I have a few problems with my Hornby 31 derailing when running through the crossover at the end of Platform 2 if set to cross over. As this is part of the run round loop and as the 31 is currently diagrammed for any loco hauled trains (parcels, engineers etc) this is a problem

The problem is caused by the fact that these points don't always close tight when thrown - arising from the fact that I used the wire supplied by Tortoise, instead of replacing it with something thicker and stiffer. I've replaced the throw wires on the other points , but it's now going to be the devils own job to do it on the platform end crossover.

However as my Airfix 31 seems to take the relevant crossover in its stride , there's an obvious fix. I always intended to detail the Airfix model at some point anyway and if I want operate a Loco Hauled Substitute set (2 coach MK1/Mk2Z rake) to cover a DMU shortage - as happened not infrequently in the 1980s - I need a second Type 2 to work it Minories-style (Platform 2 is too short to take 31 + 2 x 64' coaches - a pair of 50' vans is the limit - and this is the only platform with access to the loop. The crossover in question was originally added to the plan to give access to the fuelling point , and the fact it gave me a loop was a bonus) . I've got the blue/grey coaches - all I need is a second loco

The Airfix 31 was bought new in the late 70s for my first modern image layout, and has been stored in its box ever since that unsuccessful project was finally abandoned. It has been given a decoder , and runs well considering what it is, but is otherwise untouched. I also acquired a finished 31 body when Dapol were selling off the remaining stocks of discontinued items some years , and the intention is to detail this spare body and substitute it. Kadees must also be fitted to the chassis.

I have a further rather battered 31 body in stock, plus a rather tatty and roughly detailed specimen acquired for £15 for it's chassis (it's a runner though may need some cleaning up) , a spare Athearn PA1 chassis, and the unhappy remains of a first batch blue Hornby 31 (I seperated the body before mazak rot set in) , as well as 31 174 which is - I hope - fine. If I work my way through that lot over the years I should have quite a fleet of Brush 2s

In my teens I saw quite a lot of Immingham's 31/4 fleet on Transpennine South and Cleethorpes-Newark trains. Hornby have made a 31/1, so obviously if I was detailing an Airfix loco for myself I wanted a mid 80s IM 31/4.

Unfortunately it's not quite that easy.

There's little or no obvious external difference between a 31/1 and a 31/4. The detailed issues relate to date rather than type. Airfix produced a 31 with headcode box, bodyside steps and recessed tank filler on the roof, train heating boiler port,, nose doors and buffer beam cowls. As far as I'm aware this represents a late 60s /early 70s re-engined 31/1 from the latter 2/3rds of the production batch. A good choice for a manufacturer in 1977 but rather more problematic now

By the mid 70s all 31s had lost the bodyside steps and roof recess which had been plated over , so Airfix's blue 31 401 is wrong here . During the 1980s buffer beam cowls were removed . The first batch of ETH conversions (31 401-19) kept them when converted in the early 70s . I'm not sure about 31 420-4 , converted in the mid 70s . The second batch of conversions in the mid 80s , 31 425-469 lost them when rebuilt to ETH, and they also lost bodyside bands. I don't feel up to that level of reworking , so my target loco has to be a 31/4 from the early conversions up to 31 424.

Unfortunately IM's 31/4 fleet was basically the second batch conversions. I thought I'd found a perfect prototype and reference shot here:

http://www.class31.co.uk/picture/31408-bk-090383_t.jpg

(I originally found this on the 53A Models photo site)The train is virtually certain to be a Cleethorpes - Newark Northgate service, as it comprises four Mk1s- the Transpennine South sets were basically Mk2a s , later strengthened to 5 when the Newark sets were broken up and that service reverted to 114s

It is at this point that the problem of fan cowls rears it's ugly head - literally . Note the roof line on the loco - isn't there something projecting ?? Aren't we seeing what is horribly obvious here - a raised cowl around the roof fan grill, on the same loco at Rugby some years later:

http://www.class31.co.uk/picture/cs31408_rugby.jpg

But in 1976 she was smooth and her roofline unmarred:

https://www.flickr.com/photos/16179216@N07/5187603237/

Some internet browsing suggests that these cowls appeared in 1979-82. Many locos never got them - I have a shot of a gleaming new 31 435 at Grimsby in 1985 (I think) with no cowl, and she's plainly uncowled here (a Hull-Liverpool I reckon) http://www.class31.co.uk/picture/31435-sp-0690.jpg

I've no idea how you model such a cowl - so 31 408 wasn't a suitable target loco. And as 31 435 had uncowled buffer beams she wasn't either (Whether the buffer beam cowling has been replaced now she's D5600 in preservation I don't know). She has also had a revised smaller cab window on one side which seems to be a very unusual modification - I've not seen another photo showing this asymmetrical cab window arrangement.

As far as I can see from my surviving 1980s abcs, only five 31/4s from the first batch were allocated to IM during the period (31 403/07/08/09/20) and it appears from the class 31 photo site that all of them had roof fan grill cowling during the 1980s

So a slightly different approach is needed.

The nearest I've found to an ideal target loco is 31 415, seen at Skegness in 1982 in a photo in Diesel Retrospective Class 31. She doesn't have a roof cowl, I think the buffer beam cowling is still in place as something is going well below the buffers, and she's most emphatically in Lincolnshire . She is plain blue with no stripe but bodyside bands are still in place

At that stage she was allocated to March (the train is therefore probably the SO Cambridge-Skegness and return), and by the late 80s she was allocated to Bescot - very suitable for Blacklade. At what stage she lost buffer cowling and bodyside bands I don't know - ignorance is bliss here. (For the record a shot in the same book shows 31 414 at Wellingboro in March 1987 without either). The Hornby body from 31 270 would probably be a much better starting point for a 31 without bodyside bands as the band is done by tampo printing

So - 31 415 it will be...