Crimson Rambler

In MHO the 7ft - 6ins slide valve Spinners were the best lookers! Crimson Rambler

Spelling names was never my strong point!! Billinton was indeed recomended to the LBSC board by Stroudley sometime before he died. RB had been S's Head Draughtsman - appointed in late 1872. The two engineers enjoyed a deep friendship. Indeed it appears Stroudley persuaded Billinton to go to Derby. Of course Stroudly, SWJ, Drummond and as you point out WMS were all at Cowlairs together in the early 1860s. SWJ as Loco Superintendent, Stroudley as Works Manager, one of the foremen was a young Dugald Drummond. I guess WMS was a draughtsman (carriage or loco?). He went with SWJ to Stratford in 1866 as the GER's carriage & waggon designer. Before he joined SWJ on the E&G he had spent 1½ years at Neilsons! I read in an I Loco E paper that Drummond had to pay royalties to W M Smith for use of cross tubes. Whether that is true or not I don't know - it might be something to explore sometime. However WMS did hold British patent 6118 of 1897. This shewed several arrangements of tubes one of which (or something very similar to it) was fitted to NER No 1619 – his three-cylinder compound that became the prototype for the Midland ones. WMS was very highly thought of in Japan I believe he was given an award by the emperor (but not personally!). He arrived in Japan in 1874 and was appointed the first Loco Superintendent/CME of the Imperial Government Railways at Kobe. He also instructed engineering students. He returned to Britain in 1883. I'm far from being alone in thinking that Japanese locomotive representation Northroader has posted has more than a passing resemblence to a Midland/Smith/Johnson Compound. Certainly more in my view than Smith's later pair of 4-cylinder componds for the NER. The interconnections between nineteenth century locomotive engineers, their working colleagues/railways and their families is quite fascinating. Many years ago, I thought about producing a sort of ‘family tree’ to illustrate the various connexions as I learned them but gave up – this was long before I could use a computer drawing package. I couldn’t determine a layout that would work which would enable the inevitable changes to be made without massive redrawing. Crimson Rambler

One small correction if I may – and its one I regularly used to get wrong when he was alive! There’s only one 'e' in Btraithwaite but two in Tee! As an aside, I think we often overlook the contribution Robert Billington made to Midland (and by default S&DR) locomotives. From November 1874 until January 1890 he was SWJ’s chief locomotive draughtsman – Head Draughtsman he was originally termed. He was the man who interpreted SWJ’s requirements, delegated the work to the various draughtsmen and would have been largely responsible for the finished design together with whatever input the Master made. Thus, during those 15 years he took the ‘rough SWJ’ design such as the Class A 0-6-0T or the 0-6-0s, 4-4-0 and 0-4-4T produced at Stratford and refined them in accordance with SWJ’s wishes into the classic Johnson designs. After all the first of the 4-2-2s with that delightful ‘one-piece’ chimney appeared in the closing hours of his watch. He also would have been responsible for the various rebuilds applied to Kirtley engines. One can idly speculate how Johnson designs might have panned out had W M Smith who was at Stratford with SWJ had not gone to Japan in 1874 and instead joined him in Derby. We might surmize that Thomas Iveson, during his ten years that followed RB’s departure largely refined, honed and where necessary enlarged the existing SWJ designs. His one major piece was the 2606 Class was in effect a 60 Class fitted with a Belpaire boiler. By then J W Smith (W M Smith’s son) had joined the Derby team as a draughtsman arriving I understand when SWJ wanted to adopt Smith piston valves. Following TI’s retirement JWS became Chief Locomotive Drauightsman and the Johnson Edwardian image had its full flowering with the big 4-4-0s. Crimson Rambler

To answer your question, the evidence concerning the introduction of the reverse curve connecting the splashers is as follows:- I was told Neilson introduced it by Jack Braithewaite - the doyen of Midland locomotive aesthetics - I could claim his friendship along with that of David Tee who supported the theory. The curve first appeared on the batch of 6ft - 9ins 2-4-0s Nos 1502-1531 built by Neilsons. These engines were never given a class letter designation but were always referred to as 'Like O.232' i.e. they were similar to the corresponding Derby built engines. An early photo of No. 1527 in as-built condition appears in Summerson Vol 3 p10. Conversely the corresponding Derby built engines were to two GA drawings 79-1194 for Nos 1400-1409 (O.232) while engines Nos 1472-91 (O.273) were to GA 80-1376. As-built photos of the first batch shew a flat section as do similar condition photos of O.273. I have never seen the GA for the Neilson batch - long ago I was told it didn't exist so cannot present absolute proof but GA 80-1376 shews the flat section. If we accept the opinions of two of the most knowledgeable recent Midland locomotive enthusiasts that Neilsons did indeed introduce the reverse curve then it can explain the presence of a flat section on the next batch of Derby built passenger engines - 4-4-0s Nos 1562-71. According to Essery & Jenks (Vol 2 p86), the 1562s were the last Johnson passenger engines built with the flattened centre section. Possibly that piece of information either came from or was confirmed by David Tee. Once Derby adopted the reverse curve it was then applied retrospectively to 'flat topped' engines as they went through the works. Most likely I believe when they were re-boilered (rebuilt in Midland parlance) which possibly explains why it took about 20 years for the last of the 'flatties' to disappear. Incidentally James Johnson resigned from the GNoSR in Aug 1894 to join an engineering firm in the West of England - Scottish Locomotive History 1831-1923 - by Campbell Highet p149. Crimson Rambler

