The LNER Encyclopedia

I've honestly never gotten a clear answer about this, so I thought I'd ask the controversial question, and try to ask it in as neutral a way that I can: why exactly were the Thompson Pacifics built, and do they have their own merits they can stand on?

I know that they are generally seen as an innovative but flawed design, given their rough-riding, tendency to wheelslip, and that they tended to waste coal on lighter duties due to the higher heat of their exhaust; but I can't recall hearing anyone explain what strengths/good qualities these engines had, aside possibly from having somewhat better ease of maintenance compared to the Gresley designs.

Why were they designed? Because Thompson believed a superior locomotive would result from the mechanism principles he espoused.
Why were they built? Because the locomotive committee approved them.

There was much good content in the design: but all of it from established and proven practise at Doncaster. The novel elements in the design introduced the recognised failings. This is one of those rare occasions where the proof is very evident. On Thompson's retirement Peppercorn had the mechanical design layout of the A1 and A2 reworked, and very reliable locomotives resulted.

A good source for a balanced view, by someone who had a railway-operating background , is one of Richard Hardy’s books. He had a couple of auto-biographicals published and Steam in The Blood is the first of them (pub by Ian Allan)
The second one has a similar title.
Always a good read about locos and men generally

I think the amount of work the B1's did for the LNER perfectly demonstrates why they were built, did everything that was thrown at them, loved to see them on the Great Central, apart from the heavy coal trains that 9f's took over from the O1/4's they were equivalent of the LMS Black 5.

I would humbly submit that there are no "novel elements" in the Thompson Pacific designs. Almost everything on them is LNER outline or in keeping with conventional practice here, and abroad, including the divided drive setup and short con rods (inherited from the P2s). The double blastpipe and kylchap was standard with the A4s. Almost everything on the Thompson machines was in some way a standard component shared with other types.

The A2/2s were built to resolve the reliability and mechanical issues of the P2s they were built from.
The A2/1s were built to compare and contrast with the A2/2 Thane of Fife towards developing a new mixed traffic pacific design.
The A2/3s were built as the production variant of the A2/2 design.
The A1/1 was built as a development of the Gresley A4 design, taking the place of A10 Great Northern (and reusing some parts) towards the production of a future new A1 design.

1 A1/1, 4 A2/1s, 6 A2/2s, 15 A2/3s. 26 Pacifics.

They get a disproportionate amount of criticism compared to what the primary evidence we have shows. The A2/3 in particular was an excellent machine for availability and mileages.

The Peppercorn A2 was a development of this locomotive with a slightly revised arrangement with the length of the connecting rods and more compact wheelbase. Almost all other details remained the same, with major differences being mostly aesthetic in nature. The big changes came with a modification to a multi-valve regulator on some of the Peppercorn A2s.

My research has shown that locomotives with this type of boiler actually had lower mileages and availability compared to the Peppercorn and Thompson Pacifics with the "standard" setup and boiler types. Why - likely to be a similar problem suffered by the Thompson A2/2s in the early 50s (shortage of appropriate boilers).

The Peppercorn A1 was a 6ft 8in wheeled version of the Peppercorn A2.

Of these Pacifics, the Peppercorn A1 had the highest mileages and availability throughout their lives.

The primary evidence we have available, in the form of statistical data from the LNER's own archives (at Kew and at York) show that the Thompson Pacifics were by no means poor. Much of the stories told on them are apocryphal and should be treated as such.

S.A.C. Martin wrote: ↑Thu May 13, 2021 3:03 pm I would humbly submit that there are no "novel elements" in the Thompson Pacific designs...

The Thompson alteration in frame and engine layout from what had previously been standard on Doncaster pacifics was novel.

That a professional team of railway engineers under Peppercorn immediately reverted to a layout reflecting more of past practise, and thereby produced successful locomotives is proof that the novelty was redundant.

Hatfield Shed wrote: ↑Thu May 13, 2021 4:04 pm The Thompson alteration in frame and engine layout from what had previously been standard on Doncaster pacifics was novel.

