Saturday 29 January 2022

Battlefield Line 'Santa' Trains 2021

At the end of 2021 the Battlefield Line was able to operate a very successful service of 'Santa' trains despite government restrictions caused by the Covid-19 Pandemic. Having taken pre-bookings to fill all the advertised trains to capacity, the management was faced with a major setback at the end of October when 'Wightwick Hall', which had been planned to work all 'Santa' trains, was 'stopped' due to leaking boiler stays. The locomotive owners decided to take the engine away for repairs. With no suitable replacement on site, the railway faced the major task of hiring an alternative locomotive at short notice, triumphing in securing 7822 'Foxcote Manor' to cover the vital Christmas Season.

Origins of the 'Manor' class

In 1911, Churchward introduced the '43XX' class of mixed traffic 'Moguls'. They were well-liked, very successful and with good route availability resulting in over 300 being built. By the mid-1930s, earlier '43XX' were in need of replacement and increasing loads suggested an increase in power was required. Collett decided on a 4-6-0 which became the 'Grange' class, keeping the accountants happy by re-using the 5'8" diameter coupled wheels and motion from withdrawn '43XX'. However, the increased weight of the 'Grange' compared with the '43XX' restricted the new class (like Collett's 'Halls') to 'Red Routes'. So Collett also designed a smaller, lighter 4-6-0 design which became the 'Manor'. By 1939, 80 'Granges' and 20 'Manors' had been produced. In 1950, a further 10 'Manors' (including 7822 'Foxcote Manor') were built by British Railways at Swindon and the someimes indifferent steaming of the class was cured by a series of fairly simple modifications based on the work of S. O. Ell.


'Manor' Class from British Locomotive Types 1946 (Railway Publishing Co. Ltd.)
Click for larger view


The 2-cylinder 'Standard' designs of the Great Western have Stephenson Link motion with four eccentrics mounted on the driving axle between the frames to provide the necessary valve events. With outside cylinders and valves, as used on the 'Manors', rocking shafts are required to transmit the movement of the dieblocks between the frames (via the intermediate valve rod, rocking shaft, valve link and valve rod) to the piston valves outside the frames.

This is illustrated in the well-known diagram below, taken from the 'Handbook for Railway Steam Locomotive Enginemen', published by the British Transport Commission in 1957. Second-hand copies of the 1957 edition and the Ian Allan reprint edition of 2014 (ISBN: 9780711037946) can still be found second-hand.

Click on the above diagram to enlarge.
Layout of valve gear.


'Foxcote Manor' is currently in unlined black livery with the original 'Lion Astride a Mangle Wheel' totem on the tender. British Railways used lined black livery for mixed traffic locomotives and, in the twilight of steam, I was used to seeing Great Western engines in lined black, although I always imagined that Great Western mixed traffic tender locomotives must have felt embarrassed at being deprived of their 'locomotive green' colour carried before nationalisation. Liveries of locomotives and rolling stock is a fascinating area in itself and of particular interest to modellers. For a very quick introduction, try GWR Liveries on the GWR Modelling site.


British Railways totem on Collett 3,500 gallon tender in unlined black ('Foxcote Manor' at the Battlefield Line).

Events of Sunday 19th December 2019

I was rostered driver on Sunday, 19th December 2021 with Joe O. as fireman and Jeremy as trainee fireman. Years ago, I'd driven 'Dinmore Manor' after a restoration at Tyseley and fired 'Bradley Manor' on a service train on the West Somerset Railway but this was my first turn on a 'Manor' for some time (when 'Dinmore Manor' had previously visited the Battlefield Line, I'd managed a footplate trip but not had a driving turn). Joe and Jeremy had signed-on on time at 7.00 a.m. but fog on my road journey made me a few minutes late.


'Foxcote Manor' during preparation in the shed at Shackerstone on the Battlefield Line. Note the 'NOT TO BE MOVED' sign fixed to the front lamp bracket.

On arrival, I checked with Joe that he was taking water from the shed supply. At Shackerstone, the shed supply is a wall-mounted tap via a long flexible hose but it's a fairly low capacity feed. Normally, if there's space in the tender water tank, this supply will be left on all night but Joe told me this hadn't happened the previous night, so we'd be coming 'off-shed' with less than the 3,500 gallon tender water capacity. The water in the tender tank is indicated on a float-type gauge which rises up out of the tender in a short, vertical, cast column provided with a cast brass plate marked with readings in both gallons and water depth in feet. A small, brass indicator connected to the tank float moves up and down in a slot to show remaining water. This type of gauge isn't bad but it's always wise to allow some safety margin for lack of accuracy. The simple test of the gauge is to push the brass indicator down by hand and check that the float causes the indicator to 'bob' back to the original reading.


View of tender from the footplate, showing driver's side toolbox with the cast column of the tender water gauge in front ('Foxcote Manor' at the Battlefield Line)
Oddly, about the best water gauge I've used is the rather crude 'walking stick' arrangement favoured by Gresley on some of his engines. At the front of the tender, readily accessible to the footplate crew, a vertical pipe, connected to the tender tank, extends the full height of the tender water space. This pipe is perforated by a row of small holes, normally closed by a smaller, inner concentric pipe. A handle allows the outer tube to be rotated until the holes align with (I presume) another set of holes in the concentric smaller tube, allowing water to escape onto the footplate from all the holes below the level of water in the tender. An adjacent series of painted figures allows the indicated gallonage to be read off. I never quite developed the technique of avoiding catching some of the spraying water, often with wet boots, sometimes a wet arm as the handle was pushed closed after making a reading.
Normally, locomotives are able to replenish their water tank rapidly from a water tower adjacent to a railway line or, at more important sites, from a large storage tank serving a number of 'water cranes' or 'water columns' in strategic locations. In this country, the final stage of water delivery is normally via a short length of slightly-flexible pipe referred to as the 'bag', originally of riveted leather construction, which allows for a small misalignment between the locomotive tank filler and the delivery pipe

At Shackerstone, a large water tank is set at the top of the embankment near the south end of platform 2 feeding a water column with a swivelling arm just beyond the platform ramp. Unfortunately, this facility was not available to us because our train comprised six coaches, stabled with the northern end of the stock just clear of the barrow crossing from platform 1 to platform 2 used by staff and passengers to access platform 2. This meant the southernmost coach extended beyond the platform ramp, preventing access to the water column (unless the whole train was shunted). Shunting the train is not encouraged so, should taking water be required, the approved method is to water when returning to Shackerstone with a northbound train. This is done by stopping at the signalbox, uncoupling the locomotive and drawing forward to the column, taking water and then returning to pick up the waiting train and complete the journey to the platform.