In my opinion the 1102 class represents an example of Johnson building on Kirtley practice. I agree the detailed design work for these engines was carried out after SWJ's arrival mostly I suspect by Neilsons. Incidentally, while the 880 class was undoubtedly the precursor of the 1102s, I suggest much of the design for that class was done by Beyers. As well as being inside framed, because it used the more modern idea of shorter eccentric rods it had all of the motion arranged behind the motion plate although it used Beyer's usual swing links to support the valve spindles. SWJ used similar short eccentric rods on his 'pure' designs whereas MK had hitherto stayed with longer rods. Both 0-6-0T classes had their main frame plates spaced 4ft 2ins apart. This was the standard Kirtley dimension whereas SWJ normally used 4ft - 1.5ins. However this 'non-standard' frame spacing was retained for all the small 0-6-0Ts built subsequently - certainly as far as the Class N of the 1890s. I suspect it also applies to the Class Q but I will need to confirm this, whereas the later big 0-6-0Ts Classes S and U had their frames 4ft 1.5ins apart. Midland locomotive design is a intruiging subject even if at times it resembles a game of three-dimensionan chess! Edward Snowball was a fascinating character and thought nothing it seems of altering railway companies designs if he thought it improved things! Thus he was responsible for introducing the elegant reverse curve between the spashers of Midland 2-4-0s and 4-4-0s. He also seemingly increased the frame depth of the Class D 0-4-4T - the Derby drawing had the top frame finish flush with the top of the platform whereas as built it was proud of it by and inch or so. Crimson Rambler

Two views of the seven special chairs (and 18 keys!) needed for a 1 in 8 diamond crossing. Two chairs are identical - those receiving the nose of the point rail. At least that is the term other companies called it although I'm not sure if the Midland did. From memory the drawings just refer to it as a 'rail'. The two outermost chairs on each side are left- and right-handed pairs of one another, while at the centre there was a chair to support the check and wing rails at the knuckle. Midland outside key 85lbs chairs can be readily identified by the central rib present in the outside jaws. If the separation between diverging rails was sufficient then a rib might also be present. This can be seen in the outermost pair of diamond chairs and the first and last chairs forming the 1 in 5 common crossing. Finally specially for David(!) have started making some more masters to speed up SC production. Crimson Rambler

After a break of around 25 years I have started making some Midland chair patterns and thence moulds. I was quite pleasantly surprized by how much I had remembered and conversely comparatively how little I had forgotten of the techniques and wrinkles needed. Must be because it was for something important - unlike work for example, where I have difficulty retaining overnight what I intended to do the following day. The first of a short series of these S7 chair moulds being the 85lbs outside key 1 in 5 common crossing. This hopefully will please David Hunt of this parish because he needs some for his three-throws - as do I. Two views of the first cast:- Yes, the keys are separate because they have to be - you can't assemble the crossing without loose keys while after assembly they are needed to hold the rails in their correct positions. The 1 in 8 diamond crossing is as advanced while the 1 in 10 common is not too far behind. Then its finish off the 85lbs 15ft sprung heel points. Crimson Rambler

I will set the ball rolling if I may and David I'm sure can add some more later. Incidentally the photo of 0-4-4T No 1720 and 2-4-0 No 131 also appeared in J B Radford's book Derby Works & Midland Locomotives opposite page 49 in slightly larger format. The wagon might be slightly clearer? The tank engine was built at Derby in 1885 - part of batch O.538 which appeared May/June 1885 after which it then spent most of its life at Kentish Town. London engines made only rare visits to Derby. The 2-4-0 according to S Summerson was shedded at Derby in 1892 before being given larger cylinders in the October. During that period in the shops it almost certainly (99.9%) it received a new boiler and new splashers having the reverse curve between them as shewn on the general arrangement drawing for rebuilding the 890s - No 86-2348. Thus this suggests a possible period for the photograph extending from mid-1885 to say mid-1892, however the tank engine does not appear to have the 'London style' of painting which one might reasonably expect it to display if it had been sent to Derby for repair. For example No 1722 of this batch displayed this style c.1887. For this reason, the open smokebox and the displaced dome casing, suggests to me the engine is new thereby dating the photo to the summer of 1885. The pity is Summerson is unable to say when No 131 went to Derby only that it was there in 1892. Crimson Rambler

