This article on track circuiting (ie train on-line detection) is from a non-electronics engineer and my brush with the ‘black art’. It's a tricky subject and one which will involve branching out into electronics, a topic I have avoided for ages, but which now needs to be tackled. My Peacehaven garden railway runs quite nicely (thank you) and incorporates signalling, interlocking and block working all using electro-mechanical devices. However, the need arose to know the whereabouts of trains on the track. At one location in particular the line goes out of sight of the driver and I figured it would be nice to know when a train is occupying that section. On the layout, trains are supplied by straightforward analogue DC controllers with section switches dividing up the track power supply to particular lines on the layout.

I quickly realised the provision of track circuits was not going to be straightforward. On real railways fail-safe track circuiting has been in existence since the 1870s. In prototype practice current fed to the rails is shorted out through the wheel sets when a train is in the section thus operating a relay and giving a track occupied indication.

On the face of it this looked to be impossible with a model lines as those very rails are already used to supply current to the locomotive to drive the train. How then, could it be done? I traced the design of track circuits for model railways back to the 1950s when Linn Westcott, an American modeller, published a design. This was developed over the years and improvements were introduced and published by Roger Amos in the 1980s. I found details in his book Practical Electronics for Railway Modellers 2 which includes a description and circuit diagram for something called TEKTOR.

I am no electronics expert and I was later told that there are now much more elegant ways of achieving the same result. However, before this I had decided to try assembling a TEKTOR circuit and see whether I could make it work. Although the indication from such a system could be used for a number of applications, such as automatic signalling, I merely wanted to light LEDs on the mimic panel, showing the locations of trains.

Apparently, a software programme for component layout design is now readily available on the internet but I was unaware of this at the time.

Components were soldered to the strip board in turn, taking care with the transistors and diodes by arranging a heat sink to avoid damaging the items during soldering. This comprised a small crocodile clip placed on the leads. After each soldered joint was completed, I used a multimeter to check that no short circuit had been created between adjacent tracks on the strip board. Once, I forgot to remove the heat sink clip leading to all sorts of wrong conclusions as a result of this check!

The next step after assembly was to connect the circuit to a length of test track to see if it worked. It didn’t! Fault finding took quite a time. My limited electronics knowledge was supplemented by help from people who do know what they are doing. I was advised where to poke the multimeter test probes and what readings to look for It turned out a fault was a defective T4 transistor. I had tested this component before fitting and could only assume my heat sink arrangement had been inadequate and the item suffered heat damage. After soldering in a replacement transistor, more carefully this time, I still had no ‘track occupied’ indication. In this case, it was simply my ignorance about how transistors work which was the problem. I had failed to appreciate that a ‘track occupied’ state would cause T4 to conduct between wires 109 and 110. Once this was realised I was able to re-arrange the LED connection and hey-presto, a working module. The next stage was installation on the Peacehaven layout. By this time I had assembled two modules and decided to install them on track sections D and E, the latter being the outside line going off down the garden where the most benefit would be had from knowing if the section is occupied by a train. With D installed all went well. But on track section E the LED showed line occupied all the time even when it wasn’t. No amount of adjustment of the potentiometer would improve matters. At this point I wondered whether I had reached the design limit of TEKTOR and so decided to try to find a different solution.

Enter MERG, the Model Electronic Railway Group which I was surprised to find had been in existence for over 25 years. I was alerted to the benefits by a GOG member. MERG offers knowledge and experience on all matter of model railway electronic subjects. They also, most usefully, design and sell a range of kits for every imaginable application including one called ToTI-2 (Train on Track Indication). Each ToTI-2 module provides indication for two separate track sections, so I purchased two kits enabling the eventual application to four circuits. I also bought the MERG 12V power supply kit which provides isolated +5V and –5V supplies and is recommended for this application.

At this point it should be mentioned that ToTI-2 can be installed with the train feed coming from either of two main controllers. This is a system I use on the Peacehaven line so it would be likely to come in useful. The kits come with a specially designed printed circuit board, all the components required and a very comprehensive set of instructions. The cost is quite modest. Even a nonelectronic person can achieve a good result providing the kit is assembled with care. Video presentations of the correct soldering and other assembly techniques are readily available on YouTube and I studied several of these.

Both TEKTOR and ToTI-2 operate by passing a small electric current through the locomotive motor. Coach and wagon wheels are normally insulated and so the presence of these vehicles cannot be detected and thus the rear of a train is not ‘seen’ by either type of device after the locomotive has left the section. The solution is to modify one or more wheelsets on coaches or wagons likely to be marshalled at the rear of a train. Goods brake vans come into this category, coaches incorporating a brake or luggage compartment and a small number of other passenger vehicles which can be marshalled at the opposite end of coaching stock rakes. On the Peacehaven branch particular coaches are used at the rear of trains and as they are fitted with tail lamp mounting magnets they are easily identified. The wheelset modification is carried out by fitting a 10k ohm resistor between the wheels on one or two of the axles. Some of my coaches and EMUs are fitted with pick-ups and LED head and taillights. I found that with these the current passing through the lighting circuit is sufficient to enable them to be detected.

The outcome of all this development is that there are now six track circuits installed at Peacehaven, providing a visual indication of the presence of a train on the signal box diagram. In some ways this is no more than a ‘nice to have’ facility in that it makes the signal box look more realistic but there is a practical benefit. The driver of the train has a positive indication when a train leaves the section under control and attention can then be directed to other train movements. Of the two track circuit types tried, both work in a satisfactory way. The TEKTOR is sensitive to the calibration of the variable potentiometer. The ToTI-2, on the other hand, works more consistently and being a relatively modern design, is more stable and seems better able to cope with the variable conditions.

The TEKTOR circuit diagram is published with the permission of the designer Roger Amos. It was described in his book Practical Railway Electronics 2, publisher Patrick Stephens. Their successors, Haynes Publishing have kindly allowed the circuit diagrams to be re-used here. The ToTI-2 circuit diagram is published with the permission of the designer Ian Hart. The MERG ToTI-2 schematic used here is published by MERG under the Creative Commons Attribution-NonCommercial-ShareAlike licence. We are grateful to all for these permissions.