GWR 111 "The Great Bear" and Surrounding Controversies

A locomotive that has fascinated me, as well as I'm sure a lot of you, is GWR 111, the first UK pacific and really an enormous locomotive by any standards, being larger than any contemporary pacific on the continent and as heavy as the last South African pacifics about 30 years later. It was really a bold step in engineering considering the general conservatism in locomotive size in Great Britain especially and was the first time a locomotive with a truly large grate was tried in British practice, ten square feet larger than the Ivatt atlantics. Its long, fat boiler and massive cubic belpaire firebox more closely resembled the big Hendrie engines of the Natal Government Railways than anything on the British isles.

Whenever the 111 is brought up, it usually tethered to stories of how the locomotive was reluctantly built and essentially a failure from the get-go, becoming a dead-end of GWR and further British homeland locomotive design. I think most of these criticisms flung at the 111 were either unwarranted or relatively minor and it was a major loss at the time to abandon the 111 and lessons learned from it.

The locomotive has a bad reputation for being hard to fire, but is this the fault of the locomotive itself or the GWR stokers who were not used to the properties of a wide firebox? While I have no definitive evidence, I have a feeling this was more an issue of misuse than a flaw with the grate. Now the boiler did have some faults, including the 23 foot tube plates which were likely indeed too long for what was required and were a detriment. Even American locomotives with large fireboxes at the time didn't see the need for tubeplates that long.

This thought can be considered a general engineering concern at the time and definitely not unique to Churchward; this was right in the period (particularly in the United States) where widespread trial and error were being conducted to see how the properties of the steam locomotive were affected when enlarged and it was usually found the hard way that simple stretching and enlargening of components often hampered the performance of locomotives. This took years to sort out and is why things like the Cole ratios were created. The Great Bear was a prime example of the difficulties in building larger locomotives than what was used to. While the boiler could have had considerable improvements made, most likely in shortening the tubes by an extent and installing a combustion chamber (although that has its own range of construction issues) and greater superheat, calling it "a Star lugging extra weight around" is completely unfair in my opinion.

Now we come to the issue of the necessary size. The 111 as I said was big, almost unwieldy for the time and that was likely its main downfall. Axle loading was just too restrictive for the time for such a thing. But this still feels like another area where lessons could have been readily learned instead of passing the loco as a disappointment. The peculiar ratios of the boiler and length could have been addressed, and following designs could have likely created a monster boiler and large firebox with a much more manageable size. Even past this, it had much more route availability by the mid-twenties. I suppose it was felt unnecessary to tinker with 111 when ten-wheelers were getting the job done with relatively impressive results, but the GWR probably could have had something far better if they recognised the potential in 111, as much as I do like the Castles and Kings.

Tell me what you think about all this.

 

This could develop into an extremely informative conflab. First thought: Were combustion chambers a known quantity when 111 was built? I'd be especially interested in views, supported by citation, concerning the inside bearings on that trailing axle.

 

Combustion chambers had been used infrequently since the mid 19th century in the US and started being incorporated in larger numbers around 1895-6 (Tom Morrison's The American Steam Locomotive in the Twentieth Century). I think the early Gresley 2-8-0s had something akin to a combustion chamber, which would have been 1913.

The Hendrie engines of the Natal Government Railways had proper combustion chambers (combined with an arrangement similar to a Gaines Wall) as early as 1904. In their case it wasn't about the length of the tubeplates but rather since the bottom of the firebox was nearly level with the boiler bottom an arrangement was needed to raise the gases so they wouldn't rush straight to the tubes. The Prussian P8 had somewhat a progenitor of a combustion chamber as a small extension of the firebox. I'm sure there are a number of obscure continental locomotives with combustion chambers or similar around the time.

The combustion chamber was definitely something internationally known by the building of 111.

 

Aside from the "not invented here" syndrome (not entirely unknown in Swindon!), was there much by way of empirical evidence concerning benefits? Churchward had a progressive, scientific approach to design, so it seems unlikely he'd have overlooked such a golden opportunity, presented by a boiler of those dimensions. Was there perhaps an unfortunate earlier example in UK railway engineering? If so, was it a case of poor design , or some more practical issue?

 

Interesting subject. I will leave the experts to comment on the specific features of No 111, but draw attention to a some bits of background information.