From Friday 10th May 2019 until Sunday 12th May the Midland Railway Society will be holding a small exhibition in the Midland Hotel, Derby to commemorate the 175 years that have passed since the founding of the Midland Railway. There will be displays, items from the Society's Roy Burrow's Collection and it is hoped some models albeit sadly no working layouts. However, several members from the Management Committee of the Society will be in attendance so if you are already a member or are thinking of joining do come along and introduce yourself. If you want to know what is in the Collection or if you would like to see more of these lovely general arrangement drawings appearing on the website now is your chance! Likewise, you can meet up with fellow members while having a chat and a beer in convivial surroundings!

Please, please be careful if you are basing your model on the drawing by Ken Woodhead (Illustrated View of Midland Locomotives Vol II - p102). Drawings by K Woodhead esq, who I knew, have in my opinion a somewhat undeserved reputation for accuracy. In this particular example some of the more obvious ones are the cab controls partly include arrangements used with earlier 4-4-0 classes, the brake hangers should be the single type not the double type drawn and perhaps most important of all the end elevation is misleading because he has overlooked that the frame plates were in two parts which were bolted together in way of the motion plate. This has resulted in odd gaps appearing either side of the smokebox. Crimson Rambler

The main criticisms levelled against Midland axleboxes are that they retained a loose brass design, were ‘under-sized’, ran hot and used a poor method of delivering oil to the journal. While E S Cox may not have approved of loose brass boxes they were fitted to large European engines e.g. Germany post-war. As with short-lap valves, the overwhelming majority of the locomotives running in 1922 had inside cylinders, which necessitated crank axles and therefore limited the length of the axleboxes. Yet it is only Midland locomotives, that are vilified and singled out for ridicule by enthusiasts – in large part thanks to E S Cox. With a track gauge of 4ft - 8½ins and inside cylinders it is difficult to obtain generous axlebox dimensions if at the same time the engine is to have an adequate crank axle. Since the latter was rightly judged more important, designers sacrificed bearing area to minimize the risk of crank axle failure. To demonstrate a flaw in the enthusiasts’ criticism consider the following table:- The main criticisms levelled against Midland axleboxes are that they retained a loose brass design, were ‘under-sized’, ran hot and used a poor method of delivering oil to the journal. While E S Cox may not have approved of loose brass boxes they were fitted to large European engines e.g. Germany post-war. As with short-lap valves, the overwhelming majority of the locomotives running in 1922 had inside cylinders, which necessitated crank axles and therefore limited the length of the axleboxes. Yet it is only Midland locomotives, that are vilified and singled out for ridicule by enthusiasts – in large part thanks to E S Cox. With a track gauge of 4ft - 8½ins and inside cylinders it is difficult to obtain generous axlebox dimensions if at the same time the engine is to have an adequate crank axle. Since the latter was rightly judged more important, designers sacrificed bearing area to minimize the risk of crank axle failure. To demonstrate a flaw in the enthusiasts’ criticism consider the following table:- The figure that is being compared in the final column is the maximum piston thrust divided by the projected area of the bearing (i.e. L × d). This is the accepted method used even when, as in our case, the bearing area was nominally one-half the figure given as only ½ bearings were normally used. Inspection of these maximum loads demonstrate the flaw in enthusiasts’ reasoning since if Midland axleboxes were undersized, then so were those fitted a large number of other inside cylinder locomotives – in some instances to a considerably greater extent. So why – according to E S Cox – did Midland bearings run hot whereas the equally or more heavily loaded bearings fitted to the engines of other railways seemingly did not – or to be more accurate, at least not so frequently? To answer this anomaly satisfactorily we must consider how an axlebox bearing actually worked. We must start by immediately dismissing some widespread misconceptions concerning lubrication. It is commonly believed by many, the underpad lubrication, promoted by Churchward, Stanier et al, resulted in hydrodynamic lubrication. It does not. It cannot. This is because the pad is unable to deliver sufficient oil to the bearing for it to be hydrodynamically lubricated, furthermore, if for one moment we suppose a pad could be devised that could deliver the necessary quantity of oil, then the reservoir capacity of the bearing would be exhausted in a matter of minutes. As an aside, Richard Deeley was fully aware of underpad lubrication – Sir Henry Fowler reported in March 1922:- “When his predecessor came back from America some years ago, he was very delighted with the system of lubrication in which a mixture of waste and horse-hair was used. It was given an extended trial, but there was difficulty from the fact that the horse-hair tended to get into the oil channels and curl up into small balls.” Hydrodynamic lubrication is highly desirable in machinery because then bearing wear rates become insignificant. A car engine, by virtue of its closed circulation pumped lubrication system does operate with hydrodynamically lubricated bearings. If we assume a car travelling in top gear at 40mph, the corresponding engine speed will be around 2,000rpm or 120,000 revolutions per hour. This is equivalent to 3,000 revolutions per mile. Nowadays a life of 100,000 miles is not unusual so this equates to the engine having made a minimum of 300,000,000 revolutions yet the big end and main bearings will exhibit negligible wear. In practice, because the car is driven in lower gears for much of the time this assessment will be a gross underestimate of the engine revolutions. The goal for Cox et al was for the axleboxes to last for 100,000 miles between repairs. This, was more or less achieved latterly, but only in express engines. So, assuming driving wheels 6ft - 9ins diameter, these will make in round figures 250 revolutions in a mile; hence they were aiming for a bearing life of 25,000,000 revolutions or only 1/12 that of a car engine. Furthermore, after completing 100,000 miles the wear in the locomotive's bearings could be measured with a ruler – as well as