Comparisons in size can be made from these three locos. The G.N.R. 0-6-2 Tank is to 1/480th scale the 'Crab' and Brighton Mogul are 1/240th.
A layout (Model Railway News, June '68 p. 292) made about eight years ago in 1/240 scale (slightly smaller than the recently introduced Z gauge) by the author has been shown in operation at exhibitions every year since construction started. Another effort, this time narrow gauge (2ft 6ins) was described in the Model Railway News, Oct. '69 p. 482. In these sizes it was found that an approach to design radically different from that in larger scales is essential for satisfactory performance. The model must be made as heavy as possible to obtain good electrical contact and adhesion weight for traction; consequently only a minimum of plastic is used where essential for insulation. All the available space is filled with solid metal (mainly brass) and the front of the boiler contains a tungsten (nearly 2.4 times as dense as brass) plug. Laminations were found to be unnecessary for the armature core; the solid core shown in Fig. 1 can be made very compact as the coils do not project beyond the ends of the pole pieces, and there are no sharp edges on the core requiring thick space wasting insulation. To obtain smooth performance at respectable scale speeds a large reduction ratio is desirable. This requires very fine pitch worms and precise mesh with the worm wheel. The commonly used system of gearing between driving axles (which avoids the problem of functional coupling rods) reduces the weight since it involves cut-outs in the frames to make room for the gears. In these models the frames are solid blocks, and the coupling rods have to be fitted with precision to avoid binding at dead centres. In the step down to 1/480 scale all these problems are accentuated, and a great deal of experimental work had to be done before a successfully operating model was produced.
In commercial production manufacture of items is facilitated by dies, jigs, special tools, etc. most of which are very elaborate and time consuming to make and only justified when components are to be made by the thousands. For one off production, only the simplest of dies and jigs are worth considering. The drivers have the internal form (spokes, crank pin and axle bosses, etc.) produced by cold pressing mainly because, in this case, it is much quicker to make a steel die than to machine even one wheel from the solid; furthermore, after pressing, the die can be used to locate each wheel blank for the subsequent turning operations on the treads and flanges. it takes about six anneals to sink the die fully into the nickel silver wheel blank. The insulated wheels are made by turning off the rim, and then attaching a new rim with mica washer insulation (see Fig. 3). The cross-hatched spaces in the figure are filled with Araldite.
The
worm and worm wheel drives require high precision to give smooth motion at
slow speed. The worms are thread milled (which is the only practical solution
with such small core diameters), integral with the armature shafts. The worm
and each end pivot are hardened and tempered; the rest of the shaft is left
soft so that the inevitable distortion due to heat treatment can be at least
partially corrected by bending. The worm wheels are bronze, the teeth are
roughed out and then finished with the blank freely rotating on its own bore
against a bobbing cutter.
The
armature cores (see Fig. 1) are machined from hexagon bar to simplify indexing.
The piece is held in a special chuck (see Fig. 5) so that each coil axis
is centred in the lathe. The outer diameter is machined later, leaving a
few thou. for final cleaning up after it has been pressed on to the armature
shaft. Concentricity of this surface with the pivots is very important if
the smallest practical air gap in the armature tunnel is to be obtained;
this of course improves the power of the motor.
The
valve gear on the 'Pacific' presented many problems. The return cranks were
made integral with the crank pins; consequently some "fiddling" was necessary
to ensure that, when screwed tight into the wheels, the return crank pointed
roughly in the right direction. The expansion link axis was moved slightly
to the rear of the position on the prototype so that the link could be supported
on a long bearing between the first and second drivers. The link pivot is
threaded and rotates in a long "nut", thus avoiding the need for providing
any other means of preventing axial movement— except, of course the
few thou. due to the oscillation of the link, which is quite unnoticeable
in action. The union link is a J-shaped piece of wire with the long leg threaded
for screwing it into the cross head as the wrist pin; the short leg engages
the lower end of the combination lever.
The track for these models n ads to be laid accurately since the wheel flanges are only 0.01in. deep. The easiest solution to this problem is to make a circular track by cementing two nickel silver rings into turned grooves in an insulating baseboard with Araldite. After the adhesive has set the rails (initially over sized) can be turned in the lathe to a high degree of precision, and it is easy to give about 0.006in. super-elevation to the outer rail. The 0-6-2 has not been tried for haulage capacity, but the 'Pacific' has managed a train of seven bogie coaches with a clean dry track and clean coach bearings. Even though the track (12ins dia.) is provided with a perspex cover, muck still seems to get in and that is the major problem in running these models. It is usually necessary to inspect the model with an eyeglass when it won't go and remove fibrous fluff with tweezers. The favourite place for lodgement of this foreign matter is between the wheels and the pick-up brushes.
