of potassa colours the fibres of linen a deep yellow ; whereas its action upon cotton is very feeble. To make use of this test, a piece of the cloth to be tested is placed in a boiling solution of potassa, composed of one part of caustic potassa and one of water. After two or three minutes' immersion, the excess of alkali is expressed by means of several folds of bibulous paper, and the successive fibres of the warp and w^oof are counted ; those of linen being of a deep, and those of cotton of a light yellow, or white. 2nd. Concentrated sulphuric acid quickly attacks cotton fibres, and converts them into gum, while linen fibres remain white and opaque. By washing the gummy matter is removed, and the sulphuric acid neutralised by the addition of a small amount of caustic potassa. The 00 CHEMICAL ANALYSIS OF MATERIALS. threads of the sample having been counted, the missing ones will represent those of cotton. In paper-making linen and cotton rags are distinguished by the touch, and the workwomen very soon acquire suf- ficient skill to make this distinction rapidly. 34. Waters. — The degree of pureness of the waters, used in the different processes of paper-making, has a great influence upon the quality of the papers. Murky water, containing argillaceous and silicious substances, turns the pulp yellow ; that containing saline matters in solution, to a certain extent destroys the brilliancy of the colouring materials, or forms an impedi- ment in sizing with gelatine. If the water contains organic matter in solution, it sometimes prevents the manufacture of superfine papers. d It is therefore important to determine, at different seasons of the year, the composition of the water em- ployed in washing and other manipulation. We may obtain by filtration the amount of argillaceous and silicious matters held in suspension in the water. The calcareous salts held in solution are precipitated by means of a few grains of alum or carbonate of soda. If it is desired to make a complete analysis, the most certain method is to evaporate a quantity of the liquid and examine the residuum after evaporation, for the different proportions of the substances contained. I 35. Alkahmetrical Test. — The soda of commerce has no value except in so far as it contains soda in the con- dition of a carbonate, or of caustic soda. To determine HENRY WATSON, HIGH BRIDGE WORKS, NEWCASTLE-UPON-TYNE, GENERAL MECHANICIAN. fill iiiipiefiiii IP ill iinifii iiSff 11 piifii^ Also of improved Eevolving and Jogging Strainers in Vats complete, DOCTOR PLATES, BRASS AND COPPER ROLLS, HYDEAULIC PRESSES AND PUMPS, Gun Metal Cocks, Valves, Water and Steam Gauges, Hydraulic Rams, &c. H. W. begs to intimate that to meet the increasing demand for his improved Strainer Plates he has just completed extensive additions to his Premises and Machinery, and is now in a position to execute all orders promptly, and on the most reasonable terms. The strictest attention will always be given to maintain in the highest degree the quality of the material and workmanship, keep- ing in view also the capability of re-closing after having been worn. Raer Chopper, V. Lam., 1 ft. wide. Pinislimg Calender. These rolls are made of refined Chilled Roll Metals. BENTLEY & JACKSON, ENGINEERS, IRONFOUNDERS, AND MACHINISTS, BURY, NEAR MANCHESTER. Makers of Paper Making Machinery, Millboard Machines, Paper Cutting Machines, Ripping and Winding Machines, for preparing paper for continuous printing presses. Hydraulic Pumps and Presses. Steam Engines and Boilers. ESTIMATES ON APPLICATION. IMPOETANT TKADE PUBLICATIONS. Thirteenth Edition. Demy 8vo., 104 pages. Stiff Covers, Price 2s. 6d., or Post free for 30 Stamps, THE PAPER MILLS DIRECTORY. Price Ss. Gd., or Post free for Thirty Stamps, THE PAPER STAINERS DIRECTORY OF GREAT BRITAIN. The Seventh, a New, Corrected, and Enlarged Edition, Foolscap 8vo., price 3s. Gd., or Post free for 30 Stamps, THE STATIONERS HANDBOOK, AND GUIDE TO THE PAPER TRADE. Sixth Edition, Demy 8vo., 60 pages, Stiff Covers. Price 2s. 6d., or Post free for 30 Stamps, THE CHEMICAL MANUFACTURER S DIRECTORY. Sold by KENT & CO., Paternoster Row. Price 3s. M., or sent Post free for 30 Stamps, A MAP OF THE PAPER MILLS OF ENGLAND. Arranged by the Editor of " The Paper Mills Directory." THE EDITOR, at 24, Bucklersbury, London, E.G. i No. Price THE ART PAPER MAKING: % (§\\\)it t0 \\t %\mi anJr practice MANUFACTUBE OF PAPEE. COMPILED BY THE EDITOE OF THE "PAPEE MILLS DIEECTOEY.' SOLD BY UNWIN BEOTHEES, OXFORD COURT, CANNON STREET. B. DONKIN & CO;S SPHERICAL ROTARY RAG BOILER. To CONTAIN FROM 20 TO 25 CWT. OV RaGS. This Boiler being spherical, is twice as strong as a Cylindrical one of the same diameter and thickness. The Plates used are notwithstanding of the usual substance, thus rendering it perfectly safe, durable, and suitable for high pressure steam. The Spherical shape has also another important advantage, viz. allowing all the rags to fall out by themselves, when the boiler is revolving with its cover off. Inside the Boiler are Strainers to take off the dirt, and Lifters to agitate the rags during the process of either boiling or washing. To avoid Cement or run lead Joints, the Gudgeons and the Boiler are turned true in a lathe to fit each other, the joints being simply made with red lead. B. DONKIN & CO., ENGINEERS, MILLWRIGUTS, MACHINISTS, AND IRON FOUNDERS, Blue Anchor Eoad, Bermondsey, London, S.E. Manufacturers of Steam Engines, Turbines, Water Wheels, Flour Mills, &e., Paper Making Machines, and everything connected with i'aper Mills, Bag Engines, Kag Cutters, Cutting Machines, Glazing and Hydraulic Presses, Pumps, Hoists, &c., Steel Bars and Plates, Wires, Felts, and Deckle Straps. Machines for Printing in two Colours. Gas and Water Valves. CHEMICAL ANALYSIS OF MATERIALS. 97 the quantity of these materials present, resort may be had to tests "with the alkalimeter. The principle upon which experiment is based is as follows : Given a dilute solution of free alkali, of car- bonate or sulphate of potassa or soda, of chloride of potassium or sodium, &c. Pure sulphuric acid, diluted with water, is added to the mixture ; this acid acts only on the free alkali, or on the carbonate, and, as long as the acid is not present in sufficient quantity to produce a neutral sulphate, the liquid manifests an alkaline re- action ; but when the base is saturated, the solution becomes neutral to coloured tssts, and when this point is exceeded never so little, the liquid will redden litmus paper. We are thus enabled to determine the exact moment of saturation. Experiment shows that, if the substance analysed is pure potassa, it would take 77 grs. of sulphuric acid to neutralise 74 grs. of potassa. Supposing that we are operating on the potash of commerce, containing water, carbonic acid, chloride of potassium, and sulphate of potassa. If instead of 77 grs. 35 are sufficient, this potash contains 50 per cent, of foreign and worthless matters. To make this experiment, 741 grs. of the potash to be tested are weighed off and dissolved in such a quantity of water that the solution shall occupy 30 cubic inches. Take 3^ cubic inches, and this volume will contain in consequence 74 grs. of the potash to be tested. This alkaline solution is then poured into the vessel in which it is to be neutralised. The sulphuric acid is prepared by dissolving 154 grs. H 98 CHEMICAL ANALYSIS OF MATERIALS. of sulplmric acid in a sufficient quantity of water to complete the volume of 1-7G pt. If we now take a burette, graduated 100 divisions will correspond to 77*20 grs. of pure acid, and from this it results that, if it takes 100 divisions to complete the saturation, the potash under consideration would be pure. For 60 divisions the proportion of alkali would be 60 per 100. The number of divisions, or degrees, on the alkali- metrical burette expresses, therefore, the proportion by /veight of the alkali contained in the material to be examined. 36. Examination of Limes. — The principal substances to be met with in limes are magnesia, silica, alumina, and traces of the oxides of iron. The limes to be preferred for use in Paper-making are the rich white kinds, which contain at least 90 per cent. of pure lime. To make an analysis of any given lime, take 154 grs. of the material, and heat them in a porcelain capsule, at a temperature of 212'' to 392" Fahr. The difference of weight now represents the proportion of moisture con- tained in the specimen. Kaisiug it to a red heat in the same capsule, or better still in a platinum crucible, we determine by again weighing the amount of carbonic acid which might have been left in combination with the lime through imperfect calcination. Then take 30-88 grs. of this material and treat them with dilute hydrochloric acid, gently heating to complete the solution of all soluble principles. By pouring ammonia into the solution, the alumina, CHEMICAL ANALYSIS OF MATERIALS. 99 which might have been dissolved by the hydrochloric acid, will be precipitated. The alumina and silica are separated from the solution by filtering, and then weighed after desiccation and incineration of the filtering paper, the ash of the paper being subtracted from the weight. In the filtered liquid the lime is precipitated either by oxalate of ammonia or by sulphuric acid. In the first case by calcining we obtain caustic or quick-lime, which is weighed. In the second, the lime is treated in the condition of a sulphate, and after precipitation of the magnesia by the phosphate of ammonia and magnesia, is calcined, and the phosphate of magnesia weighed. By means of chemical equivalents the proportions of caustic lime and magnesia contained in the material are determined. As the caustic portion is alone fitted for the uses of alkaline boiling, it should alone determine the market value of lime ; the other substances being, if not inju- rious, entirely inert. For manufacturing white paper, perfectly white limes should be used. This quality, indeed, is an indication of their pureness, as poor limes are of a more or less greyish colour, owing to the large proportion of clay they contain. 36.* Chlorometric Tests. — We owe the chlorometric pro- cess generally followed to Gay-Lussac ; it is based upon the oxidising properties of chlorine salts in the presence of water ; the water in decomposing yielding its oxygen to the oxidisable body and the chlorine uniting with the hydrogen to form hydrochloric acid ; and further, upon II 2 100 CHEMICAL ANALYSIS OF MATERIALS. the instantaneous discoloration of a solution of indigo by a slight excess of chlorine. Arsenious acid taken as the oxidisable body becomes conyerted into arsenic acid. The test liquid composed of arsenious acid, which is bought ready prepared at manufacturers of chemicals, is such as to necessitate a volume of a solution of chlorine equal to its own, in order to convert all the arsenious acid into arsenic acid. This liquid contains 67 grs. troy of arsenious acid to 1-76 Joints of the solution. It is prepared by dissolving that quantity of arsenious acid in pure hydrochloric acid, diluted with its own volume of water, and afterwards adding water enough to complete the litre. To perform the experiment, take 154-42 grs. of the chloride to be examined (chloride of lime, for instance), and wash them well with water several times in a mortar ; throw all the water used in this washing into a flask and add more water to make up the 1"7G pt. Pour 0-61 cub. in. of the arsenious solution into a vessel, slightly coloured by a solution of indigo. The burette is then filled with the chlorinated solution. The scale is graduated in such a manner that a hundred divisions correspond with ten cubic centimetres and the test liquid poured in drop by drop until the colour disappears. In order better to determine the instant at which the discoloration occurs, the vessel is placed upon a sheet of paper. If 100 divisions are used the degree would bo 100. - 200 „ „ „ „ 50. " I'^O n „ „ „ 66-6. CHEMICAL ANALYSIS OF MATERIALS. 101 100 degrees indicate that 2-679 lbs. troy of the chlo- ride contains 6102'70 cub. in. of chlorine gas. The chlorides of commerce contain about 5049*67 grs. of chloride to 2*679 lbs. It is well, as in the alkalimetric tests, to perform two or three successive experiments, in order to ensure the accuracy of the first. When once accustomed to testing in this way, a per- son can make several analyses of different chlorides in an hour. The following, according to M. Payen, is the composi- tion of chloride of lime. 1st. Pulverised, at 100°, that is, containing 100 volumes of chlorine. Composition by equivalents : — 2 CI. = 886-4 4 CaO = 1400-0 4H0 = 4500 2736-4 Or by hundredths : — Chlorine 32-39 Lime 51-16 Water 16-45 100-00 2nd. Composition of chloride of lime in aqueous solu- tion. 2 CI. = 886-4 2CaO=: 700-0 4H0 r: 450-0 2036-4 1(D2 CHEMICAL ANALYSIS OF MATERIALS. Or by Imndredtlis : — Chlorine 43-52 Lime 34-37 Water 22-11 100-00 These figures clearly establish the fact, that the same quantity of lime vdW absorb twice the amount of chlorine, when the reaction takes place in the presence of water. There is then advantage in preparing the liquid chloride, when it is not necessary to transport it. The powdered chloride may be considered as retaining the chlorine by a double equivalent of lime, whereas in the liquid chloride it is the excess of water which, in effecting the solution, preseiTCS the stability of the compound. 36.** Examination of Manganese. — The market value of this substance, as applied to Paper-making, depends upon the amount of chlorine it is capable of liberating in the preparation of chlorine gas, or the decolorising chlorides. The weight of chlorine disengaged is proportional to the quantity of pure biuoxide of manganese contained in the ore. To ascertain the value of this substance, it is sufficient to act with it upon an excess of hydrochloric acid, and then determine the quantity of chlorine disengaged and collected in an alkaline solution. Precise analyses have allowed us to determine that Gl-44 grs. troy of pure binoxide of manganese disengage G102-70 cub. in. of dry chlorine at 278= ; and under a pressure of 20 inches of the barometer, or one atmo- sphere. CHEMICAL ANALYSIS OF MATERIALS. 