Driving a compound locomotive

Going back to traction engines, it's worth remembering that the axle is driven by some fairly 'Wrangly' gearing, so the crankshaft can and does rotate 'Light' on starting prior to any load coming on.

 

Further to the comment about intermediate reheaters, Chapelon's 2-12-0 160A1 had an intermediate superheater in the lower flue tubes; the smokebox looks a bit congested with all the pipework but it was a large smokebox.
Similarly to Webb's compounds, the North Eastern 2-cylinder compounds had the receiver as a large tube curling round the upper part of the smokebox, see the diagrams in Van Riemsdijk's book. This was such early days for superheating in railway locomotives at all, well before Schmidt's successes that I doubt whether either Webb or the North Eastern had conceived of an intermediate superheater and the aim was simply to keep the receiver hot and avoid the steam condensing in it. There were several attempts to develop smokebox superheaters at around this time, but all seemed to fail dismally, doubtless because the heating surface was small and the hot gases in the smokebox are a lot cooler than those at the firebox end.

 

The 160A1 provided not only good performance results but it was also used to examine the real value of the interstage superheater in conjuction with steam jacketing the cylinders with the aim of producing high power outputs at low speed, the 2-12-0 was a freight engine after all. It was found that the primary stage superheater could be done away with at truly negligible cost in fuel consumption. The avoidance condensation losses and their impact on performance and efficiency, particularly at low speed, was the object of the exercise. This experimental engine produced 3,000 cylinder horsepower at 25 mph. So it could be said that the inability of the steam locomotive to produce high horsepower at low speed was a result of the design choices taken.

 

Does anyone have a photograph of the inside of the smokebox of Aerolite, that they are happy to share?

 

Another small chapter in the ongoing sagas of the Civils vs the Mechanicals.

 

The LNER 4-6-4 had a proposed arrangement for the Superheater between the HP and LP cylinders produced in 1933, this was drawing number 15542. This was preceded by drawing 15454 which covered the arrangement of the Intermediate Superheater between the frames and 15472 which is for a Superheater Baffle Plate.

15488 is an arrangement of pipes between smokebox and cylinders for the intermediate superheater
15542 is a proposes arrangement for the intermediate superheater supports
15545 covers circulating pipes for the intermediate superheater
15547 is for mild steel supports for the intermediate superheater
15548 is for cat steel supports and guides for the intermediate superheater header

In 1934 came the following;

15562 details for drain of intermediate superheater
15564 5" bore cast steel pipe for intermediate superheater
15566 is for the supports for the 5" bore in Smokebox Intermediate Superheater

And the list continues a little ending with;

15575 Detail of Intermediate Superheater Headers


In tests dating from 30-5-35 the LP superheater temperature is being measured. The engine was worked with a HP cutoff no higher than 50% in these 1935 trials. LP cutoffs varied between 25% and 55% and are frequently equal to or higher than the HP cutoff and the maximum DBHP achieved was 1,702 which was. in terms of what it ought to have achieved quite pathetic. And so we get back to the question of how to drive a compound locomotive. Gresley was originally thinking of 70% HP cutoffs but was dissuaded by Dalby the passing of a few years and the work of a certain Frenchman proved that Gresley ha been right, he was moving on from the established view that the LP cutoff should be fixed and that demand should be catered for by varying the HP cut off and this notion was proved to be spectacularly incorrect by Chapelon. The very inverse of the established understanding did indeed hold true. 10,000 was the most powerful steam locomotive to see service in this country based on extrapolating from Chapelon's work and carrying out an analysis of the data gleaned from the set up of the double Kylchap exhaust on the 4-6-4. The LNER engine had smaller cylinders than the French engines but you cannot escape the impact of the boiler working pressure.

As a footnote 10,000 was an experiment exploring efficiency not power output. The aim was to produce a more efficient equivalent of the A1. The power that lurked beneath this intention was incidental.

 

What would have had to happen differently for 10000 to get closer to its potential before they gave up and rebuilt it?l?

 

Presumably reflecting just that point, not all double Fairlies were 'total adhesion' machines, although their inherent shortcomings mitigated against anything like the development of Garratts between K1/2 and the advent of the SAR GL, GMA/M and EAR Class 59.

 

No. 10000 as rebuilt and running as BR 60700 around 1957.