The engine had been in use the previous day and was still quite warm so Joe soon had a decent fire established. He offered to oil the axleboxes and inside motion and I gratefully accepted. I'm finding it increasingly difficult to clamber around in the inspection pit and haul myself up to perch on a coupled axle during preparation. I contented myself with going underneath afterwards for a quick inspection.


Jan emerges from the pit after inspection 'underneath' ('Foxcote Manor' at the Battlefield Line)

Jeremy assisted Joe with the oiling, in order to familiarise himself with the layout and method, which is generally similar to the 'Hall' class, which I described in the post here.

I started my own oiling by draining the previous-day's condensate from the sight-feed lubricator mounted on the boiler backhead in front of the driver and re-filling with cylinder oil.


Sight feed lubricator ('Foxcote Manor' at the Battlefield Line)

Then, I applied motion oil to the various oiling points on the outside motion and tender, examining each part for obvious damage or defect in the process.


Jan applies motion oil to the left-hand connecting rod little-end on 'Foxcote Manor' at the Battlefield Line

Finally, I walked around the foot-framing completing motion oiling 'on top', including filling the vacuum pump oil pot with an oil/paraffin mix.


Jan poses on the front framing of 'Foxcote Manor' at the Battlefield Line, having completed the oiling 'on top'.

By the time we'd finished this oiling-round, the boiler was 'brewing' nicely so we moved outside the shed to empty the ashpan and finish preparation. As you leave the shed, the track is inclined downwards and, if the brake is at all reluctant to 'bite', vigorous help from the fireman screwing down the tender handbrake handle may be needed to bring the locomotive to a stand. On this occasion, I had to place the reverser in 'back gear' and cautiously open the regulator to produce sufficient retardation to stop.
Burning coal in a locomotive firebox leaves ash and incombustible material which should drop through the firegrate into the ashpan below. The ashpan is provided with one or more hinged doors which are adjusted by the fireman to regulate the primary (or 'bottom') air. Daily (or sometimes more often) this ash has to be removed by opening the damper door and scraping out the residue with an ashpan rake. This work can be heavy and is certainly dirty. The amount of ash and incombustible clinker-forming material depends upon the source, and hence the chemical analysis, of the coal.

To somewhat reduce the amount of sheer physical effort involved, two features were introduced in 'modern' designs - the rocking grate and the hopper ashpan. The rocking grate arranges the firebars on a series of axles which can be rotated from the footplate. A slight to-and-fro shaking action can be used 'in traffic' to help break up the formation of clinker and encourage it to drop into the ashpan. Rotation through 90 degrees opens up broader spaces to 'dump' the whole fire into the ashpan. The hopper ashpan allows the fireman to open the bottom of the ashpan so that gravity will empty the contents.
Locomotives in preservation were built in different periods and some have these 'modern' features, some do not. Often, these features have been added by restoration teams to assist in locomotive management. Oddly, whereas 6989 'Wightwick Hall' had a hopper ashpan but no rocking grate, 7822 'Foxcote Manor' has a rocking grate but ashpans are scraped-out with a rake.

The fireman and I reviewed the water situation. The tender water gauge indicated 2350 gallons (capacity is 3500 gallons). We agreed that, even allowing for the extra water which would be consumed train heating, we should have enough for the three scheduled trips plus a safety margin. If we changed our view, we retained the option, on approaching Shackerstone on the second or third trip, of uncoupling the locomotive and drawing forward to the column, taking water and then returning to pick up the waiting train.

Once preparation was complete, we moved 'Foxcote Manor' onto the 6-coach 'Santa Special' train in platform 2 so that we could commence steam heating of the stock well before passengers were allowed onto the train. We were advised to not exceed 18 to 20 pounds per square inch steam heating pressure, to prevent causing leaks in the rather 'tired' coach heating installation.

Our diagram called for three round trips to Shenton during the day, leaving Shackerstone at 12:15, 14:00 and 15:45, running tender-first to Shenton and chimney-leading back to Shackerstone.


Jan, Jeremy and Joe on the footplate of 7822 ('Foxcote Manor' at the Battlefield Line)

As departure time approached, I opened the ejector steam cock to create vacuum in the train pipe. Whereas 'Wightwick Hall' had the very effective multi-cone ejector mounted on the outside of the firebox (used on larger Great Western passenger engines and described/illustrated here), the 'Manors' retain the earlier ejector (see post here), so I was expecting a delay as air was exhausted from our six-coach train. The Train Pipe indicated the correct 25 in/Hg with the Reservoir managing 23 in/Hg. I closed the ejector steam cock to check for leaks and noted that the Train Pipe held up fairly well but the Reservoir (only affecting the engine and tender) started to fall. This seemed to be in line with a verbal report I'd had so I decided to review the actual braking performance in traffic with a view to making an entry on the Inspection/Defect sheet. I re-opened the ejector steam cock so that the guard could perform his own brake test.

Station staff were still shepherding late arriving passengers to their reserved seats as departure time passed but eventually all doors were closed and I acknowledged the 'Right Away' from Martin, the guard, with a hand signal and a whistle. The handbrake was released, both the fireman and I checked that the line was clear and our signal was still 'off' and, in a cloud of steam from the open cylinder drain cocks, we gently set off tender-first. The Collett 3,500 gallon tender is not very tall so the driver's view running tender-first is superior to that with a Collett 4,000 gallon tender (which 'Wightwick Hall' had been paired with). After a few yards, I wound the reverser away from the 'full gear' I'd started in and closed the drain cocks so that I could visually check that the whole train was following and that there were no signs of alarm on the train or the platform.