103 It is evident from this that the hundredths of a Htre of chlorine will represent the hundredths of available manganese contained in the specimen of ore. To perform the experiment, therefore, weigh 61"44grs. of the manganese, reduced to powder, place it in a flask, and gradually add 1*52 cub. inches of hydrochloric acid through an S-shaped tube. The flask is supplied with another tube, through which the gas is conveyed into the alkaline solution. The mixture is heated gently, and the operation con- tinued, until the vapour of water has driven ofl" the last traces of chlorine. The resulting solution of chloride is poured into a flask, bearing an index marking the capacity of 1'76 pt. This measure is completed by adding the water used in rinsing the flask, from which the Hquid was taken. The test liquid being thus obtained, the experiment is concluded by employing the ordinary chlorometric test. The value of manganese also depends upon the pro- portion of hydrochloric acid which it employs in libera- ting the chlorine. The binoxide of manganese (MnOa) requires two parts of hydrocliloric acid to disengage one part of chlorine ; the sesquioxide (MuaOg) thi-ee parts of acid for one of chlorine ; and finally the protoxide (^InO) takes one part of acid, with which it simply forms a chlo- ride, without any evolution of gas whatever. These different reactions are expressed by the follow- ing chemical formulae : — Mn02 + 2HCl=MnCl + 2HO + Cl MnO + HCl=MnCl + HO Mn^Og + 3HC1 = 2MnCl + 3H0 + CI. The carbonates of lime and baryta, and the oxide of 104 CHEMICAL ANALYSIS OF MATERIALS. iron, ^Yhich are found in manganese ores, also combine with the hydrochloric acid, causing a proportional loss in the amount of gas evolved. The result is the libera- tion of carbonic acid^, which, in the manufacture of chlo- ride of lime, forms a carbonate which the chlorine w^ill not decompose. We thus have a loss in acid, in hme, and in the degree of the chloride. M. Pay en, in his Chimie Industrielle, gives the following process for deter- mining the amount of acid employed. To dissolve 01*46 grs. troy of the binoxide of man- ganese, and produce Gl*02 cub. in. of chlorine, a quantity of hydrochloric acid equivalent to 135*10 grs. of con- centrated sulphuric acid is required ; half of the hydro- chloric acid, or 88°, forming chloride of manganese, and the other half giving the lOO'" of chlorine. One-hundredth of the acid lost in the operation is replaced. To determine the amount of hydrochloric acid required by any specimen of manganese, treat 61*46 grs. with 1*52 cub. in. of acid, representing 250 acidimetric degrees (equivalent to 19302 grs. of concentrated sul- phuric acid) and collect the chlorine. Admitting that 100 clilorometric require 173 acidimetric degrees, it is ascer- tained by neutralising with carbonate of soda (until the carbonate of manganese ceases to be dissolved) how much of the free acid remains. Now, by adding these two quantities we determine how much is needed to complete the 200 degrees of acid employed, and the deficiency will represent the amount of loss occasioned by the specimen of manganese. 37. Antichlorine. — We saw, while speaking of the wash- ing of the pulp, that in order to remove the last traces of CHEMICAL ANALYSIS OF MATERIALS. 105 chlorine retained with much energy by a sort of special capillary attraction, it was advantageous to make use of alkaHne sulphites, or hyposulphites, which, by combining with the chlorine, annihilate its destructive effects. Antichlorine is used in England upon a large scale. It has been calculated that the annual consumption amounts to 196-85 to 321*45 tons. To ascertain whether the pulp contains free chlorine, different reagents may be used, all, however, based upon the action of iodine upon starch when expelled by chlorine from its ioduretted compounds. The test hquid may be prepared in several manners, of which we will here give two. 1st. Carry the following mixture to the boiling point : — 1 part of iodide of potassium. 2 parts of starch. 3 parts of water. This liquid, preserved in a glass-stoppered vial, will colour the pulp blue, if it contains chlorine. 2nd. Boil for about three-quarters of an hour — Starch 5 parts. Fused chloride of zinc . . 20 ,, Water 1000 „ When the liquid is cool, add — Iodide of zinc 2 parts. To ascertain the presence of chlorine in the pulp, a bolus is made of the pulp by squeezing out the greater part of the water. The presence of chlorine is then manifested by the violet-blue colour of the iodide of starch. There is a great advantage in using the hyposulphite instead of the sulphite ; indeed, 106 CHEMICAL ANALYSIS OF MATERIALS. 2-679 lbs. troy of sulphite of soda absorb 4339 grs. of clilorine. 2*679 lbs. of hyposiilpliite of soda absorb 17*666 grs. of chlorine, or about 12*6903 cub. ft. It follows from this that to counteract 11-1244 cub. ft. or 2 679 lbs. of chlorine, it would take — 9-464 lbs. of the sulphite of soda. 2-841 lbs. of the hyposulphite of soda. Since the cost is equally in favour of the hyposulphite, this compound should be considered as the practical anti- chlorine for all industrial purposes. 38. Alums. — Different kinds of alum are met with in commerce. 1st. Alum with a potassa base, of which the formula is — K0,S03 + ALO33SO3 + 24HO, presents on analysis the following composition : — Potassa 10-82 Alumina 9-94 Sulphuric acid . . . 38-77 AVater 45-47 10000 2nd. Alum with an ammonia base, of w^hicli the for- mula is — (NH3HO) SO3 + ALO33SO3 + 24HO, gives the following analysis : — Ammonia 3-89 Alumina 11 90 Sulphuric acid . . . 36 00 Water 48-21 100-00 CHEMICAL ANALYSIS OF MATERIALS. 107 Alums contain iron in variable proportions, and the presence of this metal is injurious to the colour of pulps of delicate and pure shades. For such purposes it is well to use the purified material, known as refined alum, and proof against the tests of prussiate of potassa, which re- veals the feeblest traces of iron. It is sufficient to pour a few drops of ferrocyanide of potassium (yellow prussiate) upon some crushed alum, to obtain a blue tinge, if the alum contains iron. This method is also employed for purifying alum. Into a solution of alum enough of the prussiate is poured to precipitate the whole of the iron ; after allow- ing it to rest, the liquid is decanted by means of a syjjhon, and may be used at once or reduced to the form of crystals. The precipitate of Prussian blue may be used for colouring pulp. Instead of alum, sulphate of alumina is also used, which is generally obtained by roasting together alumi- nous schist and iron pyrites, and is apt to contain con- siderable quantities of that metal ; so that it is necessaiy to purify this substance with ferrocyanide of potassium, when it is to be employed in making fine -coloured papers. Although alum is more costly than sulphate of alu- mina, it is preferable because less variable in compo- sition. The sulphate of alumina, which sometimes contains an excess of acid varying from two to six per cent, of the weight of the sulphate, gives rise, by this irregularity of composition, to considerable practical difficulty. 39. Kaolin. — Kaolin is a basic silicate of alumina, 108 CHEMICAL ANALYSIS OF MATERIALS. produced by the decomposition of feldspathic rocks. In a crude condition it contains sand of various fine- ness, of which it must be freed before using in Paper- making. This purification is effected by successive washings, which carry off the finest particles. This clay, although smooth to the touch, when dry adheres roughly to the tongue. I We give the composition of washed kaolin : — Water 12 82 Sihca 48-37 Alumina 34*95 Oxide of iron .... 1*26 Potassium and soda . . 2-60 10000 I To ascertain the pureness of kaolin, it is only necessary to wash it and then determine the quantity of quartz granules contained in the specimen. It is always prudent not to employ kaolin in the prepa- ration of size until it has been passed through the meshes of a fine sieve. i 40. Starch. — Starch is a mealy substance, generally extracted from potatoes. The chemical formula for that of commerce is Ci.,Hi„Oio+4HO. Starch contains variable proportions of water. What i is called dry starch, or that containing four equivalents, j has 18 per cent, of its weight made up of water. Placed iu a very moist atmosphere it may absorb 10 equivalents, or contain 35 per cent., while other starch which has CHEMICAL ANALYSIS OF MATERIALS. 109 only been drained contains 45 per cent, of its own weight of water. It is, therefore, important to make sure of the hygro- metric condition of the starch by desiccating it at a tem- perature of 68° to 86° Fahr. in a dry atmosphere. To increase the weight of starch it is mixed with various matters, such as chalk, plaster, sulphate of baryta, &c. The adulteration is, however, easily recognized by inci- neration. Starch burns up entirely, leaving no residuum, whereas the incombustible mineral matters are found at the bottom of the porcelain capsule in which the experi- ment is performed. 41. — Colouring Matters. — Colouring materials can scarcely be otherwise tested than by directly staining a certain fixed weight of pulp. A w^ell-supplied paper-mill ought to have a small cylinder engine, holding about 11 lbs. avoir, of dry pulp. The quality of the colouring matter may then be determined by gradually increasing the proportion and making several sheets of paper on a small mould. The results obtained in this manner are then compared. This method of experiment renders great service when a new kind of paper is to be made, and we have no very definite idea of the amount of colouring matter necessary to give the required shade. M. Liverkus, a manufacturer of ultramarine in Ger- many, recommends the following very practical method, applicable, however, to this substance alone. After having poured 46 "32 grs. troy of the ultramarine to be tried into a vessel containing 4,168 grs. of a con- centrated solution of alum, the material is agitated so as 110 CHEMICAL ANALYSIS OF MATERIALS. to cause all the colouring matters to become suspended in the liquid. After resting for some time, say half an hour, the mix- ture is again stirred. The colouring power of the ultramarine will be directly proportional to the length of time required for the solution to be discoloured. With some blues the discoloration takes place at the end of three to six hours. In others, on the contrary, no perceptible change in the colour can be observed until the second or third day. The aluminous solution is prepared by dissolving 1,135 grs. of alum in 2-679 lbs. troy of water. Ochres have a commercial value proportional to the care with which they have been w^ashed. It is an easy matter to ascertain the amount of gravelly material they contain, and the same holds good in regard to clays, which are used in considerable quantities for manu- facturing common papers and boards. It is, however, important not to pour these matters into the engine, or the vat, until they have been filtered through a long- napped felt, which will retain the coarser particles. For the finer colouring matters, fine felts or flannels are used. 42. Fuel. — The analysis of combustibles has for its object to determine the proportion of ash contained in them, and their calorific power ; that is to say, the number of degrees of heat which may be generated by a given quantity of fuel. The weight of the ash may be obtained by calcining 30-88 grs. troy of the combustible to be examined in a CHEMICAL ANALYSIS OF MATERIALS. Ill platinum crucible, and stirring the material several times, in order entirely to consume the carbon. The analysis may be considered perfect, when no difference can be ob- served between two consecutive weighings. The ash containing much carbonate of lime, which is transformed by calcination into caustic lime, a few drops of a solution of carbonate of ammonia should be added and the whole carried to a red heat. The results of this experiment are generally considered suflficient, as they allow us to form an estimate of the commercial value of the fuel. It is, however, also of the greatest importance to have correct notions of the num- ber of calorific degrees which can be obtained. The examination with litharge is the simplest method, and is based upon the following principle : — The amounts of heat emitted by different combustibles are to each other as the respective amounts of oxygen absorbed by these combustibles in burning. It will be sufficient then to compare the amounts of oxygen required for the combustion of various specimens of fuel with that already determined for pure carbon. Take 1544 grs. troy of the pulverised fuel and mix it with 308-84 grs. of litharge. This mixture is placed in a crucible, and covered over with 463*23 or 617*69 grs. of pure litharge. The capacity of the crucible ought to be such that the volume of the material should not occupy more than one-half in order to allow for the swelling which takes place As soon as the fusion is completed the substance is allowed to cool after having been heated briskly for ten minutes. The crucible is then broken, and the lump of lead which is found in it weighed. The calorific power of the fuel is proportional to the weight of 112 CHEMICAL ANALYSIS OF MATERIALS. the lead. This hypothesis, which is not mathematically exact, gives a sufficient approximation, however, for all practical purposes. It is indispensable to use litharge free from minium (red oxide of lead). Pure carbon produces with litharge thirty-four times its own weight of lead, and according to experiments made by M. Depretz each part of lead is equivalent to 230 degrees of heat. Pit coal is the fuel generally employed in France. The calorific power of this material is somewhat variable according to its source and the proportion of ash itj contains. We give, for the sake of information, the following] results of analyses : — Asb Calorific Source. per cent. power. Hartley (England) . . . 