This page aims to give a simple explanation about the way the LTMR signals trains and the past, present and future installations of the system.
The railway is split into several small sections, with entry to each controlled by the signals. Each section is only allowed to contain one train for safety and so the system must be able to detect which sections contain a train. The LTMR uses two methods to detect the position of trains: track circuits and axle counting. Currently, we use 9 track circuits and 1 axle counter.
Track circuits work by applying power through the track to a relay. The current flows through one rail to the relay and back through the other rail. When a train is in that section, its wheels join the two rails together which cause the relay to turn off and the train is detected. The sections are separated by insulated rail joiners and have the power supply at one end and the relay at the other. The rail joiners (fishplates) do not make a reliable connection between rails and so there are bonds wired across each non-insulated joiner, a lot of these on the LTMR are bright red and stand out easily. Since the relay needs to be on to say the section is clear, it is failsafe. If power fails to reach the relay through loose connections, low voltage or even a broken rail then the signal protecting the section will remain red.
Another reason for track circuits to fail is ballast resistance. Ballast resistance is the resistance between the two rails through the ballast, sleepers and other structures between them. If this resistance gets too low, after heavy rain for example, the voltage can drop below the level required to turn the relay on.
Axle counters use a form of wheel detection to count each axle as it enters the section and again as it exits. If the number of axles counted on the exit is equal to the number counted on entry, the section is clear as no axles are in the section.
The LTMR axle counter was made by one of our volunteers for his project while at college. It now serves to detect the single line section between Wesley’s Halt and Ealing End, significantly reducing the length of wiring required to detect that section.
If a signal is held at red when it is clear to go, the likely cause is a faulty track circuit. These were usually caused by loose wires or in a few cases, heat expansion had caused two sections to touch. Most recently, the voltage of one of the longer sections dropped too low after some wet weather made a conductive path between the rails. This section now features a short circuit protection module which removes the voltage drop from the previous resistor and keeps the voltage above the limit.
The LTMR signals are based off London Underground practice and therefore use two aspects, red and green. The signals tell the driver when and where he can move with a red aspect saying stop and a green aspect saying go. There are also repeater signals which show a yellow aspect instead of red. These warn the driver if the next signal is at red so that they can slow the train down in time. In certain areas, shunt signals are used. These tell the driver that they may proceed at caution, prepared to stop short of any obstruction.
The signals control the entry to each section, using the train detection methods to know when they can turn green to let a train pass. The signals also decide whether a train can pass by looking at what other signals are green. For example, if a single line is unoccupied but the signal at the far end is green (allowing a train to enter the single line), then the signal at this end will remain red. This is called interlocking and the LTMR uses relays to do this. The wires to power each signal relay pass through the track detection and other signal relays, physically preventing them turning on when he full range of conditions are not met.
The LTMR mainly uses spring held points. These points use a strong spring to hold the blades in a set direction. The train is able to push past the blades and the spring will pull the blades back to the correct position. We also have two manually operated points (on the mainline), one for the carriage siding and one for the second platform at Ealing End.
The Ealing End point has a basic mechanism which uses a physical stop to prevent the blades moving. It cannot be pushed through. There is a motor in store which was never fitted that we hope to fit in the future when a suitable position has been found. The carriage siding point uses an over-centre sprung mechanism allowing it to be held in either direction by a spring. Pushing through this point will cause the blades to switch to the direction the train has come from. The carriage siding point has a motor which was in use until the end of 2017. The motor is still in situ and we plan to use it when the project progresses further.
Until 2018, all points were detected. The signals would not turn green unless any points that sent the train on a diverging route were detected as being correctly set. Currently, none of the points are detected but they are inspected regularly throughout the day. There are plans to reinstate point detection when funds become available.
The LTMR used to use an N-Style lever frame from the museums collection. This had 11 levers which controlled the signals, carriage shed point and also a few safety functions. In addition, there were three buttons to control three signals in the station area.
The interlocking was made up of around 150 relays, including some heritage. The system was fully interlocked for the parts that were installed but unfortunately, the signals and point at Ealing end were not finished.
Control was purely manual (or “semi-automatic”), requiring a signaller at all times. There was a “mimic panel” which displayed the current state of all signals and track circuits and some other indications for points. The system was in the process of having an automatic mode installed, but this was not finished.
The railway hit a dilemma in late 2017 as it had not made enough money to purchase all of the signalling equipment, much of which was on a long term loan. Unfortunately, one of the major volunteers, and owner of the equipment, moved away and some of the loaned items were removed as they could not be paid for. The main problem was that this included the track circuit relays. The signal drawings were also incomplete, and the installation of the signalling used temporary wiring which resulted in numerous track circuit failures.
The remaining volunteers, plus a few new volunteers did not have the required experience to repair the system and so they decided to start again. The first task was to remove the lever frame to make an emergency carriage shed and dismantle the relay racks. Token working was introduced during this phase without signals
The signals were re-introduced in July 2019 with a temporary relay rack governing the single line on the lawn. This also introduced the axle counter to control entry to Ealing End. A second temporary relay rack has now been installed and is being wired up to control the Depot Approach station area. Both racks work independently of each other, with only the two repeater signals at Depot Approach requiring inputs from Harrisons Crossing.
Control is completely automatic with the exception of two switches which allow the controller to stop trains entering the lawn single line so that trains may safely enter and exit the carriage siding. Station departures are controlled by a TRTS (Train Ready To Start) button for each platform. Once the button is pressed, the rest is automatic.
In the old days, the signaller would manually record each train into a computer based register, stating the engine, platform, train type, destination and time.
Currently in development is an automated train register. Using a microcontroller, the register would record departures using inputs from the relay rack. This is a dumb system as it can’t distinguish whether a train is loaded or empty or whether the train is going for a standard journey, to the siding, or to rescue another train. Trains out of the siding are not recorded either. Using tags and a tag reader, the system should be able to record which engine has departed by reading the tag which passes over. A method to mount the tags is still in development.
The LTMR has some other major projects in store which take up a majority of the budget. Once these are complete, we will look at expanding the temporary system to provide a selection between automatic and manual control. It will also bring in more (lesser used) signals, re-instate the point motor and all points will once again be detected. There is talk about bringing the old N style frame out of storage again but discussions on where to put it are ongoing. There is also consideration being given to a computerised system to replace some (if not all) of the relays.