Aberthaw

Aberthaw Druid Brand cement logo
Aberthaw "Druid" brand.

Early Aberthaw cement logo
Early Aberthaw logo.

Later Aberthaw cement logo
1970s Aberthaw logo.

Location:

Clinker manufacture operational: 5/1914 to date

Approximate clinker production to 2022: 36 million tonnes (12th)

Raw materials:

Ownership:

As originally built, the plant represented the expansionist economy of South Wales in the period. A long series of articles in The Engineer in 1915 described it as a state-of-art plant, using home-grown equipment. Read these articles. The long period of quiescence after its initial expansion reflected the dramatic contraction of the post WWI economy in the area. Thus, although initially an aggressive and important venture, it could be described by Cook in the 1950s as small and backward. The nearby Rhoose plant was acquired in 1919, and no new kilns were installed between Rhoose A2 in 1923 and Aberthaw A4 in 1958, by which time the existing plant was obsolete. However, Aberthaw's large reserves of raw material suitable for dry process ensured its continuance. The Blue Lias here has a limestone:shale ratio of 75:25. Initially the limestone alone was used, removing the shale by rumbling. As higher C3S clinker became necessary, Ruthin quarry supplied limestone (CaCO3 92-96%) from 1939 to sweeten the mix. In 1947 Aberthaw and Blue Circle purchased the quarry as a joint venture. Subsequently, rumbling was abandoned and the entire un-sorted Lias was used.

The conversion to dry process, in advance of the other smaller companies, was encouraged by Blue Circle, which by then had a substantial minority holding in the company, and Penarth was closed to make way for the additional capacity. Kiln A5 was the third suspension preheater kiln installed in Britain (after Plymstock A1 and A2), and the 12th efficient dry process kiln. Installed in 1967, kiln A5 was briefly the largest in Britain, until overtaken by Cookstown A1 the following year. Although the plant was capable of shipping product from the port of Aberthaw, transport was from the outset by rail, in line with its relationship with the South Wales coalfield. Among British plants, only Rugby is older.

Please contact me with any relevant information or corrections. I am particularly interested in firmer dates and statistics.

Note: technical information on currently operational plants is ✄withheld in the public version of the site at present, except where already published (see references).

Power Supply

The plant was electrically powered from the outset, using its own turbo-generators. A larger power plant was installed at Rhoose which provided top-up capacity until 1939, when both plants went on grid power. The Aberthaw power plant was decommissioned at that stage, but the Rhoose power plant continued as back-up.

Rawmills

Six rotary kilns were installed:

Kiln A1

Supplier: Ernest Newell
Operated: 18/5/1914-1967
Process: Wet: a Polysius slurry dryer was added in 1937.
Location: Hot end 303216,167399: Cold end 303212,167460: originally only the hot end was enclosed, but from 1937 the whole kiln was enclosed.
Dimensions: 200’0”× 9’10½”B / 8’10½”CD (metric 60.96×3.010/2.705)
Rotation (viewed from firing end): clockwise
Slope: 1/25 (2.292°)
Speed: 0.75-1.5 rpm
Drive: 37 kW
Kiln profile: 0×2705: 6096×2705: 6096×3010: 16764×3010: 18288×2705: 60960×2705: tyres at 4267, 20117, 31699, 41453, 55169: turning gear at 29466.
Cooler: rotary 80’0” × 7'5"/6’5” (metric 24.38×2.261/1.956) beneath kiln
Cooler profile: 0×2261: 5486×2261: 6401×1956: 24384×1956: tyres at 4115, 15240: turning gear at 7925.
Fuel: 1914-1963 Coal: 1963-1967 Oil
Coal Mill: indirect: ball and tube mills: two sets common to kilns 1 and 2
Exhaust: initially via "wet bottom" drop-out chamber direct to stack. ID fan added after slurry dryer in 1937, with a cyclone. The cyclone was replaced with an electrostatic precipitator in 1946.
Typical Output: 180 t/d pre-1937, 199 t/d post-1937
Typical Heat Consumption: 1914-1937 7.6 MJ/kg: 1937-1950 6.90 MJ/kg: 1950-1956 6.70 MJ/kg: 1957-1963 6.88 MJ/kg: 1963-1967 7.2 MJ/kg


Kiln A2

Operated: 4/5/1914-1967
Location: Hot end 303224,167399: Cold end 303221,167460: originally only the hot end was enclosed, but from 1937 the whole kiln was enclosed.
Rotation (viewed from firing end): anticlockwise
Exhaust: initially via "wet bottom" drop-out chamber direct to stack. ID fan added after slurry dryer in 1937, with a cyclone. The cyclone was replaced with an electrostatic precipitator in 1947.
Identical in all respects to A1.


Kiln A3

Supplier: Edgar Allen?
Operated: 1917-1969, 1972-1975
Process: Wet
Location: Hot end 303235,167400: Cold end 303231,167465: originally only the hot end was enclosed, but from 1937 the whole kiln was enclosed.
Dimensions: Initially 200'× 10’0”B / 9’0”CD: extended 1938 with Desiccator to 216’6½"× 10’0”B / 9’0”C / 13’0”D (metric 66.00×3.048/2.743/3.962)
Rotation (viewed from firing end): anticlockwise?
Slope: 1/25 (2.292°)
Speed: 0.75-1.5 rpm
Drive: 37 kW
Kiln profile:

Cooler: rotary 60’0” × 6'0" (metric 18.28×1.828) beneath kiln
Cooler profile:
Fuel: 1917-1963 Coal, direct: 1963-1969 Oil: 1972-1975 Gas
Coal Mill: indirect: ball and tube mill
Exhaust: initially via "wet bottom" drop-out chamber direct to stack. ID fan added 1938. Fan exhaust ducted through common electrostatic precipitator with kiln 4 from 1972.
Typical Output: 195 t/d pre-1938, 215 t/d post-1938
Typical Heat Consumption: 1917-1938 7.6 MJ/kg: 1938-1950 6.90 MJ/kg: 1950-1956 6.70 MJ/kg: 1957-1963 6.88 MJ/kg: 1963-1969 7.2 MJ/kg: 1972-1975 7.5 MJ/kg


Kiln A4

Supplier: Vickers Armstrong
Operated: 1958-1969, ?1/1972-1975
Process: Wet
Location: Hot end 303248,167391: Cold end 303244,167466: mostly unenclosed
Dimensions (from cooler ports): 246’1"× 9’0¼”BC / 9’10½D (metric 75.01×2.750/3.010)
Rotation (viewed from firing end): anticlockwise?
Slope: ?
Speed: ?
Drive: ?
Kiln profile (from cooler ports): no information
Cooler: Reflex “Recuperator” planetary: 12 tubes
Fuel: 1958-1963 Coal: 1963-1969 Oil: 1972-1975 Gas
Coal mill: direct, Atritor
Exhaust: via ID fan direct to stack. Fan exhaust ducted through common electrostatic precipitator with kiln 3 from 1972.
Typical Output: 221 t/d
Typical Heat Consumption: 1958-1963 6.75 MJ/kg: 1963-1969 7.0 MJ/kg: 1972-1975 7.3 MJ/kg


Kiln A5

Supplier: Polysius
Operated: 1967-?4/1987
Process: Dry: Dopol 4-stage suspension preheater
Location: Hot end 303123,167385: Cold end 303120,167449: entirely enclosed
Dimensions: Metric 64.00×4.200
Rotation (viewed from firing end): anticlockwise?
Slope: ?
Speed: 1.36 rpm
Drive: ?
Kiln profile: 0×3800: 900×3800: 1800×4200: 63500×4200: 64000×3600: tyres at 10000, 30000, 53000: turning gear at 49000.
Cooler: Fuller 850S grate: 5° slope.
Fuel: 1967-1972 Oil: 1972-1980 Gas: 1980-1987 Coal plus petcoke from 1986.
Coal mill: direct, Babcock 7E? ball-race mill
Exhaust: via rawmill and electrostatic precipitator to stack.
Typical Output: 1967-1980 1160 t/d: 1980-1987 1230 t/d
Typical Heat Consumption: 1967-1980 3.56 MJ/kg: 1980-1987 3.92 MJ/kg


Kiln A6

Supplier: KHD
Operated: 5/4/1975 to date
Process: Dry: KHD 4-stage suspension preheater: stage 1, 2×3.175Ø: stages 2-4, 5.4Ø(2), later supplemented with preheater fuel injection.
Location: hot end (cooler ports) 303102,167394: cold end 303089,167458: unenclosed
Dimensions: Metric 65.00×4.200(2)
Rotation (viewed from firing end): anti-clockwise
Slope: ?
Speed: 2.2 rpm
Drive: ?
Kiln profile: ✄
Cooler: planetary: 10×17.50×1.750(2)
Fuel: 1975-1980 Gas: 1980 to date Coal, from 1986 replaced with up to 35% petcoke(1), supplemented by meat and bone meal from 2004, liquid wastes from 2008, and tyres from 2012.
Coal mill: ✄
Exhaust: via rawmill and electrostatic precipitator to stack(1). In 2003, precipitator replaced by bag filter(3).
Typical Output: 1975-1987 1260 t/d: 1988-2000 1520 t/d: 2016 given as 1720 t/d(3)
Typical Heat Consumption: ✄



References:

Sources::

The following is a transcript of six anonymous articles that appeared in The Engineer, CXIX, 23 April 1915, pp 398-400; 30 April 1915, pp 424-426 & supplement; 7 May 1915, pp 448-449; 14 May 1915, pp 478-480; 21 May 1915, pp 497-500; 4 June 1915, pp 552-554 & 556.

Aberthaw was the last cement plant to be constructed in the Vale of Glamorgan, and represented an aggressive independent development, with two of the largest kilns of the time.

Unlike most producers, the new company advertised its product widely in the mainstream press, and the Engineer articles were part of a publicity campaign promoted by the manager, William Alden Brown. The main suppliers, Ernest Newell, also used the opportunity to promote their products at length. The articles were part of a series of sixteen articles in The Engineer intended to promote British plant suppliers, German firms having supplied most of the equipment on rotary cement plants in the first decade of the century.

In addition to various small plants, South Wales had seen the installation of three modern rotary kiln plants in the period before WWI, of which Aberthaw was the last. The condition of the local economy is indicated by the figure to the left, which shows the population of the steam coal producing areas of the South Wales Coalfield. Welsh steam coal had been the high-calorific-value fuel of choice for railways and shipping, and a huge economy, largely based on exports, grew up. However, after WWI, Lorraine was returned to France, eliminating a major export market, and with the opening up of the Middle East, the world energy market re-aligned on oil. Thus the pre-war bonanza led to massive development of South Wales, with the building of new ports and a maze of rail links: Cardiff was Britain's second port by tonnage in 1914. But further development ceased altogether after 1918, and did not resume (and then slowly) until the 1950s.

Values of imperial units (as of 1929) used in the text (alphabetical order): 1 acre = 0.40468424 Ha: 1 ft = 0.30479947 m: 1 gallon = 4.5460756 dm3: 1 HP (horse-power) = 0.7456998 kW: 1 inch = 25.399956 mm: 1 psi (pound-force per square inch) = 6.89478 kPa: 1 ton = 1.01604684 tonne: 1 yard = 0.91439841 m.

CEMENT WORKS AT ABERTHAW.

Among the other firms which are prepared to equip a Portland cement works from start to finish is that of Ernest Newell and Co., Limited, of Misterton, near Gainsborough, the works of which we described in our issue of November 20th last. This company was, we believe, primarily formed with the express object of making cement plant, at any rate its output has always consisted largely of this type of machinery (Note 1). We cannot do better, in order to convey an impression of the scope of the firm's operations, than describe a complete Portland cement-making plant erected by it, and fortunately the opportunity presents itself, for the cement-making machinery installed throughout the recently completed works at Aberthaw of the Aberthaw and Bristol Channel Portland Cement Company, Limited, was constructed and supplied entirely by the Misterton firm. By the courtesy of the directors of the Cement Company we have been permitted to pay a lengthy visit of inspection to these works, and had the advantage of meeting on the site Mr. B. J. Day, M.I.Mech.E., the company's consulting engineer, to whose specification and under whose direction the whole of the plant was installed. Moreover, we were fortunate in having the personal assistance during our investigations of Mr. W. A. Brown, the company's general manager. We are therefore in an excellent position to give an accurate description of the works, the machinery of which, as far, at any rate, as concerns that part of it which is devoted to the actual manufacture of the cement, is, as we have said, British from beginning to end. It is true that in our description we shall have to make mention of certain other machinery which is of foreign origin, but nearly the whole of it is in connection with motive power, and from an engineering point of view there is no reason whatever why it should not have been made in this country. The Aberthaw works, therefore, offer an excellent opportunity of demonstrating the class of work which Messrs. Newell's are prepared to carry out, and we may here remark that they contain among the best, if indeed not actually the best, plants of their size and type which exist in this country - or anywhere else for that matter.

In our description it will be convenient first of all to refer to the works as a whole, and after that to deal more particularly with the various machines employed in them. To begin with, a few words may well be said regarding the raw material. The works have been erected on part of a vast deposit of lower blue lias limestone which, starting from near Cardiff, stretches westward and south-westward for many miles. Portland cement has been made from it at Penarth for a number of years, but it is a noteworthy fact that the deposit changes in composition as it goes westward. At Cardiff and Penarth it is low in silica and high in alumina, while, on the other hand, at Aberthaw it is highly siliceous. It is interstratified with beds of marl and the two materials together supply all the ingredients necessary for the production of Portland cement of the highest quality. It is a curious coincidence that the limestone is very similar in composition to the cement rock of the Lehigh Valley in the United States, where cement was first made in North America and where, at the present time, about one-half the total output of cement of the United States is manufactured. The resemblance between the two materials is shown in the accompanying table (Note 2):-

Comparison of the Raw Materials in the Lehigh Valley and at Aberthaw.
ConstituentsLehigh Valley Cement RockAberthaw
LimestoneShale or marl
Silica, SiO211.0311.0340.34
Alumina, Al2O34.421.4711.14
Oxide of iron, Fe2O31.70
Lime, CaO44.1847.8423.06
Magnesia, MgO1.180.621.42
Carbonic Anhydride, CO2, and combined water, H2O----39.0422.26
Alkalis and loss--------1.78
Calculated as carbonate of lime, CaCO379.0082.6041.18

We also give the general composition of the Aberthaw marl or shale, and it will be realised how that by a proper adjustment of the proportions the correct mixture for Portland cement manufacture may be arrived at. In fact, at Aberthaw the raw material is an argillaceous limestone having a composition which is practically ideal for making high-class Portland cement. A quarry is in course of being opened out immediately adjoining the works. The deposit is exceedingly hard and by no means easy to quarry by hand labour. Indeed, had this course been pursued the cost would have been prohibitive, and the very wise decision to adopt a steam digger was come to (Note 3). The machine in operation was supplied by Ruston Proctor and Co., Limited, of Lincoln, who have devoted their attention for years to this class of work, and it is an excellent example of mechanical construction. We have already described it in our issue of April 9th (Note 4), but we may here repeat that the machine has a digging capacity of 100 tons per hour or more, and that it has been in daily operation for months with but little cost for repairs. It says not a little for the quality of the material and workmanship entering into its construction that during the whole of the time it has been at work it has not been necessary to renew any of its coiled suspension springs nor any of the links of its various chains, and that the cost for repairs has been practically negligible.