1-50 6781 Anthracite (Wales) . . . 1-60 7406 Mons (Belgium) . . . . 216 7297 43. Examination of Papers. — Gelatine-sized papers are usually recognisable by their odour ; this characteristic is not, however, sufficient for some of those sized in this] manner but made by machinery. By incineration they are found to burn badly, leaving] a very black carbonaceous deposit ; whereas paper sized] with resin in the pulp yields rather a greyish residuum. If the results obtained are uncertain, we must then have! recourse to an elementary analysis. Gelatine will be| indicated by the amount of nitrogen collected. As resinous-sized papers always contain starch in I CHEMICAL ANALYSIS OF MATEHIALS. 113 greater or less quantities, a blue colour is obtained when tbey are subjected to the vapour of iodine. Generally it is thought sufficient to determme the amount of mineral matters which the paper contams. All that is required, then, is to incmerate 154-42 grs. ti'oy in a porcelain capsule, and to weigh the remainder after calcination. Paper made exclusively of rags yields a residuum of a third to a half or one per cent, at most, proceeding from the ash of the textile fibres, and the small amount of mineral matters introduced by the waters. By deducting a hundredth of the weight of the resi- duum, the weight of the additional matters is obtained^ The substances generally employed are — Kaolin, Chalk or carbonate of lime. Sulphate of lime. Sulphate of baryta. Clays. Ochres. The examination of these different substances necessi" tates a complete anal3'sis, which is more in the line of the chemist than the manufacturer of paper. We give the course to be pursued, based upon experi* ments made to determine the nature of substances com- posing the size used in pulps, or resinous sizing. Chemical Exainination of Paper Sized in the Pulp, *' Subjected to the following tests : Boiled with pure water, it yielded a hquid which restored the blue colour to red Htmus paper, thus revealing the presence of an I I 114 CHEMICAL ANALYSIS OF MATERIALS. alkali. An infusion of nutgalls scarcely affected its transparency, so that it did not contain gelatine ; but f iodine produced a very intense blue colour, indicating that starch formed a part of the composition. " 186"28 grs. troy of the same paper were boiled for about a quarter of an hour in water acidulated with sul- phuric acid. The liquid was expressed through a piece of line linen and the pulp well washed with boiling water. When dried it weighed only 172*31 grs. The acidulated liquid was united with the water used in washing the pulp, and saturated with carbonate of lime. After being filtered it was partially evaporated, in order to remove the greater part of the resulting sulphate of lime ; when evaporated almost to di'yness, a yellow residuum was ob- tained, having a gummy appearance, and weighing 10"33 grs. This substance, when heated in a platinum crucible, became swollen, emitted an odour of burnt bread, and gave an ash containing sulphate of lime with the sulphate of a fixed alkali, not determined. " The solution in water of this apparently gummy substance was only feebly precipitated by an infusion of galls ; but assumed a beautiful dark violet colour when treated "vsith iodine. This material was, therefore, only sHghtly modified starch. '* The 172-31 grs. of paper which had resisted tlie action of the boiling water acidulated with sulphuric acid, were treated with a weak solution of potassa. The ex- pressed liquid, while boiling, was of a transparent yellow- ish colour, but became somewhat turbid on cooling, and gave a lather like soapsuds. " A small quantity of dilute sulphuric acid was poured into this liquid to neutralise the potassa, when it became CHEMICAL ANALYSIS OF MATERIALS. 115 very milky and deposited a flaky matter, which did not redissolve by heat. This weighed about 3-08 grs. after evaporation in a capsule smeared with grease. The cap- sule, as well as the flaky matter, was washed with alcohol, which assumed a brownish colour, and became charged with fatty matter. ** The residuum, insoluble in alcohol, was composed to a great degree of starch which had escaped the action of the boiling acidulated water. The liquid, separated from the 3-08 grs. of flaky matter by sulphuric acid, also contained starch ; for when evaporated in order to crystallise the greater part of the sulphate of potassa, it yielded a yellowish mother- water, which gave a deep blue colour with iodine, and a brownish sediment was formed still containing starch. When distilled in a test-tube it gave an alkaline liquid which turned red litmus paper blue. " This appears to be due to the gluten contained in the flour of cereals used to size the paper under consider- ation. " To return to the brownish alcoholic solution resulting from washing the flaky matter. After evaporation there remained 1-54 grs. of a fatty and somewhat pitchy sub- stance, of a yellowish-bro\^^l colour, and having about the consistency of lard. Its combination with potassa was very highly coloured and had a bitter taste, which seemed to indicate the presence of a resin. To ascer- tain whether the suspicion was correct, it was treated with water and a small quantity of magnesia to neu- trahse the fatty acids, and then subjected the r^'sidue to the action of boiling alcohol, which, on evaporation, left behind a slight coating of varnish, easily recognised as a resin. i2 116 CHEMLCAL ANALYSIS OF MATERIALS. ♦' 77*20 grs. of paper yielded when burnt, 0'92 grs. of a very ferruginous ash, which also contained a noticeable quantity of manganese ; for when melted with soda in the flame of a blowpipe, it gave a beautiful blue glass. This ash did not effervesce with acids. When heated to redness with sulphuric acid, and the residuum treated with water, it had very little taste at the time of mixture, but at the end of twenty-four hours the liquid had ac- quired a very distinct astringent taste, and with ammonia produced a gelatinous precipitate of alumina ; from which it follows that alum had entered into the composition of the paper pulp." A great many machine-made papers become yellow, or present spots of a yellow rust colour. These spots are due to the action of the chlorine contained in the pulp upon the iron of the drying cylinders. Indeed the felts soon become spotted with the same colour at different points, and sometimes over their whole surface. It may be shown that these spots are really due to an oxide of iron, by employing the sulphocyanide of potassium, which produces a red colour, growing deeper as the proportion of the peroxide of iron is increased. By digesting the leaves of paper to be tested in dilute hydrochloric acid, the paper becomes white, and the presence of peroxide of iron is detected in the solu- tion. MM. Fordas and Gelis have published a paper upon this subject in the Journal de Librarie, from which we extract what follows : — " On leaving the rag engines the wet pulp is conveyed into a wooden vat, and thence to the various apparatus which constitute the paper- machine, and whose objects 1 CHEMICAL ANALYSIS OF MATERIALS. 117 are the formation and drying of the paper, and its division into leaves. " Perfectly washed pulp would be in no way changed during these different operations ; for the vapours gener- ated being only those of water, could not, by acting on the different materials of which the apparatus is com- posed, give rise to any soluble compound capable of becoming incorporated with the leaf during its manu- facture. " But instead of supposing such a case, which is never practically attained, an incompletely washed pulp is used ; a large part of the chlorine it contains will, it is true, be carried off by the excess of liquid during the first stages of the operation ; but there will always re- main a portion, which will be liberated with the vapour of water at the time of drying, attack the cast-iron rollers, dissolve their surface, and form with them a minimum chloride of iron, with which the felts support- ing the leaves wdll become impregnated, and which from them will be introduced into the substance of the pulp itself. " This impregnation of the felts with a salt of iron can- not be denied. These felts are always spotted with rust, and the yellow colour begins to be perceptible during the first days they are in use. The rust actually becomes combined with the tissue, and is the result of a maximum basic salt of iron, produced by the action of the air upon the minimum salt we have before mentioned. " This salt cannot possibly affect the paper ; it is in- soluble and combined ; but it is the free and soluble part which exists upon the surface of the cylmders, or in the substance of the felts. 118 CHEMICAL ANALYSIS OF MATERIALS. '* We will admit that the ferruginous compound enters the paper in a soluble condition, and at the minimum of oxidation, because this fact seems to us to be proved by the absence of colour in the paper at the time of manu- facture. " The complete state of dryness of the paper maintains the salt of iron for some time at a minimum, and con- sequently in a colourless condition ; but very soon the oxygen of the air, assisted by atmospheric moisture, re- acts upon this compound, and, by bringing it to a maxi- mum degree of oxidation, colours it. *' This simple reaction perfectly explains the yellow, and often nankin colour which these ferruginous papers assume. It also explains an observation made to us by a printer, namely, that this colour is frequently produced when the paper is wetted for printing, or accidentally. " As for the round spots, which have more particularly engaged the attention of manufacturers, they may be quite as easily explained. We attribute them to a phe- nomenon of crystallisation. *• They are formed through the tendency, possessed o a certain degree by the molecules of all bodies, to arrange themselves in groups when they are disseminated through- out a permeable medium. If the spots of which we are speaking are closely observed, it will be remarked that each one of them encloses near its middle an asperity, or hard body, which seems to have served as a centre of attraction. " As a result of the condition of alternate dryness and moisture to which the porous and hygrometric material of paper is exposed, an insensible displacement occurs. The molecules of the salt of iron arrange themselves CHEMICAL ANALYSTS OF MATERIALS. 119 around the most compact parts of the paper pulp, as a salt in concentrated solution is deposited around the glass rods or strings suspended in it ; or, to make use of a comparison, which seems to us more exact, these ferru- ginous molecules group themselves in the sheet of paper, as we find them doing in wet soils to form those globules of oxide of iron, or radiated pyrites, so abundant in all localities where ochrous earths are buried in organic deposits. *' Frequently when the displacement we have just ex- plained does not occur until after printing, it is the printed letters themselves which become the centres of attraction, so that the iron becomes fixed in preference upon the printed portion of the paper, which it deeply stains, while the margins appear relatively colour- less." 44. The Laboratory. — Instruments and Apparatus. Microscope. Hand lens. Scales with stand, weighing as low as 0*154 grain troy. Blowpipe. Stationary furnace. Hand furnace. Porcelain capsules of difi'erent diameters. Platinum crucible and accessories. Porcelain mortar. Spirit lamp. Instrument for ascertaining the specific gravity of salts. Thermometers, graduated upon glass. 120 CHEMICAL ANALYSIS OF MATEETALS, I Earthen crucibles. Test tubes, with stand, ■ Flasks. Glass tubes,, straight, „. ,, curved. ,, ,, S-shaped, with funnel end. Glass funnels. Earthen funnels. India-rubber pipe for jointSc Wooden tongs and holders. Alkalimetric apparatus. Chlorometric , , Acidimetric ,, Filtering paper. Reagents. Blue test paper. Red ,, „ Hydrochloric acid. Sulphuric ,, Nitric ,, Acetic ff Ammonia. Iodine. Potassa. Soda. Lime. Oxalate of ammonia. Carbonate of soda. Carbonate of ammonia. Phosphate of soda. Alum. CHEMICAL ANALYSIS OF MATERIALS. 121 Protochloride of tin. Acetate of lead. FeiTOcyanide of potassium. Sulphocyanide of potassium. Sulphate of iron. Sulphate of copper. Litharge. Tannin. Starch. Alcohol. Ether. Arsenious acid test liquid. Sulphuric acid ,, ,, Hydrotimetric ,, ,, Borax. Salts of phosphorus. Chromate of potassa. Iodide of potassium. 