John Ramsbottom


On an improved locomotive boiler. Proc. Instn. Mech. Engrs., 1849, 1, 1-11 (25 July)
Observations on blast pipe size and on great increase in heating surface and a corresponding increase in flue area.

Coking crane for supplying locomotive engines. Proc. Instn. Mech. Engrs., 1853, 4, 122-5.
A device to assist the loading of tenders with coke.

An improved piston. Proc. Instn. Mech. Engrs., 1854, 5, 70-4.
Steam leakage had been a problem and Ramsbottom devised a means for sealing the piston against the cylinder wall without increasing friction. Ramsbottom's arrangement comprised a piston with three grooves. Packing rings of brass, steel or iron could be inserted into the grooves, but the important aspect was that the rings had to be made such that they would spring outwards by a small amount. It was that spring which caused them to press against the cylinder wall and so effectively seal the piston. Over the intervening years nothing better has been found; although several variations have been developed, they are still fundamentally Ramsbottom rings. By the time he presented the paper, the first pistons as described had been at work for 16 months without any trouble and 15 other locomotives had been similarly fitted. Ramsbottom considered that a set of rings would last for 3,000 to 4,000 miles.

On the construction of packing rings for pistons. Proc. Instn. Mech. Engrs., 1855, 6, 206-08.

An improved safety valve. Proc. Instn. Mech. Engrs., 1856, 37.
The tamper-free duplex safety valve. It was not unknown for locomotive drivers to load their safety valves in order to obtain increased boiler pressure so that they could make up lost time. Such practice was dangerous and a number of boiler explosions were attributed to it. Ramsbottom's safety valve design prevented any loading which would result in an increase of boiler pressure, but did allow pressure to be released by means of a lever which had contact with both valves.

Apparatus for supplying water to locomotive tenders whilst running. Proc. Instn. Mech. Engrs., 1861, 43.
A description of a method of supplying water to the tenders of locomotive engines whilst running: water troughs

On the improved traversing cranes at Crewe Locomotive Works. Proc. Instn. Mech. Engrs., 1864, 15, 44-58.

On the distribution of weight on the axles of locomotives. Proc. Instn. Mech. Engrs., 1864, 15, 92-119.

On an improved mode of manufacture of steel tyres. Proc. Instn. Mech. Engrs., 1866, 17, 186-98

Description of a 30-ton horizontal duplex hammer. Proc. Instn. Mech. Engrs., 1867, 18, 218-31.

Mechanical ventilation of Liverpool passenger tunnel. Proc. Instn. Mech. Engrs., 1871, 22, 22-35.  ; 66-74; 184-99 + Plates 1-6; 17.
The LNWR left Liverpool via a 2035 yard Tunnel, of mean sectional area of 430 feet2, on an average gradient of 1 in 97. During the thirty-three years that had elapsed since opening this portion of line in 1837 traffic through the tunnel was worked by endless rope and a pair of winding engines at the top of the incline. All trains coming up the tunnel from the Liverpool station were attached to the rope and hauled up by the winding engines; trains in the reverse direction were controlled by the addition of very heavy brake-trucks. Delays occurred through stopping every up train at each mouth of the tunnel, to attach and detach the rope. This caused problems during the excursion season, when trains leaving Liverpool were often so heavily loaded that they were divided into two portions, each portion being hauled up the tunnel separately, and the train re-united at the top of the incline. These delays, together with the increasing requirements of the ordinary traffic, at length induced the directors to determine to remove the rope and winding engines, and to work the tunnel by locomotives in the ordinary manner; but the employment of coal-burning locomotives in a close tunnel nearly It mile long, intimately connected at each end with passenger stations of great importance, was of cotirse impracticable without a thorough and constant artificial ventilation. .

