Jump to content
Users will currently see a stripped down version of the site until an advertising issue is fixed. If you are seeing any suspect adverts please go to the bottom of the page and click on Themes and select IPS Default. ×
  • entries
  • comments
  • views

GWR 9 and 10 Dean Experimental Single Wheelers



These were Dean prototypes. Superficially rather ordinary looking 2-2-2s by the time they went out of service, they are perhaps more interesting and influential than sometimes given credit to. I don't think I really said enough about them in the printed version of the book. I shall try and produce a series of sketches, although material can be rather sparse.

No. 9 started life in 1881 as a rather absurd express 4-2-4 tank engine with 7ft 8in driving wheels and was so prone to derailment it probably never went into service at all: at times the GWR denied it had ever existed. There is a long piece about No 9 in the original form in Les Summers' book "Swindon Steam", where he draws rather different conclusions about its design than those found in RCTS. I find his arguments rather persuasive. 

In 1884 substantial parts of this locomotive were included in a renewed no. 9, which was a more conventional 2-2-2, still with the 7ft 8in driving wheels. Both iterations of no 9 had Stephenson's valve gear outside the frames with valves and cylinders inside the frames.

No. 10 followed in 1886 and was externally similar to No. 9, but had rather different internals, with the slide valves located under the cylinders. This is generally called the Stroudley layout, having been devised by Stroudley of the LBSCR.   This layout was clearly successful since it was repeated on the 3001 (Dean Single) class and all the larger inside cylinder 4-4-0s.

Both 9 and 10 were reconstructed in 1890 with 7ft driving wheels and a conventional layout, generally similar to the Queen Class. The external valve gear on No 9 disappeared, and I think that new cylinders with the valves underneath were fitted.   In 1890, or perhaps 1893 they were named Victoria and Royal Albert respectively. No 9 received a belpaire firebox boiler in 1901, together with a larger cab. They were scrapped in 1905 and 1906.

There are a few photos about of the pair in their later days - see, for example, the June 3 entry in the GWS Blog here: https://didcotrailwaycentre.org.uk/article.php/515/going-loco-june-2022. They're all a bit muddy in the shadows below the footplate though.



Edited by JimC

  • Like 3


Recommended Comments

699717919_222-DiagCNo10.JPG.e9be031991e307a2b643dcbf6883306b.JPGThis first sketch is intended to represent no 10 towards the end of her days, as in 2-2-2 diagram C. Its quite a bit more conjectural than I would like in places, most notably the brakes and what is visible of the motion and reversing gear.

Edited by JimC
  • Like 3
Link to comment



Here's my imagining of No 9 in its original form, and here are my notes I put together whilst working up my ideas. I must acknowledge various sources listed below, and assistance/ideas from the people on the GWR Elist https://gwr-elist-2019.groups.io/g/main/topic/dean_4_2_4t_no_9/95663670 and Model Engineering Clearing House forum. All errors, stupidities, misinterpretations etc are my own. 


No 9 in 4-2-4 form is a hard project. Nothing contemporary seems to have survived. We have a few sentences in Joy's (valve gear Joy) memoirs and some paragraphs from E.L Ahrons, who joined Swindon works within a few years of the locomotive being dismantled. We also have a drawing produced by Earnest Twining about 1939, for which he had sight of two original drawings which have now disappeared.


Its hard to discuss No 9 without considering Les Summers ideas. These ideas were first printed in "Backtrack", volume 18, April 2004, and in his book "Swindon Steam" (https://www.amberley-books.com/swindon-steam.html). The first thing to be said is that Mr Summers has done a good deal of primary research - and I haven't. All I'm doing is sitting on his shoulder.

Sources - Contemporary.


Joy's diaries/memoirs. Joy undoubtedly saw drawings and probably the locomotive itself, albeit under construction ('I saw drawings and all and she looked a beauty' ) . He describes it as "8.0 single and double 4.0 wheel bogies at each end". 

E L Ahrons. 

He worked at Swindon Works from 1885. No 9 had been reconstructed in 1884, so presumably Ahrons never saw the original, but may well have heard about it first hand. He also states that he had seen parts of it - an interesting statement suggesting some of what wasn’t reused in the 2-2-2 hung around for a while before being disposed of.

