09/04/08: Isn’t Technology Wonderful?

The cuckoo in the nest - click for a bigger picture

Most of my limited modelling time recently has been directed towards a layout being built for a forthcoming book. It also reduces the available working space in the garage to a little less than none at all. However I have started to revisit Redhill's baseboards again. As you may recall I adopted a modified version of one of Iain Rice's ideas to reduce the drumming noise made by moving vehicles and trains. The key ingredient is a couple of layers of foam camping mat between the baseboard surface and the frame. While I have no doubt that it works for Mr. Rice I have been having problems - every time I disconnect two boards and join them back together the road surfaces at the join end up at different heights and it takes a modicum of tweaking to get them lined up again. It is probably down to my poor carpentry skills, misinterpretation of the design or sheer ham-fistedness but up with this I cannot put, so, starting with board 6, I am putting in great big lumps of wood to support the road surface at the all-important joins. Whilst I am at it I am adding cross-members - I can do this safely as I now know where the junction mechanisms are. These are ply/timber/ply sandwiches like the outer frame and have more lumps of wood supporting the roadway.

Boards 6, 7 and 8 are first up for the work as they are due to appear at the RMweb member's day at the Chasewater Railway on July 26th (details can be found here. It seems an appropriate venue for a bit of Redhill's first public appearance!

With the ability to actually run the occasional vehicle around a reasonable circuit I have restarted my experiments with RFID vehicle identification. With a CORE-12 (used to be ID-12) reader under the baseboard connected to a PC and tags on the vehicles I was happy to see that I could identify passing vehicles with waht appeared to be 100% reliability.

Testing the first servo connection from the PC.

Bouyed up by this success I started to look at computer control of the junctions - the next ingredient of the automation scheme. When I built the control panel (see Going Dotty I was intending to use the RPC system available to MERG members - however a new system called CBUS looks promising, so I am holding fire on the computer interface proper. As a temporary solution I decided to try and use the PC's parallel (printer) port to drive the necesary junction mechanisms on boards 6 and 7. The concept is simple: the printer port is connected via opto-couplers to the inputs of the junction servos. Output an ASCII character to the printer port and the servos will be set accordingly (ASCII character 0 = 00000000 - all off, ASCII character 64 = 01000000 - one servo on, ASCII character 255 = 11111111 - all on, and so on). Of course it isn't that easy. Building the circuit was simple - I picked up a four channel opto-coupler from Maplin on the way home (just in time - they no longer stock them) and wired up to a 25-way D connector on one side and 4 servo inputs on the other. Failure - no matter what I did I couldn't get the output to change. Needless to say I was suffering from an overhelful operating system and needed to install a special driver to actually be able to control the printer port; after that I could switch outputs to my heart's content. For anyone interested in cheap and dity computer control (albeit with only 8 outputs) you can find lots of information (including the pin-out for the printer port) here and a Windows driver to control the port here - I'm using Windows XP and Visual Basic 5 - if you are on something different you may have to make your own arrangements. All I need to do now is write the program to link the input and output in a useful way!

Previous update :: Next update