The cylinder and frame arrangement was not novel on the LMS in the Princess class, nor on the GWR with their four cylinder engines, nor on the continent with a range of locomotives including Chapelon designed locomotives. It might have been novel-ish if we are talking from a purely Doncaster-centric point of view, but not LNER either as divided drive had been used before.

That a professional team of railway engineers under Peppercorn immediately reverted to a layout reflecting more of past practise, and thereby produced successful locomotives is proof that the novelty was redundant.

If you read the RCTS books you will note that several variations of the A2's layout were actually drawn under Thompson and then worked on under Peppercorn to a final drawing. There was no "immediately": variations of that layout had appeared several times over within the drawing office.

The desire to return to the compact layout was a compromise: all other aspects of the Thompson design were retained (three sets of valve gear and divided drive). The layout might have been Gresley, but the engineering was Thompson. Is that redundant, or tweaking?

S.A.C. Martin wrote: ↑Thu May 13, 2021 5:15 pm

Hatfield Shed wrote: ↑Thu May 13, 2021 4:04 pm The Thompson alteration in frame and engine layout from what had previously been standard on Doncaster pacifics was novel.

The cylinder and frame arrangement was not novel on the LMS in the Princess class, nor on the GWR with their four cylinder engines, nor on the continent with a range of locomotives including Chapelon designed locomotives. It might have been novel-ish if we are talking from a purely Doncaster-centric point of view, but not LNER either as divided drive had been used before.

And the Princess was such a success that its layout was abandoned, for the LMS' second attempt which was clean sheet. A read about the troubles of the Princess in service is instructive. The Churchward 4 cylinder layout was near the limit on the King, for a higher power output locomotive that could be built with adequate frame strength and still comply with the civil engineer's restrictions on axle load.

It's the additional length that is the killer factor, the frame structure weight increases in proportion to length. If you read Chapelon he is most eloquent on this matter, and the constraints inflicted on the designer by the typical range of maximum permitted axle load on European railways. This was not a good choice for a pacific that had to operate on the LNER system.

S.A.C. Martin wrote: ↑Thu May 13, 2021 5:15 pm

Hatfield Shed wrote: ↑Thu May 13, 2021 4:04 pm That a professional team of railway engineers under Peppercorn immediately reverted to a layout reflecting more of past practise, and thereby produced successful locomotives is proof that the novelty was redundant.

If you read the RCTS books you will note that several variations of the A2's layout were actually drawn under Thompson and then worked on under Peppercorn to a final drawing. There was no "immediately": variations of that layout had appeared several times over within the drawing office.

Please, I spent over 20 years of my career in engineering: materials research, design development, manufacturing engineering, manufacturing management; and am fully aware of the iterative nature of product design development through to final production. It's what an engineering outfit does, and I have been through it over many cycles of product development from inception through to production.

That process is a given: what it turns on when it comes to assessment, is the product's conformity to the requirements. Were the Gresley pacifics successful? Were the Peppercorn pacifics successful? You know the answer, and it clearly shows that the Thompson pacific's frame and engine layout was an unnecessary novelty.

Hatfield Shed wrote: ↑Fri May 14, 2021 10:20 am And the Princess was such a success that its layout was abandoned, for the LMS' second attempt which was clean sheet. A read about the troubles of the Princess in service is instructive. The Churchward 4 cylinder layout was near the limit on the King, for a higher power output locomotive that could be built with adequate frame strength and still comply with the civil engineer's restrictions on axle load.

It's the additional length that is the killer factor, the frame structure weight increases in proportion to length. If you read Chapelon he is most eloquent on this matter, and the constraints inflicted on the designer by the typical range of maximum permitted axle load on European railways. This was not a good choice for a pacific that had to operate on the LNER system.

The Princess Royal to Coronation development is an interesting comparison, given that the Princess layout was not entirely abandoned. A bizarre claim on your part. The Coronation is effectively an enlarged princess royal.

Having studied at length these two locomotives for some time as part of my research (they are contemporary to Gresley's works) the idea that the Coronations were developed from a blank sheet of paper is ridiculous; it isn't true.

