Steam Loco building & design in the UK and Abroad.

[Grovenor]

"Station platforms limited outside cylinders to 22" in diameter".

Subject to some qualification, the 21" cylinders had to be raised on the Crabs, I suspect Tuplin was talking about the GWR which had more generous clearances than other railways.

Also the allowable diameter depends on where the centreline of the cylinder can be located, which depends on the design of wheel and rods. Flat faced wheels allowing a bit more room than wheels with protruding bosses. Then if drive is on the leading axle the connecting rod can be inside the coupling rods bringing the cylinder centre several inches inboard and hence allowing larger diameter cylinders.

Regards

[EddieB]

 

About all this new re-creation of long lost loves - including the P2, a Patriot, even a Standard 4-6-0, I do find it a shame that no new locomotives are being constructed embodying fresh thinking about 'unexplored territory.'

The Porta project died; does anyone know anything about this link ?

 

dh

Apparently waiting for a few multi-millionaire to fetch up.

 

Please form an orderly queue.

[pH]

Given Staniers 'upbringing' at Swindon, I'm surprised he didn't use rocking-shafts on the Princess-Royals especially with out-side valve gear, and not waiting until the Coronation class.

 

Though he did try retro-fitting them on 46205 after the first Coronations went into service. It would appear it wasn't considered worth fitting them to the others, though 46205 kept the arrangements until withdrawn. (Wikipedia does not agree with pictures and several other sources about facts and dates around this.)

[asmay2002]

Subject to some qualification, the 21" cylinders had to be raised on the Crabs, I suspect Tuplin was talking about the GWR which had more generous clearances than other railways.

 

The ROD 2-8-0s had 21" cylinders and they ran on all regions at one time or another. I don't think any GWR class got as far as 20".

[jjb1970]

I'd love to see somebody with deep pockets fund construction of the Wardale 5AT locomotive. I tend to think it'd be a much more worthwhile exercise than building replicas of old designs. That said, it'd be a great exercise in seeing how far the steam locomotive could be pushed by applying more modern technologies in its design and construction but it can't address the inherent inefficiency of a steam locomotive.

[Sheffield]

The story of the class 26, no 3450 in South Africa is an interesting guide to what could have been achieved with steam traction, and for a more or less "first try" was a success. But most if not all that was done to this loco could have been done in the 1940/1950s with a bit of effort, investigation, and Imagination. It does hint at what the BR Standards could have been. Mr Cox was aware of Livio Dante Porta's work in producing 48 ton 2ft 6 in gauge locos running on 35lb rail to haul 1,700 ton trains. They could produce 1,200 HP, but Cox dismissed such things with the comment that Patagonia was a long way to go to see it.

 

I think Dr Tuplin, I believe a professor at Sheffield University rather than a railway man, is an interesting character with a some times realistic view of things, but he shared the idea that boiler pressure should not exceed 180 lbs, so his statements about cylinder diameter need to be seen in that light. His outlined design for a 4-8-0 for the GWR would have been quite something in the flesh.

[34theletterbetweenB&D]

 

 

You need to have studied value engineering under a dreadful humorist. This was always the challenge: accepted custom and practise.....

...However, none of this answers my simple initial query seeking clarification on why "under no circumstances have four...."? Or is that the product of the dreadful humorist?

 Yes.

 

I got my initial education in 'steam' from people fully brainwashed in the Churchward got everything right school, therefore Belpaire firebox, modest superheat, four cylinder engine layout for maximum power is the true and onlie path.  A little older and with some real world design and production engineering experience acquired, I realised that most people only know what they know and don't care to look any further. A brief examination of other schools of steam locomotive engineering practise revealed that none of the three items listed above are in any way binding prerequisites for extremely successful steam locomotives which outperformed Swindon's products in every way.

 

To quote Professor Tuplin ... "Station platforms limited outside cylinders to 22" in diameter"...

 Tuplin - a man often admirably prepared to think outside the box - would if questioned on this point probably would have qualified it with 'when typical UK plate frame construction is employed'.

 

The real limitations on outside cylinder diameter are:

the position of the cylinder centreline for the piston and connecting rods which must drive perpendicular to the crankpin on the wheelface;

clearance within the limits of the loading gauge, as set by fixed structures lineside with the movement of the loco and all the other variables taken into consideration.

 

So here's the stack up. Width across wheel faces roughly five feet. If the big end bears directly on the crankpin from the wheelface (this may require unusual practise such as inside coupling rods) you can add eight inches to the width (lots of power output, I have upped the big end bearing size somewhat from normal) for the cylinder centreline.

 

How big can the cylinder diameter go from those centrelines, 68" apart? Let's be conservative and say the maximum width will be limited to 8'9" across the cylinders, 105". Allowing 3" for the cylinder wall thickness and and exterior cladding, that will permit 30" outside cylinder diameter (rounding down the half inch) at those centrelines.

