Pedants’ Corner

This section is a light-hearted attempt to highlight both a) the areas where we have altered Churchward’s original design, either to suit modern manufacturing techniques or modern operation requirements (modellers beware!) or b) where we have attempted to ‘go the extra mile’ to try to retain original features partly for historical accuracy and partly for our own satisfaction of ‘getting it right’. Wherever we have made changes we have endeavoured to keep the outward appearance and sound as original as possible, while it sometimes seems ironic that the more accurately authentic we make things the less anyone notices! We hope this section may offer ideas and encouragement to fellow preservation groups faced with similar challenges.

1. MODIFICATIONS TO CHURCHWARD’S DESIGN

1.1 The locomotive is fitted with two live steam injectors. Originally the injector on the fireman’s side would have been an exhaust steam injector, being supplied by a large-diameter steam pipe, since removed, running from the cylinder block, down the centre of the chassis and feeding the injector below the cab. It would also have had a second live-steam supply, running from the central connection on the steam fountain at the top of the firebox. This change was adopted partly because the live steam injector is lower cost and also because we were unlikely to be using the engine on long steady hauls better suited to the exhaust injector. Both the preserved 2818 and 3822 retain the original feature of the exhaust steam injector. (BTW, there's a section of pipe missing in the photo which curves down to the flange on the top of the injector. It had a leak so was being replaced at the time).

1.2 As the locomotive is now to all intents and purposes a passenger locomotive it is fitted with steam heating for carriage warming. This is supplied from the central connection in the steam fountain previously used for the exhaust injector, giving the bonus that such a steam pipe ‘looks right’. However, this feeds to an industrial pressure reducing valve below the cab floor, set at a steady 40psi, with steam heat pipes running fore and aft to the shut-off cocks and hoses on the front and rear bufferbeams. The main ‘give-away’ is the steam heat pressure gauge, fitted in the GW’s standard spot below the boiler pressure gauge on the fireman’s side.

1.3 As originally built the 28XXs, together with the eight-coupled tank engines, were fitted with a Brake Moderating Valve in the vacuum brake branch pipe to the brake cylinder. Concerns were raised by the SVR in 1985 that, while this was originally provided to moderate the performance of the loco brake to be consistent with that of a rake of short wheelbase vacuum fitted wagons, it would be inappropriate for use on a heritage passenger line, particularly with lesser experienced drivers. Simple removal of the 1" dia bronze ball within the brake moderating valve reset the brakes to that of a standard passenger loco.

1.4 In 1984 it was decided to fit a hopper ashpan, designed by members of the team who did not relish the prospect of raking out the old plain type. It is based on a BR standard type, with two centrally pivoted doors and operated by a mechanism tucked away as best we could behind the fireman’s side rear sandbox.

1.5 The ashpan was fitted in 2011 with a slacking pipe fed from the injector, controlled by a GW-style valve mounted low on the cab front, fireman’s side. This is an SVRCo modification. The ashpan also has been fitted with ‘non-authentic’ pipe connections either side to allow hosing down from a shore water supply.

1.6 It was noticed during the restoration from Barry and again during the 1996-2011 overhaul that various moving parts were under-lubricated, leading either to excessive wear or to rusting and seizure. We have added grease nipples to the worst affected points, the most noticeable being the tender brake shaft and brake cylinder trunnions, with piped supplies accessible behind the front steps.

1.7 The replacement tender body, built by DG Welding at Gloucester in 1995-6, is riveted where the rivets are visible, but internally it is of welded construction. An attempt was made to incorporate (unauthentic) drains in the coal space, but these quickly blocked up in practice and have proved to be of limited value.

1.8 We have added low-level fillers to the rebuilt tender, with end fittings for a fire hose. These have been fitted on most main-line certified locomotives to assist with filling from the lineside, but have also proved very welcome at Bridgnorth for filling the tender at the start of the day, avoiding the need to shunt for access to the water column.

1.9 Various items of platework, particularly below the cab floor and around the front dragbox, were galvanised in 2001 when replacements were made, due to severe rusting of the originals. In fact ¼” (6mm) steel below the cab floor which was replaced in the first restoration had rusted through completely by 1995.

1.10 The locomotive vacuum reservoir, exposed to the corrosive conditions below the cab floor, and has been replaced in stainless steel, made in 2001.

1.11 In 2003 we made a replacement for the short stub of pipe connected to the top of the brake cylinder under the cab. Being in the same corrosive spot this had rusted terribly and so the replacement was made in stainless steel.

1.12 The pony truck lateral slides were originally fed from the oil pot below the smokebox door via two rigid pipes to two spring-loaded ‘fingers’ feeding downwards into an oil reservoir in the pony truck. The threads for securing these ‘fingers’ had got stripped before arrival at Barry. We tried to effect a repair, but on one of the first runs in steam they were found to have stripped out again. We substituted flexible PTFE hoses direct to the pony truck in place of the rigid pipes and ‘fingers’, and these have proved to be much better suited to the curvature of the SVR.

1.13 The majority of GW engines are fitted with a crosshead-driven pump to maintain the vacuum in the brakes when running. The pump is fitted with an oil pot on the top, but on the 28XXs the oil pot is separate, connected to the pump by a short pipe. In order to allow for relative movement between the oil pot mounted on the running plate and the pump mounted on its own bracket, this pipe was formed as a sort of elongated copper bellows to give a degree of flexibility. On arrival from Barry it was clear our bellows pipe had cracked, probably due to work hardening in traffic, so we decided to replace it with another flexible PTFE pipe, this being hidden behind the pump bracket, a solution which has been absolutely trouble-free.

