The rise of the 4-6-0 and why were there none on the GNR

[drmditch]

Many thanks to everyone who has posted on this thread. A lot of interesting information has been brought out.

However, I still think that there are some fundamental questions that have not been answered.

The development of high speed locomotives for long(er) distances which took place prior to and after WW1 produced Atlantics and Pacifics on the lines that became part of the LNER, and 4-6-0s on the GW and LSWR. The points mentioned above about intended distances, primary air to the grate, and ashpan capacity are all very valid. The development of long-travel valve gear and adequately-sized valves is extensively discussed in many places!

What particularly interests me however, is the secondary, shorter distance, and sometimes heavier services. Here the engineering traditions and design philosophies differed. (For ease of reference I will refer to these as 'secondary' services, although in terms of revenue-per-train-mile they probably weren't secondary at all!)

On the NER, although the Vs and Zs dominated the fast passenger (and some of the faster freight!) there was a policy of providing large strong engines for the 'secondary' services. When combined with Raven's three-cylinder policy this produced the S3/B16. (Although it is interesting that there was an outline scheme for a larger-wheeled sister of the B16 for fast passenger work.)

One could say that the situation was similar on the GCR - with a series of big 4-6-0s clearly produced for the secondary roles.

There are two interesting parallels. The B15, with a boiler similar to the very successful Q6 was not developed further, or it's reputed steaming problems resolved. The three cylinder B16 was seen as the way forward.

The B6, with an 'engine' portion the same as the very successful O4, was apparently useful and well liked - but only three were built and Robinson clearly thought the four cylinder B7 was a 'better job'.

So - even during and after WW1, neither Gorton nor Darlington was particularly worried about complexity of maintenance. Clay and Cliffe ('The LNER 4-6-0 Classes') suggest that the six eccentrics sharing the leading driving axle with the inside crank webs made things ' a little crowded'!

Both B16 and B7 were broadly similiar in wheel-size and intended use. Both lasted, despite their reputed high coal consumption, to the end of the LNER and beyond.

So, while Gorton and Darlington, both with a tradition of building robust locomotives, were building these sturdy and strong 4-6-0s, what was happening on the GNR?

Well, of course, what was being produced for broadly the same role was a large-boilered 2-6-0, the K3.
The tradition, by 1920, of Doncaster did not include any 4-6-0s, but did include 2-6-0s, and there was a logical path of development, possibly stretching as far back as the short-lived (which is was what they were designed for) American locomotives of the tail end of the 19th century.
Again, a multi-cylindered locomotive was prefered, and despite all the supposed controversy, I would see Gresley's conjugated gear as being intended to reduce maintenance rather than otherwise.

So, at Grouping, three different engineering traditions, with three types of very useful engines. The K3 had the advantage of it's superior front-end design, but the other two were possibly more robust and certainly a lot better riding.

Gresley was not a product solely of Doncaster. The LNWR and LYR made extensive use of 4-6-0s as 'secondary' locomotives. Nor was he dismissive of the traditions of the other major works. Further B7s and B16s were produced (although probably at least some of the parts were ordered by the older companies). He also selected D11s as being suitable for ex NBR lines, and A5s for NE Area. In many areas, Gresley was clearly open to new and different ideas.

Why therefore, in reviewing the locomotive stock of the new company, did Gresley not take the opportunity to create a 'secondary' locomotive which combined the best of everybody's expertise, and could have been built in quantity when required?

After all, the J38/39s were designed at Darlington with valve gear refined from the GCR A5s.
When there was a perceived need for new 'secondary passenger' locomotives in the NE and Scottish areas, the J39 boiler was used on a Darlington 4-4-0 with a refined version of the Gresley 'front end'. In the activity created by the need by a new express locomotive for the lightly-built GE lines, Doncaster asked Darlington for drawings of the B16 and the D49.

Why was the need for a durable 'secondary' locomotive not seen earlier, and the experience available not used?
The pre-grouping locomotive stock was not going to last for ever, and despite the drastic reductions in revenue, some new locomotives were being built.

Of course, in any discussion of this sort, hindsight is always easy. But I think there are four questions, one minor and three key:-

Firstly, why did no-one recognise the potential value of the B6?

Secondly, why was the K3 built in quantity rather than an 'integrated' design (which might have had better riding characteristics and perhaps wider route-availability.) Perhaps there could have been a two and three cylindered variant.

