A Coarse Guide to the Steam Locomotive for ‘N’ Gauge Modellers

[Train Waiting]

A Coarse Guide to the Steam Locomotive for 'N' Gauge Modellers - Part 62


Hello Chums

Moving the Valves

Let's start with a picturingham of a Union Mills LNER (ex-NER) 'J27' 0-6-0.  A typical British 0-6-0 goods engine, although with a larger boiler than most classes had. Two inside cylinders, with the connecting rods acting on a crank axle.





Here, from 1:00 is the crank axle from a 1:1 scale 'J27':

It might be a jolly jape to pause the film for a moment to inspect the crank axle.  From right to left we can see a driving wheel, an axlebox (one of the bearings that locate the axle in the frames), a crank, four eccentrics (with numbers 2-4 helpfully marked on), the other crank, at 90^o to the first, the second axlebox and the second driving wheel.

A busy piece of engineering.  A crank axle like this is built up from several pieces which are then secured together using a big press.  A very accurate machining and assembly job. To my mind, it is amazing that engineers were tackling this type of work as early as 1829.

The four eccentrics you can see in the video (thanks again to our chums at Grosmont) are part of the mechanical contraption that operates the locomotive's two valves (piston valves on a 'J27').

***

Right-O; lets move on to discuss these mechanical contraptions, normally called the valve gear or valve motion.

In the early days a simple mechanism called 'gab motion' was generally used. There were two eccentric rods for each valve, one for forward and one for reverse.  These went from eccentrics on the crank axle to vee-shaped ends, called gabs. The driver could select which eccentric rods' gabs operated the valve spindles, giving forward or reverse movement. And a sort of 'neutral' position. It worked but was a fairly crude piece of engineering.

Remember we discussed 'cut-off' and expansive working? Gab motion had a fixed (and long) cut-off with no provision for expansive working.

A French chappie, called Benoit Paul Emile Clapeyron, thought this was wasteful and experimented with a shorter cut-off on a locomotive of the Paris-Versailles Railway. He was impressed by the increased power and more economical working of the engine. The problem was the shorter cut-off was still fixed and expansive working became a trade-off with the engine's ability to start a train.  Hardly an effective solution, but I thought it would be nice to remember M. Clapeyron.

Efforts were made to solve the problem, one of which was the interestingly-named 'Isaac Dodds' wedge motion'¹ (titter ye not), but a proper solution remained elusive.

Can you guess where the answer, which transformed locomotive development, was found?

Newcastle-upon-Tyne, in the locomotive works of Robert Stephenson & Co..

In October 1842, Robert Stephenson and Co. built a 2-4-0 goods locomotive, No. 71 for the North Midland Railway. She isn't as well-known as she ought to be because she was the first locomotive to be built with 'link motion'.

Two clever chaps, employed by Robert Stephenson & Co., Messrs Williams and Howe, had solved the problem of a valve gear that allowed for expansive working using variable cut-offs. Howay Bonny Lads!

Properly known as the Williams-Howe link motion (and mentioned as such in Ahrons, page 63²), it became commonly known as Stephenson link motion, after the firm.

It quickly became pretty much a standard valve gear in Great Britain until the Edwardian era, and locomotives fitted with it were being built for BR as late as July, 1955 (0-6-0PT No. 9499). It also became the most commonly used valve gear in North America until around 1905 - most of these outside-cylindered locomotives with no visible valve gear that we see in Westerns were fitted with inside Stephenson link motion.

***

If you have ever wondered why an engineman might refer to shortening a locomotive's the cut-off as 'linking up', it is from the 'link' in link motion.

In the next part, we'll take a quick look at Messrs Williams' and Howe's link motion and I'll attempt to describe what the 'link' is.  We don't need to go into a lot of detail because Stephenson link motion is almost always concealed between the frames and, as such, of passing interest only to 'N' gauge modellers.

¹ Patented in 1839.

²EL Ahrons, The British Steam Locomotive 1825-1925, Locomotive Publishing Company Limited, London, 1927.


'N' Gauge is Such Fun!

Many thanks for looking and all best wishes.