Please, I spent over 20 years of my career in engineering: materials research, design development, manufacturing engineering, manufacturing management; and am fully aware of the iterative nature of product design development through to final production. It's what an engineering outfit does, and I have been through it over many cycles of product development from inception through to production.

Then you will be well aware of the limitations that Thompson and his design team had in WW2 in addition to how much development was done under Thompson too that Peppercorn took forward.

That process is a given: what it turns on when it comes to assessment, is the product's conformity to the requirements. Were the Gresley pacifics successful?

Yes; long lived despite suffering pretty badly during WW2 from poor maintenance standards.

Were the Peppercorn pacifics successful?

The Peppercorn A1s were successful. That is in no doubt.

There are question marks over Peppercorn's A2s, as I stated yesterday. Of the two designs, my considered opinion is that the Thompson A2/3 was the better A2; by way of mileages and availability. Peppercorn's A2s had their own issues (single chimney versus multi-valve regulator boilers and other issues).

You know the answer, and it clearly shows that the Thompson pacific's frame and engine layout was an unnecessary novelty.

It's only unnecessary if the types in question didn't work or didn't work well.

The primary evidence we have in the form of the engine record cards and the Use of Engine Power document from the National Archives shows they worked up to the 1960s, with decent mileages and availability too.

The frame issues are massively overplayed - and I will be reporting on why I say that in my forthcoming book.

Interesting to note that the Peppercorn A2's had higher lifetime mileages than the Thompson ones,
and the Thompson ones were according the Green books, the first Pacifics to be scrapped.

Mind you any loco that got to undertake mileages over less than 20 years of service, of about 3/4 of a million miles
must have had some good points, and I know of a couple of people who regularly travelled behind them in the early
50's and have good memories of both types.

Personally I think Thompson should be more remembered for the value of the B1, which to many were better
than a Stanier class 5, or even more a BR 4MT, rather than his attempts under wartime conditions to leave
his stamp on top link locos which were obviously not his forte.

Paul

S.A.C. Martin wrote: ↑Fri May 14, 2021 11:18 am The Princess Royal to Coronation development is an interesting comparison, given that the Princess layout was not entirely abandoned. A bizarre claim on your part. The Coronation is effectively an enlarged princess royal.

Having studied at length these two locomotives for some time as part of my research (they are contemporary to Gresley's works) the idea that the Coronations were developed from a blank sheet of paper is ridiculous; it isn't true.

'Clean sheet' doesn't mean a blank piece of paper. What it means is examining the product requirement with fresh thinking based on all the available information, and discarding anything that stands in the way of meeting the requirement, and replacement with better technique .

In this case the resulting locomotive ended up with a different frame, different engine and valve gear layout, different boiler design, compared to the previous pacific, such that there was no significant compatibility between these classes (compare for example the ability to use the A4 boiler on A3 frames).

Again by happy chance we have proof: when the turbomotive was rebuilt into a reciprocating machine using technique from the Coronation, an entirely new class designation resulted beause the hybrid was neither a Princess nor a Coronation.

The Princess Coronations (or 'Duchesses' as I much prefer to call them!) were originally intended as 'improved Princess Royals' and the numbers 6213-6217 were initially allocated to the first batch. 6201's famous. record-breaking runs to Glasgow and back on 16th/17th November (both being sub-6 hour timings with a light load) showed was the existing type was capable of, with the 'Coronation Scot' service very much in mind as the LMS's riposte to the LNER streamlined trains. Hence an 'improved Princess' seemed the obvious way to go.

However, as the design work progressed, with Tom Coleman at the peak of his powers, it became apparent that an entirely new class was emerging. There was pressure to adopt a bigger wheel diameter as part of a possible record breaking attempt (something Stanier himself was largely indifferent to), the boiler was a new design (take a bow, Eric Langridge) and the clumsy four sets of valve gear were dispensed with for the rather more elegant Horwich-style rocker lever motion for the inside valves. Accordingly, a new numbering sequence 6220-6224 was adopted, along with a new class designation, and the rest, as they say, is history. A most magnificent locomotive design, for which the only significant performance alteration was the fitting of the double chimneys following the 6234 trials in 1939. I shall champion their greatness to my dying day.