Sources - Mid 20thC


E W Twining wrote an article in "The Locomotive", published Jan 1940, which discussed some of Dean's experimental locomotives, particularly No 9. At that time there seem to have been two surviving drawings, which Collett made available to him. Malcolm Brown, from the Model Engineering  Clearing House forum, has kindly photographed the article from the Locomotive and posted it on flickr.







Its interesting and useful to see Twining's conclusions. The centreless rear bogie is agreed, but there's an issue with the support. On No 1 the bogie was hung from outside frames, but Twining considers that the tank drawing precludes any outside frames, so there must have been some sort of bracket from the inside frames. He notes that an unnamed member of staff at Swindon, who was interested in the locomotive, agrees with Les Summers that the leading bogie was surely similar to the trailing bogie in style. Twining, however, concludes that two bogies with minimal side control was surely impractical, and feels that the leading wheels must have been in the main frames (as per the broad gauge Rovers) and so no leading bogie at all. 


Holcroft states the trailing bogie was as per No 1, but that the leading bogie had inside bearings. I wonder where this comes from?  Can we regard Holcroft as a contemporary source? He was about three when the locomotive was dismantled. He wrote his book in the 1950s, and had undoubtedly seen Twining’s article since he uses illustrations from it. However he is definite about a leading bogie and doesn’t follow Twining’s supposition about fixed wheels. 

Sources - Modern


He suggests that the motion components from No 9 were used unchanged when the locomotive was rebuilt. This is a compelling theory, but there’s a problem. If we make the assumption that the tanks are accurate on Twinings drawing, which apparently was made with the original drawing available, then the eccentric rods shown on the works drawing of no 9 in rebuilt form don’t fit. There’s what appears to be a shaped recess for the link of the Stephenson motion as it moves up and down.

Summers also considers, from researching the Swindon drawings list, that the bogies were similar to the 2-4-0T no 1. This aspect of his research seems very sound. The Swindon drawing register lists a drawing of certain bogie components as covering both locomotives. He extends this to make the proposition that the recording of the leading bogie as 7’+ is incorrect, and both bogies were the same short dimension. The short leading bogie certainly gives room for the eccentrics and radius link to clear the bogie, which is not the case if the rebuilt length eccentric rods are drawn.


Mike Oxon

 in a model reported on Rmweb, https://www.rmweb.co.uk/blogs/entry/14782-william-deans-express-tank/ bases his model on the 2-2-2 motion drawing, like Summers. In order for the motion to fit he swaps the bogies round so that the long wheelbase is under the bunker and the short wheelbase under the cylinders. 


The short front bogie concept has two potential objections - firstly the statement that the long bogie was at the front, and secondly the tank cutout as drawn by Twinings. If a copy of the original tank drawing could be found that would settle the question, but it appears it has not been seen for 60 years.



Because Twining claims that the tank drawing precludes any form of outside frame and a central pivot for the rear bogie, I am inclined to believe that this was a detailed "arrangement of tanks" type drawing and the detail Twining shows of the internal cavities in the tanks for the motion are correct. If I accept this, and I freely admit that its a rather shaky foundation, then that means that the eccentric rods on the 2-2-2 were new, and longer than those on the 4-2-4T version of No 9. That in turn means that I can accept the published long wheelbase for the leading wheels. It would be interesting to see if the drawing register for around 1882-4 contains any mention of motion components for lot 54. I freely admit, though, it still seems unlikely that Swindon would have fabricated new extension rods, so I certainly wouldn't rule out Les' and Mike's belief that the rods were as per the 2-2-2 and the leading bogie must have been shorter.


Every source seems to support the concept of a similar bogie on Lot 54 as Lot 46 so I am inclined to consider that is as well proved as it can be. I think the recorded wheelbase is almost certainly correct: if I take the relative positions of tanks, boiler and bunker to be correct, which seems reasonably soundly based, then I can't make Mike's ingenious interpretation of reversed wheel spacing work on a drawing. I think either the trailing wheel ends up too close to the rear buffer beam or the first of the rear bogie ends up in the ashpan. But feel free to prove me wrong.