122 WASHING-WATER, WATER POWEE, STEAM BOILERS, ETC. Chapter VII. WASHING-WATER, WATER POWER, STEAM BOILERS, &c. 45, Washing- Water. — The water which is used for the preparation of the pulp, especially for washing purposes, is called ivash-ivater, in distinction to that which only drives and produce power. To perceive its great importance we need only consider the quantity which is necessary to wash from 400 to 500 pounds of rags in an engine. If the engine is, for ex- ample, 15 feet long, 7j feet wide, and filled about 2 feet high, it will hold, taking the rounds ends, backfall, &c., into consideration, about 160 cubic feet or 1,200 gallons, or about 10,000 pounds of water — that is, about twenty times the w^eight of rags. If enough water is used dur- ing the operation of washing to fill the engine five times, the 500 pounds of rags will be brought in contact with 6,000 gallons, or one hundred times their own weight of water. This quantity may yet be considerably increased, perhaps doubled, if we add the water which is used in boiling, bleaching, and on the paper-machine. Every pound of rags is therefore liable to be soiled during its transformation into white paper by the im- purities contained in 100 to 200 pounds, or in 12 to 24 gallons, of water. These impurities are of two kinds : those which are only suspended, floating, or mechanicallij mixed, and others fully dissolved. In most cases the latter cannot be seen, and, to make a distinction, may be called chemical impurities. WASHING-WATER, WATER ROWER, STEAM BOILERS, ETC. 123 46. Mechanical Impurities. — If a stream flows through soil composed of clay or other soft material, it will be clear as long as nothing disturbs its quiet and even flow ; but as soon as rain falls, some of the infinitely small and hght components of the earth are bodily carried along by it, and the stream becomes turbid and coloured. All these mechanical impurities are visible separate matters, and which will settle on the bottom if allowed time to do so. Mills which use surface streams as wash-water should therefore be supplied with settling ponds, of as large dimensions as the locality will permit. The water is ad- mitted into these reservoirs when it is clear and shut off when muddy ; they should therefore be large enough to hold many days' supply. If a large settling pond cannot be had, the mechanical impurities must be separated by filtration. The cheapest and most permanent materials for filters are gravel and sand. They are hard, principally quartz, and their round form prevents them from forming a mass so com- pact that water cannot pass through, while they present at the same time a very large surface for the deposition of impurities. The filters may be built of brick or stone and cement, or simply of earth ; their bottoms, according to Planche, are to be covered with coarse gravel or stone, ordinary gravel succeeding, and sand forming the upper layer. The water usually enters on the top and leaves at the bottom. In the course of time such an amount of dirt will settle on the stones that water can no longer be puri- fied by passing through them, and then they have to be thoroughly washed. 124 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. To suffer no delay, the mill should be suppHed with two such filters, or with one divided by a partition, which may be opened or closed at w^ill, so that one part can be cleaned while the other is yet in operation. A useful kind of filter is a square brick filter, divided by cross walls into four equal compartments, which are filled several feet deep with fine gravel, and connected by short pieces of large iron pipes in such a manner, that the water passes constantly through three of them in suc- cession, w^hile one can be cut off and washed out. A workman enters for this purpose, moves the gravel to one side, thus making an empty space, on which he gradually washes the gravel by mixing it with plenty of water. The dirty water escapes through a valve with which each chamber is provided. The wash- water frequently passes through additional strainers before entering the reservoir in the upper part of the mill. One, which is frequently seen, consists of two wire-cloth covered frames, with woollen rags, slightly cut in the engine, filled in between them. This double frame is fastened horizontally in a tub or vat, the water passing through it from the lower side, so that the impurities cannot lodge upon it, but will fall to the bottom and leave the wire unobstructed. If the water enters the engine from the top, a flannel bag may be tied around the outlet of the pipe, and if frequently washed out it will be of some assistance. 47. Chemical Impurities. — The chemical impurities, mostly invisible, are as numerous as the materials over which the streams pass in their courses. Some metals, but especially iron, some of the alkahes, principally lime, WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. 1Q5 and extracts of decaying vegetable matter from the drain- age of cultivated fields, are the substances which are mostly found dissolved in water. Carbonates of lime and magnesia are very slightly soluble in pure water, but dissolve freely in water which contains carbonic acid, and as the latter is always to some extent absorbed from the air, the water is capable of holding some carbonate of lime or magnesia in solution. Sulphate of lime or gypsum is another form under which this alkali is found in water. Common salt or chloride of sodium appears occa- sionally. Water which contains much lime or magnesia is called hard, and every housekeeper knows that it will not answer for washing purposes, as it does not dissolve soap until the lime or magnesia has been precipitated. If the carbonic acid, which enables the water to hold the carbonates of lime and magnesia in solution, is driven out by boihug, or absorbed by caustic lime or soda which may have been added for this purpose, the car- bonates will assume a solid form, settle to the bottom, and thus render the water soft. The sulphate of Hme or gypsum, or the clilorides or nitrate of hme, cannot be so easily eliminated, and water which contains con- siderable proportions of them is therefore called perma- nenthj hard. Hard water also affects a good many colouring materials, and its most objectionable quality is that it forms deposits in steam-boilers, which are frequently very troublesome, and may be the primary cause of an explosion. Wash-water which contains a sufficient quantity of lime to be hard, is unfit for a paper-mill. 126 WASHING- WATER, WATER POWER, STEAM BOILERS, ETC Iron salts in contact with alkalies, lime, or soda, deliver their acid to the latter, and the iron precipitates as oxide or rust, colouring the pulp until it may be re-dis- solved by sulphuric or other acids. Although the pro- portion of these salts be msignificant, the quantities of water used are so enormous that the total amount of iron is yet quite considerable. A 500-pound engme, for example, carries about 10,000 pounds of water, and if this contains only one-fiftieth of one per cent, of iron, there will be 2 pounds of it in the whole mass, and if this water is renewed five times during one washing operation, 10 pounds of iron will be brought in contact with the rags. The soda, bleach-liquor, alum, or sulphuric acid ab- sorbed or neutralised by these iron salts in the multitude of operations, in all of which water is an important factor, sum up to a large quantity in a short time. It is quite probable that the difference in the quantities of chemicals consumed for like operations on the same stock in difi'erent mills, may sometimes be traced to this source. The presence of iron can easily be discovered by the addition of a solution of yellow prussiate of potash to the water ; the iron salts will form with it Prussian blue. The total amount of mineral impurities can be ascer- tained by evaporating carefully several gallons of water and weighing the residue. For papers of a lower grade, such as wrapping, the preparation of which requires few chemicals, it is not a matter of vital importance; but as the colour of all white papers depends greatly on the purity of the wash-water, an abundance of pure, clear wash-water is one of the conditions of the successful manufacture of fine papers. WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 127 48. Sources of Wash-Water. — To determine which are the best sources for good wash-water, it is necessary to understand the manner of their formation. The water which covers the surface of the earth changes its form constantly ; it evaporates, and is taken up and carried away by the winds as vapour. The air is able to hold more water at a higher temperature than at a lower one ; any cold wind will, therefore, cause some of it to drop in the form or rain or snow. The water, on returning to the earth in this form, is as pure as it can be found in nature ; it contains no foreign matters but the gases which it takes up in the atmo- sphere, and if it flows over hard, insoluble substances, such as rocks of granite or quartz, or over sand, it pre- seiwes this purity. Streams of this kind are very valu- able, but unfortunately they can only be found in the mountains, in most cases too far from markets to be available. The purity of all surface waters, such as creeks and rivers, depends entirely on the nature of the soil over which they themselves and their tributaries pass, and should be in every case investigated. A great portion of the snow and rain filters though the ground and comes again to the surface in some lower places as springs, or gathers in large cavities below, to which access is had by means of wells. Very often these underground lakes extend from high places to low ones, and are only prevented from rising to a uniform level by a stratum of water-tight materials. In boring an artesian well this stratum is pierced, and the water forces its way upward with a tendency to reach the level of the highest point of the body of water from 128 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. which it comes. In some cases it rises through large pipes high enough to drive a water-wheel, while in others it hardly comes to the surface ; sometimes the water obtained is very pure, and at others it is loaded with foreign elements. Boring an artesian well in an untried place is like digging for hidden treasures, — a very uncertain under- taking. One or more never-failing springs of pure water, fur- nishing a full supply, are very valuable in a good loca- tion. Where good wells can easily be made, and where ex- perience has shown that they keep their supply all the year, they are, if the w^ater is chemically pure, often pre- ferable to surface water. While the latter may require to be artificially filtered, the well water has been cleared by passing through the soil. 49. Systems of Distribution. — If the wash-water can be taken from a convenient place above the mill, so that it has fall enough to run directly into the engines, a considerable amount of power and some machinery, which would otherwise be necessary to force it up, will be saved. Mills which are not so fortunate, must have receivers in some of the highest parts of their buildings, into which the water can be lifted by pumps, and from which it is distributed. These reservoirs must be water-tight, and, if of wood, should be circular ; but iron is a better mate- rial for this purpose, as it does not shrink and open in the joints, like wood, when for a time empty. If the pump has to stop for repairs, or from other HENEY WATSON, HIGH BRIDGE W0RK8, NEWCASTLE-UPOX-TYNE, GENERAL MECHANICIAN. fill liiipieipiii If III liPBOfip list f 11 Piif iSi Also of improved Eevolving and Jogging Strainers in Vats complete, DOCTOR PLATES, BRASS AND COPPER ROLLS, HYDRAULIC PRESSES AND PUMPS, Gun Metal Cocks, Valves, Water and Steam Gauges, Hydraulic Earns, &g. H. W. begs to intimate that to meet the increasing demand lor his improved Strainer Plates he has just completed extensive additions to his Premises and Machinery, and is now in a position to execute all orders promptly, and on the most reasonable terms. The strictest attention will always be given to maintain in the highest degree the quality of the material and workmanship, keep- ing in view also the capability of re-closing after having been worn. Rag Chopper, Wheel, 4 ft. diam. , 1 ft. wide. Finishing Calender. These rolls are made of refined Chilled RoU Metals. BENTLEY & JACKSON, ENGINEERS, lEONFOUNDERS, AND MACHINISTS, BURY, NEAR MANCHESTER. Makers of Paper Making Machinery, Millboard Machines, Paper Cutting Machines, Ripping and Winding Machines, for preparing paper for continuous printing presses. Hydraulic Pumps and Presses. Steam Engines and Boilers. ESTIMATES ON APPLICATION. IMPOETANT TKADE PUBLICATIONS. Thirteenth Edition. Demy 8vo., 104 pages, Stiflf Covers, Price 2s. 6d., or Post free for 30 Stamps, THE PAPER MILLS DIRECTORY. Price 2s. 6d., or Post free for Thirty Stamps, THE PAPER STAINERS DIRECTORY OF GREAT BRITAIN. The Seventh, a New, Corrected, and Enlarged Edition, Foolscap 8vo., price 3s. 6d., or Post free for 30 Stamps, THE STATIONERS' HANDBOOK, AND GUIDE TO THE PAPER TRADE. Sixth Edition, Demy 8vo., 60 pages, Stiff Covers. Price 2s. 6d., or Post free for 30 Stamps, THE CHEMICAL MANUFACTURERS DIRECTORY. Sold by KENT & CO., Paternoster Eow. Price 2s. Cd., or sent Post free for 30 Stamps, A MAP OF THE PAPER MILLS OF ENGLAND. Arranged by the Editor of " The Paper INIills Directory," THE EDITOR, at Oxford Court, Cannon Street, London, E.G. No. Price THE ART PAPER-MAKING: % ^Hik b i\t Cljeorj anir ^rattke MANUFACTUEE OF PAPER COMPILED BY THE EDITOR OF THE "PAPER MILLS DIRECTORY." SOLD BY UNWIN BROTHERS, OXFORD COURT, CANNON STREET. B. DONKIN & CO/8 SPHERICAL ROTARY RAG BOILER. To CONTAIN FROM 20 TO 25 CWT. OF RaGS. This Boiler being spherical, is twice as strong as a Cylindrical one of the same diameter and thickness. The Plates used are notwithstanding of the usual substance, thus rendering it perfectly safe, durable, and suitable for high pressure steam. The Spherical shape has also another important advantage, viz. allowing all the rags to fall out by themselves, when the boiler is revolving with its cover off. Inside the Boiler are Strainers to take off the dirt, and Lifters to agitate the rags during the process of either boiling or washing. To avoid Cemeut or run lead Joints, the Gudgeons and the Boiler arc turned true in a lathe to fit each other, the joints being simply made with red lead. B. DONKIN & CO., ENGINEERS, MILL^V1{IG^TS, MACHINISTS, AND IRON FOUNDERS, Blue Anchor Eoad, Bermondsey, London, S.E. Manufacturers of Steam Engines, Turbines, Water Wheels, Flour Mills, &c.. Paper Making Machines, and everything connected with Paper Mills, llag Engines, Eag Cutters, Cutting Machines, Glazing and Hydraulic Pressi's, Pumps, Hoists, Sec, Steel Bars and Plates, Wires, Felts, and Deckle Straps. Machines for Printing in two Colours. Gas and Water Valves, WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 129 causes, the mill has to be stopped also, unless the reser- voir holds water enough to keep it going. A strong foundation is at all times required for it, as the weight of water in even a small reservoir is con- siderable. One large mill in America is supplied with a capacious settling pond or reservoir on the top of a hill, situated higher than any part of the machinery, and is filled weekly by means of a large force-pump, driven by water-power, with a sufficient supply of water for the six following days. Two barrels of porous alum are emptied into this pond after it has been filled, in order to precipitate the impuri- ties. The lime and iron, which may be contained in the water in the form of carbonates, will form sulphate of lime and sulphate of iron with the sulphuric acid of the alum, while its other component part, the alumina or clay, is set free and carries down mechanically some of the floating impurities. 