Article of the Month

Layout Control - an alternative way.

It is inevitable that I will haul the PC into this, and my beloved Dell portable is ready for duty when that time comes ...... but now for something completely different !!

Many modellers will recall the Reverend Peter Denny and will recall a story about a device that he produced called the Automatic Crispin that took the place of his operator - although I only recall the story and not the working details of his device, I have thought of another idea that may work using modern components, as long as the trackwork has section and turnout detectors, and you run by a strict timetable.

Our amazing low-tech system uses standard sized cards such as those found in those little card-file boxes that you can buy at all good stationery shops - each card will perform an individual move in the timetable, and some work with a hole-punch is required to get the desired affect.
Anyone who knows of the original card-reading methods of computing will understand that shining a light on to a card with holes punched in it, will only allow light to pass through in certain places on to sensors that define a computer instruction - now this is our method of operation !!

Now if we have a box full of punched cards and a clever card reader gizmo, then this can deal with all of our route selection requirements while we enjoy the trains.
Each punched card has a written instruction and order number at the top, and holes punched at the bottom.

The diagram below shows the basic design of the card reader frame and the sensor array.

There are 18 sensors shown in the example although there may be many more than this depending on the layout size - each sensor is an LED on one side and a light detector on the other side (these can be bought as complete units), and our example is permitting circuit connections for sensors 1, 3, 8, 9 and 13 for card number 27. Every card has holes in different places and each sensor is related to a turnout or a power section and whether it is straight/curved or on/off respectively.
The card reader panel will have a red/green light for 'Turnout Set' status, a Turnout Setting button and another red/green for 'Card Complete' status.

Initially, the card reader will display two red lights (if the turnouts are different to the previous card) as the turnouts are wrong, and the track detectors suggest that the wrong sections contain trains or stock.
A push on the Turnout Setting button will cause the turnout switches to be compared with the card settings, and turnouts (and sections) will be changed to match the card setting until the 'Turnout Set' light goes green denoting all turnouts set correctly.

A bit of train driving will take place, although only to the extent of the set route - remembering that the sections have been set automatically, and the signals will match these settings if logically related to the sections, with the sections being shut off behind the train as it passes through - when all of the sections have been passed through and the on/off status of the sections matches the 'Card Complete' punched holes of the card, the 'Card Complete' light goes green and you're ready for the next card !

In a nut-shell, that is all there is to it.
There is no way of altering the timetable to any great extent, unless you exactly replicate the stock and section positions at the start and end of a small group of cards, so that you can use or leave these out of any sequence thereby changing the traffic pattern in some way.

There is also scope to feed sensor information without resort to a box full of punched cards by using a reading system similar to those fair-ground organs that took their info from a large continuous roll of punched paper - possibly moving a step further to automation by using stepper motors to advance the punched roll by one step whenever the end of a previous move completes successfully ....... could provide a use for that decrepit old 9-pin dot matrix printer that you can't get a new printer head for, but is working otherwise - a bit of electrical work near the Linefeed button will achieve the auto-advance mechanism.

Next month, we will take a look at a Parallel Port interface to introduce the concept of connecting up the PC, and take our first steps in getting the PC to talk to the track (and hopefully, the track will talk back as well !!)

Return to: Articles Index

February 2001