When first put to work this machine excavated the rock itself but latterly a great improvement has been made and the work of excavation rendered much more easy by the introduction of blasting operations. Vertical holes are driven into the rock from the surface near the face of the excavation by means of a specially designed drop drill which was supplied by the Cyclone Drill Company of Orrville, United States of America. Blasting charges are introduced into a series of these holes, which are taken down some 30ft. to 40ft., and fired, with the result that the rock is not only loosened and made to fall forward in the direction of the excavator, but it is broken up so that the digger has little more to do than pick up the stones. To this method of quarrying we hope to return at a later date but meanwhile we may say that the cost of drilling the holes and blasting amounts to about 4.835d. per ton of cement made, and that the total cost of blasting, excavating the material and delivering it to the crushers, including wear and tear of trucks, locomotive and permanent way, amounts to 1s. 1.14d. per ton of cement manufactured (Note 5).

The choice of a proper site for the manufacture of Portland cement is a matter calling for mature consideration. It is not enough that the raw materials of the desired composition and quantity should be ready to hand, though naturally this is a most important point. There are certainly two other essentials without which it would be difficult to compete on anything like equal terms with rival firms which possessed them. One is the facility for getting coal or other fuel, and the other is the facility for dispatching the cement when it is once made. In all three respects the works at Aberthaw may be said to be most fortunately situated. Raw material is, as we have shown, present in abundance and of ideal composition; the works are close to coal fields (Note 6) producing fuel which is suitable for use in rotary kilns; two lines of railway - the Vale of Glamorgan and the Taff Vale - pass in immediate proximity to the site, and to both of them there is direct connection by sidings, and finally the harbour at Aberthaw, which is on the Bristol Channel, provides means for the ready dispatch of cement by water (Note 7). Nor do the foregoing exhaust the advantages enjoyed by the site, for an abundant supply of water is afforded by the Rivers Kenson and Thaw, the former of which actually passes alongside the power house and flows into the latter only a short distance further on (Note 8). The position chosen could, therefore, hardly be improved upon, and when it is said that the works have been laid out with great care so that the raw materials as they pass through them are subjected to the minimum of handling, and the machinery installed is of the highest standard throughout, it can be well understood that the company is in a position to manufacture cement of the highest quality at a low cost.

The raw material as it comes from the quarry face is very nearly accurately proportioned for cement making. There is, however, at present a slight excess of clayey matter or marl and this is removed in the following manner: The stone and marl are brought from the quarry face in 6½-yard side-tip wagons running on standard gauge lines and drawn by a tank locomotive of the contractor's type. The latter hauls them up a short incline to a feeding platform arranged above two stone crushers. The wagons are discharged on to a fan-shaped grid formed of steel rails arranged so that they radiate up, from the entrance to the feeding shoot, to the level of the railway. The distance between each rail of the grid is therefore, greater at the top than at the bottom, and a portion of the surplus marl falls through the wedge-shaped spaces formed between the rails and is taken away in trucks. This marl is being employed to raise the level of an area of land adjoining the site of the works but on the far side of the River Kenson.

Figures 51 and 52

The plan and section shown in Figs. 51 and 52 will help to make the following description clear. We are indebted for the drawings from which these engravings were made to the courtesy of Mr. Day who, as we have said above, designed and laid out the whole works. We may mention here that in the course of this article it is proposed only to give the names of the makers of machines not made by Messrs. Newells. If no name be mentioned it is to be taken that the machines were supplied by the Misterton firm.

There are, as has been said, two stone breakers. These are of the jaw type with a feed opening 36in. by 22in., and they are designed to deal with 80 tons of stone per hour, and to break it up to a size sufficiently small to feed into ball mills. The crushers are both driven from a countershaft, which is in its turn driven by a 100 horse-power motor running on a 3000/3300-volt three-phase 50-period current at 580 revolutions per minute. This motor was supplied by the Allmänna Svenska Elektriska Aktiebolaget (Note 9). The crushed material as it is discharged from the jaw breakers is delivered on to two conveyor belts, one for each machine, and is raised to the top of a stone store which is divided into six compartments, into any one of which the broken raw materials can be delivered by means of delivery carriages (Note 10) which run on rails the length of the building. The stone store is designed to hold 1000 tons. Each compartment is provided at the bottom with a delivery shoot which can be opened and closed so that stone may be drawn from any of the compartments at will, this permitting of variation of the mixture as required (Note 11). The delivery from these shoots is on to a horizontal conveyor, which in its turn delivers the stone and with it, of course, a certain amount of marl, on to a tray conveyor elevator running up to the top of a two-storey building which contains the grinding machinery.

The grinding of the stone is carried out in two stages, first in ball mills and then in tube mills. There are four of each type of mill, the ball mills being arranged on the upper floor and the tube mills on the lower floor. We shall have occasion to refer more particularly to the form of the jaw breakers and these tube mills, as also to that of the remainder of the machinery later on, and for the present we will only consider the passage of the raw material through the works until it becomes finished cement, without going into the details of the various machines. A view of the grinding raw material house during construction is given in Fig. 53, and an interior view of the finished building in Fig. 54.

Fig. 53: Rawmill building under construction, from the south end. The mill piers are concrete, but the building is of rubble masonry. The slurry pumphouse is in the foreground. Fig. 54: Inside the rawmill house, from the other end The ball mills are above to the right. To the left are the tube mill outlet ends, and the slurry collecting screw runs under the floor.

The rate of feed to the ball mills is controlled by means of a revolving table (Note 12) provided with a central lifting sleeve, which allows more or less stone to descend on to the table from an overhead hopper, which is kept supplied by means of the conveyor already mentioned. As the stone is fed into the mill by means of a diverting arm arranged just above the revolving table a certain amount of water is added. The discharge from each ball mill descends to its corresponding tube mill, which is arranged on the lower floor immediately beneath it. The slurry coming finally from the tube mills has been so finely ground that when it is dried only some 6 to 7 per cent. remains on a 180 by 180 sieve (Note 13). All the eight mills are motor-driven, one motor being provided to work each grinding unit - that is to say, a ball mill and a tube mill. The motors are arranged in another portion of the building which is entirely separated from the grinding mills by a brick wall.

We may here make a digression and remark that a feature of the works is the care which has been devoted to the electrical apparatus. As far as has been possible, all the motors and their controlling gear have been installed in separate rooms, which are kept as dust-tight as possible, and are entirely shut off from the machines they are driving. The wisdom of this course will make itself amply evident in freedom from repairs and breakdowns. We feel bound, moreover, to say that though a large proportion of the electrical apparatus is of foreign origin, the whole is of excellent quality and design, and is evidently doing its work well. We must confess, however, to a feeling that it would have been more pleasing could the whole of the works have been provided with British machinery, as is the part of it which is engaged in the actual production of cement. There is no machine or apparatus in the electrical equipment, or in the prime movers driving the moving parts, which could not have been made, and made well, by British manufacturers.

- o - O - o -

To return to the motive power of the grinding mills. There are in all four three-phase 50-period motors working on a 3000 to 3300-volt circuit and designed to develope 350 horse-power at 580 revolutions per minute. These motors were, like those for the stone breakers, supplied by the Allmänna Svenska Elektriska Aktiebolaget. Each drives by two belts on to two lines of countershafting which run through the partition walls into the grinding house. All the switches and other controlling gear are arranged in dust-tight cases, and are in such places that they are readily manipulated and kept clean. The precaution of having everything in a separate room kept as free from dust as possible renders the task of cleaning all the easier. An interior view of the motor house is given in Fig. 57 in our Supplement.

Fig. 57: Interior view of the rawmill motor house.

The delivery from each of the tube mills falls into a horizontal trough in which works a worm conveyor (Note 14), whereby the slurry is led along to a pump house containing two vertical three-throw ram pumps. These pumps are driven by a motor working on to a countershaft, there being fast-and-loose pulleys for each pump, so that either or both may be worked. The motor, which was supplied by British Thomson-Houston Company, Limited, Rugby, is of 30 horse-power. These pumps are designed to lift the slurry up into three doseage tanks, arranged between and at a higher level than two large circular mixing tanks. As a matter of fact, the three doseage tanks are not at present being used, it being found quite possible to keep the "mixture" correct by means of the mixing tanks alone, but they were provided as a precautionary measure, and may be here described in some detail. The three tanks stand side by side. They are made of reinforced concrete, and are circular in plan, the top portions having parallel sides and the bottom portion being in the form of an inverted cone. Each tank is designed to hold 80 tons of slurry (Note 15), and each is provided inside near its bottom with a series of nozzles through which air can be blown under pressure. The pipes connecting the three doseage tanks together are so arranged that any one tank, if found to be too high or too low in lime, can be interconnected with one of the other tanks having a suitably lower or higher lime content. As we have said, however, these tanks are not at the moment being used, since it has been found that with the raw materials as they are at present coming from the quarry, it has been quite easy to keep the mixture correct by using the large mixers only; but the doseage tanks are there as an additional precaution should they at any time become necessary, and an air compressor made by the Ingersoll-Rand Company is installed in a house near by to supply the compressed air. A view showing the three doseage tanks and one of the large mixing tanks is given in Fig. 55 in our Supplement.

Fig. 55: Dosage tanks and slurry mixer. Fig. 64: Slurry mixing tanks under construction. In the background is the rawmill house: the structure on the top accommodated the overhead shuttle conveyor feeding the rawmill feed bins.

It is, of course, well recognised that it is of vital importance that the mixture of the slurry should be kept absolutely correct, or good results cannot be hoped for. Indeed, it must not be allowed to vary more than a fraction of 1 per cent at any time, and we understand that at Aberthaw this condition is accurately fulfilled by controlling the raw material before it enters the raw mills.

The slurry therefore is now pumped direct into one or other or both storage tank mixers which are large circular reinforced concrete tanks, each 66ft. in diameter and 12ft. deep. Each tank can hold enough slurry to make 600 tons of cement, and since the output of the works, when working to their full capacity, is about 330 tons per day, it follows that the two tanks together hold enough for between three and a-half and four days' supply. Each tank has a central brick column, on which is pivoted so as to revolve on ball bearings a two-armed cantilever lattice girder. This girder or bridge derives its motion from skeleton paddles which depend from the bridge, and which are themselves revolved by an electric motor deriving its supply of current from overhead wires. This motor is of 20 horse-power, three-phase current, and was supplied by British Thomson-Houston Company, Limited. The paddles dip down nearly to the bottom of the tank, and the combined motion of the bridge and paddles—which revolve at a faster rate than does the bridge—ensure that the mixture is kept continuously agitated and homogeneous in composition throughout. A view of the two mixing tanks during construction is given in Fig. 64. From the tanks the slurry gravitates to two sets of three-throw pumps, which are housed in a chamber formed beneath the doseage tanks.

In order to ensure that the amount of slurry fed into the kilns shall be of a regular and known quantity the following arrangements have been made by Mr Day. Each set of pumps has been designed to discharge rather more slurry than is required to provide the full amount of material for one kiln, and each pump delivers into a separate tank fixed on a platform above the feeding end of its particular kiln. These tanks have overflows so that the head of slurry may be kept constant, the surplus flowing down the overflow pipe back to the mixing tanks. For the further action of the feeding arrangement, we cannot do better than quote the actual words which Mr. Day used in describing it in a paper which he read last year before the South Wales Institute of Engineers. He said: "This [the overflow] gives a definite fixed head or pressure on the outlet pipe, and consequently a regular feed (Note 16). The feed is varied by the burner according to the working of the kiln by means of a plain sluice valve fitted with an indicator, which shows on a quadrant the exact amount the valve is open, and consequently the rate of flow. The slurry falls [from the tank] into a large cast iron cup, the outlet of which can be stopped. The quantity at any time passing can be instantaneously checked by ascertaining the number of seconds required to fill the cup to an overflow weir provided in the cup".

We have now arrived at the stage when the properly proportioned slurry is fed into the kilns. We shall refer more particularly to the construction and driving of the latter later in these articles, but we may say here that there are two of them arranged side by side—see Figs. 56 and 62 in our Supplement. Each has its own dust-arresting chamber and its own chimney. Each of the latter is circular and 200ft. high with a top opening of 8ft. diameter, and both chimneys and dust chambers, which were the work of the Custodis Company, are good pieces of work. The kilns are 200ft. long and 9ft. in diameter for the major portion of their length, with an enlarged burning zone 10ft. in diameter. A section of one of the kilns is given in. Fig. 66, and two views taken during erection are shown in Figs. 63 and 65. The kilns are fired with powdered coal, and to the coal-drying, grinding and feeding mechanisms we shall also refer later. The burnt clinker falls into inclined coolers through which is sucked by a fan a current of cold air, while just before the clinker leaves the cooler a small jet of water is played on to it. On issuing from the coolers the clinker falls through a circular riddle—which retains any large lumps which may have come through—on to a travelling plate conveyor. At this point the necessary quantity of gypsum is added (Note 17). The conveyor either discharges the burnt clinker and gypsum into the hopper of an elevator leading to the upper part of a two-storey building which contains the grinding machinery, or it may discharge it on to the ground reserved for clinker storage.