50. Quantity Eequired. — It is impossible to calculate exactly the quantity of wash-water which is required for a paper.mill, but an estimate, on which the sizes of the pump and of the distributing pipes may be based, is in- dispensable. We will indicate how the data on which such an estimate may be based can be obtained, and coupled with experience it will be a sufficient guide. The washing-engines consume by far the larger portion of all the wash-water, and we suppose, for example, that two of them, 15 feet long by 7^^ feet wide, holding each about 500 pounds of rags, are used in a mill of a capacity of about 3,000 pounds of white paper per day. It is true K 130 WASHING -"WATER, WATER POWER, STEAM BOILERS, ETC. that they do not alwaj^s wash at one and the same time, but it hai^pens so sometimes, and we must be prepared in such cases to furnish enough w^ater. If the receiver is large enough to hold all the surplus, so that no water need be wasted through the overflow, no powder is lost, but if the receiver is small the pump has to furnish an excess, wlaich during most of the time runs aw^ay. The w^ashers must be fed with as much water as they are able to discharge, and this will in most cases be amply done by a stream which will fill the empty tub in fifteen or twenty minutes. One of the engines taken for this ex- ample, holds 160 cubic feet = 1,200 gallons; two engines, therefore, require in fifteen minutes 2 x 1,200 = 2,400 gallons, or in one minute 160 gallons. The quantity of water consumed in boiling, by the beaters, and by the paper-machine, is difficult to estimate, but it is seldom as large as that required by the washers, especially at the high rate estimated in this example. We are, therefore, pretty safe in taking double the quantity, calculated for the w\ashing-engines, or 2 x 160 = 320 gallons per minute, as the wiiole supply needed. An abundance of wash-water is one of the first condi- tions in the manufacture of paper, and it is therefore wise rather to waste power than to have an insufficient quan- tity of water. We have to make allowance for deficient work of the pump and leakage in many places, and may add one-fourth of the calculated number, or ^^ ~ 80 gal- lons, and thus need a pump capable of throwing 400 gallons per minute. The power which is necessary to raise such a quantity of water can easily be calculated. If it is, for instance, taken from a well, at a depth of 12 feet, and pumped into WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 131 a receiver, the water surface of which is 38 feet above the ground, the total height through which it must be raised is 38 + 12 rz 50 feet. It takes as much power to raise the weight of 400 gallons or 400 x 8^ = 3,330 pounds 50 feet high, as w^ould be required to raise 3,330 x 50, or 166,500 pounds, one foot high. One horse-power is accepted as equal to the power which is necessary to raise 33,000 pounds one foot high in one minute, and our 166,500 are therefore equal to '330^ = 5 horse-power actual work. 51. Pumps. — Piston or plunger-pumps, with suction or pressure-valves, although the oldest style, are even at present in many cases preferred to all others. The valves must have time to open and close perfectly if a good result is expected, and their speed should there- fore always be moderate. The perfection of rotary pumps have been the study of numerous mechanics for many years^ and the patents taken out for them would alone fill a good-sized book. These pumps work mostly without valves and run fast, but their speed, and with it the quantity of water thrown, can be considerably increased or reduced at wdll ; they take little room, are operated by belts, can be easily set up, and require little care. Rotary pumps, being generally less efficient as suction than as force-pumps, are usually set as close to the source of supply as possible ; some working best, if the water is made to run into them without any suction. As force- pumps they are excellent, and some are even used for steam-fire engines. The absence of valves makes it possible for pieces of K 2 ]32 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. woocl, rags, or other solid matters, which may accidentally be in the water, to pass through a rotary, while they would obstruct and perhaps damage a piston-pump. The pressure of the atmosphere, at a low density, is equal to that of a column of water 30 feet high. If it were possible to construct a pump so perfect, that it could withdraw all the air from the suction-pipe, thus creating an absolute vacuum, the water would rise in it to a height of about 30 feet, forced by the atmosphere out- side ; but if the water had to be raised but a trifle above this limit, the atmospheric pressure could not do it, and it would never reach the pump. The height to which any pump can raise water by suction is therefore always less than 30 feet. If the suction-pipe is not perfectly air-tight, if the air enters tlu'ough a small hole or crack, feeding the pump in the place of water, a vacuum cannot be created, and the pump will run empty. The hole may be small enough to be invisible to the eye ; it may accidentally appear by the loosening of a joint, the opening of a sand-hole in the casting, or otherwise, and its discovery is often difBcult. It is most likely to be found by running the pump and creating suction, and then holding a burning light to all suspected places ; wherever there is an opening the current will change the direction of the flame and draw it in. This difficulty is a sufficient reason for such a disposi- tion as will admit of as short a suction-pipe as possible. There is another kind of pump which is frequently used in paper-mills, and recommends itself by reason of its simplicity. WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 133 It consists of a strong rubber belt carrying iron or copper buckets, which belt runs over two large flanged pulleys, one of which is located inside of a lower water-tank, while the other is fastened above the upper receiver. The shaft of the upper roll is turned by a pulley and belt, and raises the buckets which have been filled in the lower tank. They travel up and down in close-fitting wooden spouts or troughs, and are emptied, as soon as they pass over the upper roll, into a receiving channel, which connects with the reservoir. This pump is built on the same principle as a grain elevator, and works well where the water is not to be raised very high. 52. Water-Power. — The faU of a body of water, like that of any other substance, exercises a power, equal to its weight multiplied by the height of the fall, and if it is produced by the continuous flow of a stream, it can be utilized. To find out how much power there is at a given point, it is necessary to estabhsh, by a survey or by levelling, how many feet of fall can be obtained, and what quantity of water flows down the stream during a second or a minute. It is comparatively easy to measure the fall of water, but more difficult to determine its quantity. The latter is obtained by multiplying the number of square feet of a vertical section of the stream with the velocity, or with the number of feet through which the water advances in one minute. If a race is at hand it may be used for this measurement; otherwise a part of the stream with as 134 AVASIIING-AVATER, WATER PO^YER, STEAM BOILERS, ETC. straight banks and as even a width as can be found, should be selected. Measure of it a certain length, say 100 feet, and in several points of this length the vertical section, viz. its width and medium depth. From these several ones the medium size of a section is calculated. An instrument, resembling a wind-mill on a small scale, has been constructed for the measurement of the velocity of flowing waters, which, when set in a stream at any point, registers the number of revolutions of a fan, which enables us to obtain the speed. If such an instrument is not to be had, a piece of light wood may be used in its place ; it is simply thrown into the middle of the stream, at the point above where the mea- sured part begins, while the time ^vhich it consumes in flowing down through the 100 feet length is obseiTedwith a watch. The number of feet made by it in one second is the velocity, and gives, multiplied by the number of square feet of the medium vertical section, the number of cubic feet of Avater which pass through the stream in one second. If a precise calculation is to be made, the speed on the surface cannot be accepted as that of the whole body of water. The friction on the bottom and sides retards the motion, and must be taken into account. If we call the surface velocity per second, found with the floating wood V, the real velocity v of the stream is expressed, according to the best authorities, by the formula — 7-71 + V v = V 10-25 + V If we have, for example, a race or stream of 20 feet medium width, 3 feet medium depth, with a surface WASHING -WATER, WATER POWER, STEAM BOILERS, ETC. 135 velocity of 1 foot per second, the real velocity of the water will be : 7-71 4-1 77 1 zz 0-774 or feet, 10-25 + 1 100 and the volume of water which flows through it during one second : 77 20 X 3 X = 46-2 cubic feet. 100 One cubic foot of water weighs 621- pounds, and 46*2 cubic feet = 2887 "50 pounds. This quantity is available every second, or 60 x 2887'50 = 173,250 pounds every minute. If there be a fall of 15 feet, the power is equal to the fall of 173,250 x 15 = 2,598,750 pounds through one foot per minute. One-horse power being equal to the fall of 33,000 pounds through one foot height in one minute, the wat^r power amounts to 2,598,750 — 78t''o horses. 33,000 53. Dams. — The proper construction of a dam depends so much upon the location and the material which can be had for it, that general rules cannot be given. If possible, it should be so situated that a large body of water can be accumulated behind it, which may be drawn upon in dry seasons. Lakes, as sources of supply, are excellent natural reservoirs. If the dam and the mill are situated in a narrow valley where, in case of a flood, the water cannot spread over a large surface, but is stowed up high, the pressure some- 130 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. times becomes so strong that both dam and mill are swept away like chaff before the wind. Such sites on streams which are subjected to freshets, are dangerous, and require the construction of the best foundations for both mill and dam which human skill can devise. The head-race conveys the water from the dam to the point where it begins to act on the mechanism. It maybe an open conduit or a closed pipe, or a combination of both. Economy of power requires, that it should be as large as possible ; economy of first cost, that it should be as small as possible ; the right mean must be chosen. The entrance of the water is regulated by the head gates at the starting-point, w^iile the lower end pours it on a water-wheel or into the penstock of a turbine. 54. Water- Wheels. — The old-style water- w^heel is always vertical, while its more recent competitor, the turbine, is horizontal. Neither of them, nor any other hydraulic motor, is capable of returning the full actual powder of the water- fall. Vertical water-wheels are divided in undershot, breast, and overshot wheels, according to that part of the cir- cumference where the water strikes them. The overshot wheel returns the highest proportion of the actual power, while the results obtained with breast- wheels are less favourable the lower down on their outer circle the water is admitted. The overshot is therefore the only kind of vertical water-wheels which may compete with the turbine as a motor for paper-mills. If the overshot water-wheel could be constructed so, WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. 