Jeremy collected the single line train staff from the signalman and I stressed the importance of both driver and fireman checking every time that it is the correct staff for the section being entered. The cutting leading to Barton Lane bridge is limited to ten miles an hour, pending relaying, but even when our whole train had passed clear of the restriction, I kept the train running at about 10 m.p.h. During this outward journey to Shenton, Father Christmas and his helpers would work their way through the train, talking to children (and their families) and issuing presents so a very leisurely progress is appropriate. A message from the train confirmed that our speed met with approval.

But it's much harder running a train at low speed. Rolling resistance alweays seems more noticeable and axleboxes don't have the same opportunity to 'warm through'. At low speed, the vacuum pump may not maintain vacuum without assistance from the vacuum ejector. Small changes in gradient or curvature of the line can produce noticeable changes of speed which, in theory, can be compensated by lengthening or shortening the cut-off without adjusting the regulator setting but, in practice, may require alternating periods of 'steam', 'no steam' and even 'brake'. So I was concentrating fairly hard all the way but, facing the tender, I did notice how much colder it was on the exposed sections of the line even with the Collett design of cab.
Collett's deeper, windowed cab is an improvement on Churchward's skimpier cab. But originally locomotives had no cab at all and, when roofless spectacle plates were first instroduced, they were resisted by many drivers. I certainly feel a driver must be fully aware of the surroundings, listening out for all the noises a locomotive can make. Gresley and Bulleid large-engine cabs are a little too enclosed for my taste and I think Stanier's designs for the L.M.S. best handled the necessary compromises in cab design.

Collett's windowed cab, showing the boiler backhead viewed from the tender. The small, rectangular, white object dangling from the fireman's spectacle is the battery/control box for the LED lights strung along the boiler handrails for Christmas ('Foxcote Manor' at the Battlefield Line)

Although passengers were not able to join or alight at Market Bosworth, we were booked to stop so that staff in the leading coach could quickly replenish the following coaches as required. Braking involves the driver admitting air through the brake application valve to partially destroy the vacuum created by the ejector or pump. There's a delay, depending on the train length, as the level of vacuum adjusts. In early braking systems, air was introduced only through the driver's valve but improved coach designs, as in our train, added a Direct Admission valve ('DA' valve) to each vacuum brake cylinder to sense the train pipe vacuum and, as the vacuum fell, allow extra air to enter locally to reduce the delay in brake application towards the rear of the train. Travelling at low speeds with an unfamiliar locomotive and an unusual consist of coaches, it was harder to stop correctly with the leading coach in the platform and the back coach overhanging the platform at the rear, particularly with the driver on the non-platform side. With the fireman relaying the position of the first coach along the platform, we stopped as required and I selected a suitable lineside 'marker' to use on future stops.

After a few minutes of staff activity on the platform, the guard gave the 'Right Away' and we set off again, resuming our progress at about ten miles an hour. Eventually, we approached Shenton, where braking was needed to comply with the 5 m.p.h. speed restriction over the embankment slip beyond Ambion Lane bridge. Although the station platform was on the driver's side this time, I'd been told that a marker post had been erected beyond the station to assist in identifying the correct stopping point, allowing us to stop accurately. Quite a crowd of passengers appeared alongside to admire the locomotive, safely behind the fence, and both Father Christmas and The Grinch joined them as there was the usual flurry of activity. 'Foxcote Manor' was uncoupled, the Single Line Staff was transferred to Martin so that he could operate the ground frame at the north end of the station and we ran-round our train.

We were told that we could return to Shackerstone at Line Speed, which was pleasant, particularly with a tender engine chimney-leading. Of course, the initial departure from Shenton was quite gentle, complying with the 'five mile an hour slack' to Ambion Lane bridge but, once the whole train had cleared the restriction, I was able to tug the regulator to 'full first valve' and let the speed build. I'm afraid I rarely manage to justify the use of 'second valve' considering the moderate gradients, moderate loads and moderate speeds which apply at the Battlefield Line.
When I was a driver at Birmingham Railway Museum, we had a saying "Your haven't lived until you've had a 'King' in second valve" and, with a tractive effort of over 40,000 pounds and a demonstration line only a quarter of a mile long that was certainly quite an experience. Perhaps the steepest line I've driven was on 'Kaitangata' at the mining railway at Shantytown Heritage Park in New Zealand where sanding was essential to prevent slipping. Heaviest train is probably 12 broad gauge coaches (500-odd tons) with a Class 'WP' in India (brief mention here) and fastest speed around 50 m.p.h. on Pacific 'Piekna Helena' in Poland (brief mention here).
We bowled along fairly merrily, slowing to 10 m.p.h. for the 'slack' through Market Bosworth station without stopping. As expected we were routed into platform 2 at Shackerstone and I'd been told that another marker post had been erected beyond the platform end to assist in stopping correctly. The locomotive was fouling the barrow crossing used by passengers so we uncoupled and dropped forward to wait in the north end whilst passengers crossed the line. It was quite misty but everyone seemed in good spirits.


Passengers streaming off the train and back to the car park after the first 'Santa Special' of the day ('Foxcote Manor' at the Battlefield Line)

Boiler pressure had held up well during the round trip to Shenton and water consumption, as shown on the tender gauge, was as expected, suggesting we'd have enough in the tank to complete our working so we finished the run-round and continued steam heating in preparation for our second departure. Joe had agreed to let Jeremy do most of the firing on the second trip. We set off and, as before, I was concentrating fairly hard on maintaining a gentle progress towards Market Bosworth but every time I checked the other side of the cab, matters seemed as expected and Jeremy's 'shovel action' appeared good. Once again, we stopped for a while at Market Bosworth then plodded on to Shenton where we ran round.