1.14 The top oil feeds to the driving axleboxes were removed or blanked off by the SVR Engineering Department during the 2011 overhaul on the basis that they were actually considered to be superfluous. Trouble-free running since their removal would appear to support that contention.

1.15 The driving axlebox keeps have been fitted with drain plugs to allow drainage of water which is more likely to collect in the keeps of a locomotive stored stationary out of doors in all weathers.

1.16 This item really is taking logging of pedantry to extremes, but the beading down the rear of each cab side sheet was originally a hollow steel tube, which naturally rusted away from the inside. When replacing the cab sides in 2001 the replacement for this beading was fabricated from solid bar, machined with a longitudinal slot to fit the cab sheets.

2. ATTENTION TO DETAIL

2.1 When first restored in 1979 the tender was fitted with an early oval brass numberplate. While this looked very smart the feeling gradually emerged that we should no more fit a spurious random numberplate on the tender than we should fit one on the engine. Consequently we found the original tender number 2355 stamped on the brake shaft bracket and, with assistance from the ‘Foxcote Manor’ team at Llangollen, arranged for a cast iron plate to be produced in 2008 as per original fitting. This plate was then fitted using correct round-headed bolts, in place of non-authentic hexagon ones.

2.2 The front ring on the smokebox was manufactured in 2009 with the correct larger radius around the periphery. Compared with the ‘easy way’ of welding a circumferential ring onto the front plate, this necessitated an additional square bar being welded on the inside of this joint, the resulting welded structure then being machined to give the correct appearance. If you are unaware of it this is probably not too obvious, but once you are aware then it is!

2.3 We ensured correct round-headed bolts were fitted for the tender top lamp bracket.

2.4 Similarly round-headed bolts were fitted to the storm sheet brackets in place of the non-authentic hexagon ones.

2.5 When it was discovered that the chimney bell in the smokebox had become severely corroded the accepted method of repair was to produce a steel cone. We rejected this approach and instead in 1999 found a supplier capable of producing a batch quantity of properly flared bells which we were able to supply to other preservation projects.

2.6 The original blower ring was badly corroded when received from Barry and was repaired with lugs welded on the top. This just about lasted the first boiler certificate and again the accepted replacement method was non-authentic, making an LMS style blower on the top of the blastpipe. We rejected this approach also and went ahead with pattern equipment for a proper chimney-mounted combined blower/vacuum ejector ring for the 2011 overhaul.

2.7 Early in 1985 we obtained a copper-capped chimney and this was fitted briefly in April of that year. However, it was decided not to proceed with this non-authentic fitment and the loco has always run with a plain cast iron chimney.

2.8 For the new chimney, cast in 2003, the pattern was borrowed from Peter Robinson, owner of loco 3814. However, with his agreement, this pattern was modified to give the correct curved profile at the top, leading up to the caupuchon.

2.9 When re-fitting the boiler in 2011 the SVR boilersmith Graham Beddow scrutinised numerous old photos to check on whether the smokebox holding-down bolts go nuts inwards or nuts outwards? Nuts outwards was the conclusion, probably so that they end up less corroded and are thus easier to dismantle.

2.10 When the corroded parts of the running plate at the rear came to be replaced in 2000 it was discovered that the one on the right hand side had the rear corner cut as a sharp 90º corner, whereas that on the left had an elegant 2” radius. This illogical detail has been faithfully copied!

2.11 On the loco cab steps the lip at the front end of the bottom step was found to be a different shape from that at the rear. At first we thought this was a crude bodge caused by inaccurate Swindon forge tooling, but then we found that the bottom step on the other side was the same, but the other way round! For the life of us we can see no logical reason not to make both ends of each step the same, but for historical accuracy and our own amusement we have copied the original lop-sided shape when replacing them in 2001.

2.12 Some difficulty was found when assembling the cab sides in 2002 and in the end we had to resort to making one side 5/8” taller than the other, as per original!

2.13 When the loco and tender were removed from Barry efforts were made to ensure that the tender had the earlier style oil fillers on the axleboxe keeps. In the event only five were available, the sixth being one to the later cork-type design. Furthermore, one of the five was damaged and was repaired firstly by Swindon, and constantly fell apart, and then by us with a simple flap secured by two bent bolts, rather like a lever-arch file.

Unfortunately in May 2017 the tender was derailed in Kidderminster and one of these earlier keeps got broken. Currently (2018) the tender is running with three early flap-types on one side and three later cork-types on the other. It is our intention to try to obtain a full set of flap-type keeps if possible, though the trail for this has gone a bit cold.

2.14 We have a GWR Automatic Train Control unit which it is hoped we can fit under the front of the locomotive at some time in the future. However, it has to be acknowledged that, being a non-working piece of equipment, the priority for carrying out this heavy and awkward job is rather low. The corresponding battery box and cab unit, complete with brass bell, are fitted, together with the electrical conduit along the running plate, so it has to be admitted that for authentic appearance’s sake we really should complete the job.

2.15 During the 1995-2011 overhaul it was discovered that the small step plate on the top of the rh front buffer had corroded. The ‘easy way’ would have been to knock up a new step using standard chequer plate, but this would have been completely the wrong pattern, so in 2004 we took a laser cut blank that had been sourced by the SVR and, with a piece of home-made tooling in a fly-press, made the small square dimples in the top surface before the plate was then formed into shape and the corners welded up. So impressed was the SVRCo with the result that in 2006 they asked us to finish off a total of 50, this time pre-folded and welded, for the SVR to hold in stock and, we believe, to sell on to other railways.