Cheerie-bye

John

[Train Waiting]

A Coarse Guide to the Steam Locomotive for 'N' Gauge Modellers - Part 63


Hello Chums

Williams-Howe Link Motion or Stephenson Link Motion

Remember the old gab motion with two eccentric rods for each valve (one for forward and one for reverse) driven by eccentrics on the crank axle and ending in vee-shaped 'gabs' which the driver could fit onto pins on the valve spindle? Fairly crude and, with a fixed cut-off, not capable of expansive working.

Messrs Williams' and Howe's insight revolutionised steam traction.  And, with hindsight, it looked so obvious. Things often do with hindsight.

What they did was remove the gabs and join the ends of the two eccentric rods to a slotted piece of metal.  As this component linked the two rods, it was called a 'link', or more properly, an 'expansion link'.

The two links, one for each valve on a two-cylinder engine, are suspended from a common reversing shaft by 'lifting links' which are connected to the reverser on the footplate.  By moving the reverser, the driver can raise or lower the expansion link.

Fitted in the slot in the expansion link is a 'die block' which is connected to the valve spindle by a valve rod.

The position of the die block in the expansion link is controlled by where the driver has set the reverser.  If it is in the centre of the expansion link, the expansion link oscillates around the die block giving a mid-gear position.  Moving the expansion link one way or another puts the locomotive in forward or back gear.

And here's the clever bit, placing the die block right at the end of the the expansion link's slot will give either full forward or back gear.  Positions between the end of the slot and mid-gear will permit variable cut-offs and, therefore, expansive working.

How about a coarse picturingham of a diagramington, from an old volume in my library? 





The big rod at the bottom is the connecting rod, the big end of which is mounted on a crank on the crank axle.

At the top is the reversing rod, heading off to the footplate. It is connected to the curved expansion link by the lifting link.

The die block is right at the top of the lowered expansion link (incidentally, this is full forward gear) and is connected to the valve rod.

We can see the two eccentric rods which are attached to the expansion link at one end and to eccentrics on the crank axle at the other.  These eccentrics give the necessary movement to the eccentric rods.

*


A Bit More on Eccentrics

An eccentric functions like a crank and is used to convert reciprocal to revolving motion, or vice-versa.  A look back at the video of the 'J27's' crank axle will give a good view of them at about 1:00:-

For link motion, eccentrics are in two main parts.  The 'eccentric sheave' is like a metal disc placed over the crank axle. But, instead of the hole for the axle to go through being in the middle of the eccentric sheave, like in a washer, it is offset.  As the crank axle rotates, the eccentric sheave is given an eccentric motion.

The eccentric rods end with the 'eccentric straps'. These are in two parts, one fitted to the eccentric rods rods and the other, bolted to the first, is used to close the eccentric strap round the eccentric sheave. All this can be seen in the diagramington and in the video.

From the 1840s to the end of the Edwardian era, Stephenson link motion pretty much reigned supreme in Great Britain.  And, although used less and less, it continued to almost the end of steam locomotive construction for British Railways.

Needless to say, others attempted their own variations on the theme of link motion - the Gooch and Allan varieties are probably the best-known.

John W Hackworth, son of Timothy Hackworth, invented a type of valve gear that bears his name.  It was used in some industrial locomotives. Probably the best example for our inspection is 0-4-2ST No. 4, Edward Thomas on the Talyllyn Railway. Variations of it were used for some marine engine applications.

***

The next part will take a quick look at a most interesting application of Stephenson link motion.


'N' Gauge is Such Fun!

Many thanks for looking and all best wishes.

Toodle-oo

John

[Bealman]

Uber cool episode, John.

[Train Waiting]

A Coarse Guide to the Steam Locomotive for 'N' Gauge Modellers - Part 64


Hello Chums

'An apparition never seen before in modern British practice.'¹

Towards the end of the LMS railway's independent existence, Mr Ivatt, the Chief Mechanical and Electrical Engineer, undertook a fascinating experiment which would, undoubtedly, have been of great significance if the LMS had continued as a company.

The Stanier 'Class 5' 4-6-0 was an especially successful design and Mr Ivatt arranged for 30 locomotives of the class to be built with non-standard features. One of these was No. 4767, later BR No. 44767, which was built with outside Stephenson link motion, rather than the normal Walschaerts valve gear. She also had roller bearings, electric lighting and a double chimney, removed in 1953. She entered service on 31 December 1947, the last day of the LMS.