 

Many thanks for the info. I know little about traction engines, but had noticed that there are often two cylinders of different sizes atop the boiler. The designers and users of traction engines clearly considered the savings on fuel and water to be worth the slight complication of the 2-cylinder compounding system.
Many railway engineers felt the same way in the late 19th/early 20th Centuries and many thousands of 2-cylinder compound locomotives were built, particularly in Continental Europe. If compounding is wanted, the 2-cylinder option has the least impact in additional construction and maintenance costs for a budget-conscious CME. British railway engineers were generally less enthused and only a limited number were built here. I think the last 2-cylinder compounds built for use in the British Isles were in 1920 - NCC 3ft gauge 2-4-2Ts. The last was scrapped in 1954.

https://en.wikipedia.org/wiki/File:110_train_which_has_stopped_on_tracks_(13714893334).jpg

 

High suprheat and Lenz valves
Real Compound testing is possible because this compound thing runs as well

 

My take on a IMLEC winner would be a 5 inch gauge model of the 0-6-2 compound with high pressure cylinder kept at boiler temperature all the time.Mild superheat that goes to a receiver heat exchanger before going into high pressure cylinder at boiler temperature.The exhaust from high pressure has some lubricating oil but is much less hott than primary steam.
Could also work on some UK narrow gauge locomotives

 

I think most of the interesting angles have been covered here, but one other point I think might be worth making, particularly with respect to the OP's Midland Compound:

In the earlier days of compounding, when generally most locomotives had relatively short lap gear that could become decidedly constipated at short cut-offs, the fact that they ran at longer cut-offs for a given degree of expansion was a significant advantage. If your design can run expansively and efficiently at 50% cut-off on the gear, it will have a significant advantage in free-running and efficiency compared to a simple getting decidedly choked up at 25%. This is perhaps one reason why the Midland Compound is generally remembered as quite a good design, while .... that other Midland 4-4-0, the one we always end up on, no, I won't say it .... was rather less renowned. The mistaken extrapolation from this (much as with superheating), was that since compounds seemed more tolerant of insufficient valve travel, they would not derive the same benefits as simples from high superheat and long lap valves. As the very powerful and efficient later French compounds showed, for all the faults of the more numerous classes, this was wrong.

The matter of re-superheat would have been another advantage if the desire was to continue pushing steam temperatures up for greater efficiency. When, some years later, David Wardale was designing his class 26, the Red Devil, he was targeting a steam temperature somewhere approaching 450 degrees C (with 225 psi boiler pressure). This was a functional maximum partly because most common engineering carbon steels start to creep under stress above this sort of temperature, so parts like main steam pipes start to become problems. Even with that limit, he still had to put cooling steam jackets of saturated steam around the valve liners, to avoid carbonising of the oil. Because the admission side valve liners have continuous exposure to steam chest pressure, as opposed to the intermittent heating and cooling of the cylinder liners, that's where lubrication problems crop up first. The jackets were fed from the wet header of the superheater, and exhausted into the inlet side of the steam chest, mixing with the dry steam from the superheater to slightly reduce its temperature, so they didn't pull the liner temperature down below saturation.

The steam jackets tried by Chapelon, by contrast, were to hold the cylinder liner temperature up and reduce condensation losses. With steam at HP steam chest pressure venting through the jackets to the LP receiver (where it can still do useful work), the relatively low loss of performance in the configuration of 160A1 with no HP superheat makes sense. So suddenly, with resuperheat of steam at maybe 150 psi, the temperature limits of high pressure superheated steam become less of a problem.

 

https://www.zillertalbahn.at/page.c.../fuhrpark/dampflokomotiven&switchLocale=en_GB

Technical information on Unsuperheated simple and compound 0-6-2,superheated 0-6-2 running on Zillertalbahn

 

Essentially just handle, that is to say drive, the locomotive correctly.

The issue with the steam supply to the auxiliaries, particularly the injectors, I have covered. The injectors worked as they were supposed to given an adequate steam supply. If any experimental locomotive was plagued by a series of unfortunate events this was one of them. In May 1931 Gresley requested that an independent steam supply should be provided for the HP injector. This was confirmed with Stamer on the 9th, but the report by J.N. Gresham concerning the inadequacy of the supply dates back to the 12th Feb though in March and April a problem with automatic coupling on corridor tenders on the curves in Newcastle Central was being looked into and the locomotive was being weighed. 10000 was in Glasgow, for how long is unknown but there was an immediate requirement to fit the new steam supply upon the engine's return and this revealed that North Road had ran out of 1.5" pipe. All is revealed in a letter dated 29th May.

The company knew that they had work to do on the design of the boiler cladding, gas flow baffles and refractory brickwork and had been ongoing with models trials, revised ideas and drawings in 1932. A Kylchap double blastpipe was fitted in ' 35 but the choices made for the cutoffs to be used were unfortunate though. The engine at this time was not perfect but it would work well enough. If the LNER had been a wealthy company the future might have been different and they might well have built more than one prototype which would have accelerated the testing and development process. The results of these trials made after the fitting of the new exhaust reveal that if the engine had been driven at 90% cutoff on the HP the output would have been a little under 4,000 horsepower. If only those involved in the testing had a better understanding of what was needed. The engine was never given the chance to display its potential.

