Train Control

Analogue or Digital

When planning this layout, my biggest dilemma was whether to take the plunge into DCC or to stick with the DC analogue control I had grown up with. Many of the commercially available DCC systems seemed unduly complicated and more suited to the sort of person who was used to flying Battleship Galactica on a games console or who could text 1000 words a minute on a mobile phone. They are also more expensive in comparison to traditional analogue controllers. Some simpler DCC units were becoming available at a more reasonable price but they seemed to have serious limitations on functionality and expandability. The choice between analogue and digital affects the control of not just trains and points and signals, it also affects how the trains’ running and interior lighting is rigged, as well as having a fundamental affect on how the trackwork is sectioned and wired up.

As with everything in life, the cost of change has to be considered for abandoning analogue DC and adopting DCC. It’s an all-or-nothing exercise and everything has to be converted. Even with simple decoders being available at around £10 or less each, it’s still an expensive exercise for a sizable fleet. However, once I started to seriously think about the extra features I wanted such as switchable head and tail lights for the trains and interior lighting for the passenger stock, I realised that the complexity of analogue control was increasing and the cost advantage was disappearing. So I looked again at DCC.

It was at that point that I stumbled across a series of articles about DCC systems on the wonderful Australian-based DCCconcepts website and I was completely won over. The author talks about sound without going into much detail on specific systems but I had decided that the cost was probably way beyond my budget. Besides, none of the steam sounds I have heard are at all realistic and all of the background clanking, hissing, squealing and whistling call for a huge number of functions that need remembering. Even the simple function of turning of the chuff sound when coasting or slowing, which should be a simple thing to automate, has to be done manually. He writes well of the NCE control system and, for loco decoders, those made by TCS. The NCE system includes encoder modules for a variety of input switches which can be mounted in a mimic panel or anywhere else to control accessories independently of the hand controller. But, since the same control (cab) bus is used, all accessories can still be controlled from the handset when desired – truly the best of both worlds!

The NCE PowerCab starter set has a sensibly sized hand-held controller with sensibly sized buttons that can be used in one hand by people of a certain age without recourse to reading glasses or other aids. It suits the way I wish to operate my layout, I believe it represents very good value for money and it is expandable with additional handsets and boosters as my ambitions grow and my budget allows.

I concluded that sticking with analogue DC makes sense for a small personal layout with a limited locomotive fleet and where the whole layout can be seen without moving too far. For a larger layout and certainly for a garden layout, the freedom and flexibility that DCC has to offer makes the switch-over worthwhile. Since I would be starting afresh, I reckoned the overall costs would not be so different.


Perhaps “auxiliaries” would be a better word to describe such essentials as turnouts and signals. For the remote operation of turnouts, signals, and other accessories in the garden under analogue DC, some form of relay control or a local power control system would be essential to avoid lots of heavy wiring. I considered various methods but I concluded that adopting standard DCC accessory decoders would promise to keep life simple.

When I started to explore my options for accessory decoders, I stumbled across a serious problem with the vague way in which the NMRA has defined them. Their standards and recommended practices have many pages defining the behaviour of mobile decoders for locos but are frighteningly brief on accessory decoders. Added to which I believe that more difficulties are probably caused by the many differences between North American and British railway nomenclature.

When an accessory decoder is first powered up at the beginning of an ops session, it resets its outputs to “off”. For my NCE cab this corresponds with the command for “N(on)” and the command “R(off)” turns the accessory output on. Unfortunately, this is the opposite way round from British railway nomenclature.

For a British railway operator or signal engineer, the normal default position for a signal should be “on” (at danger) and for a turnout, “normal” (straight) and this is what is required at the start-of-day-switch-on. So the problem can’t be solved by simply swapping the feeds to the point motor!

I eventually found the answer in the latest version of the MERG accessory decoder. This allows the outputs to be optionally inverted relative to the power-up sense. Thus, at power on, the decoder output goes LOW; on receiving a “R(off)” (turnout reverse or signal off) command, the output goes HIGH; on receiving a “N(on)” (turnout normal or signal on) command, the output goes LOW. This gives me the expected operation with the PowerCab in accordance with British railway conventions.

Power Supply

It is perfectly possible to power a complete railway layout through one large booster and nothing else, but that would not be very sensible. For the same reason that your home electrical system has a consumer unit (fusebox) with several fuses or circuit breakers to subdivide the supply, so it is better to have the layout divided into a number of power districts. Thus, if there is a problem anywhere on the layout, only the affected part shuts down and the fault current can be limited to a safer value.

More here ...


The boasted advantage of long (4-digit) addressing is that you can use the number on the side of the loco as the address, to save having to remember or look up which is what. That may be the case for US railroad practice and a pre-nationalisation layout in the UK but, after 1948, all British locos (apart from the good old Western Region) had five digit numbers and modern multiple units have six digits. So some “interpretation” is necessary. I usually use the first two and last two digits.

And finally ...

An interesting consequence of adopting digital train control is that we have come virtually full circle from the early days. My introduction to the hobby was through Hornby Dublo 3-rail where the direction the locomotive moved was irrespective of which way round it was on the track – when you turned the controller knob “forward” or “normal”, the engine would move off chimney first or vice versa – all very logical and simple. With the advent of 2-rail electrics, “normal” and “reverse” were no longer appropriate and you had to think in terms of “left” and “right” or “clockwise” and “anti-clockwise”, even though the controller perversely still said “normal” and “reverse”. The advent of DCC has simplified all of that. The operator now simply selects which locomotive or unit they wish to drive, “forward” means forward, and off you go!

This is all well and good for a locomotive but “forward” or “reverse” for a multiple unit is often a moot point, so analogue DC control has the clear advantage there. However, when directional headlights are fitted and illuminated, there will be no doubt about the direction the train will move. I have an additional trick: I am fitting cab lights to my diesel and multiple unit cabs and these are selected by F1 for cab 1 and F2 for cab 2. Cab 1 is defined as the leading end. I realise that these function allocations would have been usurped if I had decided to go for decoders equipped with sound.


More to come on my choices and how they are implemented.


Signals and TurnoutsIndexRolling Stock