Compounds ex time machine thread

A word that I frequently use - complicated. It was and remains complicated. Just because you obtain a sample to test if you don't fully understand what you are testing you have a problem. Also if what you are testing is not a representative of the close to fully developed concept but is merely a sample of a stage in the development process and you assume it to be representative of the final stage you have a problem.
It is not a simply a case of smart enough, it is about education and knowing enough. It is also about being open minded enough to realise that there is more to learn and that you are on a journey which you will probably never complete.
The GW obtained some French compounds but didn't really understand what they were buying. How could they? Compounds had been under development on the continent for years but the GW were not a part of that process. Some on the continent knew that there was development to be made, that the journey was not complete. How to complete it was another matter. The GW were divorced from the development stream. Ideas can flow so much more easily now. Then matters were more awkward.
We didn't quite work compounding out on our railways. We came close. Know enough and soon enough, we were short of sufficiently capable people. As well as resources, time, support and more.

 

But if you can drive your working pressure up to, say 450 psi .............

 

The only British compounds which seemed to be successful were the Midland ones and they didn’t offer a massive advantage over comparable simple expansion 4-4-0s, an LNWR George would outperform them

 

Compounding or not coumpounding, GPCS, Kylpor/Lempor, roller bearings, etc.etc. all interesting huge improvements. But also complex.

And in many cases the simpel are often the best.

this link shows a photo of our class 2a No.17 "Caroline"https://dms-cf-06.dimu.org/image/02Tzm6nGmD?dimension=1200x1200
Build in 1861 by R. Stephenson she spend 92 year! in regulary service.

No superheater.
No GPCS.
No streamlined steampassages,
No Kylpor,
No high efficiency compared with modern engines.
Just traditional, simpel, robust, pure BRITISH locomotive engineering!

Regarding france, Amongst the last withdrawn steam locomotives in France, there where two 2-8-0 cylindered simpel tender engines ; built by North British.

Kind regards

Knut

 

One of the obvious reasons compounding in steam locomotives was rarely used in any country was because something like 90% of locomotives constructed were 2 cylinder machines as this is the simplest and cheapest way of building and maintaining them, (especially with outside cylinders) so if compounded you couldn't have a small high pressure cylinder on one side and a large low pressure one on the other.
3 or more cylinders on a non articulated loco only makes sense if you need very high power at high speed ( ie express Engines) and in the case of Britain the restricted loading gauge effectively put a limit on how big the outside cylinders could be, so having extra cylinders between the frames allowed you to get round this.

Compounds were tried but I think of the Midland Compounds as the only successful British production ones. As a previous poster wrote Churchward did buy 3 DeGlenn compounds from the Nord railway in France and they were tested against simple expansion locos the GWR were producing to see if there was any economy savings in coal and maintenance etc, but the differences in practice were found to be negligible so the idea of using it was dropped.

One thing the GWR did copy from the De Glenn compounds was the extension frames and front bogie design. Engines like the Saints, Castles, the original Halls all have the front bogies based on the French designs. Only with Fredrick Hawksworth did it change.

I think a GWR style DeGlenn compound would make a good new build for someone (say Didcot) and there is an original one in the French railway museum in Mulhouse which I have seen a long time ago.
Compounding ( in some cases in 4 stages) was used a lot in Steam ships of course. You can see this in the Waverley Paddle Steamer for example. ( think that is triple expansion)

 

There was quite an interesting article in one recent edition of Steam Railway on the GNR N2 class which explained it. I didn't previous know this, but they had condensing equipment fitted because they had to regularly operate along the London Underground tunnels to Moorgate station. I hadn't realized, but there was a flap in the smokebox on the blastpipe which the driver could control from the cab, so normally the blast would just go up the chimney as normal, but when the flap was operated it would be directed along pipes on the outside of the smokebox into the water tanks ( to condense), so it was only operated when approaching/inside the underground tunnels (to reduce steam and smoke in the tunnels) . Some drivers apparently didn't use it. Apparently the flaps would jam up after a while with ash so fitters had to regularly clean them to keep them working. The surviving N2 must still be fitted with this but not sure if it works still and of course it wouldn't be needed today.

In South Africa locos with Condensing tenders were built to save water as water was in short supply in desert areas where they were operating so it was done so the water could be reused.
I read at least one of these tenders has been preserved there.

 

That’s not quite true - the Wordsell- Von Borries compounds on the NER had two cylinders, HP on one side and LP on the other. They were clearly successful enough that a reasonable number were built, but not sufficiently so that they were continued. The LSWR tried one out on one of their Adams two outside cylinder 4-4-os, but clearly saw no advantage as it was rebuilt into conventional form. In an era of cheap coal, clearly the thermodynamic advantages were too marginal to outweigh the extra complexity and other difficulties.

Tom

 

The condensing systems discussed above for underground work are a very different concept from that used with compound marine and static power plants where the condenser is to create a vacuum as well as heating the feedwater. That sort of condenser was as far as I know only used in conjunction with turbines on locomotives; not entirely successfully as the boilers were then too small.

