This, of course, rules out ready to run as a means to acquire motive power, but does leave open the option of conversion, which is described in the relevent chapter.

Because of the finer wheel profiles and tolerances in the track, fitting some sort of compensation or springing is recommended for P4. The idea is to ensure that all wheels remain in contact with the rails at all times; this not only reduces the risk of derailment but, by improving the electrical pick-up, should enhance the running quality of the locomotive. The benefits of compensation apply to all scales and gauges and, although it naturally entails a bit more work, I maintain that building a reliable chassis is far easier using some form of compensation as opposed to a rigid one, especially as the number of axles increases.

Compensation can be of three main types: Beam compensation, continuous springy beams, and individually sprung axles. I hope that the principles will become clear as this story unfolds.

As a result of my research, I decided that this model will be 5236, fitted with large boiler support and short safety valve bonnet. As they were, without exception, painted black in BR days, the only other major issue is the style of crest on the tank side - clearly shown in my photograph to be the later BR one. The caption in the book says the photograph was taken in 1955, whereas I thought the later crest appeared in 1956 - an interesting anomaly! Still, the year is not material for my purpose, but importantly the photograph shows the locomotive to be in "ex-works" condition, though it is unlikely 5236 would have remained so for very long!

Because the GW brass number plate survived Nationalisation, finding an etched plate with the required number used to be a further constraint. Happily, for many years now, a d-i-y number plate set has been available from Martin Finney, so - for unnamed locomotives at least - this is no longer a problem as will be demonstrated later in this saga.

The photograph shows a fairly early stage. All the parts required for the basic structure had been detached from the fret, had their edges smoothed with a file, and surfaces cleaned with a fibre glass brush to remove any residue from the etching process. Cleanliness is vital for successful soldering

Some further etched components can been seen awaiting their turn; the cab roof, which has been rolled to shape, is also visible.

Once satisfied that all was well, construction proper commenced, starting with the footplate. I always tack solder the components initially, then check for fit and squareness before running a seam of solder along the joint.

The bolt visible in the smokebox saddle has been soldered in place to form the fixing for the chassis; there is another one at the rear.

It is a very satisfying feeling seeing the model take shape. The model was thoroughly cleaned, rinsed and dried at the end of the session.

The "skirt" under the running plate is a very neat idea to ensure that the running plate is square during construction. If this gets distorted then the tanks and subsequently the boiler will not fit true and the whole locomotive will be misshapen. Not only would this look wrong, the resulting imbalance could well have an adverse impact on the running characterstics. The "skirt" will be removed once the boiler and smokebox have been fitted, leaving just the valence; the etched join can be seen in the photograph. The frames do not sit flat in this view because the bolt, for fixing to the chassis, protrudes slightly below the bottom of the "skirt" at the rear of the body.

She still looks rather odd with the "skirt" around the valence, and there is much detail to be added, but the model is really beginning to take shape.

There are still many details to be added, including the detailed overlay on the buffer beam, lamp irons and hand rails. The holes are to locate steps, etc. at a later stage.

The curve at the front of the tanks was formed around the shank of a 2mm diameter drill; getting the bend in exactly the right place required a bit of care and the fit was checked before tack soldering the overlay in position. A further check for fit was then made, before soldering all the way round.

I have also rolled the cab roof to the correct profile, and placed it roughly in position; it will have details added once it is fixed. The problem is painting the interior - best done before the roof is fitted; I will look at making it removable.

I must now move on to the frames as I need to ensure that the boiler support lines up with the motion bracket before proceeding further with the body.

As can be seen, the coupling rods are in three separate sections, with each section being a laminate of two etched nickel silver parts; these have been sweated together but have yet to be cleaned up to remove the cusp left by the etching process. When dressed with a file they will appear to be solid castings. I formed the connecting rods at the same time.

I have also formed the cylinder block which will require soldering around the cylinder covers.

These 2-8-0T locomotives have an horrendous wheelbase which may be evident from this photograph, and will certainly become more apparent as construction proceeds.

With the boiler in place, the body is now sufficiently rigid to enable removal of the skirts; with the detail overlays added to the valence and buffer beams, the rivets along the base of the tanks and bunker, and with the front of the smoke box fitted, 5236 is really taking shape.

Now the skirts have gone, it is possible to position the body on the frames with the cylinders in place, which will enable clearances to be checked for the wider placement of P4 wheels relative to the crosshead and slide bars.

The body is substantially complete, apart from the small and delicate details; these will only be added once the chassis is built, to avoid potential damage in handling during that phase of the project. However, completion of the basic shell means that I can ensure that the motor and gearbox will fit as construction of the frames proceeds.

Sadly, they are no longer available but, happily, there are some very good alternative sources, one of which is the High Level range.

The gearboxes are available in a variety of shapes, sizes and gear ratios - the example illustrated will build a 80:1 "LoadHauler+". It consists of a fold up frame and final drive carriage, and the worm and gear wheels to give the required reduction. The long shaft has to be cut into lengths to provide the shafts on which the gears rotate. In the photograph, the axle bushes have been soldered in place in the final drive carriage, and the relevant holes opened out to give a tight fit for the shafts.

This was my first attempt at building a HighLevel gearbox, and although it was a bit fiddly, it was not difficult and the end result seems to run very nicely. I chose this model as it accommodates a nice big and powerful motor - a range of different motors is also available from HighLevel. The separate final drive carriage gives some flexibility in placing the unit, depending on the amount of space available inside the body - happily the 2-8-0T is quite commodious. The brass final gear and the driven axle will be fitted in this carriage when mounting the gearbox in the loco.

Visit the High Level website for their range of motors and gearboxes, where you can also download scale drawings to help select a suitable size, as well as find useful guidance on the gear ratio to choose. Usual disclaimers apply - this was my first purchase from HighLevel but, as can be seen, it all goes together very nicely and the mail order service was exemplary.