We can look at what compounds achieved elsewhere, draw up the diagrams and carry out calculations but people will choose to disbelieve them. Chapelon was more fortunate in first Pacific rebuild he could wave his calculations at people, put his engine out on the road and have it exceed his promises. Even then folk refused to believe the evidence.

 

Would a 4000hp loco have been of any use? You couldn't write diagrams around it use on the grounds what do you do when it is out of traffic.

I'm also curious about where the power comes from, since the sustained power output of a steam loco is fundamentally limited by the rate it can burn coal, and that is in turn limited by the rate one man can shovel it. If you built a 4000hp loco - could a single fireman be realistically called upon to fire it?

Tom

 

Well, a single 4000 hp loco, no. But given that in later years well over 4000hp has been applied to ECML expresses, I think the LNER might quite reasonably have considered, if they had one, what they could do with more of them. The P1 is an interesting comparison, since of course it turned out there was not in fact much they could do with them day-to-day that an O2 or O4 could not, but the job of that class was very different. A P1 could pull trains longer than the sidings could take, or it could provide accelerations that the rolling stock and methods of working could not accommodate. Conversely, if the LNER had suddenly realised an ability to run expresses of standard length and weight to the sort of timings they later applied to the lightweight named trains, this could be considered a significant competitive advantage. They might, of course, have decided that the W1 was too expensive, complicated and troublesome to multiply, for the advantage that this sort of power might give them. As @242A1 has said, they were not a wealthy company and in the largely contemporary A3 already had arguably the best express passenger locomotive active in that period (I will concede "one of the best" to any particularly rabid partisans). In the actual reality, of course, the A3 and the emergence of the A4 arguably finished off the W1, which wasn't demonstrating a clear advantage in any way.

As for shovelling of coal, I know mechanical stokers were tried and generally abandoned here, but there was never the duty to require them. In other countries they were used in great numbers; a certain critical mass seemed to be required for them to properly catch on and the required coal sizes be provided. The ability of a man to shovel coal need not be the ultimate limit.

So, it's a significant set of stacked what-ifs, but I think it's reasonable to say that a: if a 4000hp express passenger locomotive had come into existence in the form of the W1, the LNER might well have considered building more, and b: there is no functional impediment to a 4000hp steam locomotive in general, just that nothing ever got close within the British loading gauge.

 

The first batch of Chapelon 4-8-0 compounds had a grate area of 40.47 sq. ft, a little less than the 41.25 sq. ft of the Gresley Pacifics and these engines were 4,000 hp machines and were hand fired and so there does not appear to be any problem with regard to this.

If the design had been tested and developed properly it would, as 8126 indicates, have raised the question with the LNER's traffic department concerning what they might achieve with it and given the company's history and motto they would have wanted to use the power to commercial advantage. The LNER streamlined trains would not need to be so restricted with regard to weight, the public wanted speed and these trains made money. Quite what the knock on impact would have been with regard to other locomotive designs might have been we cannot be too sure but Gresley was working on a version of the K3 with a water tube boiler.

The LNER made use of its engines; the original A1 was designed to haul trains of 600 tons and did so when required. The A3 presented an improvement which in turn presented a problem for 10000 because the object of this exercise, as I know I have already stated, was to produce a more economical equivalent to the A1. That progress with 10000 was rather slow was a problem. The fact that no one knew how to use it was another and so the potential of the design remained hidden. The performance of the A3s and the arrival of the A4s both presented a step forward that the company was looking for and this step had to be made use of. It has to said that the LNER used these machines rather well, they not only handled the trains they also gained the company valuable publicity, even fame. Quite what they would have achieved if they had realised what might have been we can guess at. One thing is certain; if the 4,000 hp high pressure compounds had been successful the modernisation lobby might have had a few headaches and I am not imagining for one minute that development would have stopped with a production version the original design.

 

Why not have two firemen?
Wasn't the main issue with 10000 the suitability of the boiler for bumps and bangs of railway use?

 

You shouldn't need two firemen. If you have sufficiently low specific steam consumption figures then the demands made of the boiler are less, you don't need so large a boiler as you would with poor SSC figures; smaller boiler, smaller grate, less fuel needed and less demand made of the fireman.
The boiler was fine. The casing, ducting and baffles needed more development but the boiler itself does not appear to have been an issue. The paper held by the NRM are quite revealing in this respect.