 

But there's a problem, the inner part of these massive outside cylinders go straight through where the plate frames of a typical UK construction would be! The way around this is the cast frame, which is not constrained by the linear nature of the plate frame for the principal load bearing structure. Simples! Just got to be prepared to break away from a whole heap of established UK custom and practise. Take a look at the front cylinders of a N&W Y6B if in doubt. There are several other problems to solve, but this is what engineers are for.

[The Stationmaster]

Interestingly of course is that Churchward didn't seem to see any need to go that far with cylinder diameter notwithstanding having a fairly generous loading gauge in which to work and using a cast/forged front end frame extension.  

 

(Incidentally the maximum loading gauge width across the outside of the cylinder casings would be 8ft 10" on standard loading gauge)

[34theletterbetweenB&D]

Incidentally the maximum loading gauge width across the outside of the cylinder casings would be 8ft 10" on standard loading gauge

 I was being conservative on that, because a 30" outside cylinder dia loco would be an exceedingly long behemoth: better allow a tad more wiggle room for when the 6,000 dbhp output is delivering 140mph cruise speed.

[Grovenor]

 

Width across wheel faces roughly five feet.

Best to add a few inches to that figure as well, which will bring the cylinder down somewhat from 30".

Wheel faces will be 5'4" apart assuming flat face wheels.

Regards

[Grovenor]

Though he did try retro-fitting them on 46205 after the first Coronations went into service.

I always understood that this was part of the development of the Coronations. 46205 had the modified gear in early 1937, before the first Coronations came out of works but it does seem to have been way to late to actually influence the Coronation design. So maybe there was a thought of altering the others which was aborted.

Regards

[Arthur]

Thanks for the detailed replies, they are very informative.

 

A brief examination of other schools of steam locomotive engineering practise revealed that none of the three items listed above are in any way binding prerequisites for extremely successful steam locomotives which outperformed Swindon's products in every way.

 

Yes, not a prerequisite but neither a "not under any circumstances".

 

Tuplin - a man often admirably prepared to think outside the box - would if questioned on this point probably would have qualified it with 'when typical UK plate frame construction is employed'........

 

The way around this is the cast frame, which is not constrained by the linear nature of the plate frame for the principal load bearing structure. Simples!

 

Yes, I had followed that through myself. Maximized three cylinders would require adjustment of the frame width and to avoid the inherent weakness in narrowing plate frames, either by bending or fabrication, cast frames would be the solution.

 

Now earlier you commented on the extra cost of a fourth set of valve gear mitigating against four cylinders. That cost would pale into insignificance compared to that of a cast frame.

 

No British works* had the capacity nor experience to cast them and just how many locomotives for use under UK conditions would justify them? i.e. would it be worth developing that capacity?

 

Yes, they could have been farmed out but probably to the U.S. at considerably higher cost than a home made plate frame.

 

So my point is that the choice between three and four cylinders is not simple, there are many considerations, neither is absolutely right nor wrong.

 

"under no circumstances have four" is not supported by any evidence so far presented.

 

*Crewe works uniquely did, at one time, have the melting capacity when they had a three furnace open hearth melting shop but that was not supported by the experience of making such large castings. The shop opened in the early twenties, I'm not sure when it closed, pre WW2 I think. Crewe did cast trailing trucks for the two Ivatt Pacifics and if I recall correctly they subsequently required some remedial work. Embarking on casting complete bed frames was not without risk.

 

Maybe, under those circumstances, four cylinders was the more certain bet. Over engineering isn't always the right choice much though the engineers might relish the challenge.

[The Stationmaster]

The Churchward frame extensions were forged mild steel because Swindon had no capacity to cast them.  It has been suggested that the fact that they were forged contributed to their tendency to bend (upwards) under buffing forces but it would have been interesting to see what would have happened if they had been replaced by castings when the front ends started to be renewed in the 1930s.

[brack]

Given that much of our rolling stock was loose coupled, 4 wheel friction bearing stuff would it not have made more sense to address this before building a loco too powerful for the draw gear or braking force available on the majority of our stock? Building vast quantities of essentially obsolete 4w wagons postwar was daft.

 

Obviously passenger service is the exception to this, presumably why much of the technological effort that was put in went into the various big Pacifics.

[34theletterbetweenB&D]

...Yes, not a prerequisite but neither a "not under any circumstances"...

 Not from a strict value engineering perspective. If three cylinders can deliver everything - or more - that four cylinders made possible; then no four cylinder machines. And if two cylinders can do it, then no need to depart from the Stephenson format at all.  Return question, why so anxious to defend a clear blind alley?

[pH]

I always understood that this was part of the development of the Coronations. 46205 had the modified gear in early 1937, before the first Coronations came out of works but it does seem to have been way to late to actually influence the Coronation design.