Mr ES Cox's quotation, above, is rather less than complimentary.  What was behind the idea?

Some of the experimental engines had Caprotti rotary valve gear but the others, apart from No. 4767, had conventional outside Walschaerts valve gear. Mr Ivatt the Younger was a supremely practical and, in my view, under-rated locomotive engineer who decided to include Stephenson link motion in his tests for one particular reason.

Remember we discussed 'lead steam' - that is the steam admitted into the cylinder before the piston has reached the end of its stoke. I rather slapdashly likened it to ignition advance on a petrol engine for a motor-car or motor-bicycle.

The geometry of Walschaerts valve gear is such that it provides a fixed lead at all cut-offs.  Stephenson link motion has a variable lead, increasing as the cut-off reduces. This means that on starting or working hard at low speed the lead steam is less than when running easily with a short cut-off, likely at speed.  Like ignition advance increasing with a petrol engine's revolutions, the lead steam increases as the locomotive works more expansively.

All in all, this is likely to be helpful. Unfortunately, it appears that the design team at British Railways showed little interest in this experiment - I believe it was deemed to be inconclusive and No. (4)4767 remains a one-off to this day. By great good fortune, she has been preserved and is now part of Mr Smith's West Coast Railways fleet at Carnforth. I understand she is nearing the end of an extensive overhaul.

You know what it's so often like with one-off locomotives. They tend to become unpopular white elephants and see limited use, often with lots of time stopped waiting for spare parts.

But absolutely not in the case of 'Forty-Seven, Sixty-Seven'! Enginemen considered her to be an especially strong engine and she achieved some of the highest average annual mileages per locomotive for the class in the 1950s.  This points to her being used on the more exacting diagrams and having impressive reliability. No doubt, the roller bearings helped with this.   

However, I believe it is unusual for one-off locomotives to receive such approbation which suggests Mr Ivatt was on to something with Stephenson link motion for a mixed-traffic locomotive. Certainly, there were many people at Swindon who would have agreed with him.

Here are a couple of picturinghams of her outside Stephenson link motion, with fly-cranks replacing the eccentrics used with inside valve gear. They show the various components of Stephenson link motion, discussed in Part 63:-








In both picturinghams, she is stationary and is, correctly in mid-gear.  Peering through the fog of my coarse photography, you might notice the die-block is in the middle of the expansion link.


My apologies for mentioning Walschaerts and Caprotti valve gears in this context before I have described them.  Their turn will come.


¹ ES Cox, Chronicles of Steam, Ian Allan, London, 1967, page 153. 


'N' Gauge is Such Fun!

Many thanks for looking and all best wishes.

Toodle-pip

John

[Papyrus]

Thank you, John, that has given my brain some serious exercise. I have read it through several times and I think I now have the glimmerings of understanding a very complex subject. This is by no means a reflection on your writing, which is as clear as anyone could wish for, but it does explain why I never followed my father into engineering! I had heard of the Stephenson link but until now I had not the foggiest notion of what it was or how it worked. Please continue to ediccate me!

All the best,

Chris

[martyn]

Regarding the ability of 44767, I have read somewhere (so many books so long ago) that it was regarded as about 10% 'stronger' than the rest of the class, ie able to haul an extra coach or the same number at about 10% economy compared to the others of the class.

Thanks again, John.

Martyn

[crewearpley40]

Martyn

https://preservedbritishsteamlocomotives.com/44767-lms-4767-br-44767/

A clue @martyn in featuring Stephenson
Outside link motion and valve gear

[Train Waiting]

A Coarse Guide to the Steam Locomotive for 'N' Gauge Modellers - Part 65


Hello Chums

The Joy of Joy Valve Gear

As we have discussed, the last quarter of the Nineteeth century saw 4-4-0s and 0-6-0s with two inside cylinders and Stephenson link motion becoming the typical British locomotive types. Yes, there were plenty of fascinating exceptions, such as the 'Singles' for passenger work, especially on the Great Northern and Great Western, and, in the last decade, the appearance of 4-6-0 and 4-4-2 types.

We have seen how the crank axle of a locomotive with two inside cylinders and link motion was a busy piece of engineering, with two axleboxes, two cranks and four eccentric sheaves having to be accommodated between the wheels. A consequence of this, especially as cylinder diameter increased, was difficulty in making the axlebox bearings of an adequate size.