Shenton station with The Grinch and Father Christmas walking back to the platform after greeting passengers ('Foxcote Manor' at the Battlefield Line)

Once again, return was authorised at Line Speed. On the way back, I was pondering whether it would be prudent to stop and take water before we entered Shackerstone station. Shackerstone's Down Home 1 (an upper quadrant lattice post design) was off as we approached under Barton Lane bridge at the prescribed 10 m.p.h. and I realised, with horror that the brakes were dragging on. A quick glance at the boiler pressure gauge revealed the reason. Pressure was down to 160 p.s.i. and the vacuum ejector could not maintain sufficient vacuum in the train pipe. It was certain we'd come to a halt but I let the locomotive try for a few more yards before accepting our fate.


Shackerstone's Down Home 1 'off' for the 'Santa Special' ('Foxcote Manor' at the Battlefield Line)

Joe pulled one of the heavy fire-irons out and plunged it into the red-hot mass in the firebox. A bit of probing revealed quite serious clinker in the firebed. It's always problematic - failure to open the firebed up to sufficient combustion air means the boiler pressure will never recover but too enthusiastic 'action' destroys a carefully-crafted layered firebed, which will take longer to get hot again. Joe broke-up a fair amount of clinker then I took the 'bar' off him to satisfy myself that we had sufficient viable firebed to pull our train as far a the platform, Boiler pressure started to recover, encouraged by judicious use of the 'blower' (a steam jet exhausted through the chimney to enhance the draft on the fire when stationary which I often remind firemen can be the "last refuge of the desperate" and should be used in moderation). After about ten minutes, we'd regained pressure to release the brakes and, exchanging signals with the guard, we rather shame-facedly limped to the platform, uncoupled and stood down the north end, in full view of our delayed passengers who didn't seem unhappy at all.


Passengers assembling on platform 1 at Shackerstone with 7822 at the north end about to run-round ('Foxcote Manor' at the Battlefield Line)

We ran-round and coupled up to our train in the gathering gloom. Joe and I attempted to clean the fire and rebuild the firebed ready for our final trip. According to the tender water gauge, we'd ample water for the final trip so I was more worried as to whether our fire-cleaning would be adequate. Our third departure was late but we made our usual slow progress to Market Bosworth and then Shenton without incident but in the dark.

It's a different experience driving a steam train at night on a preserved line as there's little to see and hearing becomes more important in orienting yourself. There's a short article about the railway at night here. Depending on the weather, the sky may be a deep purple, with black silhouettes of trees and other lineside objects and pinpoints of electric light from houses and other features. But our last 'Santa Special' train moved through an impenetrable inky blackness with few visual clues except in the vicinity of each station where a row of electric lights threw yellow pools of light onto the platform surface and shafts or pinpoints of light from nearby houses appear. The business of running round becomes a lot harder in the dark and staff moving around on the ground need to be especially alert to the dangers of slips, trips and falls.

With running-round safely completed, Joe prepared the fire for the final run back to Shackerstone. To preserve some night vision, I averted my eyes from the glare when the firehole door was open. Joe had lit a gauge lamp and fitted it to the gauge lamp bracket provided so that the all-important water level in the gauge glass could be seen. We set off with around three quarters of a glass of water. The boiler pressure gauge wasn't very visible my side in the scattered light from the gauge lamp but the insistent hissing from the safety valves indicated that we were close to blowing-off. After clearing the Ambion Lane bridge speed restriction, the increased regulator opening silenced the safety valve hissing. We had been asked to make a special stop at Market Bosworth station to allow volunteers off, but as we waited for the 'right away', Joe reported "There's no water!". The tender water gauge still read '500 gallons' but neither injector seemed to be picking up any water. This was a serious situation.
Conventional locomotive fireboxes ('Stephensonian' fireboxes) have a double-skin water jacket construction with an inner firebox usually copper, which holds the burning mass of fuel, heavily stayed inside a larger outer firebox, usually steel, with water in between inner and outer firebox. Heat from the burning gases in the inner firebox passes readily through the copper and converts the water to steam. The level of water in the jacket must be above the highest point or 'roof' of the inner firebox at all times since the temperature of the hot burning gases in the firebox can exceed the melting point of copper (around 2,000 degrees Farenheit). As a warning, fireboxes now have a one or more Fusible Plugs screwed into the 'roof', with a hole through the plug sealed with an metal alloy solid at normal working temperatures. The alloy melts upon a dangerous rise in temperature, allowing water and steam to discharge through the hole in the plug onto the firebed. The noise and steam when this happens should alert an inattentive crew to the urgency of the situation. Allowing this to happen is called "dropping the plug" and is one of the worst sins footplate crew can commit. It's sometimes claimed that "dropping the plug" will extinguish the fire but it is intended only as a warning: indeed, there have been a few instances where the crew have not immediately realised that they had "dropped the plug". This is the 'Achilles Heel' of the steam locomotive - If you can't maintain a safe water level in the firebox, the only choice is to remove the source of heat without delay. Normally, this is done by "paddling out" the fire with the Clinker Shovel ('Paddle') and tossing the burning mass onto the ground one shovelful at a time. A locomotive with a rocking grate can 'dump' its fire more quickly (but a large mass of active fire dumped into the ashpan is not helpful). Quick action in withdrawing the fire may prevent or minimise damage to the locomotive.
We'd well over half a glass of water showing in the gauge glass so the question was, would that be sufficient to haul our train back to Shackerstone, uncouple and go for water before the level became critical? The fireman and trainee were both adamant that it couldn't be done. The fireman correctly pointed out that we would first be going uphill, which would tend to slightly increase the water level in the firebox and then be slightly downhill to Shackerstone, tending to reduce the water level in the firebox. My crew were sure we'd not be able to make it. But a locomotive footplate is not a democracy: I quietly confirmed that my assessment was that we could safely work the train back and we would depart as soon as we had the 'right away' from the guard.