Thirdly, why build the D49s at all, instead of a more versatile locomotive?

Fourthly, why did it take so long (most of the life of the LNER in fact) to create a locomotive like the B1?

Much though I would like to have seen a fleet of V4s, and much though I admire HNGs design practice, one has to wonder whether something better for a 'secondary' locomotive been acheived earlier than it was.

[earlswood nob]

Good morning all

A lot of very interesting information and opinions in this thread.

I have enjoyed reading it.

Just to prove that I've read it through, there is a slight typo; The B6 had the same boiler (5'6in diam)etc as the O5, not the O4 (5'0in diam).

I think the B6 would have been very useful if multiplied. I've also wondered about the B6 being reboilered with the O2 boiler, like the O4/5 and O4/7, but I guess the B1 coming along put paid to any ideas such as that. There must have been enough spare O5 boilers around with the conversion of O5s to O4/6s.

Earlswood nob

[Hatfield Shed]

...After all, the J38/39s were designed at Darlington with valve gear refined from the GCR A5s.
When there was a perceived need for new 'secondary passenger' locomotives in the NE and Scottish areas, the J39 boiler was used on a Darlington 4-4-0 with a refined version of the Gresley 'front end'. In the activity created by the need by a new express locomotive for the lightly-built GE lines, Doncaster asked Darlington for drawings of the B16 and the D49.

Why was the need for a durable 'secondary' locomotive not seen earlier, and the experience available not used?
The pre-grouping locomotive stock was not going to last for ever, and despite the drastic reductions in revenue, some new locomotives were being built...

Lack of money I think is the short answer.

Deliberately quoted what you posted from the sentence containing J38/39. Gresley's first proposal to the locomotive committee for this medium freight requirement was a two cylinder 2-6-0 based on experience with the K2, not a million miles distant from what would emerge 25 years later as the Peppercorn K1. Possibly one of the worst decisions the loco ctte made was to reject this in favour of the cheaper inside cylinder 0-6-0 type; I suspect the first cost savings were quickly 'eaten' by the maintenance burden resulting from J39s frequently being made to run fast - which they were perfectly capable of in power terms - and which likely would have been consideraby eased if the 2-6-0 had been the choice. (The K2 had a very successful career.)

Likewise a big 2-6-0 mixed traffic type rather than a 4-6-0. Maximum tractive wallop per pound, and the K3 did the job to the end of steam for about 10% less steel than the GC and NE 4-6-0 alternatives.

There clearly wasn't a systematic attempt at a new secondary design, until the V4. The D49 and B17 were 'custom jobs' to fill immediate needs. In hindsight a single spend of the D49/B17 outlay to earlier create a secondary general purpose loco design would have been a better move. But that's hindsight. At the time the inherited designs were still largely good enough for the secondary roles, and that problem could be left for another day.

[65447]

Accepting all of the points about the wide firebox and so on, surely an important factor was the ride quality - almost all of the locomotives lacking a trailing axle had reputations for rough riding, especially the Thompson B1?

[drmditch]

Thank you, Hatfield Shed.

One query about the 1924 proposal for a 2-6-0. According to Geoffrey Hughes ('The Gresley Influence' , with outline drawing, and his biography of Gresley) this was for a 3 cylinder locomotive, effectively a 5' 2" version of the K3, and was rejected on grounds of cost. I'm sure I've seen another reference to this proposal somewhere else, but can't locate it at present.

I do agree with your comment regarding the D49 and B17, although as regards the latter I wonder whether anyone ever made a sensible cost comparison between upgrading the GE civil engineering, and the difficulty of designing and building a bespoke locomotive.

I wish they had! Then I might have grown up with a (distant) view from my bedroom window of Pacifics and V2s on the GE main line!

Your point about the weight savings of 2-6-0 vs a 4-6-0 is well made.

In a historical perspective, the structure of the railway companies and their accounting practices underlie a number of decisions that seem strange a century later. Even as late as the 1940's we have the term 're-build' being used for what were effectively new engines. (Although I'm thinking particularly of the LMS Royal Scots there.)

In relation to Earlswood Nobs post regarding the B6s. I understand (from Jackson/RCTS etc) that the O5 and B6 shared their cylinder and valve gear design with the O4, although, of course, the boiler was larger. I have still to find a convincing argument as to why no-one on the GC didn't do more with the B6.