Being able to dispense with the eccentrics for the valve gear would create more space for the axleboxes.  David Joy, former Locomotive Superintendent of the Oxford, Worcester and Wolverhampton Railway, patented in 1879 a valve gear that dispensed with eccentrics, leaving the crank axle free for bigger axleboxes and more substantial cranks. This was known as the Joy valve gear and it was trial fitted to an old Bury-type goods engine on the Furness Railway.

The trial was successful and, in June 1880, FW Webb had Joy valve gear fitted to a new London and North Western Railway (LNWR) 'Cauliflower' 0-6-0 goods engine, No. 2365. Impressed with the results, Mr Webb became the leading exponent of Joy valve gear and the LNWR used it extensively. Incidentally, the Webb 0-6-0 was demonstrated at the Institution of Mechanical Engineers' summer meeting at Barrow in August 1880 where it attracted a great deal of interest. However, LNWR apart, at that time it didn't become widely used.




[An ex-LNWR 'Cauliflower' 0-6-0 in LMS livery. For many years, this class was associated with the SuperScenic Cockermouth, Keswick and Penrith Railway.]


As we have seen, going back to the days of the crude gab motion, a locomotive's valve gear needs a source of motion. Fundamentally, this comes from the movement of the piston in the cylinder but there needs to be provision made for forward, reverse and mid-gear, and, importantly, for expansive working.  Hitherto, the crank axle (or driving axle in the case of outside-cylindered locomotives) had been the best place from which to derive the necessary movement.

Mr Joy's patent valve gear took its motion from the locomotive's connecting rods. Here's a diagramington:- 





I don't think there is a need to include the nomenclature or a description of all the various parts, although I should be glad to do so in the discussion if anyone wishes. However, it is worth noting that a component, called the 'jack link', is attached to the connecting rod by means of a pin. The insertion of the pin required a hole to be created in the connecting rod.

On 1 October 1886, JAF (later, Sir John) Aspinall was appointed Locomotive Superintendent of the Lancashire & Yorkshire Railway (L&YR). He found some engines there with Joy valve gear and almost-identical ones with Stephenson link motion. He was greatly impressed by the average mileages between Works repairs of the Joy-fitted engines, 62,344, as against 51,319 for those fitted with link motion.  Mr Aspinall was clear in his mind that Joy valve gear wasn't intrinsically superior to link motion, but it allowed for larger bearing surfaces on the crank axle, to which he attributed the increased mileage between repairs.¹

In September 1887, Mr Aspinall met Mr Joy and agreed terms for Joy valve gear to be the L&YR standard. For about the next 30 years, Joy valve gear was the norm on the LNWR and L&YR.

*

Joy Valve Gear's Achilles' Heel

Whilst just about adequate, in my view, to cope with the power of locomotives in the 1879-1900 period, Joy valve gear was not well suited to the Era of the Big Engine. That hole to fix the jack link to the connecting rod was a source of potential failure and broken connecting rods began to occur.

A broken connecting rod on a steam locomotive is a dangerous thing as it can cause catastrophic damage. If it digs into the ballast, the result could be a pole-vaulting locomotive. Alternatively it could pierce the boiler or front of the firebox with potentially explosive consequences.

The LNWR's 'Prince of Wales' 4-6-0 class, of 1911, was the largest superheated British express passenger locomotive fitted with Joy valve gear. These engines worked hard, especially on the northern part of the LNWR between Crewe and Carlisle with its ferocious banks over the northern fells. A 'Prince of Wales' (regrettably I have not been able to trace which one) suffered a catastrophic failure of a connecting rod which resulted in the death of both enginemen.




[A pair of 'Prince of Wales' 4-6-0s. No. 5604 Enchantress in LMS crimson lake and No. 86 Mark Twain in LNWR 'blackberry black'.]


And that was the beginning of the end of Joy valve gear for new construction. Mr Bowen Cooke's large 'Claughton' express passenger 4-6-0s of 1913, for the LNWR, had Walschaerts valve gear. Mr Bowen Cooke died in harness, aged 61, in 1920 and, from 1922, his successor as Chief Mechanical Engineer, HPM Beames set his mind to how the Joy valve gear on the 'Prince of Wales' class could be replaced. I hope to return to this in the next part of this exceptionally brief mini-series. And introduce you to a racehorse that came last in the Cesarewitch in 1921 and 1922.  I hope you didn't have a tanner on her.