Safety is always the primary consideration when working on a railway but before deciding to take the train on to Shackerstone, I'd already reviewed gradient and other issues. Gradients certainly affect the level of water in different parts of the boiler but the Battlefield Line has moderate inclines (Britsh Railways was able to use the line as a diversionary route for the West Coast main line during electrification) and, to optimise boiler proportions, the Great Western was an early adopter of the taper boiler used on larger engines (including 'Manors') which also tends to reduce the water level change on gradients. Withdrawing the fire at Market Bosworth then creates the problem of a disabled train which, in our case, was fully loaded with passengers three miles from where they wanted to be. Seeking assistance involves complying with a safety protocol to ensure that one incident does not create the conditions for another.
There's a simplified introduction to the working of single line railways here.

On the Battlefield Line, Section C of the Operating Rule Book covers Single Line Working, including the procedures for providing assistance to a disabled train. After agreement betweeen the crew of the failed train and the controlling signalman, protection (a red flag or lamp as appropriate) must be set a quarter of a mile from the failed train which must not attempt to move. A member of the footplate crew must take the train staff to the end of the single line section and travel back with the crew of the assisting locomotive to the disabled train. After coupling together, the assisting driver takes responsibility for train working and carries the staff.
The rules set out that, having put protection in place, a 'member of the footplate crew' (usually the fireman) must take the train staff to the signalman by any available means which could involve a lengthy walk in the dark. Assuming a suitable relief locomotive and crew is available, the assistant engine is allowed into the already-occupied single line section, carrying the person from the failed train with the staff (these are the conditions under which "two in a section" are permitted). I was keen to avoid the delay this would incur if possible.

Although hopefully a less-likely problem, I wasn't too sure of the endurance of the electric carriage lighting system. As railways developed, gas was commonly used for the lighting of passenger areas in trains, giving rise to a number of predictable fires following accidents. The worst, at Quintinshill in 1915, involved 5 trains and resulted in 200 deaths. Electric lighting using filamentary lamps, powered from batteries of secondary cells carried in a battery box attached to each coach underframe, remains in use on preserved railways now. The batteries are charged by a dynamo on each coach belt-driven from one axle. A simple regulator (similar to that formerly used on motor cars) connects the dynamo to the battery when the dynamo output is above a charging threshold, which occurs at around 15 m.p.h. This meant that during the 'Santa Special' operations, most of the time, even when moving, the batteries were not being charged. I wasn't sure whether special arrangements had been put in place to 'keep the lights on'.

So we set off from Market Bosworth into the darkness. Keen as I was to get towards our source of water, I wanted to eke-out that precious half a glass of water, so I settled on a very modest speed with a little less than full first valve on the regulator with the 'cut-off' linked up to about 50% to extract, by expansion up to the point of exhaust, as much effort as possible from each admission of live steam to the cylinders. I hoped that this would prove an economical setting. Every time I snatched a quick glance sideways at the gauge glass, I was relieved to see the level creeping down quite slowly but the fireman was paying close attention to the level, periodically using the 'flashlight' function on his mobile 'phone to better illuminate both gauge glass and pressure gauge.

Passing Shackerstone's fixed distant signal, I started to brake, admitting only small amounts of air into the brake pipe and then allowing the action of the vacuum pump to release the brakes, keen to avoid using steam in the vacuum ejector for as long as possible. Approaching Barton Lane Bridge at the regulation 10 m.p.h., the use of the ejector was needed to augment the pump. The Home 1 signal was 'off', so we were able to drift towards the box as I asked the fireman to prepare for dismounting as soon as we stopped and making sure the guard had applied his handbrake before uncoupling the engine so that we could ease forward for water. The Home 2 ground disc signal was also off, allowing me to go towards the box so that I could explain our intentions to the signalman. I stopped with the rear of our tender still on the plain line approaching the facing points - it's bad enough uncoupling in the dark without the added complication of stumbling over facing point equipment in the 'four foot'.

Once uncoupled, I moved towards the water column, asking the fireman to 'set' me in the correct position to ease inserting the 'bag' of the water column into the tender water filler. When taking water, there's nothing more dispiriting than getting a soaking because of poor alignment, particularly at night near the end of a long shift.

Whilst we were preparing the take water, Adrian L. bounded onto the footplate out of the darkness and checked the boiler water indication in the gauge glass, where a sliver of water was still showing.
"Oh, that's alright", he commented before discussing the next moves. I explained that I'd intended to take water, set back onto our train and complete the planned working. Adrian explained that he had arranged for the Class 33 to be started-up and brought to the south end of platform 1, where it rumbled alongside us as we started to take water.


Boiler backhead with boiler pressure gauge (upper left), carriage warming pressure (lower left) and BR-type 'coupled' gauge glass and protector ('Foxcote Manor' at the Battlefield Line)

I opened the water valve on the driver's side injector and, as soon as I confirmed that water had started to issue from the overflow pipe under the cab, I opened the steam cock. The injector immediately 'picked-up', producing the distinctive, reassuring sound of a working injector. The injectors produced by the Great Western are particularly reliable which is presumably why Riddles chose that pattern for use on his British Railways standard steam classes.

Adrian had considered using the Class 33 to pick up our train and take it into platform 1 whilst we sorted ourselves out, although the stock really needed to be in platform 2 for the following day. Confirming how much water we'd taken-on by looking through the tender filler to visually check the level in the tank, we agreed that it would be a little quicker overall for the Class 33 to pick up the train and bring it into platform 2 after I'd taken my light engine through platform 2 to stand clear at the north end.

Having hauled the train into the platform, the Class 33 uncoupled and came behind us so that passengers could use the barrow crossing and walk past the 'Manor' returning to their cars. The locomotive had not been fired for a while so the boiler pressure had started to fall but the fireman and I agreed that we should have enough steam to get 'on shed' and completely fill the boiler. Despite the delayed, unusual final approach to Shackerstone, our passengers seemed happy, thanking us and exchanging season's greetings.