Perhaps our perspective is difficult. I suspect that no-one in the railway business foresaw the massive disruptions to revenue following WW1 and then the Grouping. Neither Gresley, nor Raven, nor Robinson thought that some additional complexity at the front end of locomotives would add significantly to the cost burden of maintenance.

Lastly (I must stop making these long posts!), I was speaking to the crew waiting for 61264 at Grosmont on Tuesday. They reckoned it was stronger than the LMS Class 5s, but the vibration when it runs at speed on it's trips to Battersby makes it difficult to sit down!

[earlswood nob]

Good morning all

I enjoy reading the post on this subject, so don't stop , DRMD.

Digressing slightly; I've often wondered why O5s were rebuilt with a smaller boiler to become O4/6s, and O4/1s and O4/3s were rebuilt with the larger O2 boilers.

Were the O5 boilers inferior to the O2 boilers? The O2s had slightly greater superheater area, but other boiler dimensions were comparable.

Earlswood nob

[Wainwright]

" I have still to find a convincing argument as to why no-one on the GC didn't do more with the B6."

What was the hammerblow with the B6s?

There was a lot of incipient paranoia about hammerblow in the more powerful 2-cylinder locomotives in the period, with the Bridge Stress Committee, and all...

[Wainwright]

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

[drmditch]

What was the hammerblow with the B6s?

There was a lot of incipient paranoia about hammerblow in the more powerful 2-cylinder locomotives in the period, with the Bridge Stress Committee, and all...

David Jackson's ' J G Robinson - A Lifetime's Work' page 199, states that:-
'on the basis of the Bridge Stress Committee's figures No.416's (the first 8N/B6) design showed a marked improvement in this area over all of Robinson's other 4-4-2 and 4-6-0 classes using two outside cylinders, save for the 'Fish' engines.'

Mr Jackson goes on to state (page 202) that 'if the (two-cylinder) concept had been followed it would have meant detailed and fundamental changes in boiler pressures, heating surfaces, cylinder and driving wheels, frames, bearing and balancing. Besides which Robinson had, on the building of the last two class '8N' determined on four cylinders, not two.'

I wonder if this was Robinson looking towards the heavy express design which followed, rather than the general purpose 'secondary' requirement. After all, in 1919/20 the GCR already had a quantity of such locomotives.

On the other hand the book also has several quotations from enginemen approving of the 8Ns - '...The equal of any GCR 4-6-0'.... - and 'like a Tiny (O4) in every way'.

[Hatfield Shed]

...Digressing slightly; I've often wondered why O5s were rebuilt with a smaller boiler to become O4/6s, and O4/1s and O4/3s were rebuilt with the larger O2 boilers.

Were the O5 boilers inferior to the O2 boilers? The O2s had slightly greater superheater area, but other boiler dimensions were comparable.

The Gorton boiler development stalled at a 26 sq ft grate, and I don't know the reasons why, never dug into the history of that operation. It is pretty evident that the combustion rate on this grate, and rate of heat transfer to the water and steam in their boiler and superheater layout, was performing optimally for the 3P 4-4-0, 4/5MT 5'8" wheel 4-6-0 and 7F 2-8-0 power demand, in two cylinder format. These are generally attested as good practical locos that did their work well. And they never found a way forward beyond that, despite experiments with four cylinders, larger boiler volume and much else.

I suspect a bundle of factors, starting with the need for a grate area increase and improved ashpan draughting, and also firebox circulation, increased superheater area, valve events and steam tightness, blast arrangements; all of which Doncaster was clearly ahead with. Lots of little gains in detail design in these factors, added up to better boilers with more heat supply from the grate and thus the ability to steam at higher rates and put more superheat energy into that steam than the Gorton boiler, and engine and blast arrangements that got more power and draught from a given steam supply.

[earlswood nob]

G'day all

Very interesting, Hatfield Shed.

Most of the O5 were at Mexborough, and I've read that they weren't very popular on winter trips over Woodhead as the firebox was insulated, unlike the O4s, making the cab colder.

Earlswood nob

[Hatfield Shed]

The impression most strongly given me by loco crew over the years is that provided the machine steamed freely, they were basically in a good place; and much else could be forgiven or ignored. But poor steaming was always purgatory, and nothing in the way of any other sophistication or crew comfort compensated. The problem was of course that once any loco or class got the rep of 'poor steamer', it was very difficult to shift and became self-fulfilling.

