I really couldn't come up with a better name than Helix Elevator, but this page describes how I designed and built circuits that detect a train's presence in the helix. The actual physical location of the train is shown through a set of vertical banks of LED's on the front of the helix. As part of my third N-scale layout, I had a 10-level helix the outside of which was be covered with 1/8" Masonite board. This meant that for at least five to ten minutes of train travel I would not know exactly where the train was in the helix. I have seen, and operated on, layouts where the owner had installed a closed-circuit television system for monitoring train movement in the helix. While I would have liked to have used that solution as well, I couldn't find space for the television screen. The room was just too small. Also, feedback on the train's position needs to be close to the helix for the interaction to be intuitive. A smaller, but still useful solution had to be found. The idea I came up with was to mount two vertical banks of ten LEDs to the front of the helix as shown in this diagram. One bank monitored the front edge of the helix, while the other monitored the back edge of the helix. The LEDs were positioned approximately at the vertical physical height of the monitored level in the helix. This system allowed multiple trains to be monitored; important, since my helix was double-tracked.
The key piece of technology that made this concept a reality was an electronic circuit. I found several circuits on the Internet and built two of them. The one shown here was the only one I was able to get to work, and also turned out to be the simplest to build. D1 is the infrared LED sending signals to T1, the photo transistor. For as along as light is received by T1, the indicator LED, D3, will remain off. As soon as the signal is broken, i.e. a train passes between D1 and T1, LED D3 turns on and stays on while the train passes through. Note that since I will position the infrared components across both tracks in the helix, I did not know which track was occupied. Also, since there was only one detection point at each head-end of the helix (per level), I could only infer a train's movement by watching LEDs light up and turn off. This was sufficient for what I wanted. Obviously more complex systems could be developed to monitor train travel direction and which track is being signaled. By the way, I had considered implementing a 7-segment LED display (like those found in digital clocks) to numerically report on which level a detection occurred, but those circuits are complex to build. Also, because the display can only show one number at a time, if more than one level was triggered (i.e. two or more trains are in the helix at the same time at different levels), the highest number would win out. I had a 10-level helix, which meant there were 10 levels on both the front and rear of the helix that were monitored. I therefore needed 20 independent circuits to cover the 2-times-10 detection points. The table below shows the total parts list. Note: I have listed the Radio Shack parts for your convenience, but I typically buy my components from Digi-Key. The circuits shown below were built with the parts from Digi-Key.
|R1,R4||40||680 ohm resistor 1/8W||11344827||680QBK-ND|
|R3||20||4.7k ohm resistor 1/8W||271-1330||4.7KQBK-ND|
|R2||20||10k ohm resistor 1/8W||271-1335||10KQBK-ND|
|D1||20||Infrared emitting LED (long lead is "+")||276-142||LN66A-ND|
|D3||20||LED red, 3mm in socket (long = "+")||276-068||67-1147-ND|
|T1||20||Photo transistor (short lead is "+")||276-142||PN268-NC|
|20||Printed circuit board||276-149||6008CA-ND|
|?||Wire and solder|
I decided to build a simple circuit that could help me test the infrared components. The beauty of the circuit shown here is that it also detects signals from remote controls of your television or stereo! Aim your standard remote control at the circuit (at Q1) and you can verify that the circuit itself is working (LED D1 will light up when you press a button on your remote control). The design came from a web site that is now gone. I constructed mine using a 2N2222 transistor for Q2 (because that was what I had on hand), and for Q1 I used the photo-transistor shown in the parts list above. Hook it up to a 9V battery and you're set!
My next step was to build printed circuit boards for the 20 circuits that I had to build. Although I have done this plenty in the past, the thought of etching that many boards wasn't too appealing. Then one day I was thumbing through the Digi-Key catalog when I stumbled across Surfboards by Capital Advanced. These brilliant little inventions are a big time saver. Cost-wise they are well worth it, when you compare the cost of PC board, etchant, and time spent etching. I decided that the 8-pin version (part number 6008) was just right for me, based on the prototype circuit that I had built.
The component layout diagram shows, in dark gray, where the foil sections are and, in red, where the components are to be placed. Note that D1, T1, and D3 were mounted off-board onto the helix itself. The 8 pins on the board are used to make the external connections to the infrared components, the reporting LED, and the power supply.
Here is a close-up photo of one of these circuits built.
The photo below shows the proof - I really did build 20 of these. The troops are ready to march...
For the actual installation of the circuits, the first step was figuring out how to install the infrared LED and corresponding photo transistor to the helix levels for accurate detecting. Furthermore, many web sites recommend that you place the two infrared components at an angle to the track so that the gaps in between rolling stock don't cause the LEDs to flicker (I verified that this was indeed an issue). My solution was to cut two pieces of 1/4" basswood stock to length so that they match the height of the helix levels.
I then marked off where the helix levels were on the basswood, and transferred two marks that identified where the LED and the photo transistor are to be located. The basswood is going to be glued to the sides of the luan board (sub-roadbed used in the helix) at each level, and the IR components will be installed at a 45-degree angle to the track. Be sure to mark what is the top of the basswood stock otherwise they might not line up in the future. I also marked which one will hold the LED and which one will hold the photo transistor since the LED and transistor components are of different sizes.
Using various files, I carved a 45-degree cut at each location. This angle matches the angle to which the IR components are to be mounted with respect to the track. Filing this takes some time and is tiring, but worth it. An advantage of this installation method is that the back of the photo transistor is automatically protected from stray light.
Next I drilled two holes in each carved-out section so that the components' leads could be fed through the wood. For the LEDs leads I drilled # 58 holes and for the photo transistor's leads I used a # 66 drill.
It is then just a matter of slipping the components into each slot.
Here is a photo of all the components installed. I made sure that all the positive leads for each component are on the same side. This will make it easier to install the wiring later. I built another set of these basswood sticks for the other end of the helix.
And now for the exciting step - installing the basswood strips with their components to the helix. I first clamped a matching pair to the helix and used the proof-of-concept circuit that I had built to verify that the components lined up. It turns out that these components don't have to line up perfectly. Also, I did the test with the overhead fluorescent lamps on, and the system worked flawlessly (at least, after I hooked up the wires correctly!). When the helix is fully enclosed, the system will be even better. The photo shows the LED strip being glued in place (I used yellow carpenters glue).
Another after-installation photo. Although the circuit boards and the wiring still need to be installed, a test run showed that the system worked perfectly. The IR components are quite forgiving.
Below is a close-up photo (as best as I could get it in this dark area) of the two components. Note that the "photo LED" text in the photo should state "infrared LED".
The next step was to install the 20 circuits to the helix structure. I decided to simply glue them to the vertical posts of the helix. They are glued with Liquid Nails for Foam (I had a new tube handy!). The glue is very tacky and sets in a few minutes, so there is no need to hold or clamp the circuits.
To connect all these circuits to the LEDs on the front of the helix, I decided to use ribbon cable. The cost is a bit higher than regular wire, but the professional-look of nicely routed wires is worth it. It actually made installation easier, because I glued the cable to the helix structure using the same glue as mentioned above. The inset photo shows the cable as it came from Digikey (16-wire). The photo shows the cables for the LEDs in the front that are connected to the detection circuits in the rear of the helix. The cables are nicely routed through the structure of the helix base.
The final photo of the naked helix. All the wires on the front have been prepped. The helix is ready to receive its Masonite board cover.
After installing the Masonite cover and attaching the display LEDs to this cover, this photo shows one LED being lit up by a train moving through the lowest rung in the helix.
This last photo shows the helix cover having been painted. You can see the banks of LEDs on the helix' front (lower-left in the photo). The system worked flawlessly. It was comforting to see the LED's light up as trains moved through the helix. This became even more important later on when I placed a huge mountain on top of this helix, completely hiding the interior.