7822 standing at the north end at Shackerstone after the final, eventful, round trip on 19th December 2021 ('Foxcote Manor' at the Battlefield Line)

Finally, the Class 33 moved through platform 1 to stand on the DMU siding, allowing us to follow to the ground frame giving access to the shed. Allowing for our reduced boiler pressure, a short run at the gradient approaching the shed got us inside and careful braking stopped us in the position we'd started from that morning, to ensure that the crew next day had access to the pit at the front of the engine. The engine was 'tied down', scotches set under a coupled wheel and the boiler was completely filled.

An interesting, if rather tiring, day.

Related posts on other websites

Foxcote Manor Society.

Related posts on this website


All my Battlefield Line posts.

My photograph albums

Where necessary, clicking on an image above will display an 'uncropped' view or, alternately, pictures may be selected, viewed or downloaded, in various sizes, from the albums listed:-

'Foxcote Manor' at the Battlefield Line.
All my Battlefield Line pictures.


Sunday 9 January 2022

Review of the Year 2021

Unhappily, following a miserable 2020, the Covid-19 Panemic continued to dominate 2021. The approval of new vaccines against Covid-19 has brought new hope but worldwide, the situation is very variable. In the UK, the first half of the year was in lockdown but as vaccination proceeded, some easing of restrictions was made until the arrival of the Omicron Variant of Covid-19 triggered re-imposition of certain restrictions. This is the 16th annual review since I started this blog

OVERSEAS TRAVEL

Initially, it was hoped that the 2021 expedition cruise to Iceland which I'd booked would be able to go ahead but the frequent changes to government regulations caused the operator to cancel the tour. I'd had to cancel my proposed trip to Myanmar in 2020 because of travel restrictions and these restrictions continued in place at the beginning of 2021. Following elections in Myanmar at the end of 2020, which resulted in a landslide victory for the National League for Democracy, on February 1st 2021 the election result was repudiated by the Myanmar army who seized power alleging voter corruption. The chequered history of this wonderful country has entered another tragic phase. I decided there would be no overseas travel for me in 2021.

To find more about overseas visits I've made in previous years, see:-
Jan Ford's Travels (2001 onwards).
Jan Ford's Travels - The early trips (before 2001).
Jan Ford's Travels: Around the World in pictures
(alphabetic list of countries with links to pictures).
TRAVEL IN THE UK

In a normal year, I would expect to make a number of trips (mainly by rail) to various destinations in the UK. Following the partial easing of restrictions in July 2021, I made a couple of day trips by rail to re-visit Liverpool and Blackpool and a trip to Croydon but other planned trips didn't materialise, apart from a few short rail trips in connection with my work. During the coronavirus Pandemic, rail passenger numbers were around half of normal, with 'emergency timetables' in use and increased likelihood of train cancellation at short notice as staff 'self-isolated' following positive tests for Covid-19.
The Cambrian Line in 2021 25-Jul-2021
A day trip to Liverpool 21-Aug-2021
Blackpool by Electric Train 28-Aug-2021
By Train to Croydon 12-Sep-2021
Trade Show at the National Exhibition Centre 19-Sep-2021
To Stockport by Rail 14-Nov-2021

My excursion to Liverpool included the mandatory trip on the Mersey Ferry.


Blackpool Tower with one of the preserved trams on a 'Heritage Tour'.

RAILWAYS AND PRESERVATION

All preserved railways were seriously affected by the Coronavirus pandemic in 2021, as they had been in the previous year. I continued as a member of the Old Locomotive Committee and remained a volunteer at The Battlefield Line but in 2020 I'd resigned as a working volunteer at Peak Rail although I continue to support their aspirations.

The Old Locomotive Committee

The Old Locomotive Committee (OLCO) usually organises two events each year: the Annual General Meeting, normally held at the Museum of Liverpool where the locomotive 'Lion' is on public display and the 'Lionsmeet' event. 'Lionsmeet' enables live-steam models of 'Lion' to be run and members to meet up. This event is hosted at the running track of a different Society of Model Engineers track each time. In 2021, as in 2020 Coronavirus legislation stopped both events, making OLCO's periodic newsletter called 'Lionsheart' especially important.

All my posts about the Old Locomotive Committee are here and you can find more information (including how to become a member) on OLCO's website here.

The Battlefield Line

At the beginning of 2021, Coronavirus legislation limited operations but on 19th July (optimistically called 'Freedom Day' by the government) most legal restrictions were lifted. The arrival and rapid spread of the Omicron variant of Covid-19 triggered the re-imposition of various mandatory restrictions in early December. My own involvement in public operations was initially limited to a few DMU turns and my first steam turn in the year was not until 29th August when I was re-united with 'Wightwick Hall'. I should have had a second turn on 'Wightwick Hall' on 30th October but the steam locomotive was 'stopped' for leaking stays a few days earlier so, instead, I had an interesting training day on a Class 20 diesel-electric. However, I did manage one further turn on steam with an eventful 'Santa Special' on 19th December.

Battlefield Line 'Santa' Trains 2021 (in preparation)
Class 20 Diesel Electric Locomotive 8-Jan-2022
Getting Back on Track 6-Nov-2021
A Day on the DMU 13-Nov-2021

6989 'Whightwick Hall' ready to leave Shackerstone on an earlier date.


Jan poses during preparation of 7822 for the 'Santa Specials' on 19-Dec-2021 ('Foxcote Manor' at the Battlefield Line)

You can find all my posts about the Battlefield Line here.

Incidentally, you can find all my posts about Diesel Multiple Units here (or refer to the 'Index' here).

Myanma Railways

Because of the pandemic and the military seizing power in Myanmar, I was unable to visit Myanmar in 2021. Just one new post was added in 2021, when I finally managed to describe a visit during my 2019 trip (my incorrect numbering of the workshops visited is still to be corrected):-
Railway Carriage Building and Repair in Myanmar 8-Aug-2021
You can find all my posts about Myanma Railways (in reverse date-of-posting order) here or refer to the 'Index' here.