Contributions to other's papers
Siemens, C. William. On Le Chatelier's plan of using counter-pressure steam as a break [sic] in locomotive engines. Proc. Instn Mech Engrs., 1870, 21, 21-36. Disc.: 51-4 + Plates 1-5.
Counter-pressure steam brake:. On the Tredegar and Abergavenny line in South Wales there was a descent of 1000 feet within a distance of only 8½; miles, giving a mean gradient of 1 in 45, and the difficulty of taking the trains down that part of the line was found to be practically even greater than getting them up; and in this case he was looking to the application of the counter-pressure working for surmounting to a great extent the difliculty at present experienced. In applying the counter-pressure steam for ordinary stoppages at stations, the quantity of injection water required to be turned on would of course diminish as the speed became reduced; and he wm prepared therefore to anticipate some difliculty in this application of the plan, as the regulation of the jet would probably require a, nicety of adjustment that waa scarcely to be expected from the ordinary class of engine drivers. Another application mentioned in the paper of the counter-pressure working was for shunting at stations, the regulator being kept open the whole time, and the shunting being effected entirely by the reversing lever ; and in order to carry this plan out, it was necessary not only that the screw reversing gear should be employed, but also that the engine should be fitted with good balanced slide-valves ; otherwise it was certain the men would still continue to shut off the steam before reversing, in order to render the reversing easier. In the experiments on the Buxton line it had been mentioned that the steam had been shut off in order to reverse the engine, and it was-clear the regulator had been used largely in the trials; and the men would not be got to desist from employing it extensively until the reversing was rendered as easy with the steam full on as with it shut off, which it appeared to him could not be readily accomplished.
In regard to the excess of pressure shown above the boiler pressure in the counter-pressure diagram that was exhibited from the experiments on the South Western Railway, the explanation which had been offered, viewing the column of steam in the steam pipe from the regulator to the cylinders as performing the part of a ram, appeared to him to be corroborated by the circumstance of the engine being an outside-cylinder one with separate steam-chests and forked steam-pipe, increasing considerably the distance from the regulator to either cylinder. In an inside-cylinder engine, where the total length of the steam-pipe would be less, and where also the action of one piston might perhaps interfere somewhat with that of the other in forcing the steam back into the boiler, in consequence of the . two cylinders having only a single steam-chest common to them both, it was probable the excess of counter-pressure might not be quite so great as in the diagram shown. Another reason that would account for the high pressure observed in the cylinder was the greater density of the counter-pressure vapour which had to be forced back by the piston into the boiler if it was not boiler steam alone, but a wet vapour largely charged with water from the water jet, which would therefore move more sluggishly through the passages to the boiler. That the momentum of the boiler steam rushing into the cavity of the cylinder would be great enough to produce a considerable rise of pressure in the cylinder above the boiler pressure appeared to him a reasonable supposition; and he remembered hearing a somewhat analogous circumstance, that in gunnery it was it well-known fact that when a charge was not rammed home the strain upon the gun in firing was more severe, producing a greater expansion at the breech.

Patents (non-railway inventions shown with green numbers: following Carpenter's ODNB biography: Ramsbottom's screw reverse appears to be missing). Red number indicates not in Pennie and probably by "another" John Ramsbottom. Purple indicates from Pennie's list