137 that the whole body of water would be earned by it from the highest to the lowest point of the available fall, and be there suddenly discharged, nearly the whole natural power would be reahsed, while, as it is, con- siderable losses will be suffered, some of which are here indicated. The fall is represented by the distance between the surfaces of the water in the head-race and in the tail- race. The portion of this distance between the surface in the head-race and the wheel itself acts only by impulse, but not by weight. Instead of the full outside diameter of the wheel, the distance between the point of gravity of the water in the uppermost and in the lowest bucket must be counted, as it is there where the weight of the water may be con- sidered concentrated. Over half of the depth of two buckets, or the full depth of one bucket, is thus lost. The wheel must hang free above the water in the tail- race, and the distance between wheel and water, represent- ing a small portion of the fall, is also lost. The buckets cannot be emptied suddenly at the lowest point ; they require some — be it never so little — time for it, and must therefore begin to discharge at some height above a portion of the water, the weight of which will be lost for the balance of the fall. If any water is left in the buckets on their ascent, its dead weight neutralises the same quantity of live weight on the descending side. To all the losses just enumerated must be added those from friction and contraction in the races and wheel, from leakage, and bad construction. To obtain the best possible effect, the wheel must be 138 WASH1NG-^YATEK, WATER IPOWER, STEAM BOILERS, ETC. built in such a manner that no part of it can leak or change position, and its top should be about 2^ feet below the level of the water in the race. Wood will become warped, and will rot, and is therefore inferior to iron ; but, if an all iron wheel is too expensive, the buckets alone may be of metal and the body of wood. Care must also be taken to provide an easy escape for the air contained in the buckets, through openings in the sole- I)late. A guide-board, over which the water is made to flow in the right direction and to the right spot, is used to convey it from the head-race to the wheel. The circumference of a water-wheel should, according to the best authorities, have — be its diameter large or small — a speed of not less than 4 feet and not more than 8 feet per second. A point on a large circle, if moving with the same speed as a point on a small one, requires more time to make a revolution than the latter; the larger a water-wheel is, the slower will therefore be the move- ment of its shaft. Overshot water-wheels furnish from 60 to 80 per cent. of the natural water-power, according to the amount of care and skill applied to their construction and disposi- tion. 55. Turbines. — The turbine is a horizontal water-wheel with vertical axis, consisting of a drum or annular pas- sage with a set of vanes, curved like the surface of a screw, so that the water, after having exercised its im- pulse on them, will glance ofif with as little energy as possible. While the vertical wheel is moved principally by the weight of the water, the turbine is propelled by its impulse only, and to get the best effect, the water has to WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 139 be guided so that it will strike every part of the moving vanes as nearly as possible at a right angle. The first turbine, invented by Fourneyron, was put in operation in the year 1827, at Pont-sur-l'Ognon, in France ; this Fourneyron- wheel was then, and is yet, constructed of two concentric rings, both of which are open on their vertical sides, closed on then' horizontal ones, and supplied with an equal number of vanes. The inner ring is stationary, receiving the water in the centre and acting as a guide only, while the outer one revolves and transmits the power, the water leaving at its circum- ference. Jonval made the guide-wheel and the revolving- wheel of the same diameter, and placed them, one on the top of the other, in a vertical cylinder. He was thereby able to set the turbine at any height between the head and tail-races, with equally good results, provided that the lower portion, which may be called the suction-pipe or di'aft-tube, be less than 30 feet high above the level of the tail-race. To prove this seemingly strange fact, we will take as an example a fall of 20 feet, with a turbine wheel incased in a water-tight cylinder or penstock of 20 feet height. If the wheel is placed at the lowest end, the water is forced through it with the pressure exercised by the 20 feet fall and by the pressure of the atmosphere — equal to 30 feet fall — or altogether by 50 feet. But the atmosphere has also free access to the water in the tail-race, and presses against the turbine with a force, also equal to 30 feet fall, which must be overcome. Deducting, therefore, these 30 feet from the 50 feet pressure from above, leaves only 20 feet, or the fall of the water, as active pressure. 140 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. If the same turbine is situated in the middle of the penstock, 10 feet from the surface of the water in either race, the water will be forced on to the turbine with a pressure which is equal to that of the column of water above the wheel, or to 10 feet in addition to the atmo- spheric pressure, or altogether to 10 + 30 — 40 feet. But before the atmosphere can in this case exercise any pres- sure against the wheel from the lower side, it must first overcome the column of 10 feet in the draught-tube below the wheel, and is therefore reduced to 30 less 10 or to 20 feet. The difference between the pressure above and the resistance below is therefore 40 less 20, or 20 feet, as before. The draft-tube below a turbine acts like the suction- pipe of a pump ; if air finds admittance into it, power equal to a fall of the same height is lost. There is always danger that an opening may be caused by some accident, by faults in the material, or by wear and tear ; and even the most insignificant holes, which can hardly be seen, must cause some loss of power. In most cases these pipes are located where it is very diffi- cult to examine, them, and much valuable power is often wasted before the faulty spot can be discovered. Suction- pipes or draught-tubes for turbine-wheels should there- fore be dispensed with, except in cases where for some reason the wheels cannot be set low enough to do without them. Some turbines which have within late years been much used, are remarkable especially for simplicity, and conse- quent cheapness. Every opening is provided with a gate, formed and fastened in such a manner, that it serves at the same time as a guide to the entering water. All these VVASHING-WATEK, WATER POWER, STEAM BOILERS, ETC. 141 gates can be opened and closed by one common rod, with which they are connected by levers or gearings. The water enters on the outer circle and escapes inside, but, instead of running away at once, it passes anothea* set of differently curved vanes, intended to absorb any power which may have been left in it. These horizontal wheels always run fast, and, like the vertical ones, the more so the smaller they are. They must be made of metal, because wood would take up too much space and could not be moulded into the required shape. The penstock and its enlargement, in which the wheels run, or the casing, is often built of wood ; but it is advisable to construct it of iron at all times, especially when it is exposed to the pressure of a high fall. The first expense will be larger, but the almost inevitable escape of water by leakage and the constant repairs of a wooden structure will be avoided. Turbines lose power, like overshot wheels, by friction and contraction, and because they cannot be constructed sufficiently perfect to absorb all the power. They return from 60 to 80 per cent, of the natural power ; 75 per cent, may be considered a very good result, while 80 is only obtained in exceptional cases. 56. Comparative Advantages of Overshot and Turbine Wheels.— Overshot wheels may be used when the available waterfall is such that their diameter will be reasonably large— not below 12 and not above 25 feet ; if the dia- meter would have to be beyond these limits, a turbine would be preferable. Turbines, being submerged in water, are never frozen up, and although their power will be reduced by back- 142 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. water in proportion to the diminution of the fall, they cannot be sto2)pecl by it hke vertical water-wheels. Large gearing is required to produce from the slow motion of a vertical water-wheel the high speed required by the line-shaft of a paper-mill, while one pair of com- paratively small bevel-wheels only is necessaiy with the fast-running turbine. Wherever the supply of water is either abundant or steady, the turbine will give a regular speed and good effect. But it returns the highest percentage of power only with the quantity for which it has been constructed ; and if the supply should decrease and the water fall in the race, the power produced would not only be lessened in proportion to the loss of height and volume, but the percentage obtained from the remaining waterfall would be decreased fearfully. It is therefore imperative to keep the head-race full all the time, and rather to stop and accumulate water than to use it as it comes, in inade- quate quantity. In cases where the water-supply is often insufficient, and where no very large reseiToir or pond is at hand, an overshot wheel may be preferable to a turbine, because it wiU give larger proportions of the natural power with decreasing quantities of water. 57. Steam-Boilers. — Steam-boilers form a very impor- tant part of the equipment of a paper-mill, and yet they are sometimes treated with a negligence which is crimi- nal, fi'om the danger to which every person within their reach is exposed. The consumption of fuel depends so much on the nature, construction, and treatment of the boilers, and is WASHING -\YATER, WATER POWER, STEAM BOILERS, ETC. 143 very often so heavy an item in tlie list of expenses, that too much care cannot be bestowed upon their selection and management. 58. Heating-Surface. — The gases resulting from the combustion of coal on the grate have, on starting, a temperature of about 2,400*^ Fahrenheit, and should transfer as much of this heat as is possible to the water contained in the boiler. To create a good draught in the chimney, a temperature of about 600*^ above that of the outside air is required; 1,800°, or three-quarters of all the heat created, is therefore available. The boiler must be constructed with a view to absorbing all of these 1,800°. The rapidity with which heat is transferred from one body to another is proportionate to the difference of tem- perature between them ; the gases of combustion should therefore be conducted in such a course along the boiler that this difference will be at all points as large, and as uniformly the same, as possible — an object which will be best attained, if the gases are brought in contact wdth the coldest part just before escaping into the smoke- stack, and wath the hottest part immediately after leaving the furnace. • The water being fed in at the lowest point of the boiler, it follows from the rule just given, that the gases should pass first along the upper hottest portions, descend gi'a- dually to the colder lower ones, and leave near the entrance of the feed-water. That portion of the shell which is covered with water inside and exposed to fire or hot gases outside, is the heating- surface. The capacity of a boiler for raising 144 WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. steam is directly proportionate to this heating-surface, the size of which, expressed in square feet, indicates — if the hoiler is otherwise correctly constructed and sup- phed -unth a sufficient grate-surface — its value better than a certain number of horse -power. The size of the heating surface which is to represent one-horse power has not been established by the trade, and the seller is therefore at liberty to represent the boiler as powerful as his, sometimes elastic, conscience will admit. Fifteen feet heating- surface at least should be allowed for one horse-power, although one-half of it may by forced firing be made to evaporate the same quantity of water. In calculating the power of a boiler, it is to be con- sidered, that the lower, nearly horizontal part of internal flues or tubes, owing to the difficulty with which bubbles of steam escape from under them, are found to be much less effective than the lateral and upper surfaces. To obtain therefore the real useful heating-surface, we have to deduct nearly one-third from the total one of the tubes or flues. On an average, the effective heating-surface is from I to f of the total heating- surface. It is always safer to buy a boiler of a fixed amount of heating- surface than of a number of undefined horse- power. 59. Combustion. — The useful parts of all fuel consist of the element carbon, which constitutes the solid parts, and of combinations of hydrogen and carbon in the forms of olefiant gas, pitch, tar, naphtha, &c. Both these elements, carbon and hydrogen, are, through the process of com- WASHING-WATEK, WATER POWER, STEAM BOILERS, ETC ]4'> bustion, combined in gaseous form with the oxygen of the air, and escape as carbonic acid (CO2) and water (HO). If the disengaged carbon is chilled by a cold draught or otherwise below the temperatm-e of ignition, before com- ing in contact with oxygen, it constitutes, while floating in the gas, smohe, and when deposited on solid bodies, soot. But, if the disengaged carbon is maintained at the tem- perature of ignition, and supplied with oxygen sufficient for its combustion, it burns, while floating in the inflam- mable gas, with a red, yellow, or white flame. If the boiler, or more properly its heating- surface, is small, and the firing hurried in order to produce enough steam, the combustion must be imperfect, and a loss of fuel will be the consequence. A large heating- surface may, compared with a small one, save in one year the cost of a boiler in fuel. It has been found by experiments and calculation that it takes about 24 pounds of air to furnish enough oxygen for the combustion of one pound of coal, and to dilute the gases properly. It seems evident that such a large amount of air as 24 pounds, equal to 650 cubic feet, for one pound of fael, cannot be introduced within the furnace without artificial means or draft. 