The New Front End

As a Consultant Design Engineer by profession I had undertaken projects previously with the owner of BVR No. 9 'Mark Timothy', Alan Richardson, and he invited me to design a new front end for the locomotive to improve power output and fuel economy. Concurrent with the commencement of the rebuilding of No. 9 the BVR were rebuilding No. 6 and this led to the possibility of designing and building new cylinders and draughting arrangements for both locomotives unhampered by the legacy of the existing parts.

A project of this nature is always the subject of collaboration between the interested parties and I would like to especially mention the workshop teams of both the BVR and Alan Keef Ltd who contributed their expertise to the programme.

Redesigning the front end

Hollow bronze piston valve head and stainless steel valve spindle

The hollow bronze valve heads are mounted on stainless steel valve and tail rods which are through ported to improve exhaust release and relieve loads on the valve gear. Bronze was selected for the valve heads as it provides a good wear combination with the cast iron liner. Stainless steel was selected for the rods as it had a similar coefficient of thermal expansion as the valve head and therefore would not loosen in service while also providing excellent corrosion resistance when the locomotive is laid up over the winter period preventing damage to gland packing.

Photo Alan Keef Ltd

Designers like Chapelon, Porta and Wardale advocate that design must progress considering the locomotive as a whole and with a clear understanding of the thermodynamic implications so that the technical potential of improvements can be assessed. Whether a potential improvement is commercially worthwhile must then be judged in the context of total cost of ownership and with locomotives of this size the relative balance of the various contributions is different compared to say standard gauge locomotives with different usage patterns.

In this case, despite the thermal efficiency benefits obtainable by superheating, modification of the boilers was ruled out at this stage for financial reasons as it would have meant replacing large parts of sound boilers and the fuel saving could not justify the costs involved. However, consideration was given to possible fitment of superheated boilers at some point in the future when replacement was required. Similarly, to limit the impact of the work on the rest of the locomotive, the mainframes were not to be modified and the existing cylinder mounting bolt positions, exhaust passageway cut-outs and cylinder bore and stroke were retained.

Accepting that it was to remain a saturated engine, the important point with the new front end design was still to get more useful work out of the steam before exhausting it. To do this required both expansion of the steam to the lowest pressure possible to maximise the useful work done and also minimising negative work in the form of back pressure, friction and avoidable leakage and condensation losses.

With this in mind I prepared an outline design scheme and, having obtained everyone’s approval, undertook the necessary computer modelling, design calculations and drawings to enable manufacture of the parts for both locomotives.

One of the new fabricated cylinders showing the large steam chest, exhaust ports and passageways prior to fitting the cast iron cylinder and valve liners. The grooves in the cylinder which take the Aflas O rings which prevent leakage between the cylinder liner and the fabrication can be clearly seen.

Photo Alan Richardson

One of the new cylinders

Looking at the improvements in broad terms and working through the items in a logical sequence we start at the cylinder steam chests.

One of the new valve liners

One of the new split cast iron valve liners showing the trapezoidal transfer ports and large exhaust ports. Two liners are required per valve chest one being inserted from each end leaving a gap between them in the centre of the steam chest for passage of live steam to the valve head. As with the cylinder liners they are sealed to the cylinder fabrication using Aflas O rings and are clamped in position by the valve chest covers.

Photo Alan Keef Ltd

To meet the demand at speed when the valve opens during admission you need large steam chests so that sufficient steam is available. This ensures that steam pressure doesn’t fall because it suddenly has to surge down the steam pipe from the boiler to get there. For this reason the steam chests were made as big as possible within the constraints of the loading gauge and while the volume is a little smaller than is ideal they are more than eight times larger than the originals.

The valves and ports are also important in controlling the steam flow to and from the cylinder and the transfer ports were designed with fewer but larger trapezoidal shaped ports to improve the available area during initial opening compared to the smaller but more numerous triangular form used previously. The use of narrow tapered lands on the admission and exhaust edges of the valve head enabled the valve events to be defined accurately by the valve rings under the control of the valve gear.

The valve gear was redesigned to improve the valve events by correcting geometry errors and limiting the full gear cut-off from 82% to 70% while also increasing the maximum valve travel by 10 mm to 74 mm. The overall effect is to make port openings larger for a given cut-off and is coupled with generous radii on the valve head which smooths the passage of the steam adjacent to the port.