BREWOOD HALL

The committee which organises the Brewood Garden Party, held in the gardens at Brewood Hall, decided at the beginning of 2021 that, for the second year running, it would not be possible to stage the event because of the Covid-19 Pandemic.

There's an album of pictures showing the 1-acre garden through the seasons in 2021 here with a separate album (covering a number of years) showing the fungi which appear annually here. There's one post from July with musings on the garden in 2021 here.


Brewood Hall had some light falls of snow in November 2021.

During the year, as always, various repairs were carried out around the property. In 2021, I documented some of this work, together with work from previous years, including a domestic disaster from 2018 when a burst water pipe resulted in expensive repairs.
Brewood Hall Maintenance (3) 1-Aug-2021
An unfortunate occurence 31-Jul-2021
Brewood Hall Lantern Roof 18-Jul-2021
You can find all my posts about Brewood Hall (in reverse date-of-posting order) here.

TY GWYN

Ty Gwyn is a small commercial woodland around 27 hectares in area near Corwen in Wales.

In 2021 I finally achieved my ambition of taking a boat on the small pool when, during a visit with Ann and Dean on 12th June, they surprised me by suddenly producing a new inflatable boat. The weed growth made manoeuvring the craft with the oars rather difficult but I had a great time and the weather was really warm. I had good weather again for a visit on 17th July with my friend Geoff.

There's a report covering both visits here


Jan relaxing on the pool at Ty Gwyn 2021 (Photo: D Mould)

To see all my posts on Ty Gwyn, click here.

'ROAD TO MANDALAY' SOCIAL CONTRIBUTION

Since 2008, I've supported these educational and medical charitable initiatives in Myanmar, formerly known as Burma. I'm also currently a trustee of the Belmond Myanmar Foundation in the U.K. which also provides support to this work. Educational Support is provided to a wide range of schools and orphanages and Medical Treatment to all is centred on the Bagan Medical Clinic which was opened in 2011. Until the Covid-19 pandemic, I made at least one visit each year to see as much as possible of the work but, in 2020 and 2021 this has not been possible. On 1st February 2021, following elections at the end of 2020, the army seized power alleging election fraud, compounding the problems of the Covid-19 Pandemic as the country descends into deadly civil war. Because of the combination of the pandemic, the actions of the army and the civil unrest provoked, the United Nations estimates half the population in Myanmar will shortly be living in poverty, making the need for relief work greater whilst the risks involved in supplying that help increase. Despite the risks and difficulties, help continues to be provided by dedicated local people with some overseas financial support.

You can find all my earlier posts on Educational Support here.

You can find all my earlier posts on Medical Support here.

OTHER BLOG POSTS IN THE YEAR

This review covers trips and events during the year but doesn't always include other types of post such as:-
Technical articles;
Historical articles;
Work-related topics;
Recollections of past events;
PREVIOUS ANNUAL REVIEWS
2020
2019
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
2008
2007
2006

Saturday 8 January 2022

Class 20 Diesel Electric Locomotive

I was booked to drive the 'Hall' at the Battlefield Line on Saturday, 30th October 2021 but a few days earlier, the locomotive had been 'stopped' due to leaking boiler stays. The railway put together a 'scratch' crew with a driver passed to drive the replacement diesel electric locomotive Class 20 110 but kindly invited me to have a training session on the diesel-electric if I wished.

I'm passed to drive on the diesel multiple units at the Battlefield Line but not passed on any of the various other diesel locomotives. About 30 years ago, when I was a regular volunteer at Birmingham Railway Museum, I was passed out by their Traction Inpector (whose 'day job' was Traction Inspector on British Rail) to drive their '08' shunter and visiting preserved Class 47 'Ixion' which was used for driving experience courses. I also did a bit of shunting with a visiting Class 31. I've always been a fan of the Class 08, a rugged and reliable machine, and my brief 'User Notes' (with a link to training notes prepared by British Rail, York) are here. Under supervision, I've since had 'hands-on' on a number of other diesel locomotives at various locations with mechanical or electric transmission (and shunting locomotives with hydraulic transmission) but never a Class 20, so I accepted the offer of a training session with Graeme as instructor driver.

Origins of the Class 20

Following Nationalisation of the railways in 1948, British Railways initially committed to building a range of 'Standard' designs of steam locomotives. By the time the last was built (Class '9F' 'Evening Star'), the 'Standard' stud numbered 999 locomotives. But difficulties in recruiting staff to operate and maintain these locomotives, combined with problems in the mining industry extracting the coal to fuel the fleet induced politicians to embrace a change of direction and replace steam with diesel and electric traction. This transition is briefly introduced in the post The Modernisation of British Railways. A series of 'Pilot Scheme' diesel locomotives were ordered from various manufacturers in four power ranges increasing from Type 1 (a general-purpose freight type) through Type 2 and 3 up to the express passenger Type 4.

English Electric were chosen for the Type 1 design. They had been building rugged diesel engines for marine and other applications for some time and their 'straight six cylinder' had given good service in the LMS diesel shunters. In 1947 they introduced a 'Vee' type of engine with pairs of cylinders set 45 degrees apart driving a common crankshaft. The 16-cylinder version (16SVT) of this engine had been used by Ivatt in the well-known LMS prototype main-line diesel electrics 10000 and 10001 and also in the Southern Railway's candidates emerging in the British Rail era as 10201, 10202 and, finally, the uprated 10203. This proven engine series was adopted for the 1000 h.p. (750 kW) Type 1 locomotive using their 8SVT Mark II diesel engine.
8 - number of cylinders
S - supercharged (actually turbocharged)
V - Vee form
T - adapted for railway use
Mark II (4-valve cylinder head)
The engine crankshaft was bolted directly to an English Electric 319-3C main generator with an output of 1070 Amps at 600 volts d.c. An English Electric 911-2B auxiliary generator 'overhung' on the main generator, producing 110 volts d.c. for the compressor, exhauster, traction motor blowers and battery charging.