60. Draught. — This draught is usually produced by a chimney, and sometimes by a fan or other blowing machine. The gases inside of a chimney are expanded by heat, and therefore lighter than those outside, and the draught is proportionate to the difl'erence in weight between the L 146 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. column of gases inside and that of an equal column or volume of air outside. The efficiency of a chimney depends, therefore, principally on the height of its crow^n above the fire-grate. Several formulae have been pro- posed by which it is to be calculated, but local experience has usually the deciding voice. It is, however, advisable to build the smoke-stack high enough to answer, not only present demands, but also increased ones which may be made in the future. Small pieces of coal which have escaped through the chimney, can frequently be found in the screens, and sometimes in the paper, but if the stack is very high, the smoke wall be carried off to a distance before its floating solid parts can I'eacli the ground — a great advantage, especially where soft or bituminous coal is used. As a rule w^hich will answer in most cases, a chimney may be built to a height of twenty-five times its inside diameter or width in the clear. The area of the width of a chimney can be made 0"16, or -^ the area of the fire-grate, if the latter is of the ordinary construction, or equal to the sum total of the area of the flues in any one place on the course of the hot gases. The hot gases cool off, contract, and consequently re- quire less room as they ascend through the chimney, and many scientific writers therefore recommend building the stacks conical or pyramidal inside and outside, that is, narrower towards the top than at the bottom. It has, however, been lately found that smoke-stacks, constructed as inverted cones, or wider at the top than at the bottom, give a better draft. This contradiction of the long-followed tlieory is in practical use on most loco- motives and in the brick chimnevs of numerous fiictories. WASHING-WATER, WATEK POWER, STEAM BOILERS, ETC. 147 and it may be stated, that such stacks are much lower than anybody would dare to build them on the contrac- tion plan, and that they generally give satisfaction. They are perfectly perpendicular outside, and are made funnel-shaped inside by means of a double wall. The outside wall may, for instance, start at the bottom with a fall brick of 9 inches, and run out at the crown with one- half brick, or 4|- inches, while the inside wall is only half a brick, or 4^- inches, thick, parallel with the outside one, connected with it at intervals by a brick or binder, but leaving a few inches, distance between the two. This inside lining is only carried up for a part of the height, and then broken off, thus leaving the top considerably wider than the lower part. The foundation of a chimney should be as solid as a rock, as it has to sustain the enormous weight of the bricks which are piled upon it. The slightest sinking of a part of the foundation may cause the top to lean over, and perhaps to fall. Access must be provided to the inside of the stack, through an iron door or through an arch near the bottom, for the removal of the ashes which will gather there in the course of time. The hot gases always carry some fine dust or coal along, and deposit them on parts of the boiler over which they pass ; the portion of the heating- surface thus covered is ineffective, and it is therefore imperative that the flues and boiler-surfaces should be frequently cleaned, and doors must be provided for that pui-pose. 61. Grate-Surface and Firing.— The ordinary rate of combustion for factory boilers is, according to Rankine. L 2 148 WASHING-WATER, ^YATER POWER, STEAM BOILERS, ETC. from 12 to 16 pounds of coal per hour on every square foot of grate-surface, the size of which may be approxi- mately determined from this. The same WTiter not only recommends a sufficiently large grate-surface, but warns us against any increase of its size beyond the prescribed limits, bacause too much air would then be admitted, and absorb a great deal of the heat without being of any use. The admission of air can, however, be well regulated by the damper, and it cannot be conceived why a large grate-surface, w^iich permits slow combustion with little draft and a light covering of fuel, should be objectionable. We have lately seen a system of boilers which con- firms this opinion. They are twdn boilers, consisting of a lower cylinder, fitted with 2 or 3-inch flues, and an upper plain cylinder. They are not walled-in in the or- dinary manner ; the lower cylinder, resting with its ends only in the two short walls, hangs free, so that the fire can play all around it. The upper cylinder rests above the lower one in the same two end walls, and the long side w^alls are arched up, so that they join it all along on both sides. This brings the upper half, or its steam room, beyond the reach of the flames, while the lower half is open to the fire, like the lower flue-cylinder, from which it is not separated by any arches or partitions. The grate occupies the w^hole space between the four walls under the boiler, and is three or four times as large as usual. The fire-doors are on the long side of the boiler, and as numerous as the room permits. When two such boilers are required, they are walled-in between the same four walls and above one common grate, without any separating walls. The grate being very wide, it is WASHING-WATER, WATEPw POWER, STEAM ROLLERS, ETC. 149 then necessary to have fire-doors on both long sides. The fire or hot gases, after having passed around the lower cylinder and below the upper one, are conducted to the outside at the end, which in ordinary boilers contains the fire-doors, and descend from there through iron conduits into the flues of the lower cylinder, through which they pass to the stack. The grate should be at a convenient height above the ground for the work of the fireman, and as the lowest part of the boiler must be at some distance above the fuel, the whole structure becomes necessarily very high. The large grate-surface is slightly covered with coal, which is tlirown in through each door in regular succession, and thus kept uniformly spread. Very little draught is required, and the combustion seems to be perfect. We have been informed that these boilers furnish in one case, with 7 tons of coal, as much steam as others, which had been previously used, would produce with 10 tons ; and that in the other the consumption of coal had been reduced by their introduction from 20 to 9 tons for exactly the same work. If a boiler is provided with only a small grate-surface, the necessary amount of combustion of fuel cannot take place without a strong draught, or in other words, the air must be forced through the coal and along the boiler with great rapidity, sometimes so fast that it can neither become well heated or thoroughly deprived of its oxygen. The necessity for a strong draught indicates, therefore, that the grate-surface is not sufficient ; and it would pro- bably result in a great saving of fuel for many boilers if their fire-places were extended and their draughts reduced. 150 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. Smoke is not only a nuisance, but also a loss of as much unburnt coal ; it escapes in thick clouds when the fire-doors are opened and fresh, especially finely divided, coal is thrown in. The cold air reduces the temperature, while the draught carries off small particles of coal un- touched. A simple way to prevent this loss, to some extent, which is applicable to any common steam-boiler, consists in the division of the grate into two separate parts by a brick wall in the middle parallel with the grate-bars. Each division has a separate door, and while the heat is greatest in one of them, fresh coal is thrown into the other ; the unconsumed coal v/hich is carried off by the draught mingles with the flames from the hot division, and is thus burnt up. The sudden changes of tempera- ture which are so injurious to the boilers, caused by the opening of the fire-door, are thereby also in a measure prevented. A constant stream of fresh air is often conducted into the fire-hearth through channels in tlie brickwork, where- in the air on its passage is partly heated. Ordinary grates are composed of straight, narrow cast- iron bars, laid alongside of one another, and leaving, for a coal-fire, openings enough between them to amount altogether to one-fourth of the area of the grate- surface. These bars when heated expand, and become frequently warped into such crooked forms, that they have to be re- placed by new ones. Many patent grate-bars have been constructed with a view to prevent any change of their outer form, or of the open space for the admittance of air, by providing room for expansion and contraction in the bars themselves. Some of them are quite successful. WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. 151 and not only save coal, but also last much longer than the common bars. It is a mistake to suppose that the sprinkling of water over coal will improve combustion ; the water must be evaporated and transformed into steam, absorbing thus a great deal of heat at the expense of the fuel. The coal must be spread on the grates uniformly, and not too thickly, and if stirred at all, it is to be done from below. A good fireman can economise more than all the in- ventions which have been made for this purpose are able to do, while a careless or ignorant one may waste many times the amount of his wages in fuel. Several years ago prizes were offered to such firemen as should prove themselves most efficient at a com- petitive trial, at Miihlhausen, then in France, now be- longing to Germany. Fourty-four offered themselves, and the best eighteen were selected from the number ; each one fired up during ten hours with the same boiler, fuel, &c. and it was found that the best fireman could evaporate nearly twice as much water with the same amount of coal under exactly the same circum- stances as the worst one. To appreciate this result, it must be considered that only experienced men offered themselves for the trial, and that only one-half of these were admitted. 62. Construction of Steam Boilers and Test.— The de- signs for the construction of steam-boilers are so numer- ous that a description cannot be attempted in this work, but a few words may be said which will apply to them all. It has been found by experience that a thickness of 152 ^VASHING-^VATER, WATER POWER, STEAM BOILERS, ETC I inch is the most favourable to sound riveting and caulking of boiler-plates, and they are therefore seldom used much thicker or thinner. If a cylindrical boiler is required to endure an un- usually high pressure, the necessary increase of strength must be attained, not by increased thickness of the plates, but by diminution of the diameter. The flat ends of cylindrical boilers are given about one and a half the thickness of the cylindrical portions ; cast-iron should not be used for them nor for any other part of the shell. These flat ends are usually connected with each other by longitudinal stays, and sometimes with the cylinder by means of angle iron, but such rings are liable to split at the angles, and it is therefore preferable to bend the edges of the flat ends and rivet them to the cylinders. Plates which overlap one another should have the overlapping joints facing upwards, on the side next to the water, that they may not intercept bubbles of steam on their way upwards. The joints in horizontal flues should be placed so that they do not oppose the current of the gases. Those parts of boilers which are exposed to more severe or more irregular strains than the rest, or to a more intense heat, should be made of the finest iron. Lately there have been some steam-boilers built of steel plates. As they have about one -half more tenacity than iron ones, they can be made lighter, and it is not unlikely that steel boilers will take the place of iron ones at some future day. In paper-mills the demands for steam are often so sudden and large, that some of the water will be carried WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. 153 along mechanically through the violence of the motion, unless a large storeroom for both water and steam, but especially for the latter, is provided. A steady pressure is particularly required for the paper-machine, and large boilers, wdth plenty of steam room, will be more apt to furnish it than small ones. Every boiler, before being put into operation or walled - in, should be tested with a hydraulic pressure twice as great as the highest steam-pressure allow^ed to it. Water-pressure is used on account of the absence of danger, in case any part of the boiler should give way. 63. Feed-Water. — If possible, the boilers should be fed with hot water ; not only because as much heat as it contains is directly saved, but also for the reason that the injection of cold water chills and may injure the hot plates. There are plenty of sources in a paper-mill from which hot water may be obtained, but the dryers of the paper-machine are the principal ones. The con- densed steam is conducted through a pipe into a reser- voir or tub, and if too hot to be pumped, it is therein mixed with fresh water and forced into the boiler by the feed-water. These pumps refuse to work with highly heated or boiling water, and large quantities of escaped steam from steam-engines, by which the temperature of the feed-water might be raised to the boihng-point, are often allowed to blow out into the open air, because they would heat the water too much. Such steam can, however, be utilised by being con- ducted through long coils of pipe, fastened in an upright steam tight cylinder of boiler iron, interposed between 154 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. the pump and the boiler. The feed-pump forces the hot water first into this cyHnder or heater, where it acquires a very high temperature in contact with the steam coil, and thence through the usual check- valve into the boiler. The feed- water does not pass through any pump after it has left the heater, and can therefore be raised to any temperature. It is of the greatest importance that the water in the boiler should be kept at the right height all the time, and since the regular feed-pump may get out of order, there should be a second one, or some other means provided, by which the supply can be kept up in such a case. Giffard's injector or steam-pumps may be recommended for this purpose on account of their independence of any gearings or motors, but as all of them consume steam, and some wull not