6644: 12 July 1834: Construction of power looms for weaving cotton and other fabrics materials into cloth or other fabrics.
6975: 6 January 1836: Machinery for roving, spinning and doubling cotton and other fabric substances.
12,384: 21 December 1848: Running wheels and turntables; application to shfts or axles driven by steam or other power.
13,781: 22 October 1851: Machinery for measuring and registering the flow of fluids; application to registering the speed of and distance run by vessels in motion.
767/1852. [packing ring for cylinders]
via Dickinson
309: 9 February 1854: Hoist for raising and lowering railway rolling stock and other articles.
408: 21 February 1854: Welding.
322: 12 February 1855: Construction of certain metallic pistons.
451: 1 March 1855: Steam-engines: obtaining motive power more economically.
1299: 7 June 1855: Safety valves, feeding apparatus for steam-boilers.
2700: 30 November 1855: Machinery or apparatus for measuring and registering water and other fluids, and obtaining motive power from the same.
Wrought iron rail chair
1157: 9 May 1859: Machinery for printing fabrics.
1527: 23 June 1860: Supplying the tenders or tanks of locomotive engines with water.
2460: 10 October 1860: An improved mode of lubricating the pistons and valves of steam-engines and other machines activated by steam.
White, John H. Some notes on early railway lubrication Trans Newcomen Soc., 2004, 74, 293-307 states that there is an ealier British Patent of 27 November 1858, but Peter Skellon's Steam locomotive lubrication (p. 57) claims that there is no earlier Patent. Displacement or hydrostatic lubricator.
2252: 12 August 1862: Machinery or apparatus for measuring and registering the flow of water and other fluids.
3372: 17 Decembert 1862: Machinery or apparatus for measuring and registering the flow of water and other fluids.
924: 13 April 1863: Machinery for hammering, rolling and shaping metals.
2790: 10 November 1863: Machinery or apparatus for measuring and registering the flow of water and other fluids, and in obtaining motive power from same.
48: 7 January 1864: Improved machinery and apparatus to be employed in, and improved modes of manufacturing hoops, rails and other articles of cast steel.
3073: 12 December 1864: Improvements in the manufacture of steel and iron, and in the apparatus employed therein.
3171: 21 December 1864: Construction of hydraulic and other engines for obtaining motive power, and for compressing and measuring fluids.
89: 11 January 1865: Improvements in steam hammers and in appartus employed in combination with steam hammers.
375: 10 February 1865: Machinery for rolling and hammering iron and other metals.
736: 16 March 1865: Machinery for rolling and shaping metals.
1425: 25 May 1865: Machinery employed in the manufacture of hoops and tyres.
1975: 31 July 1865: Improvements in the manufacture of hoops and tyres and on the machinery employed therin.
342 7 February 1867 Machinery and apparatus for supporting, moving and forging heavy masses of metal.
386 12 February 1867 Machinery for transferring engines, carriages and waggons from one line of rails to another.
1300 3 May 1867 Construction of turbines and other hydraulic engines or apparatus for obtaining motive-power and for measuring the flow of fluids.
15: 1 January 1868: Improved engines and generators for the same, to be worked by heated air, steam, or other gases or vapours.
2956: 26 September 1868: Apparatus for communicating between the passengers, guard and engine driver of a railway train.
820 18 March 1869 An improved mode of ventilating railway tunnels
1405 7 May 1869 Steam and hydraulic apparatus for pumping, forcing and measuring water, and for raising weights.
1740 4 July 1871: Motive power engines, pumping
1060 11 March 1880: Working expansion valves of engines


John Pennie has produced an excellent biography. John Ramsbottom was born in Todmorden, Lancashire, on 11 September 1814, the son of a cotton spinner who owned the only steam-driven mill in the area. He was educated by local school masters and Baptist ministers. His practical training started at his father's mill, where he was given a lathe and built small working steam engines. He then rebuilt and re-erected the beam engine at his father's mill and invented the weft fork, later adopted universally, which enabled looms to be worked at high speed. He was an active and enthusiastic member of the Todmorden Mechanics' Institute, showing quick perception and ability in all mechanical matters.

In about 1839 he went to Manchester and joined Sharp, Roberts & Co., who had a high reputation as builders of locomotives and cotton spinning machinezy. Here he gained experience in locomotive building and his ability sufficiently impressed Charles F. Beyer, in charge of the locomotive department, so that in 1842 he recommended Ramsbottom as locomotive superintendent of the Manchester and Birmingham Railway, which in 1846 became part of the London and North Western, of which he was then appointed district superintendent of the North Eastern Division.

In 1857 he was promoted to locomotive superintendent of the Northern Division, covering all routes of the LNWR north of Rugby, at Crewe works where in the following year the first of almost 1000 of his 0-6-0 freight locomotives were built. Some of these were still running over sixty years later.

Ramsbottom was a first-rate organizer, a brilliant inventor, and he had an endearing personality. His first important contribution to the locomotive engineering world was the double-beat regulator valve. This was introduced in 1850; it was followed two years later by the split piston ring and the ancestor of the modern mechanical coaling plant; and in 1856 (see Patents above) by the displacement lubricator and the safety valves which bear his name. Towards the end of 1858 appeared the first of his celebrated 'DX' 0-6-0 goods engines, of which altogether 943 were built, representing one of the earliest examples of mass production. They were provided with screw-reversing gear, yet another Ramsbottom invention.

Griffiths tells a story (which he stated is almost "foklore") of a meeting between Bessemer and Ramsbottom in which the former advocated the use of Bessemer steel for rails, to which the response was reply is supposed to have been 'do you wish me to be tried for manslaughter?' And it was only after rigorous testing that Ramsbottom permitted trials of steel rails (earlier steels had been too hard and brittle for use in rails). Following this a Bessemer plant was installed at Crewe Works.

In 1859 he built the first of his 2-2-2 type express locomotives, two of which took part in the railway 'race' between the Eastcoast and Westcoast companies from London to Edinburgh in 1888, and in 1863 he developed the 2-4-0 type for use over more heavily graded routes.