Fig. 56: View along kilns from the firing platform (from the south). Fig. 62: View along kilns from the feeding platform (from the north). Evidently No 1 (right) is stopped, and No 2 is running.
Fig. 63: One of the rotary kilns (No 2) being erected. The No 2 fine coal hopper is also in place.
Fig. 65: Rotary kilns showing riding rings and carrying rollers. View from the west of the cold ends: No 1 was evidently the last to be installed.

The grinding equipment consists of four ball mills and four tube mills arranged in two floors in one building, so that each tube mill comes directly under its corresponding ball mill. A view showing the building in course of construction with the various mills in position is shown in Fig. 58 in our Supplement. The eight mills are driven by four motors, which are an exact counterpart of the motors which drive the raw materials mill, and which are contained in a compartment of the grinding house, but completely cut off from the mills by a brick wall. All the motors are on the ground floor and each drives by belts on to two countershafts, the upper of which drives a ball mill and the lower a tube mill. The motors are all of 350 horse-power and they were supplied by the Allmänna Svenska Elektriska Aktiebolaget. We shall refer in more detail to the grinding mills later. Meanwhile we may say that the plant has been designed to grind the cement down to such a degree of fineness that there is only a residue of 7.5 per cent. on 180 by 180 mesh sieve. The ground cement, as it is delivered by the tube mills, is taken by a travelling band conveyor and delivered into the hopper of an elevator, which lifts it to the top of the building and discharges it on to a second belt conveyor which runs up a covered-in flying bridge connecting the grinding house with the top of the cement silo. The angle at which this bridge is arranged is 27 deg. (Note 18) from the horizontal. Arrived at the top of the silo, the cement is transferred to a third band conveyor which is horizontal and which traverses the whole length of the building and can discharge it by means of a travelling throw-off carriage, which runs on rails laid on a gantry carried on columns resting on the walls of the main structure, into any one of the twenty compartments or bins into which the building is divided.

Fig. 58: View of the finish mills under construction.

The silo building is constructed of reinforced concrete and is a fine example of work of this character. Each of its twenty bins is designed to hold 500 tons, so that the total capacity is 10,000 tons, which permits of a storage of some five weeks with the works operating to their fullest extent. The building, which is situated between two sidings communicating with both the Vale of Glamorgan and the Taff Vale railways, covers 6700 sq. ft. of ground and rises to a height in the centre of 77ft. It is supported on fifty-five concrete columns, eleven of which measure 36in. by 24in. in cross section, the remaining forty-four being 24in. square. The roof, which is of corrugated iron laid on felt boarding is alone not of reinforced concrete. Each of the twenty bins has a width of 10ft., a length of 30ft., and an average depth of about 45ft. The bottoms are hoppered at an angle of about 60 deg. so as to form a V-shaped trough, and each bin is provided with four outlets each fitted with a sliding gate. There are thus no fewer than eighty points from which cement may be drawn off as required.

The headroom below the lowest point under the bins is 6ft., and, as the whole of the floor is clear, saving for the supporting columns, there is a large space available for laying out and packing the cement. The loading floor is extended on either side of the building for a distance of 22ft. so as to form platforms for the two sidings, both platforms and sidings being entirely covered in by lean-to roofs. The platforms are long enough to permit of twelve trucks being loaded at one time.

Some further particulars of the building will doubtless be of interest. The division walls between the bins are 5in. thick and they have vertical beams spaced at 6ft. centres. These beams are supported by 1¼ in. diameter tie-bars arranged at varying pitches, being spaced at 9ft. centres at the bottom and at 11ft. centres at the top. The longitudinal walls taper in thickness from 8in. at the base to 6in. at the top, and are made sufficiently strong not to require any vertical beams or tie rods. The aggregate used for the concrete was limestone obtained locally and crushed on the site. The reinforcement consisted of indented steel bars. The concrete mixture used generally throughout the building was of the strength of 1-2-4, but the columns were given a greater proportion of cement. A machine mixer was employed. No scaffolding was used for erecting the building, the work being done from platforms fixed to the centering and tie rods. The quantities of materials required in the construction, which we may mention was carried out by Messrs. Lewis Rugg and Co., of Westminster, were 380 tons of cement and 109 tons of steel. We have dealt at some length with this structure as it forms quite a feature of the works both in design and construction. It is shown during construction in Fig. 59, and completed in Fig. 61 of our Supplement.

Fig. 59: Cement silos under construction. View from the north. Fig. 61: Completed cement silos to the left in a northeast view of the plant taken from the Vale of Glamorgan trackside. Centre view is the finish mill house and the coal preparation plant is to the right.

Although we propose to deal in detail with the coal-grinding and handling plant in a future article, a few words may now be said regarding the coaling arrangements generally. The coal is brought on to the site on two sidings, which are quite independent of the cement sidings already referred to, and one of which is taken to the power-house and the other to the coal-grinding house. Near the latter there is a large area of land devoted to the storage of coal. This is of considerable advantage because it enables the firm to buy coal in large quantities when it is cheap and to store it on its own land. The importance to the cement manufacturer of having cheap coal may be realised from the fact that for each ton of cement produced a quarter of a ton of coal is burnt in the kiln. For a works of this size, therefore, when in full operation, some 500 tons of coal are required per week for the kilns alone. Having regard to the large quantity of coal which it would use, the company went into the question of what it would cost to handle the fuel by various methods. It was found that if dealt with by hand the cost would be not less than about £800 per annum. It was therefore decided to purchase a number of 12-ton hopper-bottomed wagons which could be run over underground bunkers and discharged. Alongside the coal-grinding house are arranged four such bunkers, each of which contains 40 tons. These bunkers are designed to deliver the coal automatically into a band conveyor, which in its turn discharges into the feed hopper of an elevator. The latter raises the coal and feeds it into the hopper of a pair of crushing rolls, in which the large lumps are broken up.

As delivered to cement works, coal always contains moisture, the amount varying in degree up to some 10 per cent. If stored in the open small coal may absorb more than this should the season be rainy, as the past winter has been. In order to grind it satisfactorily, coal should not certainly contain more than 2 per cent. of moisture, and to reduce the moisture to this figure special means of drying have to be provided. The dryer at Aberthaw is of the inclined revolving tube type heated by a coke fire. It is driven by an electric motor of 20 horse-power, which was supplied by the British Thomson-Houston Company. The dried coal is taken by a conveyor to an elevator, which lifts it to a conveyor running over the tops of hoppers above two ball mills. Coal is by no means easy to reduce to a fine powder owing to its soft, silky and flaky nature, but the combination of ball mills and tube mills perform the work adequately, and this is the combination employed at Aberthaw, where the coal is ground finely enough to show a residue of not more than 5 per cent. on a 100 by 100 sieve. The ball mills are on one floor and the tube mills are on the floor below it. Each pair of ball and tube mills is driven by belt by a 150 horse-power motor supplied by the Swedish firm already mentioned, these motors being contained in a portion of the building which is entirely separated from the grinding machinery. The ground coal discharged from the tube mills is taken in a spiral conveyor to an elevator, which raises it so that it can either be delivered to a large coal storage bunker or else to one of the two feed bunkers which are situated one over each of the two rotary kilns at the firing end. The remaining illustration in our Supplement Fig. 60 gives a general view of the works. In the middle distance on the right is seen the dwarf chimney of the electric power-house, to which we shall refer in our next article.

Fig. 60: Southward view of the plant from the embankment of the Taff Vale railway, north of the bridge over the River Kenson. The mill leet runs left to right across the picture, running behind the pile of waste shale being used to "raise the level" of the marshland, and in front of the steep bank. Across the leet on the left is the raw material bunker, with a sloping conveyor feeding the rawmill behind. The small building to the right is the pump house abstracting water from the leet to feed the power house.

- o - O - o -

Attention may now be directed to the power plant, which is of considerable interest, though unfortunately much of the machinery it contains is not of British manufacture. A glance at the plan Fig. 51 ante, will show that the power house is close to the bank of the river Kenson. It will have been gathered by the foregoing description that the whole of the machinery throughout the works is driven electrically—in fact, the only steam engines at work are those used on the steam digger, the drill and the locomotives. This means that a large electrical central station had to be provided. Fortunately, the proximity of the two rivers and the consequent abundance of water, easily got, allowed of the employment of steam turbines. Steam is obtained from five Lancashire boilers, 30ft. long and 8ft. 6in. internal diameter, supplied by Tinker, Shenton and Co. These are fitted with Bennis stokers, the hoppers of which are each separately fed by automatic elevators, which deliver the coal from bunkers below the siding, into which it has been discharged from the hopper-bottom wagons above referred to. The boilers work at a pressure of 180 lb. to the square inch and are fitted with Sugden's superheaters in the down-take flues. The final temperature of the steam on leaving the superheaters is somewhere in the neighbourhood of 600 deg. Fah. Between the boilers and the chimney is a Green's economiser with 640 tubes, which raises the temperature of the feed-water to 150 deg. Fah. Quite a short chimney is employed, it being only 55ft. high, and draught is provided by an induced draught fan and compressed air on Bennis's system. The boiler feed is pumped into the boiler by means of a three-stage centrifugal pump, which is driven by a high-speed steam turbine of the Pelton wheel type. This combined set, which runs at 3000 revolutions per minute, was supplied by Sulzer Bros., of Winterthur, and it has we understand, worked exceedingly well. In case of its breakdown, however, a single ram pump, made by J. P. Hall and Sons, Limited, Peterborough, has been installed. The turbine feed pump exhausts into a condenser fitted in the hot-well tank, and as the latent heat of the steam is thus recovered the combination is found to be very economical. The river water, it is not extraordinary to find, seeing that it comes from limestone country, contains 30 deg. of hardness, and consequently the make-up water, which amounts to about 10 per cent., has to be softened, a Lassen and Hjort apparatus being installed for the purpose.

The generating machinery consists of two 1500-kilowatt turbo-alternators—see Fig. 67. Both the turbines were supplied by Escher, Wyss and Co. They are both of the Zoelly type. One of them drives a Brown-Boveri generator and the other a Siemens generator, both being direct-coupled. Both alternators are designed to generate three-phase 50-cycle current at 3000 to 3300 volts when running at 3000 revolutions per minute. One of these generators will furnish current sufficient to work one-half of the plant at present installed, that is to say, plant which is producing 1200 tons of cement per week. They are excellent machines in every way and are, we gather, working with high efficiency. The generators are kept cool by air circulation, a wet filter of the Balcke type being furnished for cleansing the air.

Fig. 67: The two 1.5 MW generators. Fig. 68: The power house switchboard.

Each generator has its own condenser, which is arranged in a basement below the engine-room floor. The whole of the condensing plant was supplied by Escher, Wyss and Co. The condensers are of the counter-current surface type, and all the condenser pumps are of the rotary type. The circulating pumps are contained in a separate building, which is slightly nearer the river than is the power-house. Each is driven by a 40 horse-power Siemens motor. This power is only required at starting, as an arrangement has been made by which the outlet of the condensers is connected to a water seal at the same level as the pump-house. The pumps, therefore, only have to deal with the friction in the condensers and piping plus a short suction head. It is calculated that this arrangement ensures a saving of at least 200 tons of coal per annum. Each condenser has its own condensate pump and dry air pump and both are electrically driven. The condensate is returned direct to the hot well, over which is a Lea notch tank connected through a float tank to a Lea recorder. The latter not only indicates and records the total boiler feed, or steam consumption, of the plant, but also indicates and records continuously the quantity of water used. In this way not only can the rate of evaporation in the boilers at any moment be read, but also the total quantity over any period can be ascertained. In the house containing this instrument there is also a Sanders-Rehders CO2 recorder. We may here say that the system of record-keeping not only in the power-house but throughout the whole works is of the most detailed and useful character (Note 19). There is no process or part of a process of which the fullest records are not kept, and these records are not only of extreme interest but of the utmost value. Among other records kept are accurate figures regarding the power required per ton of output with varying degrees of fineness from the different units.

In the circulating pump-house are two Rees Roturbo pumps, each driven by a British Thomson-Houston motor. These are installed for providing a general water supply to the works, and for the purpose of arranging for a constant head of water a tank has been erected between the two kiln chimneys at a height of some 40ft. from their base. A large quantity of water is used in a cement works. The amount required to mix the slurry is considerable (Note 20), and in the Aberthaw Works advantage has been taken of a copious supply of water readily obtained to cool all the bearings of the mills, kilns and crushers with water and thus to economise lubricants.

The main high-tension switchboard—see Fig. 68—in the engine-room was supplied by the B.T.H. Company. It consists of thirteen panels controlling both generators and two 400-kilowatt oil-cooled transformers. From the board also run separate circuits to the crusher house, raw mill, clinker mill, coal mill, and to each kiln.

Throughout the whole works there are in all forty-two motors of various sizes down to 5 horse-power. Of these we give a list below with the names of the makers. Some of them have been alluded to in the foregoing:—

No.Output H.P.MakerLocation
8350Allmänna Svenska Elektriska Aktiebolaget4 Rawmills
4 Finish Mills
2150ditto2 Coal Mills
375ditto
240Siemens Brothers' Dynamo Works, Ltd2 condenser water pumps
220ditto
250British Thomson-Houston Company, Ltd
230dittoRawmill slurry pumps
?
325ditto
820ditto
59ditto
27.5ditto
35ditto

There is also a certain number of motors smaller than this. All the motors above 50 horse-power are supplied with current at 3000 volts pressure For those below this power current at 220 volts is employed and this pressure is also used for the lighting. The kiln motors are each supplied direct from the switchboard in the power-house. Distributing boards made by the Union Electric Company, Limited, are fitted in the raw mill, clinker mill, and coal mill, so that each unit in each of these mills is under separate control and can be immediately stopped by pressing a button on the feed platform. All motors over 5 horse-power are of the pipe-ventilated type and all under that power are of the totally enclosed type.

There is in the engine-room in the power-house, in addition to the high-tension board already alluded to, a low-tension board which comprises ten panels with twelve power currents and nine lighting circuits. This board was supplied by Switchgear and Cowans, Limited.