[Wainwright]

Thanks for the info about the Bridge Stress committee, Drmditch.

I guess my query was more about the difference in hammer blow between the B6s and their 4-cylinder equivalents, although it is interesting that they were an improvement over the previous 2-cylinder types...

I had read that the Bridge Stress Committee clarified a lot of things that were not fully understood previously (ie before 1920) so it is possible that Robinson's choice for multi-cylinder designs was based on concerns about hammer blow, but in a 'hunch' sort of way, not based on rigorous testing?

Also regarding hammerblow, 2-cylinder 4-6-0s, and later LNER developments in the 20s, Wikipedia in its article on the N15 / King Arthurs (haven't checked the sources yet) has this to say about Gresley's involvement in post-Sevenoaks derailment panic - apparently comparative trials of Urie/Maunsell King Arthurs and Maunsell 'K' and Maunsell/Holcroft 'K1' 2-6-4Ts were carried out on the Great Northern:

"No. E782 Sir Brian was used on the former Great Northern main line for performance trials against the SECR K and K1 class tanks following a railway accident at Sevenoaks, Kent in 1927. The tests were supervised by the London and North Eastern Railway's CME, Sir Nigel Gresley, who commented that the class was unstable at high speeds. The instability was caused by motion hammerblow and exacerbated by irregularities in track-work. This caused excessive stress to the axleboxes and poor riding characteristics on the footplate. Despite this, the class benefited from an excellent maintenance regime."

(I'l have to get Holcroft's autobiography out of the library again to check his account of this!)

This is written as if the issues were with the N15s and not the Ks, which may be incorrect, but does (if further evidence were needed) suggest Gresley had a suspicion of hammerblow in 2-cylinder locos.

These trials of Southern locos are also interesting in that they may have led to the dumping of another LNER proposed 'secondary passenger' class we haven't discussed, the possible passenger 2-6-4T for the Southend routes to be built as the same times as the B17s. In the end, more B12s were built, but what an LNER3-cylinder 2-6-4T for fast and semi-fast usage in the Thames estruary would look like is interesting, given the LMS had some, and The K1s already referred to were active on the south bank of the Thames...

[Hatfield Shed]

...Also regarding hammerblow, 2-cylinder 4-6-0s, and later LNER developments in the 20s, Wikipedia in its article on the N15 / King Arthurs (haven't checked the sources yet) has this to say about Gresley's involvement in post-Sevenoaks derailment panic - apparently comparative trials of Urie/Maunsell King Arthurs and Maunsell 'K' and Maunsell/Holcroft 'K1' 2-6-4Ts were carried out on the Great Northern:

"No. E782 Sir Brian was used on the former Great Northern main line for performance trials against the SECR K and K1 class tanks following a railway accident at Sevenoaks, Kent in 1927. The tests were supervised by the London and North Eastern Railway's CME, Sir Nigel Gresley, who commented that the class was unstable at high speeds. The instability was caused by motion hammerblow and exacerbated by irregularities in track-work. This caused excessive stress to the axleboxes and poor riding characteristics on the footplate. Despite this, the class benefited from an excellent maintenance regime."...

Pretty sure - from memory - that the wikipedia article has a poor redaction from the original text, and the instability and track sensitivity was attributed to the 2-6-4Ts

[Mickey]

"No. E782 Sir Brian was used on the former Great Northern main line for performance trials against the SECR K and K1 class tanks following a railway accident at Sevenoaks, Kent in 1927. The tests were supervised by the London and North Eastern Railway's CME, Sir Nigel Gresley, who commented that the class was unstable at high speeds. The instability was caused by motion hammerblow and exacerbated by irregularities in track-work. This caused excessive stress to the axleboxes and poor riding characteristics on the footplate. Despite this, the class benefited from an excellent maintenance regime."

I am not overly knowledgable about the technicalities of the workings of a steam locomotive but with regards to the Sevenoaks high-speed derailment of a 'Rivers class' tank loco in 1927 from memory after reading about this accident very many years ago (45-50 years ago) wasn't there also an issue with regards to 'water surging about in the locos side tanks' at high speed a common feature of tank locos in the early years before side tanks were fitted with 'baffles' inside there side tanks limiting 'water surge' making them more stable at running at higher speeds?.

The GNR 4-4-2 Atlantic wheel arrangement when 'in motion' always had a pleasing look on my eyes.


Mickey