Back pressure is minimised by allowing some of the exhaust steam to pass through the hollow valve head to make use of the other exhaust passageway branch. The combined effect of these changes is to promote good breathing while also reducing loads on the valve gear by lightening the valve head and balancing the load created by the exhaust back pressure.

The transfer passageways between the valve chest and cylinders and the exhaust passageways have also been carefully designed with rectangular sections to maximise the area for flow. They are as straight as possible but, where necessary, bends are flowing and even the valve chests and cylinders are smoothed internally. Compared to the original arrangement the passageways are enormous and one look at the illustration shows the difference in size between the original cylinders and the replacement units.

The new cylinder on the left shows the large inline transfer ports and exhaust passageways compared to the original cylinder on the right. The relief of the face which bolts to the locomotive mainframes can also be clearly seen and this reduces thermal conduction losses through the loco frames.

Photo Alan Keef Ltd

New and old cylinders

Sealing of the valves and piston has been improved by using Clupet rings which are shaped rather like a key ring and have no gap between the ring ends for steam to leak through. In addition the valve heads now have three rather than two ring grooves as the increase in lap has created enough space to fit them. The valve head is made of bronze giving good wear characteristics in combination with the cast iron valve liner and is supported by a tail rod to improve guidance.

Table 1. Comparison of the original ZB cylinders with BVR valve modifications as fitted to No. 7 and the new cylinders for No. 9.

The new cylinders are designed to allow the locomotive to breath freely and this is assisted by the large steam chest volume, increased valve travel coupled with reduced full gear cut-off to improve valve events, increased lead steam coupled with larger steam and exhaust lap, and larger transfer ports and exhaust passageways to reduce back pressure.

Notes: [1] The original pipework in the smokebox is the limiting factor rather than this value which is given for completeness. [2] This is the area in one end of the valve liner plus (on No. 9 only) the area through the ports in the hollow valve head. [3] These figures are not known for No. 7.

Item DescriptionNo. 7No. 9
Bore, mm176176
Stroke, mm280280
Steam chest inlet passageway area [1], cm²19.6419.64
Steam chest volume, cm³688.75886
As a % of swept volume of cylinder10.186.4
Valve diameter, mm8888
Area of transfer port in valve liner, cm²38.936
Max. valve travel, mm @ % cut-off64 @ 82%74 @ 70%
Lead steam, mm12
Steam lap, mm1319
Exhaust lap, mm02
Min. area of transfer port in cylinder liner, cm²8.534.32
Area of exhaust port [2], cm²28.377.99
Min. area of exhaust passageway in cylinder, cm²45.3799.36
Clearance volume [3], cm³-752.7
As a % of swept volume of cylinder [3]-11.04
Maximum expansion ratio @ 15% cutoff [3]-4.26

The new Lempor exhaust system installed in the smokebox showing the combined blast nozzle and blower assembly mounted on top of the tapering exhaust stand. The spark arrestor is not fitted and the copper pipe connecting to the right hand side is the blower supply.

Photo Alan Richardson

The new Lempor exhaust system

Avoidable heat loss has been lowered by the use of better and thicker ceramic insulation and reduction of contact area between the main frames and cylinders. Heat loss is further minimised by arranging for all coasting to be done in mid-gear as this prevents cooling of the cylinders by cold ambient air drawn in through the snifting valves. To enable mid gear coasting, without abrasive smokebox combustion gas and ash being drawn into the cylinders via the blast pipe, requires the provision of a small quantity of steam to the steam chest. By using this steam to atomise the cylinder lubricating oil as it enters the steam chest the lubrication of valves and pistons is also improved.

A view of the exhaust stand from above showing the divider of the Kordina immediately below the flange. The purpose of the Kordina is to reduce the negative effect on the other cylinder during an exhaust event by avoiding transfer of backpressure. During the dynamometer trials which were carried out the benefits were shown to be significant.

Photo Alan Keef Ltd

The exhaust stand from above
The four nozzle blast cap

The four nozzle blast cap with integral blower. The blast nozzles are of the convergent divergent type to maximise the coefficient of discharge and minimise the exhaust back pressure. The nozzles are also angled outwards to ensure that the blast fills the mixing chamber of the chimney evenly.

Photo Alan Keef Ltd

Better expansive use of the steam can now be made as a result of the improved valve events and with a small clearance volume a maximum expansion ratio of 4.26 is achievable at 15% cut-off.

