Detailing the boiler - Below the footplate

Reverser mechanism

First, a bit of explanation.

The reverser is the mechanism used to control the direction of travel of the locomotive and the amount of steam admitted the cylinders. Think of it as the accelerator pedal in a car.

On Dutch steam locomotives, the reverser is generally of a fairly simple design: a lever in the driver’s cab, with notches to lock the cylinder cut-off, and a linkage to the valve gear mechanism.

The reverser is so simple that it is rarely shown in full in illustrations: a lever in the driver’s cab, a rod running alongside the boiler and a cantilever attached to the valve gear.

With the Garratt locomotive

• to begin with, two engines must be controlled instead of one;
• a greater distance must be bridged. After all, the engines in the power units are much further away from the driver’s cab;
• the entire mechanism has many more pivot points because the rods must be articulated to rotate with the bogies.

This quickly makes the reverser on a Garratt very heavy to operate. It is likely that Hanomag opted for a steam-powered gear lever for this very reason, a sort of power steering, in other words. To the best of my knowledge, the LTM 51 was the only Dutch steam locomotive to have a steam-powered reverser.

From: ‘De Ingenieur’, 26 August 1932, Fig. 10.

The essence of the steam powered reverser lies in the addition of two cylinders to the reverser mechanism. The steam cylinder is responsible for actually moving the entire linkage system to the power units. However, a steam cylinder alone would shift due to the vibration of the locomotive. That is why a ‘locking’ cylinder was also added.

The whole system worked as follows.

• The driver unlocked the mechanism and moved the gear lever. The reverser acted via a linkage on the valves of both cylinders.
• On the one hand, steam was admitted to the steam cylinder so that force was exerted on the reverser rods towards the power units in the desired direction.
• At the same time, the valve in the ‘locking’ cylinder was opened. That cylinder contained glycerine, and opening the valve allowed glycerine to flow from one side of the piston to the other.
• The steam cylinder was filled until the desired position was achieved.
• The steam was then shut off and the mechanism was locked by closing the valve in the ‘locking’ cylinder.

As the reverser mechanism is located beneath the footplate, it is usually difficult to see in the few photographs we have of the LTM 51 (click on the image to enlarge).

Here is a diagram of how it looks (click on the image to enlarge).

The parts of the reverser are shown in yellow and red. The parts I have replicated are shown in yellow. The parts I have omitted are shown in red. After all, you can’t replicate everything in 1:87 scale.

Building this mechanism has been the biggest challenge of the locomotive’s construction so far. Fortunately, I’ve already had the chance to practise the basics of all the techniques involved when making other parts, including the four small forks I need. But here, everything has to be scaled down a notch. I’ve never done it before, so I can do it!

I’ll start by making a head of it with the cylinders. I know myself: if I don’t get started, I’ll keep mulling over it endlessly. From the general layout drawing of the LTM 51, I derived

• The length and diameter of the cylinders.
• The distance from the cylinders to the base plate.

In Fusion 360, I created a sketch to determine what dimensions the square valve housing above the cylinder would need to have in order to achieve the correct distance to the base plate.

To create two small blocks with a round recess into which the cylinder would fit, I soldered two strips of brass together. The required 2.5 mm hole was drilled slightly off-centre because, as the drawing shows, I didn’t want a semi-circular recess but only part of one. The blocks were then milled to length. Sawing and filing would, of course, also work. It might even be quicker, as handling such small parts in the milling machine isn’t exactly easy.

The cylinders themselves were turned from 3 mm brass rod. The cylinders and valve housing were then soldered together using 240C solder.

Next, the clevises were made which are attached to the ends of the actuator rods.

Inclusing there connecting pins.

To give you an idea of the size, I’ve placed a matchstick behind it. The connecting pins have a diameter of 0.45 mm. The head has a diameter of 0.7 mm and a thickness of 0.1 mm.

Fitting the cylinders under the base plate.

That turned out to be a fiddly job, as the cylinders need to be parallel to the frame and also at a specific distance from it. To achieve this, I milled a small aluminium plate to the required thickness. That made it easy to gauge the correct distance and also keeps the cylinder parallel. A small scratch on the frame plate indicated where the first cylinder should go. When fitting the second cylinder, I almost forgot to insert the connecting rod first. I remembered just in time.

The left-hand cylinder is slightly askew in this photo. So, based on this photo, I’ve corrected that.

To shape the end of the long rods towards the bogies into a flat head, I first annealed the ends. This makes the brass very soft. I then flattened the end in a machine vice. I cut a small strip of brass of the desired thickness to fit into the vice, so that the jaws would stop closing at the right moment. This ensures a consistent result.

That’s how it all fits together on a Garratt.

The rods were held in place with BluTack whilst they were being soldered.

The observant reader will notice that I didn’t use the connecting pins I mentioned earlier. It all became too fiddly for me.

Another job done.

What’s still nagging at me is that I haven’t modelled the rods leading to the driver’s cab or the supports that hold the long rods in place. I might do that after all. But I’m done for now.

Air tank

The brake air reservoir is positioned across the frame, with its ends visible beneath the boiler cradle. In itself, it is a simple part to make.

A simple turning job.

A tight fit.

When I was fitting the steam pipes, I didn’t give a second thought to how the air tank would fit. Well, I got away with it in the end. Once it was placed on its feet, it turned out to be wedged against the steam pipes. It fit but only just.

Pipes on the left side

Two pipes had to be fitted on the left. The yellow one is the air pipe from the pump to the tank mentioned earlier. I’m not sure what the red pipe is for.

As these pipes are long, they cannot be left suspended in the air at their ends. So I had to create some mounting points for them. I quickly drew up a sketch in Fusion.

Turning on the lathe isn’t really a problem, but drilling a 0.5 mm hole through it afterwards is a challenge. So I changed the usual order of operations: first I drilled a row of holes in the parent material and then did the turning work from there.

The drilled hole served as the reference point for all subsequent machining operations, so prior to turning, the tool was set to zero relative to a thread passed through the hole.

Before long, I had my first long bracket, and shortly afterwards three short ones.

The photo on the left shows them side by side for comparison. The three short ones aren’t quite the same size, but the differences are only about a tenth of a millimetre or so, which is a lot at this scale.

On the far right of the photo is a commercially available handrail knob by Alan Gibson.

Well, a few holes strewn across the appropriate places, a bit of soldering, and another job was done.

Lever for the ash pan

The ash pan is opened and closed using a lever operated from the cab via a rod.

Once the disc was prted off, I drilled a hole in the base material, turned the disc over and inserted it back into the base material with the spindle. A bit of of CA glue secured everything in place. I was then able to continue milling out the lever.

Ready for assembly.

The ash pan lever and its pull rod are in place.

The side hatches of the ash pan have now been sanded smooth, and the injector overflow tube has also been fitted.