Apart from the Ivatt GNR and Marsh LBSC Atlantics, there were a few other early examples of large wide fireboxes in Britain, pre-dating the Great Bear. The GER built an experimental large tank engine:

https://www.gersociety.org.uk/index.php/locomotives/j-holden/a55

The GWR itself had built a small-wheeled 4-6-0 goods engine No 36 with 35 sq ft grate in 1896, followed by the "Krugers" in the early 1900s:

https://en.wikipedia.org/wiki/GWR_2602_Class

The Great Bear was the third Pacific class in Europe, following the Paris-Orleans Rly Class 4500 and Baden Rlys Class IVf of 1907. By 1914, Pacifics were widespread across France, Belgium and South Germany, mostly around the same size as the Great Bear. The numerous French Pacific designs failed to achieve the expected performance until Chapelon sorted them from 1929 onwards. The Baden IVf was not a great success and the class was scrapped in the 1920s, but Germany's next Pacific design, the Bavarian Rlys Class S3/6 of 1908, was much better and worked until the 1960s.

https://en.wikipedia.org/wiki/Baden_IV_f
https://en.wikipedia.org/wiki/Bavarian_S_3/6

 

Beattie had fitted what were in effect combustion chambers to some of his locos in the 1850s ...

Tom

 

Forgot about his complexities. Do I recall those were later removed?

 

The Krugers had combustion chambers, and apparently trouble was experienced with them, so Churchward was very much against trying one again. Oddly though some of the Kruger boilers, working at lower pressure, served as stationary units in Swindon works until the 1950s.

Axle loading is an interesting one. The GWR limit at the time was 19.5 tons, and the Bear stretched it to 20. I believe the Gresley A1s were also 20 tons, increased to 22 with the A3s. Partly because of the long tubes the Bear boiler in original form had a very large superheater area, slightly greater than an A1, but I've seen it said that in practice the steam temperature was lower than that from a Std 1 boiler.

The Bear was similar in several major dimensions to an LNER A1, (90 against 91 tons for instance) and as we know that was not as good a locomotive as a Castle, although the Bear had better valve gear than an A1.

I don't think anyone rational could accuse Churchward of suffering from not invented here.

 

Logical, perhaps, but I've never claimed to be rational. I've seen far too many who do !

 

As in all these cases irrespective of if the loco was good or bad, was there a job for it to do?

If no, however good it was then thats it.

The one job it did seem to be good at, fast freight, resulted in the GWR building the limited number of 47xx's then the Granges, which suggests that The Great Bear was capable of replacement by a 2cyl 4-6-0

 

Boiler length lessons took a while to sink in. Raven did the same with his Pacifics - they were sometimes referred to as "Skittle Alleys", and for good reason.

 

The Great Bear gets a lot of unfair press.

I personally think it was a success: it was the first pacific, proved the wheelbase, showed there was capacity for larger locomotive thinking and it worked revenue earning trains regularly, and over many years. So it passes the first test that so many of the experimental steam locomotives fail at.

I have noticed that between big four enthusiasts there is a trend to be more critical in some areas and more lax in others where locomotive design is concerned. If we were being consistent, we'd acknowledge that really, The Great Bear was a locomotive ahead of its time on the GWR in terms of track and axle limits and that is really what counted against it. Nobody would have batted an eyelid at that a decade later.

 

The key thing with boiler length is the A/S ratio of the tubes (mean hydraulic depth, see A.F. Cook's book Raising Steam on the LMS and elsewhere), i.e. if you increase the length of the boiler you need to increase the diameter of the tubes in proportion. Churchward was clearly aware of this and installed larger tubes in the Great Bear than the Star, but I will need to check up on the actual dimensions.

 

2.5in against 2 in. The other factor that occurs to me, though, is that the plain wide box has much less heating surface in proportion to the grate than a wide box. This too seems to have been considered in the first iteration of the Bear boiler with arch tubes, but it seems to me that a combustion chamber must provide much more capture of radiant heat.

Success is too strong, but it certainly compares favourably with the least successful designs of some other designers /design teams.

There were basically 3 iterations of upgrading the Star chassis with more boiler capacity. The Bear was the first, the second was the Standard 7 boiler Star - a shortened Bear barrel on a bigger narrow firebox, which didn't get off the drawing board due to weight, and then the Castle, which had a Std 7 boiler with a smaller barrel to keep the weight down. The Castle seems to have surpassed expectations, and, at least with reasonable coal, negated any real need for a 4-6-2 on the GWR. What was needed, one may guess, was a grate rather wider than a narrow firebox, but still in a rectangular rather than square box. Unfortunately the need to fit the rest of the locomotive into the design - those pesky wheels, frames and things - precludes such! A study of Beyer Garrett boiler proportions might be interesting as there were presumably fewer constraints.