The locomotives seen in my well-dodgy picturinghams are SuperSpiffing Union Mills models.

¹ HAV Bulleid, The Aspinall Era, Ian Allan, London, 1967, Page 91


'N' Gauge is Such Fun!

Many thanks for looking and all best wishes.

Tickety-tonk

John

[port perran]

Just as a complete aside and not really to do with the wonderful coarse guide but does anyone know why CJ Bowen Cooke is buried with a rather elaborate headstone at St Just in Roseland in Cornwall?

He certainly died in Inverness but what was his Cornish connection?

[crewearpley40]

Just as a complete aside and not really to do with the wonderful coarse guide but does anyone know why CJ Bowen Cooke is buried with a rather elaborate headstone at St Just in Roseland in Cornwall?

He certainly died in Inverness but what was his Cornish connection?

Couple of thoughts on Charles Bowen Cooke
https://newprincegeorgesteam.org.uk/the-family-of-charles-john-bowen-cooke/

https://www.falmouthpacket.co.uk/news/19738022.charles-john-bowen-cooke-memorial-stone-st-just-roseland/

@port perran Martin

Maybe a family connection

[Papyrus]

Another very interesting discourse on a valve-gear design I had never previously heard of. However, one seemingly obvious question entered my naïve non-engineers brain:

Why was it necessary to make a hole in the connecting rod itself, which was obviously asking for trouble? Wouldn't it have been better to add a small bracket to either the top or bottom of the casting and fix the link to that, and thus leave the connecting rod intact throughout its length? I'm sure there must have been a reason otherwise somebody would have done it.

Cheers,

Chris

[Jim Easterbrook]

I think connecting rods would have been forged, not cast. Bashing a piece of hot metal to make it longer and thinner aligns the "grain" of the metal, making it nice and strong. It would be hard to add a bracket as part of this process, and bolting one on afterwards means making holes and we're back where we started. (The hole would be forged rather than drilled, so the grain flows around the hole, but it's still a place where stress would concentrate.)

[Train Waiting]

A Coarse Guide to the Steam Locomotive for 'N' Gauge Modellers - Part 66


Hello Chums

Other Valve Gears

We discussed earlier the dominant position of Stephenson link motion as a valve gear for locomotives in Great Britain at the end of the Nineteenth century. The LNWR and LYR used Joy valve gear as standard but, as we have seen, its flaw of a hole in the connecting rods made it unsuitable for the era of the Big Engine. But a change was about to occur which would result in another valve gear coming into prominence in the new century.

Walschaerts Valve Gear

Invented in 1844 by Egide Walschaerts, a Belgian engineer, its acceptance was slow but it gradually became widely used in Europe.

The first example of its use in Great Britain was in a 0-4-4T single Fairlie locomotive, built by Avonside of Bristol and sold to the Swindon, Marlborough and Andover Railway¹ in 1881, becoming No. 4 on that line. The locomotive had outside cylinders and valve gear. Unfortunately, the valve settings were badly arranged and the little engine had a voracious appetite for coal. As such, it saw little use. The locomotive was scrapped in 1892.








The private locomotive building firms gradually built up their experience with Walschaerts valve gear as it was increasingly specified for engines being built for abroad.

Mr Ivatt the Elder fitted outside Walschaerts valve gear to his experimental four-cylinder 4-4-2 of 1904 for the Great Northern Railway.

From 1905, Dugald Drummond fitted his series of spectacularly lacklustre LSWR four-cylinder 4-6-0s with Walschaerts valve gear for the outside cylinders and inside Stephenson link motion for the inside cylinders.  The different characteristics of the two valve gears placed all these massive locomotives at a serious disadvantage.

Incidentally, William Pickersgill's unbelievably inept three-cylinder '956' class 4-6-0 of 1921 for the Caledonian Railway had outside Walschaerts valve gear for their outside cylinders and Stephenson link motion for their inside cylinder.

I think the real impetus for the use of Walschaerts valve gear in Great Britain came from Mr Churchward obtaining permission from the GWR's Board to purchase three French four-cylinder 4-4-2s, for comparative purposes, between 1903 and 1905.  These were fitted with Walschaerts valve gear and when Mr Churchward designed his four-cylinder 'Star' class 4-6-0, introduced in 1907, he used inside Walschaerts valve gear.