feed with hot water, the regular feed- pump is always to be used in preference. A valve must be provided at the lowest point of every boiler, through which it can be emptied ; this valve or blow-off cock must be frequently opened to enable the deposits of salts or mineral matters to escape before they have had time to solidify. According to the quality of the feed-water this has to be done several times, or only once a day, and the boilers should in all cases be emptied entirely at regular intervals. Some substances occurring in feed-water seem to stick so closely to iron that they cannot be removed by blowing ofi water ; they must be chemically dissolved, and numer- ous powders and liquids are sold at high prices for this purpose, but the introduction of one or more gallons of common coal oil or petroleum will answer in many cases just as well or better. WASHING -WATER, WATER POWER, STEAM BOILERS, ETC. 155 64. Explosions. — We quote here from remarks made on this subject by W. T. Macquorn Rankine, Professor of Engineering and Mechanics in the University of Glasgow, in his Manual of the Steam-Engine and other Prime Movers — a work which has furnished much data for this chapter : — " Explosions result : *' I. From oiiginal iveahiess. This cause is to be ob- viated by due attention to the laws of the strength of materials in the designing and construction of the boiler, and by testing it properly before it is subjected to steam- pressure. " II. From u'eahiess prjduced hij gradual corrosion of the material of which the boiler is made. This is to be obviated by frequent and careful inspection of the boiler, and especially of the parts exposed to the direct action of the fire. " III. From wilful or accidental obstruction or overload- ing of ike safety-valve. This is to be obviated by so con- structing safety-valves as to be incapable of accidental obstruction, and by placing at least one safety-valve on each boiler beyond the control of the fireman. " IV. From the sudden production of steam of a pressure greater than the boiler can bear, in a quantity greater than the safety-valve can discharge. There is much difi'erence of opinion as to some points of detail in the manner in which this phenomenon is produced, but there can be no doubt that its primary causes are, first, the overheating of a portion of the plates of the boiler (being in most cases that portion called the ' crown of the furnace,' which is directly over the fire), so that a store of heat is accumulated ; and, secondly, the sudden contact of much 150 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. overheated plates with water, so that the heat stored is suddenly expended in the production of a large quantity of steam at a high pressure. " Some engineers hold that no portion of the plates can thus become overheated, unless the level of the sur- face of the water sinks so low as to leave that portion of the plates above it uncovered ; others maintain with Mr. Boutigny, that when a metallic surface is heated above a certain elevated temperature, water is prevented from actually touching it, either by a direct repulsion or by a film or layer of very dense vapour, and that when this has once taken place, the plate being left dry may go on accumulating heat and rising in temperature for an in- definite time, until some agitation or the introduction of cold water shall produce contact between the water and the plate, and bring about an explosion. All authorities, however, are agreed that explosions of this class are to be prevented by the following means : — 1. " By avoiding the forcing of the fires, which makes the boiler produce steam faster than the rate suited to its size and surface. 2. *' By the regular, constant, and sufficient supply of feed-water, whether regulated by a self- acting apparatus or the attention of the engine- man to the water-gauge ; and 3. " Should the plates have become actually overheated, by abstaining from the sudden intro- duction of feed-water (which would inevitably produce an explosion) and by drawing or extin- guishing the fires, and blowing off both the steam and the water from the boiler." To overcome this difficulty several inventors have con- WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 157 structed boilers which are combinations of cylinder-boilers with systems of wrought-iron tubes. But whatever device may be used, and however excellent it may be in other respects, a steam generator cannot be considered a safety-boiler, if fire or hot gases are allowed to come in contact with a cylinder of more than 4 inches diameter, which contains water and forms a part of the boiler. 65. Steam-Engines. Expansion. — The steam-engine has been much improved since its invention about 100 years ago, but the fundamental principles governing its con- struction are yet the same as laid down by Watt. The power of the steam-engine is derived from the alternate action of the steam upon the two sides of a piston, which is thus moved from one end of a cylinder to the other, the reciprocating motion being changed into a rotary one by means of a crank. We suppose the piston of a high pressure engine, for an example, to have arrived at one end of its course, and to be on the point of starting to return, forced by steam admitted into the narrow ^pace behind it, while the empty cylinder in front communicates with the open air. If the steam is of 60 pounds, equal to 4 atmospheres over-pressure, its real pressure will be 75 pounds, or 5 atmospheres, which, being opposed by one atmosphere, or 15 pounds only on the other side, pushes the piston forward with 60 pounds to every square inch of its surface. If fresh steam is admitted constantly during the whole course of the piston, the largest amount of power of which the cylinder is capable will be produced, but the steam leaves the engine with nearly its full pressure. 158 WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. If we take, for a second example, a cylinder of tlie same diameter, but of twice its length, and admit only the same amount of steam for each stroke, as in the first example, we find that, after the fresh steam has been shut off, the piston is yet moved forward to the other end by the steam, which filled one-half of the cylinder. This second half of the movement of the piston is produced by expansion, and through it, if extended far enough, can the pressure of the steam be utilised and reduced until it is nearly equal to that of the atmosphere, or 15 pounds. It is evident that all the power produced by expansion in the second example, with the same quantity of steam as was used for the first example, is clear gain as com- pared with the latter, and though this is not exactly so in practice, it yet explains the economy of expansion. These examj^les show also, that a larger C3-linder is capable of producing the same power with less steam than a small one, or that engines of ample capacity are the most economical, if provided with proper arrange- ments for expansion. 66. Condensation. — Another way of increasing the powder of an engine is to reduce the counter-pressure by the creation of a vacuum. The steam, instead of escaping into the air, is for this purpose conducted into an apparatus, wdiere it is suddenly condensed by contact with cold water finely divided by a sprinkler. The water used for this condensation is there- by highly heated, the counter-pressure reduced consider- ably below^ 15 pounds, and the power exercised on the piston increased as much. WASHING-WATER, WATER POWER, STEAM BOILERS, ETC. 159 The use of these condensers thus enables an engine to work with very low steam-pressure, and reduces the danger to which high-pressure in the boilers exposes it. 67. Different Systems of Engines, and Utilisation of Escaping Steam. — All engines may be divided into — 1. Non-condensing or high-pressure engines ; 2. Condensing engines. High-pressure engines are simple in construction, easily managed, and therefore generally used whenever steam is only an auxiliary power, to be stopped and started according to the state of the water-power. The valve which admits the steam is usually regulated by a gover- nor, which is set in motion by the line-shaft. If the shaft turns too fast or too slow, the governor closes or opens the valve, letting in less or more steam, and in- creasing or decreasing the expansion. The escaping steam should never be allowed to waste directly into the air ; it can be made useful by passing through a coil or other system of pipes immersed in water, or it may be conducted through large pipes, and heat the building, or into the mixing-pans to boil the liquors. In some mills, which run by steam-power altogether, the escaping steam is thoroughly used up for boiling waste-paper in tubs, and by heating the mill and the feed-water for the boilers. One mill, which from waste- paper and with steam-power only, produces 5,000 pounds of good printing paper per day, economises so well that direct steam is not used anywhere except in the steam- engines. It has been stated that the estabhshment con- sumes only 1,500 pounds of coal per day over and above 160 WASHING- WATER, WATER POWER, STEAM BOILERS, ETC. the amount which would be required if the power were furnished by water. The dryers form a natural condenser for the high- pressure engines which are used for driving paper- machines. Condensing engines require large quantities of water, are complicated and expensive, but they furnish more power with the same amount of fuel than any other kind. This is especially the case when high-pressure steam first acts in a small cylinder, from which it passes into a larger one, where it propels the piston by expansion and condensation. The escaping steam can, however, be so well utihsed in many paper-mills that the simpler high-pressure engines answer often as well as more complicated ones. It is indifferent whether an upright or horizontal engine is selected, provided it be a good one, fastened on a solid foundation. 68. Power of Engines. — The statement that a steam- engine gives a certain amount of horse-power must be made in connection with that of the dimensions of the cylinder, steam-pressure, speed, expansion, (tc. if it is to be of any value. A steam-engine will give nearly double power if its speed is doubled. Fast-running engines not only wear out soon, but get more easily heated and out of order : and while it is the seller's interest to speed them high and represent them more powerful, the purchaser's is just the reverse. Steam- engines should therefore be purchased according to their size, but not by the horse-power. The power of an engine is the total mean pressure on HENRY WATSON, HIGH BRIDGE WOREH NEWCASTLE-UPON-TYNE. GENERAL MECHANICIAN. iiii ■iiiiisfiiii if ill ■PMiii liiffii piitii^ Also of improved Revolving andJogging Strainers in Vats complete, DOCTOR PLATES, BRASS AND COPPER ROLLS, HYDRAULIC PRESSES AND PUMPS, Gun Metal Cocks, Valves, Water and Steam Gauges, Hydraulic Rams, &c. H. W. begs to intimate that to meet the increasing demand for his improved Strainer Plates he has just completed extensive additions to his Premises and Machinery, and is now in a position to execute all orders promptly, and on the most reasonable terms. The strictest attention will always be given to maintain in the highest degree the quality of the material and workmanship, keep- ing in view also the capability of re-closing after having been worn. Rag Chopper, Wheel, 4 ft. diam., I ft. wide. Finishing Calender. These rolls are made of refined Chilled Roll Me'als. BENTLEY & JACKSON, ENGINEEKS, lEONFOUNDERS, AND MACHINISTS, BURY, NEAR MANCHESTER. • Makers of Paper Making Machinery, Millboard Machines, Paper Cutting Machines, Ripping and Winding Machines, for preparing paper for continuous printing presses. Hydraulic Pumps and Presses. Steam Engines and Boilers. ESTIMATES ON APPLICATION. IMPOKTANT TEADE PUBLICATIONS. Thirteenth Edition. Demy 8vo,, 104 pages. Stiff Covers, Price 2s. 6d., or Post free for 30 Stamps, THE PAPER MILLS DIRECTORY. Price 2s. 6d., or Post free for Thirty Stamps, THE PAPER STAINERS' DIRECTORY OF GREAT BRITAIN. The Seventh, a New, Corrected, and Enlarged Edition, Foolscap 8vo., price 3s. 6d., or Post free for 30 Stamps, THE STATIONERS' HANDBOOK, AND GUIDE TO THE PAPER TRADE. Sixth Edition, Demy 8vo., 60 pages, Still Covers. Price 2s. 6d., or Post free for 30 Stamps, THE CHEMICAL MANUFACTURER S DIRECTORY. Sold by KENT & CO., Paternoster llow. Price 2s. Gd., or sent Post free for 30 Stamps, A MAP OF THE PAPER MILLS OF ENGLAND. Arranged by the Editor of " The Paper Mills Directory." THE EDITOR, at Oxford Court, Cannon Street, London, E.C yZ^jo- J i(r^ ^,.^^/A/.^^> .o^^ I S \ Washing 'WATER, water power, steam boilers, etc. 161 the surface of the piston, less the pressure against it in pounds, multiplied with the velocity of the piston in feet per minute. This product must be divided by 33,000 if the theoretical horse-power is to be obtained ; and from it we have to deduct for condensing engines 25 per cent, and for high-pressure engines 13-1 per cent, loss from friction and pumps, in order to find the actual horse- power. While the steam is expanding in the cylinder, its pressure decreases constantly, and to make an exact calculation its mean pressure must be determined. If we have, for example, a high-pressure engine of 15 inches diameter of piston, 3 feet stroke, 60 pounds of steam, 50 revolutions per minute, and an expansion of ^, the steam acts on a piston surface of — 15 X 15 X 3-14 . , J = 176-6 square inches, and the mean pressure on the piston for a steam-pressure of 60 pounds above the atmosphere, or for 60 + 15 = 75 pounds, is 63 '487 pounds. The atmospheric counter- pressure, which is equal to 15 pounds to the square inch, must be deducted from these 63*487, and leaves 48'487 pounds as the available pressure. The theoretical power is therefore — Square Pressure inches per Diston square surface. » mch. Strokes Feet tfon^Mr