60019 Bittern

Its odd about these classifications though. The GWR, in its idiosyncratic way classified number series of different classes with the same wheel arrangement together, but I've never seen anything that suggests these groupings were actually used in any way. I suppose it perhaps kept the filing of record cards more organised. As I type that it occurs to me that the admin organisation back in the office probably had a much bigger influence on this sort of thing than we might realise.

 

You would think that their workmanship on gresley would speak for itself

 

LSL are more than happy with SNGLT's overhaul capabilities and processes, hence why 60019 is going to them for overhaul. It's not a spur of the moment decision, discussions have been ongoing for years.

 

They also broke the system with the B1 & B2, by bumping the originals (GCR engines) to B18 & B19 in 1943 & 1945 respectively. And in the case of the letter X, it covered two different wheel arrangements. So, not that logical really.

 

B18 and B19 were going to be withdrawn imminently when the decision was taken to recategorise them. Perfectly logical to replace their numbers with new locomotive classes. X was considered an open category for “everything else”. Of which there were not many in that category, so that’s fine as a catch all.

I’ve yet to come across a basic classification system from any other railway. With the first letter you know the wheelbase instantly. The number defines the exact class.

 

The GSR in Ireland used the same system, though (according to Wikipedia) only at Inchicore. Anyone able to tell us how that came about?

 

That's not an issue with the system though. More an issue with when people decided they weren't going to follow it as strictly as perhaps they should have...

 

I think, picking up on something @Jimc said earlier, you probably need to go back a step and ask "why do you even need class designations at all?" In other words, I suspect the system you have probably depends in a large part for whose benefit it was designed.

As an example, the LMS system seems to me to be a Shedmasters' system. You look at a shed, look at all the duties, and come up with a requirement that that shed will need so many Class 8P locos, so many 6-7P, so many Class 5 and so on, in order to cover all the duties. At that point, as locos go from shed to works for repair, and re-emerge from works to be allocated back out to sheds, you just need to check each shed has a suitable complement. It's pretty irrelevant the subtle distinctions between normal, Caprotti and Stephenson black 5s; or Fowler, Stanier and Fairburn 4MTs - you just need to ensure a shed that requires ten 4MT tanks actually has ten 4MT tanks. The LNER would probably have invented class distinctions to cover the various forms of Black 5 based on wheelbase variations, boiler variations and so on, but that is of little relevance to a shed.

(An early version of that was Stroudley's system, which initially classified both the "Richmond" 0-4-2s and "Gladstone" 0-4-2s as class B because they were both designed for the same duties, despite being demonstrably very different. Initially some of the "Grosvenor" 2-2-2s also got classified as class B because they were also passenger locos.)

The LNER system by contrast seems to be more of an accountants system, particularly the way it splits similar locos into sub-classes. Operationally, the difference between a Gresley A1 and a Gresley A3 is largely irrelevant, as demonstrated by the fact that for many years they worked side by side on the same duties. But for an accountant, being able to split them for statistical purposes is probably quite useful if you are interested in answering questions like whether the coal saving from the revised valve gear and boiler justifies the cost of conversion.

FWIW, the SR system of class designation never moved on from the pre-grouping class names, so you have four different systems all overlaid seemingly without much rhyme or reason. The fact that it was never rationalised is I suspect evidence for the contention that it wasn't seen as very important: the fleet was small enough that class sizes were generally small and there wasn't much mixing between the former groups; the number of "new" SR locos was fairly small; and the pre grouping locos were largely just being maintained until electrification swept them away. Interestingly, early in the grouping Maunsell got all the locos classified on a complex system that was based on power, range (coal / water capacity) and I think braking capacity, but that was I believe just to be able to quickly get a grip on the inheritance from four main companies, rather than with the intention that locos were redesignated into new classes as the LNER did.

Tom

 

An interesting question perhaps is why? The wheel arrangement is not precisely the hardest thing to spot across the yard. Clearly other lines didn't see a system of classification incorporating wheel arrangement as all that critical, although the GWR had numbering systems that incorporated wheel arrangement (and due to the GWRs dislike of wholesale renumbering there were three different systems at the same time!). However I do suspect filing. Used as we are to spreadsheets we can re-sort in seconds its easy to forget that there was a time when changing the order of a list might take a junior clerk half a day...