Although never that common, apart from on the Great Western, inside Walschaerts valve gear was used in some early Twentieth century designs, Peter Drummond's '131' 4-4-0 of 1913 for the Glasgow & South Western Railway being a good example.

It was in August 1912 that the soon-to-be-familiar combination of outside cylinders and Walschaerts valve gear became part of the mainstream of locomotive design. Mr (later, Sir Nigel) Gresley's 10 'H2' two-cylinder 2-6-0s for the Great Northern brought the type into prominence.





They were followed by Mr Bowen Cook's four-cylinder 'Claughton' of 1913 for the LNWR, Mr Gresley's 'O1' 2-8-0 of 1914 for the Great Northern, Mr (later, Sir Henry) Fowler's 2-8-0 class, built by the Midland Railway for the Somerset & Dorset Joint Railway from 1914 and Mr Urie's 'H15' 4-6-0 of 1914 for the LSWR.

The New Look had well and truly arrived and, by the outbreak of the Great War, outside Walschaerts valve gear was no longer a curiosity and it gradually became commonplace. Special mention ought to be given to Mr Maunsell's 'N' 2-6-0 for the SE&CR. Design work on this class was delayed by the War and the first example entered service in 1917. With two outside cylinders and Walschaerts valve gear, high-degree superheat and long lap/long travel valves, I like to think of this class as the first truly modern British locomotive.

The distinguished locomotive historian, Philip Atkins, observed that, by 1954, roughly 30% of the 18,500 steam locomotives at work on BR had Walschaerts valve gear.² Whilst at first glance this seems impressive, inverting the figure is especially interesting. As late as 1954, forty years after the New Look arrived on the scene, roughly 70% of BR's steam locomotive stock still did not have Walschaerts valve gear.

Making allowances for some ex-LNWR and LYR engines with Joy valve gear and a few locomotives with poppet valves probably doesn't change the figure that much and we can assume about two-thirds of BR's steam locomotives were fitted with Stephenson link motion. And, as we have seen, in 1954 new engines so fitted were still being built. 

*

Walschaerts valve gear first appeared in the North America in 1876 but didn't really catch on until the Baltimore & Ohio Railroad fitted it to the 2400, a Mallet, in 1904.  After that, its rise was unstoppable and it became the most common valve gear in use in North America. The second most popular was Abner D Baker's Baker valve gear  of 1903, which is, essentially, a development of Walschaerts valve gear.

*

The next part will discuss the 'Prince of Wales' 4-6-0 conversions to Walschaerts valve gear.


¹  The Swindon, Marlborough and Andover Railway and the Swindon and Cheltenham Extension Railway amalgamated in 1884 to form the Midland and South-Western Junction Railway.

² Philip Atkins, Edwardian Steam A Locomotive Kaleidoscope, Crecy Publishing Ltd, Manchester, 2020, ISBN 9781 9108 09655 Page 11.


'N' Gauge is Such Fun!

Many thanks for looking and all best wishes.

Cheerio

John

[Bealman]

Excellent, John.

[martyn]

Another UK user of Joy's gear was TW Worsdell on both the GER and later the NER.

The GER M15/LNER F4 were direct descendants of the Worsdell 'Gobblers' but with valve gear changed to Stephenson by his successor, James Holden. I presume the GER thought that the Joy valve gear was the reason for the original locos' high coal consumption, and hence their nickname. Even though the Worsdell series had gone by Grouping, the F4s and F5s retained the nickname.

However, when he went to the NER, Worsdell developed a cousin to the M15s, the NER class A/LNER F8, which retained the Joy valve gear until withdrawal in the mid 1930s. Presumably Mr Worsdell had learned lessons from the GER series, and improved the details of the valve gear. Without going through all my RCTS books, I don't know if other NER classes used Joy's gear, but I think TW Worsdell's successor, his brother Thomas, didn't use it.

The Baker valve gear was used on at least four New Zealand Railway classes of locos, the highly successful J/Ja/Jb 4-8-2s, and some Ka 4-8-4s. I don't know if there were others. It is possible that this valve gear was used on some UK traction engines. 

Thanks yet again, John.

Martyn