During the construction, I took Liverpool in whatever its current state was to several shows, where it attracted some interest. Usually that took the form of "What on earth is that?", although a few clever people were able to identify it immediately. As a result of one such showing, I was asked to make another model. It was just a pity that I was not able to start the two models together, as there would have been a considerable time saving that way. As it was, from that point forward in the construction of the first Liverpool, I made two of everything and put one away safely. So when the first one was done and off to the paint shop (I prefer not to do my own painting), I had some of the parts I needed for the second. However, a lot still had to be made.
I also took the opportunity to revise the construction in some respects, based on experience. The main problem with number one was that dismantling was a fiddle. With a prototype like this one, the assembly is never going to be the orthodox split between body and chassis, but I decided that I could make number two a bit simpler. So what I have done here is to summarize the changes.
The biggest change was to allow the wheels to drop out of the inner frames, so that they can be removed wihout dismantling them from the axles. This may seem obvious now, but for some largely forgotten reason I did not do that for number one, probably because of all the different levels involved. All the axles except the leading axle run in plain bearings (the leading axle is in a sleeve bearing again). The driving axle is rigid and held in place using a simple screw jack arrangement that can be seen in the left photograph.
The two centre axles have some vertical float so that the leading and driving wheels always stay in contact with the track, and will be prevented from falling out by means of a wire keeper. Since they should only ever have to come out for painting, that should suffice.
The right photo shows the majority of the inner frame assembly, upside-down. The change here has been to extend the stretchers supporting the boiler outside the frames in order to mount the cylinders, rather than putting the cylinders on separate extensions as before. The frames were slotted where the stretchers intersect them, using a slitting saw of the appropriate width, and when soldered up, the whole assembly is very rigid.
The next big change was the eccentrics. Previously, they were fixed to the wheels, and the eccentric sleeves were each divided and assembled by means of a bolted flange, just like the prototype. The bolts were 16 BA (even that is slightly over scale), and proved to be so fiddly that, once assembled, I did almost anything to avoid having to undo them ever again.
This time, only the inner of the two eccentrics is fixed (using Loctite) to the wheel. The other is loose, and is held in place by means of the crank pin, which passes through a clearance hole in the outer eccentric and screws into the inner. This means that the sleeves can be solid, and simply driop into place on the eccentrics, then everything is secured by the crankpin.
On the inner eccentric, opposite the crankpin hole, is a little pin that locates into a blind hole in the outer eccentric. It is not strictly necessary because the wheel centre keeps the eccentrics in the correct relation to one another, but it was useful during manufacture because I could check the assembly of the valve gear away from the model.
The sleeves were later scribed across the flanges to represent the split line, and dummy bolts from
Scale Hardware added to finish them.
One final detail to aid assembly and disassembly was to make the pin holding the connecting rod in the crosshead screw in, rather than pushed in. That allows the connecting rod to be released entirely, and means that the slidebars, crosshead, and cylinders can be assembled without having the connecting rod flapping about. A slot for a screwdriver was cut on the inside end of the pin, and this is visible in the photo below, which shows a view of the slidebar assembly not normally seen when the boiler is in place. This is a nice little feature, and I intend to use it in future models where there is enough meat in the crosshead to tap a thread.
Previously I remarked on how difficult it was to make cutouts in the boiler to leave space for the carrying wheels, and this new model gave me the chance to do something better. In this case, I filled the lower half of the boiler with a mixture of "liquid lead" (fine lead shot) and epoxy resin. The lead was for traction purposes, and the epoxy was to stiffen up the skin of the boiler so that it could be machined. The photo below shows the boiler set up on the milling machine, in order to make the cutouts. A single point tool was used for this, taking fine cuts. The result was a very satisfactory set of arcs in which the splashers for the carrying wheels could be located and soldered to the boiler.
Incidentally, the idea seems prevalent that liquid lead can be fixed in place using PVA adhesive, and I have even seen it on packets of the stuff that I have bought. Avoid it like the plague. As many have discovered to their cost, there is a reaction between the lead and the PVA causing it to swell and, if it is tight packed, burst its container. My method is to pour in the liquid lead dry, mix up ordinary 2-part epoxy, and spread a layer on top of the lead. It then goes in the oven at about 100°C for an hour. The heat first makes the epoxy mobile so that it penetrates throughout the lead, then causes it to set. If you allow it to set at room temperature, it stays as a layer on top of the lead.