The switch gear is all of the dust-proof type. The low-tension distributing hoards are contained in cast iron boxes mounted on brickwork bases. These boxes are dust-proof. All the cables are of the three-core paper insulated and lead-covered type and were supplied by the British Insulated and Helsby Cables, Limited. They are laid underground in brick ducts covered with cement slabs. There is one main cable trench carried throughout the works. Of the whole electrical equipment it can certainly be said that it is well designed and carried out. Ample power seems to have been everywhere provided. In a cement works where all classes of machines are subjected to very heavy usage this is a most important point.

In addition to the various contractors already alluded to we may say that the masonry buildings were erected by E. Turner and Sons, Limited, and A. G. Collins and Co.; the structural steel work was provided by Rees and Kirby, Limited, and Walker Brothers, Limited; and the conveyors and elevators were jointly supplied by Ernest Newell and Co., Limited; Babcock and Wilcox, Limited; Fraser and Chalmers, Limited; and the New Conveyor Company, Limited.

A description of a cement works is not complete without some reference to the testing arrangements, and at Aberthaw these are of a most elaborate kind. Every device and machine necessary for making analytical tests of raw material and finished cement and for carrying out physical tests of every possible description have been provided. The laboratory is a detached building in which the various investigations are carried out in separate rooms, the whole department being under the able control of Mr. O. J. Corelli (Note 21), who was good enough to carry out numerous tests for our benefit. In all cases the samples showed very good results indeed, far above those required by the British Standard Specification. We have also been provided with copies of analysis of the cement and of tests carried out by independent experts. These we give below:—

Analysis of "Druid Brand" Cement made by the Aberthaw and Bristol Channel Portland Cement Company.
%
Silica, SiO225.38
Insoluble residue0.22
Alumina, Al2O33.73
Oxide of iron, Fe2O32.43
Lime, CaO65.26
Magnesia, MgO1.03
Sulphuric anhydride, SO3 (Note 22)0.56
Loss on ignition (CO2 + H2O)1.18
Alkalis and loss0.21
Total100.00

Physical tests and other particulars.
Nature of TestDruid BrandBritish Standard Specification, minimum requirements
Neat cement:— average of six briquettes:
At seven days608 lb.450 lb.
At twenty-eight days753 lb.674 lb.
3 parts sand. 1 part cement:— average of six briquettes:
At seven days217 lb.200 lb.
At twenty-eight days340 lb.263 lb.
Note that in terms of modern EN 196 compression testing, this is 6 MPa at 7 days and 14.5 MPa at 28 days.
Fineness:—
Residue on 180×180 sieve9.5%≤14%
Residue on 76×76 sieve0.4%≤1%
Soundness:—
Le Chatelier expansionNil≤10mm.
Specific gravity3.160≥3.10
Hydraulic modulus2.54≥2.0, ≤2.85

It will be appreciated from these figures that the Aberthaw cement is of excellent quality, nor are its full properties expressed by the foregoing. The pat tests also give very good results; indeed, we believe that no single failure has ever been experienced. Throughout the works there are numerous cases of brick walls having been served with a thin rendering of cement and sand. We examined several of these and in no case could any hair cracks be observed.

In conclusion, we desire to express our thanks to the directors of the company for permission to visit and describe these interesting works and to Mr. B. J. Day, Mr. W. A. Brown, and Mr. O. J. Corelli for courteous assistance in the preparation of this article.

In a future article we propose to discuss in more detail the various cement making machines used in these fine works, which machines, it will be remembered, were all made by Ernest Newell and Co., Limited, of Misterton.

CEMENT MACHINERY AT THE ABERTHAW WORKS, &c.

In the three preceding articles we have given a general description of the works at Aberthaw of the Aberthaw and Bristol Channel Portland Cement Company. We now propose to refer in more detail to the various machines which are in operation there and which have all been supplied by Ernest Newell and Co., Limited, of Misterton, near Gainsborough. We shall also take occasion to describe other machines which are used in cement manufacture and which this firm is prepared to supply, so that the present article may form a complete compendium of the cement-making machinery constructed by it. Messrs. Newells have, as we have already said, specially laid out their works to manufacture Portland cement machinery, and they have been most successful in their efforts. The firm started business with a full knowledge of what it had to face in the way of competition from abroad. It realised to the full what was the hold which continental firms had obtained on the British market, yet it was confident in its ability to hold its own, and one cannot but admire the pluck of Mr. Ernest Newell, the founder of the firm, for thus fearlessly entering the lists. For, it must be remembered, as we have pointed out before, that the makers abroad—especially the German makers—had busily spread and fostered the idea that unless foreign machinery, the makers of which had had so much more experience than our own people, were employed, failure must necessarily follow. Moreover, they were ready to equip works in their entirety. Then, again, foreign rivals were prepared to cut prices almost to any extent so as to keep British makers out of the market. Yet, in spite of all this, Messrs. Newells have made a success of their venture, and no small proportion of this success is undoubtedly due to the excellent quality of the machinery they supply, and to the thorough study they have made of the various processes carried out in the manufacture of Portland cement. The importance of this last attribute is very great; in fact, it is paramount. Unless our makers have an exact knowledge of the raw materials employed and of how these materials will behave under varying conditions, the hope of success is very small. Messrs. Newells have been investigating the problem of Portland cement manufacture in all its aspects for many years past, and they are now prepared, on being provided with an analysis of the raw materials, to put up works fitted with the machinery best suited to manufacture cement of the highest quality from those materials, provided, of course, that the latter are suitable for the purpose.

The firm naturally found at first that the task which it had set itself to perform was by no means easy of accomplishment, but it persevered, and as giving some idea of the measure of the results which have accrued, we may say that during recent years it has managed to obtain, in spite of keen German competition, a considerable share of the cement engineering business not only in this country but abroad. It has supplied several plants to the Associated Portland Cement Manufacturers, Limited, to the British Portland Cement Manufacturers, Limited, &c. For a branch of the latter company it has recently delivered a two-kiln plant for a works in Vancouver (Note 23). Then there are the Aberthaw Works, and we are prepared to say that it would be hard to beat any single item of the cement-making plant there. Finally, it is interesting to note, as a measure of the esteem in which this firm's products are held in this country, as contrasted with plant of German origin, that it informs us that it has during the last two years sold more cement machinery in the United Kingdom than all its German competitors added together have done (Note 24).

In describing the machinery at Aberthaw, it will be convenient to follow the same plan as in describing the works, and to deal with the various machines as they are employed in the course of the manufacture of the cement. First in order, then, come the two stonebreakers, which are of the swinging jaw or Blake type. This machine is illustrated in Figs. 69 and 70, and, as will be observed, it is of particularly solid construction. It has a feed opening measuring 36in. by 20in., so that it can take in large lumps of stone. The frame is made up in sections, which is, of course, convenient from the point of view of transport. The side frames are of cast iron thickened at the ends, and where the two shafts come, while the end pieces are of cast steel. The side frames are recessed where they join the end frames, and are tied together by heavy mild steel bolts which pass through from side to side. The general construction is very similar to that of other jaw breakers which have previously been described. The machine operates on the toggle principle, the pitman being raised and lowered by means of a cam on a main eccentric shaft. The pitman is made of cast steel and its head is water-jacketed. It is lined in the bearing with steel alloy, which is itself lined with white metal. The eccentric shaft is of mild steel and it is furnished with two heavy fly-wheels outside the frame. At Aberthaw the machines are provided with fast-and-loose pulleys and an outboard bearing. but if desired rope driving, as shown in Fig. 70, can be applied. The swinging jawstock is made of mild steel and it is faced with two corrugated reversible manganese steel jaw-plates provided with zinc backing. The end-piece opposite the swinging jaw is similarly furnished with manganese steel plates, which form the other face of the jaws, and are also reversible. The toggle plates and the grooved cushions in which they work are of manganese steel, and the former are provided with shrouds to prevent lateral movement, so that there may be no loss of power due to friction between the plates and the side frames and no injury to the latter by grooving. The toggle plates are also provided with large grease-boxes at each end to ensure a good supply of lubricant for the toggle joints. There is an arrangement by which the cushions can be withdrawn without the necessity for disturbing the pitman, jawstock or toggle block. There is the usual method for regulating the size to which the stone passing through the machine is broken by means of a toggle block and adjustable wedge, with which the readers of these articles are now familiar. Both toggle block and wedge are of cast iron and they are anchored by heavy bolts. A drawback rod with four rubber discs to furnish the necessary tension is provided with the object of ensuring that the jawstock shall swing back to its full extent on the return stroke and keep the various portions of the machine in their correct relative positions. A great deal of attention has evidently been paid to the question of lubrication, and, as will be seen, numerous sight-feed and other lubricators have been provided, all of them in accessible positions. In addition to these, there is, of course, the water cooling, so that it is most unlikely that these machines would ever have hot bearings.

Figs. 69 & 70: Newells jaw crusher of the type used at Aberthaw.

Each of the two jaw breakers at Aberthaw has a capacity of from 40 to 50 tons of hard limestone broken down to 2 in. cubes per hour, and each requires about 50 brake horse-power to drive it in full work. Both machines were in operation at the time of our visit, and both were operating most successfully. As a matter of fact, they were not set to break up the stone into very small pieces. and since there were no crushing rolls to follow them, it says much for the grinding mills—which we shall mention later.

The article goes on to describe Newell's "Lion" gyratory crushers (not installed at Aberthaw). Refer to The Engineer, 14/5/1915 for details.

Fig. 73: Newells crushing rolls as used for coal crushing at Aberthaw.

For breaking up lumps of clinker, shale, marl, chalk or coal, or for reducing the broken stone delivered from either type of stone breaker just described, Messrs. Newells supply crushing rolls such as those shown in Fig. 73. As illustrated, the rolls are provided with fast and loose pulleys, and this is how the machines are usually sent out, but if required friction clutches can be fitted. The driving shaft is furnished with a fly-wheel and the rolls are driven through spur reduction gearing. The two rolls are coupled together with coupling wheels with deep teeth so as to permit of adjustment apart of the rolls and hence of the size to which the material is crushed. Six sizes of these machines are made. Two of these have rolls 20in. diameter and either 16in. or 24in. long; in two the diameter is 28in. and the length either 16in. or 24in.; while the third pair have rolls 36in. long (sic) and again either 16in. or 24in. in length. The approximate respective weights of these six machines is 95, 100, 110, 118, 130 and 140 cwt. The centres or hearts of the rolls are of tough cast iron keyed to hammered steel shafts. The shells of the rolls may either be of chilled cast iron or of manganese, chrome, or high carbon steel, according to the work required of them, and they may be either smooth, fluted, corrugated, or toothed to suit the nature of the material on which they have to operate. They are fitted with spring cages, which are adjustable, so as to yield when stresses greater than those necessary to crush the material are encountered. Ordinarily the main frames are of cast iron, but rolled steel girder frames are supplied if required so as to facilitate transport or for other special reasons. It will be seen from the foregoing that Messrs. Newells are quite prepared to supply stone-breaking and crushing machines to suit any sort of material or to fall in with any special predilection (Note 25) on the part of the prospective user.

- o - O - o -

To return now to the actual machinery at Aberthaw. The broken stone as it comes from the two-jaw crushers falls on to two inclined band conveyors, which take it up to the top and deposit it in the raw material silos. These silos are, as a fact, constructed of ferro-concrete, but Messrs. Newells have frequently supplied silos made of mild steel for a similar purpose. For the process of grinding the stone there are four wet ball mills and four wet tube mills. The broken stone is brought up by a conveyor from the storage silos and is delivered into four mild steel bins arranged one over each of the ball mills. We have already referred to the automatic means of feeding these mills, but we may now give the following amplification of our former references. Newell's improved feeders consist of a rotating table, the edge of which projects over the feed hopper of the ball mill. The table is revolved by means of bevel gearing from the main shaft of the mill itself, and is surrounded by a casing. Over the centre of the revolving table is a vertically adjustable pipe connected to the feed bin overhead. The vertical height of this pipe can be varied. In its lowest position its lower edge is close to the table, so that no material can escape; in its highest position a quantity of stone in excess of that necessary for the full output of the mill can pass on to the table. The man in charge, therefore, has it in his power to arrange the quantity of feed to exactly the correct amount. The speed at which the table is revolved is slow, but it is high enough to permit of the material spreading over its surface. At the point where the table projects over the feed hopper of the mill a scraper is fixed, and is set at such an angle that the stone on coming against it is directed into the hopper.

Ball and Tube MillsFig. 74: The four Aberthaw ball-and-tube rawmill sets.

The arrangement of the raw material grinding mill grinding house is shown in Fig. 74. The four wet ball mills are of the sieveless type and are 9ft. internal diameter by 6ft. long. The body of each mill consists of a mild steel weldless tube ¾in. thick, while the end-plates are of rolled steel 1¼in. thick. The end-plates are secured to the bodies by 5½in. by 4½in. by ⅞in. welded angle flanges and 1in. rivets. Arranged round the internal circumference of the mill are chrome steel hunch-plates and the inner surfaces of the end-plates are also protected with chrome steel lining plates. the hunch-plates and end protecting plates being all held in position by bolts taken through to the outside. The feeding hopper is of cast steel, as are also the inlet and the driving naves. The drum shaft on which the mill rotates is of mild steel and is 10⅝in. in diameter in the largest part. The feed of the mill is in the centre of one of the end-plates through the hollow feeding nave. The outlet for the ground material is through a series of grids formed in the outlet end-plate, each grid being a separate steel casting covering an aperture cut through the end-plate of the drums. The grinding balls are of hardened forged steel, and their size varies from 4in. to 5in. in diameter. The mills are driven by machine-moulded spur gearing, and the main bearings, which are lined with white metal, are provided with grease caps. The driving pulleys are 8ft. in diameter and 22in. in the face.

Along with the stone the necessary amount of water is fed into the mills, each of which is surrounded by a drip casing constructed of mild steel plates and angles, and is carried on a supporting structure consisting of rolled steel joists and channels. The whole arrangement is well shown in the engraving, in which will be seen the feed bins, the revolving feed tables, and the four ball mills, each immediately over its corresponding tube mill, to which the discharge from the former flows by gravity. The ball mill is designed to reduce the raw material to the size of grits, which is the size best suited for feeding the tube mills.