 

In GNR days it wasn't uncommon to see an 0-4-2 as mixed traffic, but a 2-4-0 as express passenger. 0-6-0? Definitely goods!

 

More prosaically, the Railway Companies (Accounts and Returns) Act 1911 required the loco fleet to be set out and analysed between tender and tank locos by quantity and wheel arrangement. The MoT returns also required average weight for the various types. Whether this was of any use to anyone is debatable.

There was in fact a Board of Trade Committee appointed in 1910 to hear evidence on the whole subject of accounts and statistics. I cannot find anything that suggests the issue of recording the wheel arrangement was discussed but Churchward was a witness and there is a lengthy record of his testy exchanges with the committee, the railway executives generally tending to come from a point of view that most statistics, even the more dynamic ones, were too generalised and/or fairly or totally useless in helping a layman understand the business.

The attached is the first page of Churchward's evidence, which was in the form of a pre-prepared memorandum.

 

There were/are lots of loco classification systems, often tied in with the numbering system by making the class type a prefix to the number (As on the GWR - after a fashion). The SNCF took its system from the PLM. If the engine's ID started with 141C for instance, you knew that was a specific type of 2-8-2. Other railways, in France and elsewhere, had classification and numbering systems that started with the number of coupled axles. For example, a New South Wales C38 and a Japanese C53 announce themselves as six-coupled types.

The LMS was unusual in using a power classification as its main loco class system. For a more complex system devised to suit railway operators, consider Czechoslovakia:
(1) 1st digit is number of coupled axles.
(2) 2nd digit indicates max permitted speed (Add 3 then multiply by 10 to give km/hr).
(3) 3rd digit indicates max axle-load (Add 10 to give tonnes).

On the whole, I think the LNER loco class system was as good as any. You can argue whether they should also have renumbered the stock in 1923, and whether or not Thompson adopted the best renumbering system in 1946.

 

Also with GN 2-8-0 O1 to 03, and GC 2-6-4 L1 to L3.

 

The issue with the LNER system was the numbering, not the lettering. Having adopted a logical system, common sense would have been to add the next new class on the end of the series. So, for example, the GCR B1s and B2s should have been left be, and the new Thompson class numbered B18.

Instead of which, in a system where the class name became a descriptive noun, the renumbering of classes introduced confusion.

The impression left - fairly or otherwise - is that the numbering was as much about pride as about ability to refer to all locomotives of a type consistently. That's not just about trainspotterish desire to classify things, but must also have had an impact at sheds and on clerical staff, relabelling records, parts, etc. to suit the new convention.

Edit/PS: BR learned from this in their TOPS classification. Having applied a rough order by power classifications to the classes, new introductions have been added to the number series as next in range. It's not perfect (look at cl. 47 sub-classes, and why the leap from class 56 to class 58, backfilled by class 57 as some of the class 47s were re-engined?) but it avoided a lot of the pitfalls of the LNER system as implemented on the ground).

 

That logic only works if you believe classes B1 and B2 were going to stay in work forever for the LNER.

On a railway intending to remove vast swathes of older locomotives, reusing number classifications is perfectly legitimate and happens regularly in many industries.

Only the enthusiast would complain about a railway company wanting to put its newest locomotives front and centre of its own classification system.

 

I disagree; the logic of reclassification before the "old" class is fully withdrawn, and while records, parts, etc. may all need to be updated is not an economically rational way to proceed. It is, if not enthusiast logic, then certainly the logic of Top Trumps - not of a large, rational, business.

Whatever the operational purpose of a classification, that label takes on meaning in peoples' minds, and in operational systems. Where a number series is extensible, any decision to reuse an in-use category is economically irrational.

You make a very good case for Thompson in your writing, based on economic, labour, and engineering necessity, as against the allegations of personal motives. The reuse of classifications is inconsistent with that logic, and risks reopening a question that I believe is rightly closed.

 

Or perhaps a historian looking at records and trying to work out whether its old A1 class or new A1 class being talked about in this particular casually dated register, and whether any careless clerks had put any of them under the wrong category?
And yes, as 35B suggests, a storesman doing a stock audit and trying to figure whether this dusty piece labelled A1 is actually A1 or A10?

 

The complete failure though is the current Class 70, which by rights should be a low-voltage DC electric locomotive. There are lots of numbers in the 60s that were not used but obviously were pre-allocated to something else.