That leaves me with Twining's interpretation of a non bogie leading bogie. That implies Ahrons got his description wrong. I really struggle with that, Ahrons was just too near the time, and its not as if a fixed pair of carrying wheels was an unfamiliar concept to him. Holcroft didn’t repeat the fixed wheels theory when he wrote his book too. I think Les must be correct, and there was a bogie of the same style and many of the same parts as the rear one, albeit with the longer wheelbase. The obvious objection is Twining's - that such an arrangement would scarcely be trackworthy. Well, of course it wasn't. I wonder if the arrangement with fixed leading wheels is perhaps too close to an 0-6-4T to be quite so spectacularly keen to fall off the rails. Maybe Dean simply over estimated the centring effect of the swinging links.


I have chosen to ignore Holcroft's statement about inside bearings for the leading bogie, and have made the assumption that he is following Twining's drawing rather than stating something he knew. Arguably this is the most unsafe conclusion I have drawn, and I am not altogether comfortable with it. However carrying wheels with inside bearings were rare on the GWR, and no other bogie had inside frames until the Churchward standards. I also have trouble working out where suspension components would actually fit bearing in mind the presence of the cylinders.


So altogether I come up with a different conclusion to others who've trod the same path! That's at least consistent with the shadowy nature of the whole locomotive. Feel free to rip my conclusions to bits folks, that's what they are there for! I really ought to go up to the NRM and look at that drawings register, but its a big trek for me, and I'm not a keen traveller these days. Also, I presume it means much eye straining deciphering of Victorian copper plate handwriting.


Things that worry me.
The frames

With inside frames only and the bogies to clear the frames get awfully shallow. Perhaps there was no real footplate and the tanks were outside the frames and overhanging them.

The Tanks

The tanks get horribly complicated around the driving wheels and axles. Indeed to all intents and purposes they are in two halves each side. How did water flow between? I’m making the assumption Twining had a detailed drawing of the tanks. Why doesn’t he show the water filling location? His text implies a back tank under the coal, which is likely enough, How did the water flow to the other tanks?

It seems fearsomely complicated to build, although most of that seems like platework. I wonder how the costs in the account book Mr Morgan found for Les Summers compare with others?



Twining has drawn clasp shoes on the driving wheels. I can’t find any evidence of such a fit out on other GWR locomotives. The conventional arrangement on singles was brakes on driving and trailing wheels. An express tank - no tender brakes remember - and just a single wheel braked sounds decidedly sketchy. Is this an argument for Twining’s fixed front wheels? Then it could have had 6 braked wheels. Or did they arrange bogie brakes. It was possible I suppose, done on carriages. I've just drawn a single brake shoe, but surely there must have been more.

There’s an awful lot of gubbins in a very small space between driving wheels and the trailing wheels of the bogie. We have the sanding for sure, then presumably suspension arrangements for the bogie and the brakes, Can it all fit? Quite a complicated area.

Argue away!

Edited by JimC
  • Like 2
  • Informative/Useful 1
Link to comment

I found the article from 'The Locomotive', 1940 very interesting, as it indicates that Twining had access to more information about the prototype than I gave him credit for, when I made my conceptual model. 


I had not recognised those cut-outs in the tank that do point to shorter eccentric rods than on the first 2-2-2 re-build.  It does still seem rather odd, however, that the re-build should have reconstructed the outside motion with new rods when there was no apparent need to do so, especially as there were no access problems after the tanks had been removed.


Like you, Jim, I am greatly indebted to the research that Les Summers has done.  I strayed from his deductions only in allowing that Ahrons was correct about the two bogie wheelbases but might have placed them at the wrong ends!  It is always extremely difficult to decide which 'facts' to accept and which to ignore, when so much of what has been written was based only on personal recollections.  I think Les uncovered more information than anyone else seems to have done in recent times and made several well-considered deductions.


It seems very odd to me that Twining asserts that there were no outside frames and also that the drawing he had access to showed nothing about tank filling and balancing arrangements.  The drawing of the 4-4-0 'No.1' shows very shallow frames at the points where the front bogie 'E' brackets were attached, so why could not 'No.9' have had a similar type of frame extending the full length to carry brackets for both centre-less bogies?.