In 1862 Ramsbottom became all-line chief mechanical engineer of the LNWR, Britain's largest railway, with locomotive construction concentrated at Crewe works. His mechanical and organizing abilities were given full scope during the rapid development of these works, where he introduced Bessemer open-hearth steel-making. In 1868 he installed Siemens-Martin furnaces. He reduced the number of individual locomotive types and standardized their components to the greatest possible extent, and new machine tools and equipment ensured accuracy in manufacture and facilitated transfer of components within the works The result was a more than twofold increase in the rate of locomotive production.

He was the inventor of water pick-up troughs laid between the rails, whereby a scoop lowered into the trough from the locomotive or tender allowed additional water to be picked up whilst the train was running, thus making possible much longer non-stop runs. First applied in 1860, this was rapidly extended to most British and some American and French main-line railways. In 1870 locomotive haulage replaced steel-rope haulage of trains up the steep gradient from Liverpool, and Ramsbottom designed ventilating extractor fans to remove foul air and smoke from the tunnel.

Ramsbottom retired in 1871, having been responsible for the construction of 1,035 locomotives at Crewe since his appointment fourteen years previously. In 1871 he retired from the LNWR. In 1883 he became consulting engineer and subsequently a director of the Lancashire and Yorkshire Railway, where he was responsible for the design and construction of their Horwich (Bolton) locomotive works and was chairman of the rolling-stock and locomotive workshop committees. He was also a director of Beyer, Peacock, & Co., locomotive builders, a firm in which his two sons held important positions. He died in 1897, when many of his inventions were still in use on the more modem engines designed by his famous successor, F. W. Webb.

A modest and kindly man, he took a great interest in technical education and was a governor of Owens College, Manchester, where in 1873 he endowed the Ramsbottom scholarship for young men in the locomotive department of LNWR. He was a founder member of the Institution of Mechanical Engineers in 1847 and its president in 1870-1. He was a member of the Institution of Civil Engineers from 1866 and received the honorary degree of M.Eng. from the University of Dublin. Ramsbottom died on 20 May 1897 at his home in Alderley Edge, Cheshire, and was buried in Macclesfield cemetery.

Reed's major account of Crewe Works amplifies the above concise notes:

Ramsbottom was possibly the most remarkable engineer ever to be connected with Crewe works. He was born at Todmorden, the son of a small cotton spinner who owned the first steam-driven mill in the valley. John had little in the way of schooling, being trained by and worked for his father, but he schemed-out and made other machinery for the district, and at the age of 20 took out a patent in conjunction with his uncle, Richard Holt, for the improvement of power looms to weave two pieces of fabric at one time. His first 24 years were spent in his birthplace, after which he went as a journeyman on textile machinery to the Manchester works of Sharp Roberts, where he came under the influence of two outstanding men, Richard Roberts and Charles Beyer. From them probably came his appreciation of accurate measurement and its application to batch production. Eight of his testimonials from residents of the Todmorden area dated 1839 still survive, and everyone mentions the reliability of his character .

At Sharp Roberts he must have been drafted early onto locomotive work, and there he attracted the notice of Beyer, who had the locomotive side in his hands. Some three years later a recommendation from Beyer helped Ramsbottom to get the job of locomotive running foreman, under the engineer, of the newly-opened Manchester & Birmingham Railway, though he had no experience of railways at all. He took up this position in May 1842, and was promoted to take charge of the locomotive and rolling stock department of the MBR at œ170 a year in November 1843 when that section was separated from the chief engineer's department.

He retained that position when the MBR was absorbed into the LNWR, and when the lines between Manchester and Leeds were opened in 1849 the engines also came into Ramsbottom's charge as part of the new North Eastern Division stock - his salary was advanced to £500 a year from £300, and in 1853 was stepped up to £850. The LNWR board thought so well of him that in 1856 he was given charge of the rail rolling mill at Crewe, though his headquarters continued at Longsight, Manchester. During his time on the NED he took out nine patents, including the split piston ring (1855), safety-valve (1856), feed pump, turntables, hoists for raising and lowering locomotives, and a coal hoist.