 

You'll forgive me I hope Jim for disagreeing - looking at it in the round, it basically did the same sort of work that a King did later on. If we say that the King class is a success, despite its red route availability, then The Great Bear surely with 527,272 miles done between 1908 and 1924 (16 years work) on similarly limited work is at the very least successful from the point of view of the railway's operations? Daily revenue earning service for 16 years is a very good return on investment for a locomotive prototype, I dare say. If we're arguing it didn't set the world on fire with its performances, I can see that most certainly, but success must surely be defined in its return on investment for the railway? In that, both for publicity and the actual work done - it is a successful locomotive, don't you think?

I do agree with you on the latter point - it puts Leader, Fury, the Paget locomotive and other similar engines absolutely in the shade - but is comparing it to them fair? I rather think The Great Bear cannot be classed alongside those engines. It was good at the day job and the others didn't get to that stage!

 

My reading of "Raising Steam on the LMS" gave me a strong impression that the LMS were all at sea in boiler design, and spent a lot of time with revisions to tube layouts, diameters and superheater / tube ratios before getting close to success. (Heresy alert): an awful lot of Stanier's designs seemed to be pretty poor straight off the drawing board and only became successful after a lot of iterative modification.

Caveat: We have two very good sources for how the design process worked on the LMS - Langridge and A.F. Cook. If I am critical, it should at least be tempered by the fact that it is based on a review of LMS design that is maybe not as clearly documented for other companies: perhaps they also struggled but without the subsequent insider accounts to lay it bare. Even so, reading Langridge / Cook on the LMS; KJ Cook on the GWR and Holcroft on the SR, you get three very different pictures of what was exercising the respective drawing offices.

Tom

 

I'm not sure the Bear did the work of a King. I understand it was on a link with the prototype 47. It certainly seems to be true that it was a competent locomotive at fast freight, but it wasn't really used on the top express trains. All this backs up the thought that the boiler wasn't producing adequate steam when under high demand. At best, perhaps the equal of the Star, but with a lot more reserve when the steam required was within the boiler's capacity. Yes it could do useful work - but work that was within the reach of the surely much cheaper 47s. So not by any means a waste of money, but not a good return on the investment either.

Tom, I've heard it said the design department of the LMS wasn't well regarded by the staff of the other companies. I tend to agree with you about boiler design. Its well accepted that the LMS couldn't get the Red 5s to perform adequately until they were given high superheat, with all the extra expense in lubricants and maintenance that entailed, and yet the GWR boilers that they had drawings of to refer to performed more than adequately with moderate superheat. So surely something went badly wrong in the implementation. Maybe it was "Not invented here" in the LMS drawing office? I'm probably a biased onlooker, but I seem to see a fair bit of that in Cox.

 

Not sure I agree with that, Tom. The first two Pacifics, 6200/01, were not the best steamers in the world but adequate. The three-cylinder 4-6-0s were the problem children, and all sorts of reasons were given for this, and it was a combination of all these - tube diameters, draughting ratios combined, while the low superheat, often blamed, didn't cause the problems but did exacerbate them when they occurred. True to tell, the 5Xs were never reliable steamers right to the end, improvements and all. But the two cylinder engines were good from the start. Lessons learned in getting the 5Xs to steam and applied to them made them better, but they were good enough to meet traffic needs straight from the drawing board. The lost cause was the Class 3 2-6-2T, which defied all efforts to make it go.

The LMS wasn't static in design; it learned as it went along and applied the lessons. Hopefully, the others did the same. An example was the Class 3 boiler family, which started out with vertical throatplates. A later improvement to be made was the sloping throatplate with a small combustion chamber, and so this was adopted even though these boilers were not directly interchangeable with the earlier ones, so there were two types of boiler for the Black Fives, 8Fs and 5Xs. This of course made a nonsense of standardisation and would never have been allowed on at least one other railway at the time. But which was the better option: to retain a good design throughout, or 'upgrade' to a better one when discovered?

And be a bit careful with Eric Langridge. Having started on the LSWR, he was very anti-GWR. And where did Stanier come from? Swindon! Langridge goes to great lengths as to how Derby weaned Stanier away from every unsuccessful GWR feature, not mentioning the many that worked, worked well and were retained.

 

On the 5Xs, did the two which were rebuilt with bigger boilers, 5735/6 do any better with steaming?

same comments on the class 3 2-6-2 tanks as some of those were also fitted with larger boilers?

 

No doubt Didcot will eventually produce a new build recreation to provide some answers.