On my model, I placed the fillers at the front, on the grounds that they would be easy to reach from the front footplate, in view of there being no foot-plating outside the tanks, which I had assumed spanned the full width of the shallow frames.


I am very cautious about any later writings, such as Holcroft, because I feel they were probably influenced by Twining's interpretation, as well as other 'legends' that had grown up over the years.  Twining points out that there is no evidence for centre bearings on either the front or rear bogies (if that's what they were)


My model has centre bearings for the bogies as, otherwise it would have been completely unable to negotiate any curves. This failing appears to have been the principal failing of the prototype!



  • Informative/Useful 1
Link to comment
9 hours ago, JimC said:

I can't make Mike's ingenious interpretation of reversed wheel spacing work on a drawing

As you will know from my blog, I rarely make detailed measurements, so you may be correct.  I simply took the Twining drawing and used Photoshop to swap the bogies.  It seemed to me, by eye, that there was sufficient clearance (just) but it is possible that accurate measurements could prove me wrong.




  • Like 1
Link to comment

That photo is lifted from Les Summers' book, its his photoshop recreation of No 9.  I recognise parts of it as being lifted from a photo of the 1813 class - know the one?

I've heard it described as good looking, but not to my eyes I'm afraid.


Edited by JimC
  • Like 1
Link to comment

To me, it has all the charm of a Bo-Bo diesel. 


As I wrote in my Railway Modeller (Special 2019) article, 'The Swindon Ghost':

"The appearance of subsequent express railway locomotives could have changed substantially, if this design had proved successful. I feel that the box-like structure. formed by the long side tanks. gives something of the appearance of a modern power car, carried on bogies at each end"


What @Anniecalls a "dismal" 🙂

  • Like 2
  • Funny 1
Link to comment


No 9 again, this time in her 1884 incarnation. This retained the external  Stephenson's gear, in my opinion probably slightly modified, but was given a new domed boiler. In 1890 she was to be reconstructed again, to much the form of No 10 as shown with Diagram C above. In 1902 she was reboilered with a Belpaire boiler - domeless with a raised firebox, which was a Class N type. She was withdrawn in 1905, whilst 10 followed into oblivion in 1906. 
This drawing is pretty well founded. I have a copy of a part Swindon works drawing which was published in the Engineer and shows the locomotive from the boiler centreline down and driving wheel forward, and I was given an original Swindon blue print of a weights diagram of the 1884 configuration by Chris Hext, ex Swindon drawing office, which is now with the GWS at Didcot.

Edited by JimC
  • Like 4
Link to comment

I've had sight of the GWR drawing register which gives the titles of drawings. Its evident that there were definitely two bogies on No 9, and that the trailing bogie had brake gear fitted.


Having worked through the drawings list its slightly frustrating. Nothing new about that when dealing with No 9! Most drawing titles don't distinguish between leading and trailing bogies, and those that do (frame plan, life guards (=guard irons?)  include both on the same drawing. There's also no evidence from the drawing list that the axle boxes were different on the leading and trailing wheels, but this is not definitive. As is often repeated, absence of evidence is not evidence of absence. They might have had two entirely different styles of axlebox on the same drawing. There is certainly no evidence that the two bogies were radically different, but there isn't any firm evidence that they were fundamentally similar either. There were parts in common with the 4-4-0 No1 (lot 46), but not many and not major. The two such  drawings are entitled "Spring for bogie" and "Bogie brackets & slides" The two designs were being worked on at the same time (1879-1880 drawing dates for Lot 56, 1879-1881 for lot 54 which I suppose is circumstantial evidence that they would have been similar, but they don't appear to have shared major components. I don't think I've really discovered anything that Les Summers hadn't spotted before. I'm perhaps less confident that the No 1 and No 9 bogies were similar than I was before I started the exercise, but I think the balance of probabilities has to be that the two bogies on No 9 were much the same with outside bearings. I wonder where Holcroft got inside bearings from. As far as I can see Twining drew inside bearings because he knew there were no fixed outside frames, but we now have proof positive he was mistaken about the possibility of fixed leading wheels.  So did Holcroft get inside bearings from Twining, or some piece of 60 year old office gossip? 

Edited by JimC
  • Informative/Useful 2
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...