He was made locomotive superintendent of the combined Northern and North Eastern Divisions of the LNWR in 1857, and in 1862 became the first all-line locomotive superintendent at a salary of £2,000 a year, increased to £3,000 as from 1 January 1864 and to £5,000 from 1 January 1869, a reflection of the greatly increased responsibilities, including the steel plant, he was holding and the amazingly good job he was making of them. These salaries set the pace for the 19th century chiefs at Crewe; Cawkwell, the LNWR general manager from 1858 to 1874, never had more than £3,000 a year, nor did his successor, George Findlay, ever rise to the £5,000 level in the next 18 years. Ramsbottom's salaries also began the enormous gap between the remuneration of No 1 in the locomotive department and those of the various Nos 2, none of whom had more than 20 percent of the chiefs salary. This big gap continued to the end of the LNWR and in later years caused trouble in the line of succession. Much of the work done by Ramsbottom at Crewe is detailed in Chapter 5; extensions of it included the mechanical ventilation of Edge Hill tunnels at Liverpool in 1871 when endless-rope haulage was replaced by locomotive power, the solution being based on a Ramsbottom patent of March 1869.

Largely through the enormous amount of work that he did for over a dozen years, and the strain of accomplishing it to the satisfaction of himself and Richard Moon, he began to wear out in 1870, and the stress became more intense in the summer and autumn by his presidency of the Institution of Mechanical Engineers, though he gave no presidential address, and by the death of the then Crewe works manager. In September 1870 he told the Board he would have to retire at the end of twelve months, and his absences became more prolonged in 1871. For some years after the end of September 1871 the LNWR board retained him as consulting engineer at £1000 a year.

The complete release from the immense day-by-day administrative task and the large extensions of Crewe Works and the locomotive department made the cure. By 1875, at 60 years of age, he was back to full health and vigour, and remained a hale and hearty man with firm lines for another 20 years. He took up outside consulting work, though he was already wealthy from his large salary and patent royalties. Altogether he took out 40 patents, but only three of them came after his retirement from Crewe. Nine were related to steel and steel-making; six, over 14 years, were to do with fluids. The first of them in 1851 resulted from his suggestion to the LNWR board in January 1849 of fitting a cheap apparatus in the nature of a gas meter, to record mileage and speed run by each engine."

In 1883 he was appointed consulting mechanical engineer to the LYR, where in collaboration with Barton Wright, the locomotive superintendent, he planned the new Horwich works and specified its equipment. On the death of Lord Houghton in 1885 he was elected a director of the LYR and became chairman of the locomotive committee, in which position he greatly influenced and supported his old apprentice John Aspinall, who had succeeded Barton Wright. [This period is covered in great depth by H.A.V. Bulleid in his biography of Aspinall). He resigned only in 1896 on account of age. Also in 1883 he became a director of the Beyer Peacock private company, formed some years after the death of Charles Beyer had terminated the original partnership. Here he strengthened the close relations he had maintained with Gorton Foundry from its inception in 1855.

One cannot record that Ramsbottom was a nepotist, and he promoted no one above his capacity; yet he had a strong family feeling and helped his sons, cousins and nephews into jobs, but thereafter left them to make their own way. His sons, John Goodfellow and George Holt, were trained at Gorton Foundry in preference to Crewe from 1880 and 1884. Both remained with Beyer Peacock until February 1900, attaining increasingly responsible positions.Ramsbottom's younger cousins, Frank Holt (1825-93) and Charles Holt (1830-1900) also were in locomotive work. The former began an apprenticeship with Sharp Roberts and Sharp Bros before Ramsbottom left, and later went to Crewe, and while there was put in charge of the South Staffordshire Railway locomotive working as Trevithick's representative while the ND had oversight. Later he was for a short time in India, and then was with Beyer Peacock, R. & W. Hawthorn, and the Midland Railway, and was works manager at Derby until he died. He was credited with the initiation of steam sanding. His brother Charles was works manager at Gorton Foundry from 1877 to 1900. Robert Ramsbottom Lister, son of one of John Ramsbottom's sisters, was taken on as an apprentice at Gorton Foundry in 1869 and remained with Beyer Peacock all his working life, being chief draughtsman 1890-1900 and works manager April 1900. A second nephew, Frederic John Ramsbottom Sutcliffe, became a Crewe apprentice in 1860, and as a result of his experience at the original 'melts' he went into the private steel industry and became well known.