The draughting arrangements in the smokebox were improved by the fitment of a multiple blast nozzle Lempor exhaust system which creates sufficient draught for efficient combustion with lower exhaust back pressure. A Kordina was also incorporated in the exhaust passageways from the two cylinders where they combine below the blast nozzles. This device prevents the exhaust release from one cylinder increasing the back pressure in the second cylinder.

The four nozzle blast cap is effectively 44.5% larger than the original single nozzle blast pipe taking into account the improved flow coefficients and has an integral four nozzle blower. This was machined rather than fabricated because of the small physical size and accuracy required which made this the easiest option. The Lempor mixing chamber and diffuser is concealed within a chimney of classical external shape but is larger in diameter as the exit area of the chimney was increased by 123% compared to the original. A further benefit is that the revised layout distributes the draught more evenly across the boiler tube plate and it is hoped that this will reduce the incidence of tube leakage in service from the upper rows of boiler tubes.

Table 2. Comparison of the original ZB stovepipe and single blast nozzle exhaust system as fitted to No. 7 and the new multiple blast nozzle Lempor system for No. 9.

The new exhaust system is designed to reduce back pressure and incorporates a Kordina in the blast pipe stand coupled with increases in both blast nozzle and chimney exit areas.

Item DescriptionNo. 7No. 9
Number of chimneys11
Total blast nozzle tip area, cm²15.9120.91
Total chimney choke area, cm²181.48166.64
Ratio chimney choke area : blast nozzle tip area11.48.0
Choke diameter, mm152146
Mixing chamber length, mmN/A291
Mixing chamber length : diameter ratioN/A2.0
Diffuser length, mmN/A452
Length of bellmouth, mm047
Chimney length inc. bellmouth, mm375790
Total chimney exit area, cm²181.48404.81
Chimney exit diameter, mm152227
Diffuser included angle10.3°
Diffuser area ratio1.02.43
Overall chimney length excluding bellmouth : mixing chamber dia ratioN/A5.10
Number of blast nozzles14
Individual blast nozzle tip area, cm²15.915.23
Blast nozzle exit dia, mm4525.8
Blast nozzle throat diameter, mmN/A23.9
Ratio of blast nozzle exit area : throat areaN/A1.165
Kordina crossectional area, cm²N/A20.91

Rebuilding gets underway

New cylinder in place

One of the new cylinders in place with the steam pipe connected but without the lagging or end covers. The large rectangular exhaust passageway sweeping back from the front of the valve chest can be clearly seen.

Photo Alan Richardson

During the later half of 2002 work got underway in earnest at Alan Keef’s works. Under the supervision of works manager Phil Kent, the frames were lengthened and strengthened at the front and rear to suit the new design. During this period the pony truck and rear bogie were extensively rebuilt to provide adequate movement and side control. The successful 'shoe horning' of the alternator and rotary air compressor onto the rear bogie was particularly difficult with very limited space.

While the boiler was removed from the frames the opportunity was taken to carry out minor remedial work on it. The washout sockets, dome and one longitudinal stay which had curiously been found to be welded in position under compression all receiving attention.

Meanwhile, engineer Alice Keef had been looking at the typical weight distribution of the ZB design and confirmed that it was excessively back heavy. To correct the balance it was decided to fit a heavier front buffer beam and make the smokebox from 40 mm instead of the usual 6-10 mm thick plate. Alice is confident that this is one smokebox which is never going to rust through!

New coupling rods were designed by Alice to clear the lower running boards required by the new Leek and Manifold outline and Phil and his team in the works made these items along with replacement or remedial work on nearly every other item of the motion and valve gear.

Setting the eccentric rod length during rebuilding of the redesigned valve gear which enabled the valve travel to be increased by 10 mm to 74 mm. The maximum full gear cut off was also reduced to 70% to improve the valve events. Note the thickness of the smokebox which was increased to 40 mm to improve the weight distribution of the locomotive which was rear heavy as originally built.

Photo Alan Richardson

Setting the eccentric rod length

As work on the chassis progressed, attention turned to the design and fabrication of the new tanks, bunker and cab. The tanks and bunker were fabricated first and then a wooden mock-up of the cab was prepared to check the sightlines and positions of controls and equipment before finalising the drawings for construction.

The new cab is extremely spacious and is fitted out to a very high standard internally with opening windows, a sunroof, and a wooden lining to the cab roof from which is suspended an instrument binnacle with integral LED spot lights for illumination of gauges. This is definitely the loco for winter nights on Santa trains!

The article continues with part three: Commissioning Trials.