being heard! Locomotives employed in lower speed traffic never attained 100,000 miles before the equivalent wear appeared maxima of 50,000 or 60,000 miles was more their lot. Ironically, the car engine derives its lubrication system from a design patented in 1890 and 1892 by Albert Pain the Chief Draughtsman of Bellis & Morcom, makers of stationary steam engines. The first example, built in 1890 continued in use until 1919 producing 20 horsepower at 625rpm. Journals and bearings shewed little signs of wear after 29 years of service during which the crankshaft made more than 4,000 million (4 billion) revolutions. To appreciate why locomotive axleboxes did not behave in this way we have to remember the lubrication system was completely different. In the Bellis & Morcom system or a car engine there was a closed-loop oil circulation system, which enabled a high flow of oil to be delivered to, and then pass through, the bearings. It was this sufficiency of oil delivery which enabled the bearing to operate hydrodynamically and thus with negligible metal-to-metal contact. As the oil escaped from each bearing it drained down into the sump before being pumped round again. In direct contrast, in a locomotive axlebox, because the oil was consumed on a total loss basis the bearing could only receive a meagre quantity of oil so that the very limited supply would last the journey. These two approaches may be demonstrated in a Stribeck diagram. ZN/P is a mathematical relationship tying bearing load P, speed N and oil viscosity Z to the coefficient of friction m in the bearing. It is important to note the scale for the friction is logarithmic so each division changes by a factor of ten-fold. At the point of minimum friction, hydrodynamic lubrication has been achieved and bearing wear becomes negligible. the coefficient of friction m is synonymous for bearing wear and also for the running temperature of the bearing since it is the friction that gives rise to both. Locomotive axleboxes operate in the thin-film (or mixed-film) region to the left of the point of minimum friction. The oil pad, irrespective of where it is placed, or a mechanical lubricator, cannot deliver sufficient oil for the bearing to operate hydrodynamically. Instead, because of the far smaller quantity of oil present there was some metallic contact between the journal and the bearing – hence the higher coefficient of friction and the resulting wear. Since the ability the lubricant/metal combination possessed in forming this layer was affected by the metals used in the bearing as well as the choice of lubricant, this gave rise to the need to be able to measure it. Mountford Deeley not only invented a machine for doing just that, but also introduced the term ‘oiliness’ to describe the effectiveness of the oil/metal combination. The machine was designed primarily for the study of the ‘oiliness’ of lubricants and the variation in frictional resistance produced by different oils and metals. This is demonstrated by the fall in the static friction recorded by a Deeley machine as the percentage of rape oil was increased. Eric Langridge observed:- “The bearings that were sufficient for MR days were not equal to the rough and tumble of LMS handling. Quality of materials went down with economic pressure from commercially-minded managers and oil became poor. All this can be false economy from an engineering point of view; cheapness does not really pay. However, I sometimes wonder if critics of bearing sizes ever had a go at designing themselves!” The LMS attempted to save money by substituting cheaper lubricating oil, which caused the bearings to run hotter. This in turn gave rise to another problem, also prompted by its penny-pinching. The Midland along with other pre-grouping companies used for its most highly loaded bearings e.g. locomotive big ends and axleboxes a white metal alloy having a high tin content but which contained no lead. Leaded white metal was reserved for lower loaded bearings. The presence of lead formed a eutectic with the tin, by which is meant that the resulting white metal alloy will melt at a lower temperature than the melting points of the pure constituents. The LMS adopted leaded white metal presumably under the influence of George Hughes who had used leaded white metal on the Lancashire & Yorkshire - yet ironically E S Cox reported in his paper on locomotive axleboxes, the L&YR had had problems with hot ’boxes! As a consequence of these two cost saving exercises by the LMS, not only did the company increase the likelihood of its axleboxes running warmer through substituting a poorer oil, but it then actively encouraged the affected bearings to fail by lining them an alloy that softened and melted at a lower temperature. It should be appreciated that a ‘hot ‘box’ was an ailment, which could afflict all locomotives although those engines whose axleboxes were the most heavily loaded meant they were more sensitive to the factors that could prompt heating viz insufficient lubrication, incorrect oil, badly made trimmings, presence of dirt or sand, misalignment etc, because they had less reserve. During the Second World War, the LNER developed its ‘W’ oil at the request of Sir Nigel Gresley in response to the high number of hot boxes that company experienced under wartime conditions. The LMS adopted ‘W’ oil in 1943 applying it to four important classes fitted with highly loaded bearings viz the Standard 4F 0-6-0 and 7F 0-8-0, plus the ex-LNWR G1 and G2 0-8-0 classes. Before concluding I would like to refer you to a memorandum dated 9th June 1931 and addressed to Syndham Symes - it was reproduced in full in LMS Journal issue N°7. In it the author, one E S Cox, reported that in 1930 the Standard Class 7 0-8-0 suffered 53 hot boxes out of 120 engines, which is the equivalent of one hot box per engine every 27.2 months. Figures for the ex-LNWR G1 and G2 0-8-0 classes, also appeared and although they carried similar sized bearings the difference in their performance was startling. For the Gs class, out of 60 engines there were 22 hot boxes, giving one every 32.7 months – little different from the ‘Austin Sevens’. However, it is the G1, which is the most revealing, out of 401 engines, there were 40 hot boxes. The equivalent of one hot box per engine every ten years, which put these inside cylinder engines on a par with the 1939 figures E S Cox proudly stated (and others have repeated) was the mean performance of all of the taper-boiler classes plus the ‘Royal Scots’, which by then also carried Stanier ’boxes. In other words, it was perfectly possible to obtain what LMS engineers considered to be axlebox perfection from an engine fitted with ‘under sized’ bearings.