John Ramsbottom's principal characteristic was supreme competence. He was a completely objective man, and though well aware of his worth he was not egotistic. His manner was pleasant; he had no vanity, and no one laughed more heartily than he at Wodehouse's quip at the farewell Crewe dinner in 1871 that the new home of Mr Ramsbottom had hitherto been the property of a Mr Sidebottom and was situated at Broadbottom. Cusack Roney in 1868 described him as "the earnest, persevering, never-tiring John Ramsbottom." He had the ability to pick good subordinates from the choice available, and once he had trained them to his ways and became confident that they could stand the pace, he did not trouble them from day to day, and he saw to it that they had regular increases in salary. From his humble beginnings he rose not only to pre-eminence in the world of mechanical engineering, but also one day entertained the Prince of Wales (later King Edward VII) to luncheon in his house near the works after Edward had made a tour of the shops in January 1866. This was at a time when there was more social protocol than in later years, and before Lord Richard Grosvenor (later Lord Stalbridge) had joined the LNWR board and brought a succession of distinguished visitors to Crewe.

On his retirement Ramsbottom was entertained formally by the LNWR board at Euston Hotel. Among those present at the dinner were the Lord Mayor of London, the Duke of Buckingham, who (as the Marquis of Chandos) had been LNWR chairman 1855-60, a critical period in Ramsbottom's career, Charles Beyer and Sir Joseph Whitworth. In 1873 Ramsbottom donated £1,000 to Owens College, Manchester, to found a two-year scholarship for LNWR locomotive department employees under 21 years. Another event ofhis time at Crewe was the naming ofa new street after him; in 1872 Webb named one of the new express 2-4-0s in his honour. Ramsbottom was a founder member of the Institution of Mechanical Engineers, and when he died at Alderley Edge on 23 May 1897 only two founder members remained: his friend Peter Rothwell Jackson of the Salford Rolling Mills and Richard Williams of Wednesbury. A year or two after his death a memorial window was put into the then new chancel of Christ Church, Crewe. Ramsbottom's estate was proved at the large sum of £144,372, but he made no bequests outside the family, nor was he known to contribute to organised charities. The silver plate given him by the company when he retired was much prized, and he left it specifically as a family heirloom.

In 1860, at the close of his career, John Ramsbottom was awarded an honorary Master of Engineering degree by Dublin University. The tendency of outstanding British locomotive designers to get their engineering degrees after their life's work was finished, and honoris causa too, explains a lot about British locomotive design. Or rather, it explains why Britain fell behind more academically-conscious nations as soon as the pioneering phase was over. Ramsbottom might in fact be described as the last of the great British locomotive engineers. By the time he retired most of the inventions and innovations which could be developed by shrewdness and mechanical intuition had made their appearance, and it was the theoreticians armed with an academic approach who would make the next real advances. England's Churchward was perhaps an exception, but apart from him it was non-British engineers with years of academic study behind them who set the pace: the American Professor Goss, the German Professor Schmidt, and that keen student of thermodynamics, Andre Chapelon. What happened when a theoretician without academic training got into the saddle would be shown by Ramsbottom's successor, Webb, whose pursuit of a valid improvement (compounding), combined with an inability to make use of evidence in a scientific way, led to the construction of hundreds of deficient locomotives on the London & North Western Railway. Ramsbottom was born near Manchester in 1814 and acquired his initial engineering experience working for Sharp, Roberts, & Company; the same firm which had set Charles Beyer on the locomotive-designing road. The early forties were a good period for engineers in search of rapid advancement, and at the age of twenty-eight he secured the job of managing the Longsight Works of the Manchester & Birmingham Railway, a company which soon afterwards became part of one of Britain's major railways, the London & North Western. Thus in 1846 Ramsbottom was distinct locomotive superintendent of the LNWR.' s North Eastern Division. The following year he gained additional distinction by reading a paper on boilers to the first meeting in Birmingham of the newly-founded Institution of Mechanical Engineers. Although he had introduced a new type of regulator (the L.N.W.R. 'double-beat' regulator) in 1850,

Ramsbottom's first major invention was the split piston ring. This was astonishingly successful and was adopted throughout the world. The problem of establishing a reliable and long-lasting steamtight fit between the piston and the cylinder wall had hitherto been unresolved. Ramsbottom's solution was to cut circumferential grooves in the piston, into which were fitted narrow cast-iron split rings. The rings were split to enable them to be opened out when being fitted to the piston. They had to be held in the closed position as the piston was reinserted in the cylinder, after which their natural elasticity held them tight against the cylinder wall, their natural diameter being one tenth greater than the piston. Several rings were fitted to a piston, in order to cope with any slight irregularities in different parts of the cylinder. On trial, Ramsbottom found that his rings could do over three thousand miles before renewal. Moreover, with these rings the weight of the piston could be halved.