It is rather unkind to do this to so small a loco, but we thought that you would want to see the 1/480th scale Class 'N2' in detail.
The shaded area shown in Fig. 2 indicates what is meant by filling up all available space with metal - the alnico magnet and the tungsten plug would also be shaded if the exact centre line section were drawn. The motor casing is a silver soldered assembly of brass and steel arranged to satisfy the requirements of the magnetic field circuit. Note—in any silver soldered joint between brass and steel, the brass must not be of the free-cutting variety; 70-30 Copper-Zinc (ordinary sheet brass) composition is satisfactory. The large side tanks of the prototype provide space for a relatively large alnico magnet. Each of the three coils on the armature consist of 250 turns of 0.0014in. dia. wire. Due to the restricted length of the armature shaft a face type commutator had to be used. Drum commutators are preferable when there is room for them. The worm is 0.046in. dia. and 60 t.p.i. meshing with a 24 tooth worm wheel. The core dia. of the worm is very small—about 0.02in. dia. The brushes consist of small copper--carbon blocks soldered to thin phosphor -bronze strips. The configuration of these strips is the result of an attempt to get the maximum flexibility in the available space.
The R.H. drivers are solid (i.e. not insulated) and current is conducted through these and the axles to the frames to supply the grounded R.W. motor brush. The L.H. drivers have insulated tyres (see Fig. 3) which contact a shim brass sprung strip housed in the channel shaped recess in the top half of the main frames. This recess holds a shim brass channel sec-tion insulated from the frames by paper and held in place by Araldite. This pick-up strip conducts current to the L.H. motor brush which is insulated from the frame.
The coupling rods are of nickel silver, mounted on crank pins 0.016in. dia. with 0.031in. hexagon heads. The rear pony truck is not pivoted; the axle is allowed enough side play to negotiate curves. The outer casing is built mainly from 0.005in. nickel silver sheet, the main structural joints being silver soldered. The boiler is turned from brass and bored to accommodate the tungsten plug. The dimen-sions of the components may be assessed from the information that the length over the buffer beams is 0.925in. The "wires" in the area of the smoke box and side tanks are not supposed to be hand-rails; they are part of the condensing system used on these locos when working on the London Underground track.
Another example of overdoing the reproduction size, but again to let you see the animal, in this instance the USA Reading G3 Pacific (1/480th).
This model was made about two years after the G.N. 0-6-2. The main trouble with the latter is the frequent cleaning of the commutator required. The G3 'Pacific' was chosen as the next prototype because it had space for a long armature' and the wide Wootten firebox would admit of a larger diameter armature core. Since the average USA loco has no well defined footplate it is not convenient to separate the functions of "outer casing" and "mechanism". In this design (see Fig. 4) the boiler is integral with the motor casing, and again consists of a silver soldered assembly of brass and steel. The bracket which supports the tail bearing of the armature projects beyond the cab into the tender space. The pin extending down-ward from this bracket serves for the draft connection to the tender. The motor brush pivots are pins projecting from the "backhead"; the R.H. pin is insulated. Spring pressure is applied to the brushes by a hairpin spring anchored by the screw under the cab roof. Current pick-up is taken from the R.H. insulated tyres via brushes bearing on the treads—in a larger scale these could be made to represent brake shoes and hangers.
For some approximation to the proto-type, this model had to have outside cylinders and valve gear; not an easy job with a coupled wheelbase of only 0.35in. The best that one can say is that the expansion link, eccentric rod, union link and combination lever all have (more or less) the motion expected of them. The smallest threads I have been able to make are 0.016in. dia.—this is equivalent to building a full size loco using no bolts less than about 7½ins dia.! The photographs in this article will probably be reproduced three or four times full size, so some of the details are bound to look a bit on the rough side!
The Great Northern (USA) Mikado seen above is one of the 1/240th scale live steamers referred to by the author in his last article (MR November, 1973). The Pennsylvania RR streamlined K4 is an even more recent example of this breed of small scale steamers. Advantage was taken of the casing to hide the overscale length of the cylinders on this model, permitting the fitting of long pistons and piston rod guides, thereby greatly im-proving wearing qualities and reducing steam leakage. By inclining the cylinders upwards at the front, clearance for the leading pony truck was provided—the truck on this model is true to scale for both wheel diameter and wheelbase. The skyline casing was also made use of; it serves to give a little more water capacity, and, since there is no salient stack, the oversize 'exhaust hole' is not very noticeable, but it greatly improves the flow of hot gases through the flue. The performance of this locomotive is the best yet in the series; it has achieved runs of 140 laps of the track (about 840 ft. or 40 scale miles) which is nearly twice as far as any previous loco in this scale. Both pictures by Arthur Gee