Auxiliary generator, main generator, 8SVT diesel engine forming power unit of Class 20 locomotive.

The locomotive underframe was two longitudinal, parallel box sections (each formed from two channel sections plated-over top and bottom to produce a box section), with welded 'transoms' forming a rigid frame. Part of the two underframe box sections were sealed off to provide the modest fuel capacity of 390 gallons and two of the 'transoms' also served as balance pipes between the two tank sections.

The power unit was mounted on the underframe together with auxiliaries (including engine-driven fan, oil pump and water pump and electrically-driven air compressor, vacuum exhauster, traction motor blowers, fuel pump), all protected by a long 'bonnet' or hood , inset to give a 'running board' each side. A series of hinged access doors along each side allowed staff ready access to the equipment. At what was regarded as the rear of the locomotive, a driving cab was provided, with duplicated driving 'desks'

The assembled locomotive was mounted on two 4-wheel bogies using welded-frame equalising-beam design with helical springs and 3 foot 9 inch diameter wheels. Overall length was 46 feet 9.25 inches and weight 73 tons.

The four axles were each fitted with an English Electric 526/8D (/5D on first 50 off) nose-hung 6-pole series-wound, force-ventilated d.c. motor rated at 600 Amps, 300 volts d.c. driving through a single reduction gear.

The locomotives were arranged so that up to three units could be interconnected for multiple working from the leading cab using Electro-Pneumatic (E.P.) controls, allowing heavier trains to be worked.


Class 20 locomotive General Layout
Click for larger view


With the English Electric pedigree, the design proved reliable and did all that was asked of it so between 1957 and 1968 a total of 228 locomotives were built, originally numbered in the D8000 series but renumbered with the '20' prefix when the 'TOPS' system (see description on Wikipedia) was introduced in 1971.

The improved cab environment proved popular with some drivers and the good view of the line when running 'cab leading' was appreciated. The more restricted view ahead through the smaller window looking along the 'bonnet' when 'nose leading', although not that dissimilar from that on a steam locomotive, was not favoured and it became commonplace for the locomotives to 'hunt in pairs', coupled nose-to-nose, allowing drivers to adopt the preferred 'cab-leading' driving method in both directions. The class gained an enthusiast following, who dubbed the class 'Choppers', leading to a number of units having been preserved. Perhaps more signicantly, examples still remain in commercial service in a number of countries. I think there are still occasional passenger railtours using the class 20. I wrote about one such tour I witnessed (but didn't travel on) in the 2015 post here.

Class 20 at the Battlefield Line

Graeme W. was booked driver with Jamie W. as Secondman. We were using the locomotive with final running number 20 110 which was stabled down the North End at Shackerstone, currently painted green and carrying its original number D8110.


Class 20 stabled at the North End at Shackerstone.

We all clambered over the locomotive as Graeme demonstrated the pre-start checks.


L: Air Compressor (with Traction Motor Blower behind) R: Vacuum Exhauster (Class 20 at the Battlefield Line)


View of 8SVT engine: Class 20 at the Battlefield Line


Main Generator (Class 20 at the Battlefield Line)


Overhung auxiliary generator and fire extinguishing installation (Class 20 at the Battlefield Line)


Left side of locomotive showing 'bonnet'. Note handrail above engine/equipment access doors and underframe-mounted battery box/battery isolating switch painted black (Class 20 at the Battlefield Line)

Graeme started the diesel engine. Electric start is provided by connecting the 110 volt battery to an extra winding on the main generator so spinning the engine crankshaft. We waited for the compressor to produce sufficient control air pressure and then Graeme shunted to our 6-coach train waiting in platform 2. Jamie 'hooked-on' ready for our first round trip to Shenton. Graeme drove the first trip then, for the rest of the day, I shared the driving with Graeme and (on trip 3) Adrian.


Class 20 110 arriving back at Shackerstone on 30-Oct-2021

The two driving desks are positioned for left-hand drive. Layout is standard with brake controls on the left, power and reverser controls on the right, with an inclined instrument panel facing the driver. Fairly crude swivel seats are provided fixed to the floor at each driving position.

Light engine movements are controlled by the 'straight' Air Brake but, when working vacuum-braked stock, control is via the separate Vacuum Brake Application Valve which is intended to apply the air brakes on the locomotive proportionally to the train vacuum but I need more training to properly get the 'feel' of this combination, tending to underbrake or, more often, overbrake and then having difficulty "blowing off" the vacuum to keep rolling.

The Driver's Surveillance Device (DSD) or 'deadmans' is in the form of an inclined, hinged metal plate on the floor which must be kept depressed when in motion or otherwise, after a short delay, all brakes are applied. The back wall of the cab has a handwheel for applying a 'parking brake', requiring numerous turns to apply or release the brake through a chain drive.

I did notice that, having left Shenton 'nose-leading' and mounted Shenton Bank, advancing the power controller to produce line speed in the cutting did produce a 'business-like' exhaust noise. At the end of the service schedule, I was able to leave the locomotive with Graeme as it was being used again for an evening special. All-in-all, a good day.


Class 20 at the Battlefield Line showing cab and Number 2 Bogie, with Jan in earnest conversation.

Related Posts on this site

The Modernisation of British Railways.
Class 08 User Manual.

Related Post on other sites

English Electric diesel engines.

Book references

[1] 'BR Class 20 Diesels' Michael Oakley (Bradford Barton) ISBN 0 85153 419 8.
[2] 'Description and Fault Finding Instructions for Drivers' (English Electric/BRB Book FS227)
[3] 'Diesel Traction Manual For Enginemen' (BRB 1962)

My pictures

Class 20 at the Battlefield Line

Depending on the display device, the right hand edge of pictures may not display. To see an uncropped image, click on the picture. Alternately, you can find the image by following the 'My pictures' link and display or download the image in various resolutions.