The four tube mills are each formed of a welded steel body 27ft. long and 6ft. 2in. in diameter. The end covers, which are of cast steel and of the domed pattern, are provided with hollow trunnions and are lined inside with manganese steel. The mills are divided into two chambers of unequal lengths. The smaller chamber, which is at the feeding end, is lined with corrugated grinding plates, and is loaded with hard steel balls. In it any large nibs which may have passed through the ball mills are reduced to a size in which they are most efficiently dealt with in the larger or final grinding portion of the mill. The latter is lined with silex flint blocks and the grinding media are flint pebbles (Note 26). The partition between the two chambers is in the form of a grid with slots only large enough to permit of the passage of quite small grits. The trunnion bearings of the mill are 21in. in diameter and 20in. long. They are babbited and water-cooled. Both the feeding and discharging are effected through the trunnions.

On leaving the mills the slurry is found to be very finely ground, and when dried from 90 to 93 per cent. of the powder will pass through a 180 by 180 mesh sieve. It falls down into a trough, in which works a screw conveyor, which leads it to a sump, from which two slurry pumps take their suction. The mills are revolved by gearing, each mill having bolted to its end angle ring at the feeding end a heavy spur wheel, which is made in halves so as to permit of removal or repair without having to lift the tube from its bearings. With it meshes a cast steel shrouded pinion, which is carried on a mild steel shaft revolving in three bearings. This shaft is driven by belt from an electric motor, one motor driving a ball mill with its corresponding tube mill. The pulley required to drive the mill is 21in. wide on the face.

Messrs. Newells are prepared to supply tube mills for wet grinding in a large number of different sizes. First of all there are six diameters, these being designated as sizes 0, 1, A2, 2, 3 and 4. Then the tubes in each of these can be made of varying lengths, and the different particulars are set out in the following table:—

General particularsSize 0. Size 1. Size A2. Size 2. Size 3. Size 4.
Diameter inside steel tube3'6"4'0"4'6"5'0"5'6"6'2"
Length inside steel tube12'-20'13'-20'13'-20'16'-26'20'-26'20'-27'
Approximate brake horse power required18-2824-3430-4248-7465-8584-112
DAWN Horsepower (Note 27)16-2725-3934-5457-9392-120124-169

It will thus be seen that a very large number of sizes and powers are available. As showing the nature of the weights involved, we may say that in the largest size—that at Aberthaw—the weight of the machine without its charge of pebbles is 24¾ tons, while the charge of pebbles is 14 tons, a total of 38¾ tons without taking account of the charge of slurry. We may add that in addition to wet tube mills. which are made with welded tubes all in one piece, Messrs. Newells are prepared to supply mills with built-up riveted tubes if desired. One of the latter with its driving gear is shown in Fig. 75.

Fig. 75: Newell Riveted Tube Mill - not installed at Aberthaw.

The slurry pumps do not require much description. A view of a pair of them is shown in Fig. 76. They are of the three-throw vertical type, with plungers 8in. in diameter and 16in. stroke. Each is self-contained on a cast iron bed plate and each is driven by belt through machine-cut double gearing. They receive their power from a motor through a counter-shaft, and fast-and-loose pulleys with hand wheel worked striking gear are provided. The three pump cylinders of each set are mounted on a rectangular pump box, to the bottom of which are attached three suction pipes, which are taken to the sump, each pipe having its non-return suction valve. The discharge is through delivery valves of the disc type into a chamber common to the three pump barrels, and the outlet from this chamber is fitted with a pendulum type of retaining valve. The pumps are very stoutly made and designed for working pressures up to 200 lb. on the square inch. A lever safety valve is attached to the pump box, which can be adjusted to suit the working pressure. As a matter of fact, the actual working pressure at Aberthaw is nothing like 200 lb. on the square inch. In lift alone, even to the doseage tanks, the maximum pressure is not a quarter of this, but with a substance like slurry it is best to be on the safe side and to provide plenty of margin.

Fig. 76: Slurry Pumps.

The delivery from the pumps is, it will be remembered, not ordinarily to the doseage tanks, but to the slurry mixers. The mechanical stirring mechanism, which was provided by Messrs. Newells, has already been fairly fully described in the account of the works. It may, however, be here repeated that a lattice girder, made up of heavy rolled steel joists and angles, and supported at its centre on a ball bearing so as to be free to revolve easily, carries depending from it four large skeleton paddles—two on each side of the central bearing. These paddles are made to revolve by gearing driven from an electric motor working on to a shaft running along the length of the girder, and the action of the paddles, which are separately rotated, is such that, owing to the consistency of the slurry, the girder, and with it the paddles, revolves slowly about the centre axis of the girder, the motion of the mixing apparatus as a whole being on the sun and planet principle. The pivot of the lattice girder rests on a pillar built up in the centre of the tank, and consists of a cast iron foundation plate on which is mounted a large ball bearing, protected from contact with the slurry, and provided with a cast iron hollow pillar or central sleeve secured at the top to the girder. The paddles are built up of rolled steel joists and angles, and each paddle revolves round its own mild steel centre shaft, which shaft is itself carried by suspension bearings attached to the main girder. The engravings—Fig. 64 ante and Fig. 78—will make the foregoing description perfectly clear.

Fig. 78: Dosage tanks, slurry mixers, kilns and coolers. This inexpertly modified image was probably based on a photograph taken before the nearer (No 2) mixer had been completed. Despite its faults, it was used in Newell's advertising, presumably relying on the prospective audience's lack of appreciation that the centre post of a mixer is, as the term implies, in the centre.

The feed of slurry from the mixers to the kilns is brought about by two sets of three-throw plunger force pumps, which are exactly similar to those already described, and hence call for no further description. There is also no need to refer again to the method of feeding the kilns, which was described in detail in the general article on the works—see above—and so we come to the kilns themselves. With regard to them we may say at the outset that we have never anywhere seen a better pair of kilns. Both in construction and in mounting they are excellent examples of high-class engineering work, and would more than hold their own in competition with any foreign-made kilns. British firms may have had to pass through troublous times (Note 28) in learning how to build rotary kilns successfully, but they have certainly acquired the requisite knowledge, and in this department as in the case of other branches of cement machinery they have already equalled, if, indeed, they have not gone ahead of their foreign rivals. The two kilns at Aberthaw are not the longest made (Note 29), but they are each of them 200ft. long by 10ft. diameter in the burning zone, and 9ft. diameter in the remainder of their length (Note 30). They are inclined at an angle of 1 in 25. The steel shell of the tube is ¾in. thick, and each kiln is carried on five wide cast steel riding rings or tires, which rest on two roller bearings mounted on bed-plates. These riding rings are not attached directly to the shell of the tube, but are carried on chairs, which are themselves bolted at intervals round the tube. Moreover, the rings are not a tight fit on the chairs, for a distinct clearance space has been provided, so that the rings themselves can revolve on the kiln, and, as a fact, do so to a slight extent, the bearing surfaces of the chairs and the inner peripheries of the rings being lubricated to facilitate motion. In this way any ill-effects due to expansion—which were among the troubles met with in the early days—are entirely counteracted. Fig. 77 is a view of part of one of the kilns, showing one of the riding rings and the rollers on which it rests. Where the roller rings come the kiln shell is reinforced under the chairs by means of ¾in. mild steel wrapper plates. In fact, the kilns are as complete as present-day science can make them, and we understand that they have given no trouble whatever from the time they started to work until now. They are, moreover, entirely free from "twist", which is a defect from which kilns made by the best of foreign makers have been known to suffer, and one of the effects of which is to wrench the lining out of position and reduce the length of its life.

Fig. 77: View of Kiln 2 back end showing tyre and roller.

For some distance down the interior of the kiln at the feeding end, buckets and lifter plates are arranged so as to lift up the slurry and to allow it to fall down through the upcoming hot gases in streams, and thus to expose the maximum quantity of material to the drying action of these gases (Note 31).

The kiln is revolved by means of a motor acting through gearing on to a heavy cast steel spur-toothed ring mounted in an elastic manner on the kiln. This mounting consists of plates riveted to the kiln shell and projecting tangentially from it. The toothed ring, therefore, is not in direct contact with the kiln, and is hence protected from excessive heat and from the heavy expansion which would result from such contact. The cast steel spur gear wheels which transmit the power of the motor to the kiln and reduce the speed are mounted on the kiln foundations and practically work immersed in oil. They are a fine bit of work and show no signs of wear whatever. They are quite silent in operation. It will be realised, of course, that the reduction in speed from the motor to the kiln is very considerable. The speed of the motor, which is of 50 horse-power, can be controlled between 365 and 730 revolutions per minute, the corresponding speeds of the kiln being from ¾ to 1½ revolutions, and the normal speed one revolution per minute. So as to prevent longitudinal movement of the kilns thrust rollers which bear against the sides of the riding rings nearest the firing end are provided (Note 32).

Both kilns are lined from end to end with specially refractory firebricks arranged in varying thicknesses according to the temperatures met with. To these firebricks. which are known by the name of Adamantine, and which were supplied by Charles Davison and Co., Limited, of Ewloe, near Chester, we allude in a subsequent article.

It may interest our readers if we state that what was, we believe, the first rotary kiln put to work in this country was that erected at Mitcheldean. It was 15ft. long and 3ft. 6in. in diameter (Note 33). A few years ago Messrs. Newells took out this kiln and replaced it with another which was 120ft. long and 7ft. 6in. diameter, and which was at the time the largest in the United Kingdom. As showing what vast strides have been made since the small kiln was built, we may say that whereas it consumed about twice the weight of coal that it produced clinker, large modern kilns with good coal only use a weight of fuel representing some 27 per cent. of the weight of clinker burnt. Messrs. Newells are prepared to make kilns from 7ft. 6in. to 9ft. in diameter, with or without enlarged burning zones up to 10ft. diameter, and from 120ft. to 200ft. long (Note 34).

At the lower or firing end of each kiln is arranged a hood, made of mild steel and lined with refractory firebrick, which fits over the end of the kiln and can be run back on travelling wheels, and a hole in the centre of which the burning nozzle projects into the kiln. There is an aperture in the bottom of this hood to allow the calcined slurry or clinker to fall down in the cooling drum. A clinker cooling drum is provided for each kiln and each is arranged immediately under its own kiln. These cooling drums are very similar to the kilns themselves, both in construction and operation. Each is a long tube or cylinder built up of ½in. steel plate, the length being 82ft. and the diameter 6ft. 6in. for the greater part of the length, but increased at the upper end to 7ft. 6in. (Note 35), where there is a lining of refractory firebrick. For the remainder of the length the lining is of cast iron. Cast iron Z section lifting plates are arranged longitudinally around the internal periphery of the coolers, and these lift the clinker nodules and let them fall in showers through the current of cold air which passes up through the cooler on its way to the kiln. Each cooler is arranged at a slope of 1 in 14, and each is revolved at a speed of about one revolution per minute by cast steel gearing, exactly as are the kilns, the necessary power being derived from an electric motor. As in the case of the kilns thrust rollers bearing against the riding rings are provided to counteract any tendency of the coolers to move longitudinally. At the delivery end of each cooler, which is immediately above a plate conveyor, is arranged a circular grid. This, while allowing the well-burnt clinker to pass through, prevents large imperfectly-burnt pieces from doing so. However well a kiln may be managed it is practically impossible to avoid having these partially burnt lumps sometimes, though it is by no means always that they are removed as at Aberthaw. In a good many works we have seen they are allowed to go on to the grinding plant, it being considered that, representing as they do a comparatively small proportion of the total volume of the clinker, their inclusion will have no noticeable effect on the finished cement. It is, however, undoubtedly the best practice to remove them. It may here be said that the lifting plates, which we mentioned as being arranged near the feeding end of the kiln, are provided very largely with the view of keeping the drying slurry well broken up and hence of preventing the formation of lumps which the revolving motion of the kiln must tend to produce (Note 36).

In Fig. 78 is given a comprehensive view which shows on the right hand side the two slurry mixers with the three dosage tanks between them and on the left the kilns with the coolers beneath them, the feeding platform and the two chimneys coming in the centre. The plant has been designed to produce 120,000 tons of cement in the course of a year, which means a continuous daily output from each kiln of nearly 165 tons, or, say, 1150 tons per week. We may add that the smallest cooler of the type used at Aberthaw which Messrs. Newells ordinarily make is 3ft. 6in. in diameter and 40ft. long (Note 37), and the largest is 6ft. 6in. and 7ft. 6in. in diameter and 82ft. long.

The kilns at Aberthaw are coal fired. The arrangement for bringing the coal on to the site have already been described, and the passage of the coal through the various processes has been briefly alluded to in a foregoing article. It now remains to describe the different machines which are employed. The coal is first of all raised from the underground bunkers, in which it has been deposited by railway trucks, by means of a bucket elevator which discharges into a mild steel hopper arranged immediately over a set of crushing rolls. These rolls are of Messrs. Newell's standard design, which has already been described and illustrated in Fig. 73. They break up the coal into quite small pieces, which are discharged into a feed hopper over a table feeder. The latter is of exactly the same type as that employed for feeding the raw material grinding mills. The table delivers the coal in a regular stream into the coal dryer.

Fig. 79: Newell's Double Tube Coal Dryer.

The coal dryer is of the double drum rotary type, It is 50ft. long and is made up of an outer mild steel shell 6ft. in diameter and an inner steel tube 2ft. 6in. in diameter. It is mounted in brickwork in such a manner that the hot gases from a furnace are first of all made to travel inside the brick setting and to be in contact with the exterior of the outer cylinder, and then to return along the inside of the inner tube to the chimney, which is situated at the feed end of the dryer. Meanwhile, the coal is fed from a feeding hopper through a shoot into the annular space between the two tubes so that it is acted upon by heated surfaces practically all round it. In its passage down from the top to the bottom of the dryer it is being continuously lifted and showered by lifting plates attached to the interior wall of the outer tube. It thus obtains the maximum possible effect of the heat from the hot gases on either side of the annular passage. It may be explained that the inner drum is attached to the outer cylinder by means of suspension links so as to allow for any unequal expansion of the two drums. The furnace is built on to the brickwork setting which surrounds the dryer, to which it is connected by flues. The outer tube is carried by cast steel riding rings mounted on the tubes, which rings rest on cast steel rollers the bearings of which are fixed to cast iron bed plates. A thrust roller is supplied, as in the case of the kilns and coolers, to limit the axial travel of the drum, which is set at an angle of one in 20. The rotation of the tube, which is at a speed of about 5 revolutions per minute, is effected by means of belt pulleys and bevel gears with a spur pinion engaging a toothed gear ring attached to one of the riding rings, the necessary power being obtained from a 20 brake horse-power motor. The whole arrangement is well shown in Fig. 79. This size of dryer is capable of dealing with about 10 to 12 tons of coal per hour, reducing an original water content from some 10 per cent. to about 2 per cent. or under (Note 38). The sizes of dryers of this type ordinarily made by the firm vary from 3ft. 6in. in diameter and 20ft. long to 6ft. in diameter and 60ft. long. At Aberthaw the coal is, as we have explained, dried by gases from a separate furnace. Messrs. Newells are however, prepared to supply dryers heated in other ways, such, for example, as by means of producer gases or by hot gases from the kiln.