 

I too thought 46205 had been altered to test the arrangement before it was used on the Coronations. I'm sure I have read that. However, I checked the date on which 46205 was modified before I wrote that previous post. "British Pacific Locomotives' by CJ Allen says it happened "in 1938". "Locomotives Illustrated #76 - The Stanier 'Princess Royal' Pacifics" gives a date of March 1938. The first 5 Coronations were put into service during 1937.

[bike2steam]

As a final note on this subject, as far as I'm concerned, are words of O.S.Nock, who more than many others had a vast array of footplate experiences, and the class that he rates as the best smooth, and steady ride, 'for many hundreds of miles' without the merest hint of hunting, and undue oscillation was the 'Duchess' class.

Advances in ideas, and technology accounts for nothing if the final product doesn't come up to expectation.

[Arthur]

Not from a strict value engineering perspective. If three cylinders can deliver everything - or more - that four cylinders made possible; then no four cylinder machines. And if two cylinders can do it, then no need to depart from the Stephenson format at all. Return question, why so anxious to defend a clear blind alley?

I'm not defending anything, I really couldn't care less whether it's three or four. I do want to understand the reasoning behind your assertion otherwise it's a misleading statement in an otherwise interesting post.

 

 

You made an unequivocal statement "under no circumstances four...." I had never heard that before, a number of successful four cylinder designs existed so my curiosity was piqued and I asked for clarification. To support such an unequivocal statement I expected an equally unequivocal answer.

 

All I have had is obfuscation.

 

It is quite clear from the discussion so far that there is, as I suspected, a number of considerations to be taken into account and that in some circumstances four cylinders was a reasonable design choice.

 

Tuplin's statements may well have been made in the context of traditional plate frames. Nothing wrong with that balanced against the cost and risk of moving to a whole new technology (for UK builders) of cast frames. A costly technology which otherwise was probably not justified under UK conditions.

 

So, it might have been a choice of compromise, probably a very reasonable compromise, but that is very far from "under no circumstances".

 

Sometimes the Best is the enemy of the Good.

[Talltim]

On a tangent, why do cylinders have to be cylindrical? (Apart from the obvious name issue!)

[cheesysmith]

Having cylinders cylindrical creates even forces over the piston and the cylinder itself. Honda tried to make a super bike with oval cylinders for racing (basically a v8 with the cylinders squashed together into a oval shape) and even they could not get it to work. And before you dismiss this, remember the physics relating to a piston in a cylinder are the same for both internal combustion engines and external combustion ones.

[Sheffield]

Given the sort of equipment in a typical steam workshop an oval cylinder would be interesting to machine, and then re-machine to restore after wear. Some of them seemed to have trouble making circular ones stay steam tight. 

[34theletterbetweenB&D]

You made an unequivocal statement "under no circumstances four...." I had never heard that before, a number of successful four cylinder designs existed so my curiosity was piqued and I asked for clarification. To support such an unequivocal statement I expected an equally unequivocal answer.

 

 I have given you an unequivocal answer, from the stance of value engineering. If you can do something with fewer components, then on no account use more. As a parallel, you will notice that this process is now well underway on i/c engines. Once the four cylinder was 'standard' for smaller cars, but three cylinder designs are becoming more common. This is how engineering works 'to do for five bob, what any fool can do for a quid'; as Nevil Shute Norway memorably put it .

 

On a tangent, why do cylinders have to be cylindrical? (Apart from the obvious name issue!)

 In specialist applications - not related to power production as far as I am aware - square is utilised. You can pack in more swept volume within typically cuboid constraints of most constructions this way.

[Arthur]

from the stance of value engineering.

 

Anything stated from a 'stance' is not unequivocal because there are other stances.

 

Thank you for your accepting that 'under no circumstances' was, therefore, open to disagreement.

[LNER4479]

I too thought 46205 had been altered to test the arrangement before it was used on the Coronations. I'm sure I have read that. However, I checked the date on which 46205 was modified before I wrote that previous post. "British Pacific Locomotives' by CJ Allen says it happened "in 1938". "Locomotives Illustrated #76 - The Stanier 'Princess Royal' Pacifics" gives a date of March 1938. The first 5 Coronations were put into service during 1937.

I believe that the rocking lever arrangement to actuate the inside valves on the Duchesses was a close copy of the arrangement fitted to the L&Y 4-cylinder 4-6-0s. The fact that Tom Coleman, the LMS chief designer who oversaw the development of the Duchesses, had spent time at Horwich works might not be a total coincidence in this regard...

 

(edited to correct factual error)

[Grovenor]

"British Pacific Locomotives' by CJ Allen says it happened "in 1938". "Locomotives Illustrated #76 - The Stanier 'Princess Royal' Pacifics" gives a date of March 1938.

And Essery and Jenkinson in LMS locomotives vol 5. said "early 1937". Either way it was clearly after the Coronation design was done and manufacture of the first almost complete.

Regards