Rob Paisley has a lot of model railroad circuits on his website, complete with detailed diagrams and notes for electronics beginners. For some projects, he also provides a parts list with DigiKey part numbers, and for some, he sells kits that include printed circuit boards.
I’m going to try his Infrared Proximity Detector Circuit to use with my signal system. I’ve looked at a few other block occupancy detection (BOD) methods, but the others have what I see as downfalls. With DCC, you can use a system that detects the voltage draw from the track power, which tells the system where on the line the locomotive is. The problem with this, for me, is that the rest of the train doesn’t normally draw current, and so in order to make the rest of the train trigger this type of occupancy detection, you have to install power-pickup wheels on at least some of the cars, with a resistor between the two wheels. That sounds problematic and like a lot of unnecessary hassle.
With an IR proximity detector circuit, an IR LED, hidden inside a tube below the track, in between two ties, shines invisible light straight up. An IR receiver is located under the track right next to the IR LED, but the position of the LED and the tube around it prevent the sensor from seeing the IR light, until a train comes along and the IR light gets reflected back down to the sensor. The circuit can also be used with an across-the-tracks installation, where the presence of a train is detected not by the reflection of the IR light, but by the lack of it, as the train blocks the beam from passing across the track. Rob explains on his website that this installation type is necessary in tunnels, where the low ceiling would reflect the IR light back down to the detector, causing a false occupied condition.
This will be used in conjunction with this Three-Light Model Railroad Signal Driver. One question I have right now is, what happens if the train is short enough that the end of it clears the detector at the beginning of the block before the front of the train reaches the next detector. At that point, no detectors would see a train, but the block is still occupied, so does the signal driver circuit hold the occupied status until the next detector is triggered? I’ll have to look into this.
UPDATE: Rob confirmed via email that a short train located between sensors would result in a clear signal. He sent me a schematic that shows how to connect multiple sensors in parallel, so that a single block can have up to four sensors. That sounded like a lot of work at first, and I thought maybe I should instead go with voltage-sensing BOD circuits for DCC. I don’t have a DCC system yet, nor any DCC locomotives, so I’m wondering if I can still build a prototype, using an AC transformer to apply 14.5-15 volts, and make some resistor wheel sets for the few cheap freight cars I have. At this point, I think I’m still going to at least experiment with the IR sensor circuit.
UPDATE 2: After prototyping these circuits on breadboards, I feel pretty confidant that these circuits will work well for me. I’ll definitely need more than one detector per block, possibly at least three for each (beginning, middle, and end of block). I need to do more testing with the breadboards, since some of the detectors stopped working reliably after a while, possibly because I might have accidentally shorted/damaged some components. I’m going to try to get them working again, and will then try adding the circuits that make the LEDs turn on and off slightly more gradually, like incandescent lights.