In 1856 Ramsbottom introduced his sight-feed lubricator, a boon to LNWR. enginemen, and the safety valve still known throughout the world as the Ramsbottom valve. The latter con- sisted of two parallel pipes which projected vertically from the top of the boiler and, between them, a coiled spring which pulled down a beam on to the two valves at the top of the pipes. These valves lifted equally and were so well balanced that when blowing-off pressure was reached in the boiler the initial reaction of the valves was to vibrate, producing a recognizable hum.

In 1857 Ramsbottom was appointed L.N.W. locomotive superintendent at Crewe. In 1859 he appointed Webb as his chief draughtsman, and it was Webb who was responsible for most of the design work on Ramsbottom's locomotives. Rams- bottom realized that a 2-4-0 passenger and 0-6-0 freight locomotive could handle the bulk of the Railway's traffic, and built large numbers of these types, which continued to be built, with alterations, by his successor Webb. It was these basic 2-4-0 and 0-6-0 types that were still doing much of the work in Edwardian times, Webb's later designs having proved so unsatisfactory. Ramsbottom's standard freight machine (class DX) was produced in numbers very large by contemporary standards; 943 units were built, and Webb based his own 0-6-0 on this design. In general, Ramsbottom did not believe in building engines large enough to handle the biggest loads likely to be encountered. He preferred to design for the average task, with the result that his locomotives were often worked much harder than on other lines. This set up something of a tradition, subsequent LNWR locomotives being liable to severe 'thrashing' by their drivers, yet designed to stand up well to such treatment. His locomotives, while setting a tradition, themselves owed much to the tradition of their predecessors; the 'L.N.W.R. took' had already been established and the simplicity and exterior balance of Ramsbottom's designs did nothing to detract from it. However, he did spoil the appearance of his machines by his fondness for crazy fretwork; the chimney caps of his locomotives had pieces cut out of them for the sake of what Ramsbottom called decoration. The pieces which he cut out of his driving wheel splashers were less hideous, and were sometimes appreciated by enginemen looking for a place to direct their oil-cans. In common with most, but not all, of his contemporaries he offered little comfort to his enginemen. The very successful class of 0-6-0 saddle tank that he designed for yard work had screw reverse: for most purposes this was superior to the conventional lever reverse but, because it needed more time to operate, it was quite unsuited for locomotives whose main work was stopping, starting, and reversing. Again, Ramsbottom refused to provide his enginemen with a roof over their heads at a time when other designers had accepted the possibility that wet men might not work so well as dry men. The truth is that Ramsbottom was a hard man in a hard age. Indeed, his predecessor Trevithick had been pensioned off precisely because he was too kind-hearted. It is likely that Ramsbottom bears the responsibility for the autocratic and repressive tradition which established itself at Crewe in the last decades of the century.

In many ways Ramsbottom's most spectacular invention was the water trough. He liked small and rather flimsy tenders, so when he was faced with an acceleration of the Irish Mail, which was to run non-stop between Chester and Holyhead ( eighty-five miles) in 125 minutes, the idea of a device to pick up water at speed had an obvious attraction. He decided that a long track pan between the rails, with the water regulated to maintain a 5in. depth, would be the best solution. A scoop lowered from the tender would dip 2in. below the water surface, leaving a 3in. clearance for contingencies ( like stone ballast in the trough or weak springing on the tender). Ramsbottom estimated that at 15 m.p.h. the scoop would lift the water 7ft 6in., and on the first trial, in which the vertical pipe from the scoop into the water tank ended 7ft 6in. above water level, he had the satisfac- tion of noting that at 15 m.p.h. the water was propelled to the very top of the pipe, and stayed there, without overflowing into the tender tank. These tests showed that 40 m.p.h. was the best speed for pick-up, 1,150 gallons being taken; faster and slower speeds resulted in a smaller intake. In order that the engine crew might maintain the optimum speed over the quarter-mile trough which he had laid at Conway in North Wales, he fitted a locomotive speedometer (or 'Velocimeter' , as he called it), that he had earlier devised. This consisted of oil inside a glass tube, mounted vertically. The tube was rotated by a cord actuated from the rear axle. As the axle rotated faster, so did the tube, causing the top surface of the oil to become more and more concave. A scale was engraved alongside the tube, by which the bottom of the concave surface marked the speed in m.p.h.