All of which has got nothing to do with Thompson pacifics! I have to say that I'm largely indifferent to the class and their reputation, just not really on my radar as being of any great interest. However, I was with my Dad on Tuesday, who clearly remembers the A2/3s first coming out as a 9 year old trainspotter, looking rather striking in their lined green LNER livery - he used to stay with his cousin in Little Bytham in the summer and they'd both be keeping a sharp lookout for the next new one to appear.

On Tuesday, he told me that, when he was doing his 'square bashing' in East Lincolnshire, (1956-58), when returning back to his base after a weekend's leave, he would regularly travel on a late night train from King's Cross as far as Peterborough and that was almost invariably a A2/3. And - so he says - the runs were frequently 'lively', the crews enjoying themselves on the well-known racing stretches beyond Hitchin, speeds in the 80s not being uncommon. Not very scientific, I know, but he certainly reckoned that the A2/3s at least were a decent enough engine - but not so the A2/2s, far less successful, in his view (he never particularly saw or experienced the A2/1s so reserves judgement on those four 'orphans of the storm'). Sounds to me then like a classic case of the A2/3s being an improvement on the original rebuilds, learning the lessons, etc.

Given that the Peppercorn machines have also been mentioned, I do think that the Peppercorn A2s were highly capable locos and - significantly - were THE answer to the heavy trains on the Edinburgh-Aberdeen road in the BR steam era, the Aberdonian being a regular roster for them over that stretch. Which is of course where the story all starts in terms of the P2 design of 1934. And a route that the rebuilds did not prove to be a success on.

I do hope, therefore, that Simon is not only using availability / mileage data as the sole criterion for measuring 'success' or otherwise of the different loco types in his new book. At the end of the day, any measure of 'success' has to consider the trains the locos worked and the ease - or otherwise - with which they worked them. Unfortunately, a lot of that might appear to be subjective (although the facts of what trains they worked are facts) but it is undoubtedly the case that there are a lot more variables at work with a steam powered machine, where its performance is very much at the mercy of the skill of the crew and their willingness to put the effort in, compared to a diesel / electric driver who just has to waggle a handle. A steam locomotive type that responds willingly to its crew's attentions is ultimately going to be looked upon more favourably than one that doesn't. That in turn DID influence what trains were worked as, ultimately, the job of the operating department is to deliver the timetable plan with the minimum of fuss. A wise motive power superintendent knew his men and knew his machines and married them up accordingly so much as it was in his powers to do so - with a fair bit of skulduggery thrown in for good measure. By such nefarious means were many locomotive reputations made and broken.

Please excuse me if this has been discussed above.
To me the real mystery is why Arthur Peppercorn did not, apparently, value the double kylchap of the A2/3 design, when the rest of the steam circuit on the A2s was pretty much the same.

This is discussed in 'Thompson and Peppercorn - Locomotive Engineers', Col.H C B Rogers.
Mr Harrison did ensure that the presentation model of No.525 for Mr Peppercorn's retirement did have a double chimney.

According to the relevant "green book" the drawing office at Doncaster was unable to fit both the kylachap and the self cleaning
smokebox gear in the same space, during the design stage.

Since the loco was designed at a time when they knew that ongoing staff problems, the less cleaning that was needed the better.

Paul

john coffin wrote: ↑Thu May 20, 2021 11:45 am According to the relevant "green book" the drawing office at Doncaster was unable to fit both the kylachap and the self cleaning
smokebox gear in the same space, during the design stage.

Since the loco was designed at a time when they knew that ongoing staff problems, the less cleaning that was needed the better.

Paul

That makes absolutely no sense, given Bronzino was later fitted with the kylchap, and all of the multi valve ones were too. And the fact the smokebox isn't that much shorter than the Thompson equivalent. They managed it in Humorist which has the smallest smokebox of all of the kylchap fitted types - so what on earth was it really about?