As the opening sentance in the Go-Between says:- ‘THE PAST IS A FOREIGN COUNTRY: THEY DO THINGS DIFFERENTLY there.’

Hello David - no I'm not going to Warley this year - pity really because it means I will miss Love Lane. Instead I plan to move the lathe and other workshop bits next weekend. The conversion of the Dapol Jinty (am I allowed to call it that?) to S7 is nearly complete. Just need some brass rod & tube for the replacement brake stretcher rods - then its final assemby (part glued). Rather than reprofile I used Slater's wheels. They are set up with little side play, but with gauge widening it should be OK for a little under 7ft radius i.e. 4.5 chains. While awaiting the brass stuff cast an eye over the Slaters Open Cab 0-6-0T kit. As you know it represents the Class N version, but only No. 1997 was shedded down south, - Wellingborough - and then only briefly soon after building. The Class N had a different frame design from the majority of the 'Half Cabs' with the profile differing above and below the platform. I have a note (possibly it was following a conversation with David Tee) to the effect the 1903 diagram book is incorrect and the engines to O.991 were built to the same drawings as the Class Ns. Photos suggest the previous batch (O.968) also had the same frame design. I rather liked the idea of one the latter as they were fitted with low cabs and the Salter springs arranged athwartships and lived at places like St Albans. In the end, as it necessitated me guessing the cab radii I have reluctantly abandoned the idea. Thoughts then turned (very briefly!) to whether I could build it as a Class A as a few were at Kentish Town at the right time. It would however demand a lot of new parts so decided it would be better tackled as a scratchbuild project. Interestingly Summerson suggests that none of the Class A 0-6-0Ts had new frames, but I think this is wrong. There are photos of several engines fitted with replacement frames. So, unless you happen to know please where there is a cab drawing for the O.968 batch or alternatively the relevant dimensions, it looks like it will have to be No. 1997! Rambler

So David you have succumed and joined us! Great! This is my first posting so, although I have been reading other people's input for ages, never had anything to say that could be contained in a single posting until now! Thank you for you for encouraging me - your friend in west Wales.