Condensing just for feedwater heating was in vogue for a time in the Victorian period, on a few Dean engines it can be identified by exhaust pipes on the back of the tender. It wasn't persisted with for very long.

The alternative to compounding, for reducing the losses due to temperature changes within the cylinder on expansive working was the uniflow cylinder, which again added complexity and weight.

The DeGlehn compounds available to Churchward were not as efficient as they could have been, according to Chapelon's later investigations and analysis. Had those improvements to compounding been implemented a few decades earlier perhaps GWR locomotive design might well have followed a different course.

 

Firstly what we are looking at is reducing costs, and of course you can attempt to reduce your coal costs at the costs of significantly increased capital or maintenance costs.

Secondly as has been pointed out coal was cheap in the UK unlike France where it was expensive and often imported. In addition to this you have a more generous loading gauge and there may well have been operational differences which may have worked in favour of compounding.

Finally of course there is a significant difference between the training of French & UK footplate crews - the French crews were all trained engineers capable of getting more out of their locos and the availability of capital - or lack of in the UK

 

And were paid accordingly...

 

Outside the UK Wordsell-von Borries compounds were very successful in various places with poor or no local coal supplies. There were lots on the British railways in Argentina, although most of them moved away from compounding in the 1920s or 1930s. But some were in service (as compounds) until well after WW2.

 

What, very badly...?

 

That's very true in theory. However, with a conventional locomotive valve gear (Stephenson's, Walschaerts, etc) the point of cut off and point of exhaust are inextricably linked and the earlier the cut off, the earlier the point of exhaust so you can't get that dame amoun of energy. If you go to a Caprotti or other gear that gives independence to to point of exhaust it becomes more of a possibility.
The Midland compound was relatively successful but it proved possible to only fit three cylinders so the power output was never going to approach that of a simple locomotive and, as UK steam locomotives developed, power was generally considered to be a prime consideration.

 

W.M.Smith also produced two four-cylinder 4-4-2s for the NER. Fowler nearly produced three (or more) four-cylinder Pacifics for the LMS. The issue was fitting the two LP ones, they had to poke through the frames in a rather an alarming manner. The four-cylinder compound Dreadnaught was perfectly successful.
Cox (in his 1947 paper and reproduced in one of his books) has a diagram showing that the LMS compounds (Midland 4-4-0s and compound Dreadnaught 4-6-0) were only efficient on front-line work, and lots their advantage in fuel consumption as they were downgraded to secondary work. Whether this was due to less sustained high output work (where they could be run expansively) or less highly trained drivers is an interesting unknown.

 

Although it is undoubtedly possible to run a simple loco expansively, it often wasn't done as much as it should have been in practice, not least because on many two-cylinder engines it made the footplate uncomfortable!
Unfortunately the comparative tests in the UK (basically the GW ones by Churchward and the early LMS ones) were extremely flawed and there isn't enough data to prove anything at all...

 

In fact du Bousquet, who died in 1910, anticipated Chapelon to a great degree (which the latter readily acknowledged) with his smaller wheeled 4-6-0 version of the Atlantic. As @242A1 has said, the 70 ton 4-6-0 became at least the equal of the 89 ton King but much of the development work was done in du Bousquet's lifetime. He had spotted the need to open out steam passages. Churchward was trying to prove that "his" simples were equal to the French compounds. He did not realise that the standard had been upped.

Much has been made of loading gauge advantages in France but there is little reference to the restricted axle loadings permitted. The need to improve adhesive weight without adding to axle loading had led to the Atlantics being developed into the 4-6-0s. Much later the style of the splendid 232U1 was cramped by the heavy axle load brought about by its tough build restricting its route availability.

PH

 

The LMS compounds had 21" by 26" LP cylinders and 200 psi boilers, giving 22,650 lb nominal TE. If you keep the 21" diameter and increase the stroke to 30" and the boiler pressure to 280 psi, the nominal TE becomes 36588 lb, not much different from an A1. Similar proportions to 242A1 would imply a single HP cylinder about 19" by 25" or 18½" x 26", similar to the A1's cylinders. So you end up with similar power to an A1 and potentially slightly higher efficiency. But I am inclined to agree that the improvement in efficiency would be too little to justify the extra complexity and the extra reciprocating inertia.

 

The longer wheelbase would probably preclude the use of cylinders as large as on the 4-4-0s

 

the Teutonics did fine work on the Corridor . a 46 ton engine taking 300 tons + day in ,day out ,let alone 1890 , is good work in any language . you would have been happy to see a class 5 do that in 1960

the 1st 5 Smith Johnson Midland compounds were also very good - fast and strong . the rebuilds were definitely a de-tuned version for ease of handling . Definitely a candidate for a new build for the S&C imo

the W.Worsdell 3CC ,3 cylinder compound was also a good loco .certainly the best of the NER designs and probably on a par with the original Midland locos.

 

my understanding is that although the Midland compounds were 200psi only 175psi was available to the cylinders . Why ? I have no idea