Most of Ramsbottom's period as top man at Crewe was devoted to improving the organization of the works. He introduced strict standardization of parts and of gauges. A Bessemer, and later a Siemens-Marhn, steel plant was introduced, and an 18in. gauge steam railway linked the various parts of the works. Among other innovations were cast iron locomotive wheels with H-section spokes, much used on successive generations of LNW. freight locomotives, a horizontal steam hammer, and various improved tools. He retired in 1871 at the age of fifty-seven, in the belief that his health was failing. In fact he lived another quarter-century, during which he became a director of Beyer, Peacock and of the Lancashire & Yorkshire Railway. He did much of the planning in connection with the latter company's new locomotive works in Horwich.

His two sons reached high positions in Beyer, Peacock. It has already been remarked that Ramsbottom was virtually the last of the great British locomotive engineers, because after him the lack of academic training began to hold back men who might otherwise have made great contributions. There is a lot which can be said against the British tradition of training engineers on the job. Years of rather boring 'experience' on the shop floor must have driven many bright young men away from the railway service, and dulled the brilliance of those who slogged it out. Lack of academic training, other than attendance at night courses at the end of a hard day's work, must have meant that good ideas lacked the theoretical underpinning that could have ensured their success. On the other hand, it is also true that the better British locomotive designers, because of their type of training, were excellent organizers of production. In this sense Ramsbottom is again something of a landmark. His inventions were almost entirely devised before he became locomotive superintendent, and after coming to the top position his most important work was not design and invention but production and administration. In this he resembled most chief mechanical engineers, and for this work the traditional non-academic training of British engineers was probably the most advantageous.

In 1868 Sir Cusack Roney wrote respecting Mr. Ramsbottom (cited Rly Mag., 1899, 5, 232)
"At the head of the mighty establishments at Crewe. . . . is one man who, if he had been in Egypt, with works not a quarter the size and not half so ably carried out, would have been at least a Bey, or more probably a Pacha; in Austria a Count of the Holy Empire; in any other country in the world, except England, with crosses and decorations, the ribbons of which would easily make a charming bonnet of existing dimensions. But in England, the earnest, persevering, never tiring JOHN RAMSOTTOM is – John Ramsbottom."

Rogers (biography of Chapelon): John Ramsbottom of the London & North Western Railway was one of the greatest locomotive engineers of his day. His DX class 0-6-0 goods engines, which were built from 1858 to 1872, numbered more than any other steam locomotive class in British history. The DX was an extremely light engine which could run anywhere on the system and it had the screw reversing gear which Ramsbottom was the first engineer to use. He produced a noteworthy class of 2-4-0 express locomotive which as continued and modified by his successor Webb made the fastest running in the railway races from London to Aberdeen in 1895. Because, at the time these engines first appeared in 1866 many engineers held that only engines with single driving wheels were suitable for fast running, it is likely that Ramsbottom had discovered that a good steam circuit was of greater importance.

Carpenter, George W. biography Oxford Dictionary of National Biography
Pennie, Robin. John Ramsbottom: a Victorian engineering giant. Lancashire & Yorkshire Railway Society, 2007. 96pp
Proceedings of the Instn Mech. Engrs. 1897, 52. 236-41 Obituary.
Min. Proc. Instn Civ. Engrs, 1896/97, 129, 382. Obituary

H.A.V. Bulleid, The Aspinall Era.
Nock, O.S. Railway enthusuast's encyclopedia
Talbot, Edward An illustrated history of LNWR engines. 1985.

W. A. Tuplin, North Western Steam (1963);
Hambleton, F.C.John Ramsbottom: the father of the modern locomotive. 1937.
Ottley 2863: notes that partly a reprint of material in J. Stephenson Loco. Soc., 1937, February. Only 30 pp..
John Ramsbottom.
Locomotive Carr. Wagon Rev.,1941, 47, 143-7; 178-82.
Supplement to earlier work??

Nature of Ramsbottom household (including 6 children - one aged two - the great man was 66) in Alderley Edge from 1881 Census: Backtrack 14, 637.

Updated: 2011-01-06

Home page