The dried coal leaves the outlet end of the dryer by means of a gravity shoot and is taken by a spiral conveyor to the coal grinding mills. Arrived there it is lifted by an elevator and delivered by another conveyor into two feed hoppers which are arranged immediately over two feed tables, each of which feeds a ball mill. The feed tables are exactly similar to those used for the raw material and for the undried coal.

- o - O - o -

The two coal-grinding ball mills are of Messrs. Newell's standard type, known as "Lion" Ball Mills. They are illustrated in Figs. 80 and 81, the latter of which shows half of the mill in section. The drum is mounted on a steel shaft to which it is secured by two naves, the nave on the driving side being solid, while that on the feeding side is hollow and is provided with arms or ribs. The shaft is carried in bearings fitted with gun-metal steps, which are secured to sole plates fixed by bolts to the foundations. The mill is driven by spur reduction gearing. The countershaft carries fast and loose pulleys and a shrouded pinion which gears with the drum shaft. The latter forms the axis of the revolving drum, which latter is keyed to it as is also the large spur wheel for receiving power from the pinion shaft. A feed hopper, which is secured to the sole-plate of the drum shaft bearing, is fitted to the hollow nave at the feeding side of the drum. This hopper may, under certain circumstances, be furnished with a regulating feeding device.

Fig. 80: Newell's Lion Ball Mill. Fig. 81: Part section of ball mill.

The side or end plates of the grinding drum, which are secured to the naves and carried by the drum shaft, are of rolled steel and are lined with hard steel protection plates. The stepped grinding plates, which are of cast steel, are arranged circumferentially round the drum, as may be seen in the section Fig. 81. The faces of these plates are partly flat and partly radiated, one half of each plate being solid and the other half perforated with taper holes. Scoops and guard plates are fitted to each grinding plate. At a short distance from the grinding plates are arranged perforated screens and a little way further out still there are fine sieves fitted to steel frames, which are secured to the circumference of the drum, which latter is covered in by a dust casing constructed in a number of parts and provided with doors for the examination of the sieves and screens. The lower part of the dust casing is formed into a hopper, into which the ground material which has passed through the screens and sieves drops and from which it is discharged through a flanged outlet. The coarser material which has passed through the holes in the grinding plates, but which cannot get through the holes in the screen, is automatically returned by the motion of the drum, and the action of scoops to the interior of the drum, for further grinding, the pieces falling through the holes in the stepped part of the plates. In the same manner the tailings from the fine screen are also returned to the drum. The material delivered from the machine therefore must all have passed through the sieve, which is constructed to pass only fine grits. The grinding media are hard steel balls.

Messrs. Newell's make for stock six sizes of this type of mill, the particulars of which are given in the following table:

General particularsSize 0. Size 1. Size 2. Size 3. Size 4. Size 5.
Diameter of grinding drum6'3"7'6"7'6"8'10½"10'0"10'0"
Width of grinding drum3'3"3'11"4'6½"4'6½"5'6"6'6"
Revolutions per minute272525212118
Brake horse power122027374550
Weight of mill with balls, cwt148202230310335360
Weight of balls, cwt142227405870
DAWN Horsepower (Note 39)142530456376

The discharge from each ball mill falls by gravity to a tube mill arranged immediately below it. These two tube mills are somewhat similar to the wet tube mills already described; they, however, differ in certain particulars. Thus, there is only one chamber instead of two and the grinding media are all flint pebbles. The tubes are 24ft. long by 6ft. in diameter and they are lined with silex bricks. The standard arrangement of one of Messrs. Newell's tube mills for dry grinding is shown in Fig. 82. These mills are supplied in exactly the same diameters and lengths as are the firm's wet tube mills, the only difference being that the horse-powers vary slightly in the two types. The end plates are domed in shape and are of cast steel. They are lined inside with hard cast iron wearing plates. Cast in one piece with the end plates are hollow trunnions, which serve the purposes of carrying the tube in its bearings and feeding and discharging the mill. The trunnion bearings are 21in. in diameter and 20in. long. They are babbitted and water-cooled. The inlet trunnion is furnished with an internal automatic screw feeding device for regulating the feed to the mill, while the outlet trunnion is lined in the bore with hard cast iron liners. A cast iron slotted grid plate is fixed near the delivery trunnion which, while it allows the powdered coal to pass, retains the grinding pebbles. The pulley driving the mill is 7ft. in diameter and 22in. in the face.

Fig. 82: Tube mill arranged for dry grinding.

After being treated in these mills the coal is ground so finely that 95 per cent. of it will pass through a 100 by 100 mesh sieve (Note 40), and it is then in a condition to be used for firing rotary kilns. The pulverized coal dust on being discharged from the tube mill falls into a trough containing a spiral conveyor by means of which it is taken to the hopper of an encased 15in. bucket elevator which lifts it up and delivers it into the enclosed trough of a 12in. spiral conveyor which runs horizontally over the centres of two feeding hoppers, one of which is situated immediately over the centre line of each of the two kilns and a little to the rear of the kiln hood.

Fig. 87: Kiln coal feeding arrangements. See Note 41.

The arrangements for firing the kilns are shown in Fig. 87. The coal dust hoppers are shaped like inverted truncated pyramids. At the bottom of each is fixed a plate valve the opening of which can be varied by means of a chain which passes over a wheel and is taken down to the firing platform so as to be readily manipulated by the burner. Fixed immediately below this valve is a horizontal trough containing two 5½in. screw conveyors, which revolve in opposite directions. These screws are driven by motor through a double cone variable speed gear, which is also controlled by a chain going down to the burning platform. The burner has, therefore, two means of controlling the amount of coal being supplied to his kilns. As a matter of fact, two such means are not an absolute necessity and the burner is usually found to content himself by manipulating one of them, but it is, of course, a safeguard to have two.


Fig. 83: Coal dust feeding device. The firing pipe is under high pressure to impart a large momentum to the flame in the kiln. The venturi puts the fine coal feed pipe under suction, preventing fine coal from blowing back through the screws and leaking out at the joints.

The discharge from the trough containing the twin spirals falls into a pipe which delivers it into the top of the firing nozzle pipe some little way back from the point where the latter is taken through the hood of the kiln (Note 41). Just below the pipe coming down from the spirals is arranged a distributor as shown in Fig. 83. This is in the form of a taper tube fitted inside the main air pipe in such a position that the smaller end of the tube is just in front of the point where the powdered coal falls in (Note 42). Along the main air pipe and through the distributor tube is passed a continuous blast of air furnished by a centrifugal fan, and in this way the coal dust is taken along by the air current and injected into the kiln. The centrifugal fan draws its supply of air from the coal mill house. This arrangement not only helps to keep the latter free from impalpable coal dust but ensures the burning of the dust and hence increases the efficiency of the plant, though possibly only by a very little. The speed of the fan may be controlled by the burner from his platform as well as the speed of rotation of the kiln, so that any necessary adjustment can be made by him without his having to leave his post. The injection nozzles of the kiln are provided with spherical joints so that the burner can direct the flame to any desired part of the material coming down the kiln. This again furnishes another means of controlling the burning. It will be understood that there is a complete coal feeding plant, including extractors, variable speed gear, centrifugal fan, &c., for each kiln.


Fig. 84: Clinker grinding plant.

The grinding of the clinker to produce the finished cement is, of course, a most important operation. At Aberthaw it is carried out in two stages, and the plant employed consists of four steel ball tube mills and four tube mills. The steel ball tube mills are as their name implies, tube mills in which the grinding media are steel balls of various sizes. They are, however. shorter and of stiffer build than the tube mills already described. The steel ball tube mills at Aberthaw are 17ft. long and 6ft. 2in. in diameter. The shell is made of a seamless steel tube and the end covers are dome shaped and of cast steel. The lining is of cast steel plates. The arrangement of the clinker grinding plant is shown in Fig. 84. The feed to the steel ball tube mills may be either by means of a table feed as shown, which has already been described, or by a jigging or shaker feed arrangement, such as is shown in Fig. 85. The apparatus consists of a mild steel trough suspended from a cast iron frame and capable of being shaken or jigged by means of an eccentric. To the end of the trough is attached a second and similar trough which projects into the hollow trunnion at the feed end of the mill. The clinker gravitates from the storage hopper through the aperture in the cast iron frame on the upper trough and, when the latter is stationary, the flow of material is checked, whereas, when the shaking operation is in motion, the material progresses along the troughs into the mill. The stroke of the eccentric is adjustable, and hence the rate of feed can be varied at will while the mill is operating. Some such devices as the foregoing or their equivalents are absolutely essential to the proper working of a cement grinding plant, for, unless the feed is steady and regular, the resulting cement will not be round regularly and evenly.


Fig. 85: Shaker feed for tube mills. The early mills, being comparatively small, had correspondingly small trunnions, and getting the feed through them was a major difficulty.

The clinker-finishing mills to which the discharge from the steel ball tube mills falls by gravity are very similar indeed to the finishing mills for coal grinding and their chief point of difference is that their lining is of hard cast iron, whereas that in the coal mills is of silex. The clinker finishing mills are 26ft. long and 6ft. 2in. in diameter (Note 43) and their methods of feeding and discharging are absolutely identical with that of the coal plant and need not be again referred to. The grinding media are flint pebbles.

The product of these mills is the finished cement and it is taken away by a system of conveyors and elevators to the cement silos as already explained. The mills were, at the time of our visit, operating extremely well, and we gathered that they had always done so and that there had never been any trouble or difficulty in getting them to grind the cement down to any degree of fineness. The residue on a 180 by 180 mesh sieve at the time of our visit was some 7½ per cent. At that time the British Standard Specification allowed a maximum residue of 18 per cent. This has since been reduced to 14 per cent., so that the residue at Aberthaw in ordinary working is only just over half of that maximum.

The grinding of rotary kiln clinker in two processes by the method just described enables steam or water to be introduced into the finishing mill. This is, in some cases, not only advantageous but necessary in order to produce cement which will conform with the requirements of the British Standard Specification (Note 44). There are cases, however, when this is unnecessary, and in such instances a combined tube mill can be economically and efficiently used. Messrs. Newells manufacture such a mill under licence from a patentee (Note 45) and an external view is given of one of them in Fig. 86. It consists of a long steel tube, the first and shorter portion of which is lined with steel plates and charged with steel balls. The clinker is reduced to a coarse powder in this portion of the mill and is then discharged into a screening chamber. In the latter the more finely ground material passes through a screen and is led to the second and longer portion of the tube mill, which may be lined either with flint stone or hard cast iron and which is charged with flint pebbles. The tailings from the screening chamber travel back and are delivered through openings at the feeding end of the first chamber.


Fig. 86: Newell's chamber grinding mill.

Messrs. Newells point out that when it is possible to use a mill of this type the horse-power per ton of cement ground is materially reduced and there is also a sensible economy, not only in the foundations, but also in the height of the building, as well as a reduction in the length of the elevator, and only one belt is necessary instead of two.

In the foregoing we have sought to give some idea of the machines in operation at Aberthaw, but we have been forced by reasons of space to pass over, with a bare mention, many pieces of apparatus and devices which are employed. It may be here repeated, however, that the whole of the cement-making machinery at these works with the exception of the feeding mechanism to the stone breakers, which was specially designed by Mr. Day, the consulting engineer, the raw stone conveyor, and the clinker conveyor were supplied by Messrs. Newells. Into the question of why the two conveyors were not supplied by this firm we need no go, but we can say this, that it was certainly not because it was unable to make them for its conveyors which are in operation are of first-rate quality and workmanship.

At the risk of being accused of repeating ourselves we feel that we must say again, in conclusion, that the machinery in operation throughout the Aberthaw works—which, with the exceptions we have mentioned, is entirely British—is of such quality that it could not be beaten by any foreign competitor; nor could any foreign machinery turn out better cement. It is interesting to be able to state, moreover, that a record was created in the erection of the works. At the end of January, 1913, the site was absolutely untouched (Note 46), yet the works were in operation and manufacturing cement in the beginning of May, 1914, since which time they have continued to run with uniformly good results.

The remaining article discussed other Newell's equipment not necessarily used at Aberthaw. I have included those parts that are interesting as showing the state of engineering at the time.

SOME ADDITIONAL CEMENT MAKING MACHINES.

In the previous articles, we have described machines made by Ernest Newell and Co., Limited, Misterton, for the manufacture of Portland cement by one particular process and with one set of raw materials, that is to say, the wet process with limestone and shale. It should not be imagined, however, that this firm is only prepared to supply machinery for working on these lines. On the contrary, it is willing and able to equip works for operating on any system and with all types of raw materials, whether these be the soft chalk and clay of the Medway or Thames valleys, the marls of Cambridgeshire and other districts, the limestone deposits of various parts, or, indeed, materials of any description which are suitable for making cement. In many cases, of course, machines which are used in one process are suitable, perhaps with minor modification, for employment in others. Thus, the crushing machinery for hard materials is the same whether the wet or dry process is employed. The grinding machinery of one kind and another is practically identical, and, of course, the kilns are the same no matter whether the raw material fed into them be in the form of slurry, or damped powder (Note 47). It is not our purpose therefore, in what follows to give a detailed description of all the machines with their modifications which are made by Messrs. Newells for the various processes. It, is sufficient to say that the firm quite well understands how to adapt the different machines we have already described for any particular purpose.

