Home | Design | Construction | Track Laying | Control & Electronics | Rolling Stock | Operations | Links | Contact
An often-overlooked issue of model railway wiring is that it needs to be maintainable over the lifetime of the railway. Ideally, if it is well enough installed no attention will be necessary. But if a wider does suddenly appear hanging out from the bottom, where did it come from? Will it be possible to work it out?
It is widely recognised that DCC does NOT eliminate track wiring - it just changes it. My railway needs block detection, so I've got a lot of track power feeds just as would be needed for "cab" control. Using accessory decoders located out around the railway cuts down on wiring lengths to points and signals, but it doesn't remove the wiring completely.
I wanted to have a certain "modularity" to the wiring, so that wiring lengths were not unavoidably long. With that in mind I looked at the locations of track feeds, signals and points and planned where the accessory decoders, block detectors & signal controllers. These were then clustered into several areas, where hinged panels have been provided for ease of access.
The CAD design allows a lot of the wiring to be captured into the
design, at least as far as knowing which "block detection"
section each piece of track is connected to and which accessory
numbers are associated with each point & signal. This allows the
source of most wires to be found very quickly.
DCC Boosters
The Digitrax boosters can each provide up to 5A of track current. That is sufficient to drive around 10 "N" gauge locos, and on the face of it would be sufficient on its own. In practice there are two reasons why this isn't the case:
-
The power needs to be broken up into zones, so that a short in one zone won't kill operations in another;
- Accessory decoders always need "live" power.
My approach has been to plan to have three boosters:
-
one to provide the command station, and power to one end of the layout;
- one to provide power to the other end of the layout;
- one to provide power exclusively to accessory decoders.
Beyond that, I have planned to use the Digitrax PM42 to split the
track power into a number of "zones" each of which has its
own current limit. This allows the effects of shorts to be localised.
In particular, track areas where trains are driven by humans are
isolated from areas where automatic control takes place so that a
train driven onto a wrongly set point won't trip the whole area.
Point Control
The point motors will all be DCC controlled. I plan to use SEEP point motors, driven by CML Electronics DAC10 accessory decoders. Both are reliable, well known systems. The point frogs are all electrofrog, and need to have switched power. The SEEP PM1 motor has a built in switch for this purpose. At crossovers, I will need to use additional relays for crossover frog power.
Signalling
Signals will be controlled automatically, using the CML Electronics SIGM20 controllers. These are automatic control units: the signal states are set automatically by train movements, and by setting the pointwork. If the signal ahead is red- the next one goes amber; if the block ahead is red the signal goes red; if a point is set against the track, the signal will go red. All of this is automatic - no user intervention.
Block Detection
To allow automatic control, the train positions need to be known. By far the best way to do this is with track current detection. Optical sensors are great for stopping a signal at a specific location, but not so good at detecting a whole track zone that may be many times bigger than a train.
Forward to LocoNet Architecture
Page Last Modified on: