GWR four-cylinder arrangement?

Around twenty years ago I was on the footplate of Evening Star exhibited alongsite a Star/Castle or King thing in Swindon museum.
I also went under the 4-6-0 and found it complicated but it was before mobile phone cameras and light was to low for silver iodid photograhy.
My dream before sleep is to modify a Manor into a two cylinder compound by scrapping one outside cylinder and put a new big low-pressure cylinder where King etc have two simple inside cylinders.
Can I be guided to a decent drawing or photo of GWR four-cylinder ,frames,stay and cross-head fixation?
If not ,then the three cylinder LMS/BR arrangement can also show how to do it.
Or the LNER footballers.B17

 

That makes Francis Webb's three-cylinder compound system seem quite sensible! It's basically the same but with one less High pressure cylinder.

 

I would stick with the day job!

 

It is tempting to think Webb Three cylinder compound system on the many simple three and four cylinder locomotives that followed on UK and German rails.
The smaller outside high pressure cylinders makes placement within UK loading gauge easy.They can be very well insulated.They can be steam jacketed.
The low pressure low temperature inside cylinder is a solid rock for rest of locomotive.
Newest german three cylinder Pacific 03 was compared to a norwegian designed 2-8-4 fourcylinder compound in 1941 and used 18% more steam at high speed and power.
The total mass of steel and coal wasted in Germany and UK by not compounding is alarming.

 

Not so the second driving axle having to deal with offset torque and massive torsional stresses.

Not a huge problem in a country like Britain sitting on massive reserves of cheap coal, so the building and maintenance costs became the dominant financial factors.[/QUOTE]

 

I have tried to show the consequence of putting Webb on P2 compared to what was not working.
The two small outside high pressure cylinders can be spread wider and allow a Krauss-Helmholtz or Zara truck arrangement

Webb P2 by Hermod posted Jun 23, 2026 at 9:30 PM

 

I'm reminded of Adams' trial of the Wordsell-von Borries system (a two cylinder compound) on one of his 4-4-0s. It worked OK, and saved a small amount of coal, but that was more than outweighed by the additional lubrication demands, such that it was overall more expensive to run.

British engineers were not stupid, (and I don't suppose Norwegian ones were either). If there had been a saving to be had with compounding in this country, the CMEs would have used it. Most railways tried it towards the end of the 19th century, and most reverted to simple propulsion - that tells you something about the total costs, not just coal and steel. By contrast, the rapid uptake of superheating and improved valve events for express locos in the early years of the twentieth century shows you that if there was a genuine improvement to be had, the major CMEs were perfectly willing to adopt it. But you have to judge their work against the economic and social restraints of the era in which they were working, not of today.

Tom

 

[/QUOTE]

Locomotives can be 18% cheaper in first cost and boiler maintenance is also 18% less?

Webb P2 by Hermod posted Jun 23, 2026 at 9:30 PM

Webb cylinders versus Gresley on P2
High pressure could have sat farther out so that a Krauss-Helmholts or Zara could guide locomotive

 

The first thing to point out is that reduced coal costs bring costs elsewhere

Secondly the UK loading gauge is much more restricted than the European making it easier to fit the larger cylinders a compound requires.

Finally as the UK had large indigenous coal resources fuel economy was not the priority it was in Europe.

 

From memory, the valve gear for Castles/Kings is inside Walchaerts with a modification.
Two cylinder compounds follow the Von Berries system as per a number of NER 19thC locomotives. The preserved NER 2-4-4T is AFAIK the only survivor.

 

Which raises the question of how successful were the LMS Compounds rated as 4P ?

 

The Midland (from W.M. Smith (North Eastern) Compounds system worked well and they were, for their size, good and strong machines. I don't have figures for their coal consumption but since these would be comparisons against, say, Claughtons and Dreadnoughts, both bigger engines and suffering from severe valve leakage at the time, and Derby was conducting the tests, I wouldn't draw too many conclusions from them. But they were hampered by the usual Midland short lap, short travel valve gears which the compound system had first to overcome.

Henry Fowler attempted to design a four-cylinder compound from them but ran up against James Anderson, who was opposed to big engines, and finished up with the simple Scots instead.

 

AFAIK one got as far as Bournemouth and so the story goes an ex Saltley fireman had to give the driver a very quick demonstration of how to use the regulator as it limped rather slowly out of Bournemouth - if you keep it on full open then it works simple and takes off nicely

 

Sorry, but wrong way around. The Compound regulator was an ingenious bit of kit and allowed the unskilled driver to drive them effectively. At small openings live steam was fed to the two low pressure cylinders to allow starting, at the same time the exhaust from these was fed to both ends of the inside high pressure cylinder so the piston therein was in equilibrium and producing no output. As speed rose the regulator would be opened wider and less steam go to the LP cylinders and an increasing amount to the inside HP cylinder. Once the regulator was fully open all boiler steam was fed entirely to the HP cylinder and only its exhaust to the LP cylinders so it was working fully compound.

Therein lay a problem: when on top link, long distance work the regulator would be fully open for long periods; it would work compound and all was well, but when they were demoted to secondary work with frequent stops and restarts they would work longer with a partially open regulator and therefore only semi-compound, so most of the advantages were lost.

 

As is well known, Churchward compared the performance of French designed compounds against his four cylinder simples , decided he was right and the rest is history!

 

You just removed Manors from my perverse dreams.
Why ,by the way ,was this special GWR method of uniting cylinders and frames not used on the modified Halls and anywhere else in UK?

 

I don't think that it was "special" as it was a commonly used arrangement in the U.S. but the downside was a 'weak' front end requiring the use of braces to the smokebox. With the use of continuous frames and separate cylinder castings on the Modified Halls (and subsequently Counties) this weakness was avoided.
Ray.

 

The LMS Compounds were the best express locomotives the LMS had before the Royal Scots. They were better than the Claughtons and LYR Class 5 all the other express passenger locos the LMS had.
This isn't to say they didn't have their weaknesses. Fowler had simplified Smith's design by only having one regulator to control both the high and low pressure engines (Smith had separate regulators) as well as the usual poor Midland front end.
Nowadays, 100 years on, people tend to look down on short travel/short lap valve advocates but valve design was very controversial in the early 20th century as a peek at the Proceeding Of The Institute of Locomotive Engineers will show. Advocates of the long travel/long lap arrangement were quite often considered to be dangerous revolutionaries.
After the visits of Castles to the LNER and LMS and the success of the Royal Scots and Fowler 2-6-4T became obvious, supported by Maunsell's use of it on the SR (which dated back to the First World War) the long travel/long lap arrangement became standard to the extent that the FR has modified its Victorian engines to suit (as well as superheating).

 

Save yourself some time - just remove all preserved locos from your dreams.