There are, however, certain other machines constructed by this firm to which reference has not yet been made. Some of these are for special employment in the wet process, some for use in the dry process, while the others are suitable for either. We shall describe these machines later in this article, but before doing so we propose to set out the views held by Messrs. Newells (Note 48) regarding the comparative applications of them wet and the dry processes. This information the firm has courteously communicated to us, as it has everything else for which we have asked, and it is of particular interest as representing the mature decision of experts who have for years past been giving special attention to Portland cement making as carried out both in this country and abroad.

Except in cases in which it is impossible to obtain a supply of water in sufficient quantities (Note 49) and at a reasonable cost, Messrs. Newells always recommend the wet process. They point out that even in the United States and Germany, where the dry process is extensively employed—largely because the raw materials are easy to deal with by that process, since they are generally more uniform in character than they are in this country—the wet process is slowly but surely making headway, and several of the largest and latest works have been equipped for working on the latter system. Messrs. Newells argue that, whilst it is doubtless true that in the wet process heat is lost in driving off the 40 per cent. or so of moisture contained in the slurry, the loss thus occasioned is to some extent counterbalanced by the fact that, in the dry process, the raw materials must themselves be dried before they are ground and that fuel must be used in doing this. There is, moreover, the amount of water used for damping the material before it is fed into the kiln, which has also to be driven off. Probably, therefore, there is but little to choose between the two processes when regarded from this point of view (Note 50). In the dry process, however, the problem of mixing the raw materials so that they may be perfectly combined is exceedingly difficult. Messrs. Newells affirm, and, in our opinion, rightly, that however efficient the mixing plant may be, dry materials can never be mingled as intimately as they can be when made into a slurry, and they also maintain that should any " doctoring " be required it is much easier to add the necessary quantity of either of the raw constituents to slurry than to a dry mixture, so as to be certain that there is intimate association (Note 51).

This opinion, it may be said, is shared by many of our cement makers, though it is only fair to add that there are others who hold contrary opinions, and that some very large works on the dry system, in which, we understand. some excellent results are being obtained, have recently been put in operation (Note 52). In fact, as in so many other departments of cement manufacture, a good deal depends upon the taste and fancy of the designer or manager of the works (Note 53); some like one thing and some another. Those who have had a larger experience with the dry process than with the wet naturally understand it better and would probably get better results from it. Messrs. Newells are perfectly alive to this fact, and when necessity—such, for example, as a scarcity of water or the desire of a customer—requires the erection of plant to work on the dry system, they are quite prepared to equip it. At the present moment the firm, as a fact, is engaged in designing a works for India in which the raw materials, after being dried, will be ground in ball and tube mills, then mixed in double-bladed mixers, and damped before being fed into the kilns. The firm is also prepared to quote for briquetting machines and for shaft kilns.

In the dry process the material must first of all be broken into small pieces. For this purpose a jaw or gyratory stone breaker may be necessary, or the raw material may be of surfs size that only crushing rolls may be required. In either case one or other or a combination of some of the machines already described and illustrated in Figs. 69 to 73 ante inclusive may be employed. Then the raw material must be dried, and for this purpose the firm supplies rotary dryers very much of the same pattern as the rotary coal dryer illustrated in Fig. 79 ante. It is, however, of the single drum and not of the double drum type, and the hot gases may be made to pass either inside or outside the drum—preferably the former. Arrangements may be made for utilising the hot gases from the kilns or from a separate source. For grinding, ball mills followed by tube mills are recommended, and these may be practically identical with the machines made by the firm and already described. After this point the methods of procedure in the wet and dry processes are quite different . In the wet process, it will be remembered, the slurry from the tube mills is taken to the mixers, where it is kept continually agitated by the revolving stirrers, and any adjustment necessary in the chemical composition made. In the dry process the usual method is to deliver the raw ground meal into storage silos, and to draw off from the hoppered bottom of these by means of an extracting screw arrangement.

This apparatus works virtually on the same system whether it is used for powdered coal, as for kiln-firing purposes, for raw meal, or for unground but broken material, the arrangements being modified to suit each particular operation. Fig. 88 shows the general arrangement of 8in. diameter twin extracting screws specially designed for dealing with raw meal. The extractor consists of a cast iron box for attachment to the underside of a conical storage bin. It is furnished with two cast iron pipe extensions in which twin-plate spirals rotate and deliver the material from the bin to the outlet. It will be readily understood that unless the spirals are revolving no material can be delivered from the bin, and that the amount extracted will vary with the speed of revolution of the spirals (Note 54). One of the latter is arranged with an extension for the keying on of a belt pulley, which is driven from a variable speeding device, and the two shafts are coupled together at the ends remote from the pulley by means of equal-sized coupling spur wheels. It will be seen, therefore, that the two spirals revolve in opposite directions. To mention one of the modifications of this apparatus. it may be said that, when coal is the material being handled the spirals are often formed of cast iron segments mounted on a square shaft.


Fig. 88: 3-inch twin extracting screws.

When dealing with raw meal it is frequently the custom to arrange that the extractors shall remove from the silos more material than is fed into the kiln. The surplus is then taken by an elevator and re-delivered into the storage silos. This is done in order that the material may be the better mixed (Note 55). The result is that material from the bottom of the bins or silos is always being added to that which is at the top so that it mingles with the newly-ground meal coming from the grinders. This method is practically the same in principle as that employed in the wet process, in which more slurry than is required is pumped to the kiln-feeding platform, where it enters a tank with it fixed overflow, the surplus flowing back into the mixers. Of course, a good deal of mixing goes on as the dried raw material passes through the ball and tube mills (Note 56), but this cannot be entirely relied on, and hence some additional mixing arrangement becomes necessary, since the importance of adequate mixing cannot be too strongly insisted upon, and it is impossible to give too much attention to it (Note 57).

An installation of raw meal mixers and storage bins, in which this principle of mixing as arranged by Messrs. Newells is shown, is given in Fig. 89. In this case there are three storage bins, which are some 30ft. high and which measure 13ft. 6in. by l0ft. 6in. in cross section. They are, as will be seen, constructed of wood. The raw meal is brought from the grinding mills by the conveyor A, which discharges into an elevator B. The latter raises the material to the top of the bins and delivers it into a spiral conveyor C, which transfers it into at further spiral conveyor D arranged over the centres of the three bins and having outlets through which the material can be discharged into each or any of the bins. The bottoms of the bins are hoppered and in the constricted part of each is fixed a pair of revolvable wooden rollers arranged horizontally and parallel to one another. The rollers are 2f t. in diameter and 13ft. 5in. long, so that there is only a clearance of a quarter of an inch between their ends and the sides of the bins, since the latter are 13ft. 6in. wide. There is a space of 2in. between the peripheries of the rollers and when the latter are revolved material is extracted from the bins and falls in each case into a spiral conveyor E, which carries it to an elevator F. By this the material is again lifted to the top of the bins and is discharged into a spiral conveyor G. The latter can either deliver into another spiral conveyor H, which takes the meal to damping mixers, where water is added before the material is fed into the kiln, or by means of an outlet at the point where the conveyor G passes above the top of the conveyor D, the material may be again delivered to the latter conveyor and hence into any one of the bins to be again circulated. If necessary, too, other connections can be made by which each conveyor G can return the material direct to the bin it passes over (Note 58).


Fig. 89: Raw meal storing and mixing silos.

The object of silos and mixing arrangements such as those just outlined is to keep the meal in a constant state of circulation and hence to have the whole body of material of the same composition throughout. Moreover, by storing large quantities of raw meal opportunity is given to correct or " doctor " the mixture to keep it uniform.

It will be understood that each bin has its own extracting arrangements, elevator and conveyors, so that it is possible for the contents of one bin to be (a) transferred to and mixed with the contents of any other bin; or (b) continually kept in circulation by being lifted from the bottom to the top. In some of their plants Messrs. Newells carry this idea still further by introducing a return conveyor, by means of which the meal can pass through the complete series of bins and back again from the starting-point. It will be observed that by means of the arrangement illustrated it is possible either to use one or all the bins together to supply the kiln, or, if necessary, one can be filling, another emptying and another simply circulating (Note 59). A plant such as this will ensure as perfect a mixture of the ingredients in the raw material as it is possible to obtain in the dry process.

When mixed and passed by the chemist as being correct in composition, the raw meal is ready for firing in the kiln. Whether it is to be burnt in a rotary kiln or in a shaft kiln a certain amount of water has to be added to it to damp it. The amount so added should be only just sufficient to make the particles of material adhere together and to render the mixture somewhat plastic. For this purpose Messrs. Newells supply what they term their mixing and damping worm, an example of which is shown in Fig. 90. The machine consists of a steel plate trough, in which rotates a square shaft having secured to it a series of steel knives or arms, which propel the material from the feeding to the delivery end of the trough. The water is supplied, through a perforated pipe, and in the case of a rotary kiln just enough water is added to prevent the material from being blown out of the kiln as dust with the draught. The trough may be some 12ft. or 13ft. long and a certain amount of additional mixing of the material takes place during its passage through the machine. The delivery may be either direct, to the feeding hopper of the rotary kiln or, if the vertical kiln be employed, to the feeding hopper of the briquetting press.


Fig. 90: Damping and mixing worm.

The subsequent treatment of clinker made on the dry process is precisely the same as in the wet process, though, if the vertical kiln be used, the lumps of clinker will have to be broken up in crushing rolls such as are shown in Fig. 73 ante.

There follows a description of a Carr disintegrator (see The Engineer , 18/6/1915, p 597)

In previous articles we have referred to the wash mills used for dealing in the wet and semi-wet—or thick slurry—processes, with the softer varieties of raw materials for cement making. Messrs. Newells supply wash mills of various sizes for different outputs, and the general arrangement is shown in section in Fig. 93, while a plan view to a reduced scale is given in Fig. 94. The illustrations represent what is known as the firm's 18ft. mill. In the particular mill shown the harrow frames are suspended from arms secured to a vertical shaft, but another design is supplied in which the tynes are held in a frame rigidly secured to the shaft.


Fig. 93: Section through washmill.


Fig. 94: Plan of washmill.

The pit in which the harrows revolve is of brickwork or concrete and Messrs. Newells recommend that it should be made octagonal in plan, as shown, as giving a better grinding effect since that shape favours the formation of eddy currents and prevents the tendency which there is, where the pit is circular, for the material to be driven round in a regular flow. A dumpling is built up in the centre of the pit and on it is mounted a foot step bearing for carrying the bottom of the vertical shaft. Cross girders carried on pillars from the side walls support the upper end of the shaft and the driving bevel gear. A short distance above the foot-step bearing a casting is secured by bolts to the vertical shaft, and to it is in turn secured horizontal arms which extend nearly across the full diameter of the pit. These arms are of mild steel channels arranged back to back and may be either two, three or four in number. They are rendered more rigid by the aid of adjustable tie rods. From the arms are suspended by chains cast iron or cast steel harrow frames to which are secured tynes of flat or square section, and either made of mild steel, or, if desired, of manganese steel. The tynes are fixed to the frames in such a manner as to allow of their being lowered as they wear or of any one of them being removed for renewal. Gratings or screens are arranged in the walls of the pit and through these the slurry, as it is formed, passes to be dealt with in other machines for final grinding. For the latter purpose Messrs. Newells recommend the use of tube mills.

Wash mills are arranged in groups of any number, depending on the output required, and are usually fitted up as coarse or fine mills, according to the size of the grids with which they are provided. The power required to drive them varies, of course, with the size and number of harrows employed; but, roughly speaking, with a group of three washmills for an output of 500 tons of finished cement per week, when working night and day, each mill will require about 15 brake horse-power. It will be observed that in Messrs. Newells' design fast and loose pulleys are arranged on the main driving shaft, and that the ratio of reduction between the latter and the vertical shaft is as 6 to 1.

There follows a description of a tyre-borne wet ball mill (see The Engineer , 18/6/1915, p 597)

As an adjunct to their cement or raw meal grinding plants Messrs. Newells are sometimes called upon to supply air separators (Note 60). With such machines it is possible to dispense with secondary or fine grinding machines. The firm makes air separators of various sizes, the largest being 9ft. in diameter. The construction and working of the machine is well shown in Fig. 96. The principle on which it works is akin to that employed in the separation of chaff from grain in a corn dresser. In the latter case the corn to be dressed is thrown by lifters against a forced air draught. The berries being heavier than the chaff pass through the draught, but the chaff is caught and carried away by the latter and perfect separation results. There is practically the same action in the air separator. The material to be sifted—for that is what it amounts to though no sieves are used—is fed into the machine at the centre of the top cover and falls on to a quickly revolving plate, where, by centrifugal force it is dispersed in a fine spray equally over the edge. In falling it meets a strong draught of air, which is created by the fan which, as will be observed, forms part of the machine and is mounted on the same spindle as the revolving shaft. This draught carries the lighter and consequently finer particles along with it into an outer annular chamber, in which they are deposited as the velocity of the air decreases owing to the area of the chamber being much larger than that of the orifice from which the air first issues, whereas, the heavier and consequently coarser particles sink into the inner circular chamber. These two chambers are, of course, separate and each has its own outlet from which the " fine meal " and the " tailings " may be removed from their respective outlets. The air draught of the fan creates a constant circulation of air to the outer chamber and back through the stream of material to the fan. The degree of fineness of the product may be regulated by altering the volume and velocity of the air current, which is controlled by means of a damper inside the machine and by an adjusting device for the fan plates (Note 61).


Fig. 96: Section of air separator.

The output of machines of this class depends upon the size of the apparatus, the nature of the material, the condition in which it leaves the grinding machine, and on the fineness of the product required. It can reach 5 tons per hour. The wear and the driving power of the machine are both small. The 9ft. apparatus in full work dealing with Portland cement will take about four horse-power.

It will have been seen from the foregoing that Messrs. Newells' productions are of the most varied description and that the firm is quite able, as it claims to be, to install plant for Portland cement works operating on any system. We may add that, from personal experience, we are able to say that the firm's machines are of first-class workmanship; and that the reports reaching us from cement manufacturers using them have been uniformly good.

Notes

Note 1. Newell supplied equipment for other mineral processing industries.

Note 2. The comparison with Lehigh cement rock is somewhat absurd, and rendered all the more so by the lack of separate alumina data. The alumina would in fact have been about 1% in the limestone and 7% in the clay. Here, as elsewhere in analyses of the period, "alkali and loss" is not a determined quantity - it is just a balancing term. It should not be imagined that the material was low in alkalis. W A Brown had worked for some time in the USA, and was keen to promote American methods.

Note 3. All other Blue Lias plants at this time were quarrying by hand, manually separating the layers. Eventually all the surviving Lias plants changed to the Aberthaw method, crushing the mixed rock and scalping out the softer clay fractions.

Note 4. Ruston Proctor was founded in 1857, making locomotives and agricultural machinery. It merged with Richard Hornsby & Sons in 1918, and was in 1930 taken over by Bucyrus-Erie of Ohio, forming Ruston-Bucyrus, and was the main producer of excavating equipment in the first half of the 20th century.

Pictures of the Aberthaw excavator - the largest of Ruston Proctor's range, and certainly the largest in the UK cement industry at the time. Many plants - even "modern" ones - were still quarrying by hand. These appeared in The Engineer, CXIX, 9 April 1915, p 358 and Supplement, and were described on p 362.

Note 5. Brown was a big fan of blasting, and this is an early example of its modern use, essentially as the first crushing stage. He also maintained a detailed system for measurement and control of departmental costs, and could quote these data with some precision - after less than a year's run! The two costs mentioned are £2.04 and £5.56 in 2015 money.

Note 6. John Wyndham Beynon was the major shareholder of the company, and the family firm, T. Beynon & Co., owned a number of collieries, mainly in Monmouthshire. However, the technical sponsors of the new plant were a Thames-side consortium, headed by Philip Tolhurst. The Directors were listed in the Western Mail, and were:

It is interesting to note that, during the General Strike in 1926, the plant imported coal from Spain (Liverpool Journal of Commerce, 20/10.1926, p 5).

Note 7. The Pontyclun-Aberthaw branch of the Taff Vale Railway was opened in 1892, having been built on the speculation that a port would be opened at Aberthaw, allowing coal from throughout the coalfield to be exported from there. The Vale of Glamorgan Railway was opened in 1897, and connected the Ogmore and Llynfi valley collieries to the port of Barry. Although the building of docks at Aberthaw might still have been considered a possibility in 1914, the site was upstaged by the construction of Barry Docks in 1889-1914, and although marked on maps as the "Port of Aberthaw", no dock was ever constructed. Given that the tidal range at Aberthaw exceeds 12 m (39 ft), a locked dock is essential for serious shipping. Nothing larger than a sailing barge might have taken cement, and any cement despatched by the plant by sea went from Barry or Penarth.

Note 8. The "River Kenson" mentioned is in fact the artificially constructed leet of the Burton Wood Mill, which functioned on the plant site until rendered unserviceable by the construction of the Cowbridge-Aberthaw railway in 1892. Since the leet originates 300 m upstream, above the spring tide limit, it always yields fresh water.

Note 9. Usually abbreviated to ASEA, Sweden's biggest electrical firm founded 1883. In 1988 it merged with Brown Boveri to form ASEA Brown Boveri (ABB).

Note 10. This is a rather clumsy description of a traveling belt tripper discharge.

Note 11. Only if the composition of the stone in each compartment is known. Since each rawmill was fed with a single feeder, control of the composition of their feed hoppers was the last chance to correct the mix. The only control over the chemistry of the stone was the riddling process before the crushers. It may be that one or more compartments contained recycled shale ("marl").

Note 12. Feed Tables

Feed Table. The feedrate can be varied by raising the collar, increasing the width of the conical pile of material; by moving the plough ("knife") to intercept more of the pile; or by changing the speed of rotation of the table. A later (1930s) and larger (40 t/h) feed table on a wet rawmill. Feed tables were standard for feeding ball mills from the 1890s until WWII, when belt weighers became available.

Note that there was only one feeder per mill at Aberthaw, and the only function was to control the throughput (and therefore fineness) of the material. The right hand picture shows an installation with two feeders - one for limestone and one for shale, so that the slurry composition can be controlled.

Note 13. Around 85-90 μm. This represents a fine slurry, something of an overkill with this raw material.

Note 14. Slurry was usually allowed to run down sloping troughs ("flumes") to a pump in a pit, and the use of a screw here seems redundant, especially since the slurry was evidently very thin.

Note 15. From the plan, 12'6" diameter and 20' deep - 13' cylindrical and 7' in the cone. The casual dismissal of these tanks would cause raised eyebrows in a Technical Survey! The size perhaps provides the reason that they were not in use. 80 t of slurry makes 29 t clinker. At a clinker production of 330 t/d, this represents just over 2 hours run. During times when the rawmills were running, each tank would probably have to be turned around in an hour. As pointed out in Note 11 above, the composition of the rawmix was controlled at the crusher silo extraction points, several hours previously, so it was quite impossible to make a sensible batch control decision on a one-hour timescale. They needed to be at least three times this size.

Note 16. Only if the viscous properties of the slurry remain constant, which is never the case. This feeder principle was obsolete by now.

Note 17. The plan shows no gypsum silo and gives no indication how this was done. Since only around 300 kg/h of gypsum was needed, it might just have been a man with a shovel and a wheelbarrow.

Note 18. This is an insane angle. Cement on a belt easily becomes fluidised, and will slide back down a slope. For cement, a 12° slope is about the limit.

Note 19. Details of Brown's production reporting scheme are to be found in his book.

Note 20. The moisture content of the slurry is nowhere stated, but from the capacities of the slurry tanks given, it can be assumed that a very thin slurry with about 42% H2O was being produced. This means about 1.1 tons of water was required for each ton of clinker made.

Note 21. Octavian Julius Croft Corelli, b 8/1/1886, Hampstead. Went on to manage plants in Gwalior and Coimbatore, India.

Note 22. This low value is confirmed by other analyses of the time: clearly less than 1% gypsum was being added.

Note 23. As regards kilns, the only APCM installations were the three at Burham and one at Vectis, and for BPCM, Kiln 4 at Johnsons, although Jarrow and Mitcheldean both had kilns already installed when acquired by BPCM. The Canadian plant was Bamberton on Vancouver Island. The majority of plant supplied was for firms outside The Combine. Probably their biggest project was the six Burham-sized kilns for Poliet & Chausson at Gargenville on the banks of the Seine near Paris, in 1924.

Note 24. During this period, Newells supplied eight kilns. Edgar Allen supplied two, and FLS (Danish) supplied five. Among German firms Fellner & Ziegler supplied one, Krupp supplied three, Pfeiffer supplied two, and Polysius supplied two. Obviously orders on German firms dropped off after mid-1914.

Note 25. One might suggest with historical hindsight that "falling in" with their customers' eccentric predilections was the main cause of the extinction of the British equipment supply industry.

Note 26. Although flint pebbles are harder than steel, and much cheaper, they have the disadvantage of a lower density (2650 compared with 7000 for steel), so a mill of given volume produces correspondingly less grinding effect. The early use of flint (which died out after WWI) was mainly because small cast steel balls were at the time not readily available.

Note 27. My calculation, assuming 28.3% volume loading and 70% critical speed. With the ball mill (DAWN power 164 HP) the pair should draw 333 HP - the motor power is 350 HP. The much lower powers quoted imply that the larger diameter mills were run much slower, and that optimum speed was mistakenly assumed to be inversely proportional to diameter.

Note 28. In view of the fact that there were only two such firms, and that nearly all the kilns were supplied by Newells, it is interesting to speculate which of these were "troublous".

Note 29. Eight kilns of 200 ft or slightly longer had been installed before those at Aberthaw, but FLS had installed a 70 m kiln at Wouldham as early as 1912, and a second one was put in at Penarth nearby at about the same time as the Aberthaw kilns. Many more 200' by 9' kilns were installed after the war.

Note 30. The diameters are external: the internal diameters, as normally quoted, are 9'10½" and 8'10½".

Note 31. It was common to have some sort of lifter arrangement in wet process kilns, to improve heat exchange. They had the disadvantage that, if they worked, the dust loss increased considerably, but they would usually simply block up and have no effect. Such arrangements were immediately discarded when chain heat exchange systems were introduced in the late 1920s.

Note 32. This was an odd choice. Thrust rollers were usually placed on the tyre nearest the drive (the third tyre in the case of Aberthaw) in order to keep the spur ring in mesh as the kiln expanded.

Note 33. See the article on early rotary kilns. The Mitcheldean kiln did not operate, and so discussion of its performance is somewhat specious.

Note 34. Newells subsequently installed another eight kilns in the UK, the last being Ellesmere Port No. 6 in 1927, and none of these exceeded 202' long, although longer kilns may have been installed abroad.

Note 35. Once again these diameters are external: internal diameter 6'5" and 7'5".

Note 36. The formation of lumps is in general unrelated to conditions in the drying zone of the kiln. See also Note 31 above.

Note 37. The coolers installed at Mitcheldean and Jarrow in 1908 were both 3'6" × 50', while the earlier Kirtlington cooler was 3'0" × 30'. The only other small kiln (1911) was Rugby No 1, which had a concentric cooler.

Note 38. This depends on the nature of the coal. It can be assumed that Aberthaw was using a fairly high rank coal with low inherent moisture content.

Note 39. My calculation. The mills run around 83% critical speed. The volume loading averages only 5.7% - it may be that a higher loading was not possible with this peripheral discharge design.

Note 40. 5% at 100 μm: a reasonable fineness for a high rank coal.

Note 41. The placement of the fine coal hopper and feeder almost on top of the kiln hood was extremely dangerous, and led to a much-publicised fatal accident at Aberthaw during the commissioning of Kiln 1 (26/5/1914). Subsequent designs placed the hoppers further from the kiln. The accident occurred when attempting to unblock the twin screws. To do this, a slide must be inserted above the screw to hold the coal back while the screw casing is split open. The slide was not quite fully in, and fluidised coal dust flooded both the access platform and firing floor, and instantly ignited as it entered the hood. Six workers' escape routes were cut off and they had to jump through flames to ground level.

Note 42. The combination of twin screws and venturi injector was a feature of coal firing systems from the Hurry and Seaman patents.

Note 43. It's interesting to note that, although these were the largest mills on offer, they are exactly the same as mills supplied to Johnsons seven years earlier. Two such mills were provided for an installation with about half the thoughput. So although Newells were trying to give the impression of continuous development, nothing had happened in the intervening period, and the Aberthaw mills were "old technology".

Note 44. The main reason for two-stage milling was that the engineering of mills was still in its infancy, and many structural problems had still to be solved, while good small media was still unavailable. The use of flint in the second stage meant that it was necessary to screen the first-stage product, since the flint would not grind oversize particles. However, by taking advantage of the two-stage system, water injection could treat the symptoms of another unsolved problem - the occasional or continuous production of high-free-lime clinker. Manufacturers remained completely in the dark about this because the measurement of free lime content had yet to be developed, and they could only react to sudden appearance of expansion in the finished product. Hydration of free lime with water addition to the crushed clinker went some way towards ameliorating the problem.

Note 45. The mill pictured looks very much like the Polysius Solomill.

Note 46. The plant celebrated its centenary of operation on 21 July 2012. Time elapses slightly faster in South Glamorgan.

Note 47. The fact that kilns were still mostly short and with little heat exchange at the cold end accounts for the lack of differentiation between the wet and dry processes: the hot gas from the calcining zone was either cooled with slurry in the wet process, or wasted up the stack in the dry process.

Note 48. "We", here, is probably Bertram Blount, and it is never entirely clear whether the views expressed are those of Aberthaw, of Newells, or his own.

Note 49. This was the reason given when the first modern dry process plant - Cauldon - was built.

Note 50. This stupid argument was still being wheeled out in the 1970s, as an excuse for not installing efficient kilns. Compared with the "40% or so" (but often much more) in the slurry, the 5-10% in the raw material or the damped meal was minuscule.

Note 51. The admission that dry process blending was poor or non-existent was a much more persuasive argument for preferring the wet process, but emphasis on this had to be tempered by the fact that new dry process plants were still being built, and these potential customers had to be mollified.

Note 52. He is talking about Ellesmere Port. Previous dry process plants (Norman, Kirtlington, Southam, Premier) all had argillaceous limestone raw materials, and even with poor blending, the chemical variability of the mix was manageable. At Ellesmere Port, on the other hand, relatively pure raw materials were selected, and rawmix variability undoubtedly caused major operation difficulties.

Note 53. In the English of 1915, "fancy" was practically synonymous with "fantasy". In this respect, little has changed in the last 100 years. The customer, whether fantasist or not, is always right.

Note 54. With all the materials handled, the common problem in this system was the tendency of the material to pack into the neck of the silo, causing it to "bridge over", and cease to flow, or to pack into the screw, forming a rotating plug, and again stopping flow.

Note 55. This is really not the case. The power required to move material around was more-or-less proportional to the amount moved, and operators would always try to minimise this. The real reason for moving an excess amount was because the rate of flow (whether in dry or wet process, but much more in the former) is variable, and the extraction rate is set so that, even at lowest flowrate, there is still usually enough to keep the kilns fed.

Note 56. Since the residence time of the material in such a grinding system is only 15-20 minutes, the mills can only blend out variations during that time scale, but systems were indeed designed on the assumption that this was sufficient - see, for example, Premier.

Note 57. No, it isn't!

Note 58. As mentioned in the article on raw material preparation, simple recirculation of material to the top of a bin produces no blending effect, because the stream of material "rat-holes" through the bin.

Note 59. The considerable flexibility of operation suggested could be seen as a way of dealing with eventualities as they arose (which they no doubt rather frequently did), or perhaps as a recipe for total chaos.

Note 60. I know of no air separator classifier installed at this time in British cement plants using ball mills. The Askham air separator was patented as early as 1885 and was used at a number of locations using flat-stone grinding to make finer grades of cement. It was first used in conjunction with a ball mill for cement grinding in Germany in the 1890s. However, in Britain, they were accepted only slowly on modern grinding systems, beginning in the 1930s, and were still uncommon in the 1960s, although they are now considered indispensible in grinding circuits. A classifier promptly removes the fines from a mill system: if allowed to remain, the fines "over-grind" (get finer than they need to be) and by padding-out the space between the coarser particles, make grinding less efficient. An explanation for reluctance to accept such sensible technology - at least in 1915 - is that after half a century of very coarse grinding, the introduction of the tube mill had so revolutionised grinding that it was felt that no further improvement was necessary.

Note 61. He neglects to mention that the tailings are returned to the mill inlet for further grinding.