Traffic Lights Part II - Traffic Light Controller
By Ty’s Model Railroad - 8/08/2012 10:39:00 PM
A note about this post: I do not take any credit for the design of this circuit. I instead turn the spotlight on Rob Paisley and his website, who has without a doubt spent an unimaginable amount of time, effort, and talent developing dozens of model railroad circuits that ultimately make our layouts achieve things we never thought possible. The intent of this post is to only describe my experience of applying and constructing Rob’s designs on my own layout. Please visit Rob’s website, which has a wealth of wiring information and useful model railroad circuitry.
-Ty
If there is one thing I've learned from building my layout thus far, it’s
that the most time consuming, frustrating, problematic, and complex projects
usually are the ones that are taken for granted. Often, these fade into the
background and go mostly unnoticed but play such a huge role in making your
model something extra-special. My traffic lights were no exception to this.
Now, I’m not exactly saying my traffic lights go unnoticed; they are
definitely a nice feature. It's what controls those tiny little LED lights
that I'm referring to. I myself took for granted the electrical complexity
behind making all the lights synchronized, especially as I was building the
entire system completely from scratch. I know commercial controllers are
available from $50.00 to several hundred dollars, but doing it myself was a
challenge I was up for.
Last year while searching for methods to synchronize model traffic lights,
I came across Rob Paisley’s 20 output sequencing circuit. Rob had designed a circuit that essentially created 20 separate outputs
that progressed one at a time from 1-20 in a continuous loop. When combined
with a novel lighting circuit, synchronization of an entire intersection in
both directions is accomplished within the 20 output steps, then repeats. I
am going to refrain myself from even trying to go into detail on the
particulars of this circuit, as Rob Paisley has already explained it in
depth on his website. Rob’s site includes detailed diagrams and parts lists,
as well as in-depth explanations of how the circuits work. Rob also offers
commercially built circuit boards and kits for this and other circuits he’s
designed, many for model railroad use. Nevertheless, I was determined to
build it myself.
After placing my parts order with Mouser Electronics (using the parts list
off of Rob’s website), I went to work putting together my own wiring
schematic for this circuit. I made my own schematic for 3 reasons. First was
because I wanted a plan of the entire circuit, from the controller right to
the lights on the layout. The second reason was to plan the actual layout of
the circuit board. I drew the schematic in a way that I could literally plan
exactly where every connection and lead would go when building the actual
circuit, ensuring that everything was spaced correctly so I didn’t run out
of room on my PC board. The third and final reason was to better understand
how the circuit itself worked. Even though Rob explains the function and how
this circuit works in detail, if you do not have a good understanding of how
it work, then it will be a lot more difficult to build and even harder to
troubleshoot. I am no electronics expert myself, so I spent a lot of time
researching each component and IC, finding and reading the datasheet for
each one as Rob recommends on his website.
Once all of my components arrived from Mouser, I purchased a 2200-hole PC
board from an electronics store and eagerly went to work putting it
together. I used my schematic almost exclusively to construct the circuit
on the PC board, marking off each completed component and section of the
circuit with a highlighter to ensure I didn’t get lost or miss a
connection, or worse, make a wrong connection. Double, triple, and
quadruple checking my work against my schematic as well as Rob’s original
wiring schematic almost guaranteed no major errors were made.
I used my soldering iron to make all the connections on the back of the
PC board, and used bare steel wire for the leads between connections.
Special consideration needs to be made for spacing of connections that
pass over each other, which requires the steel lead to pass up through the
PC board, over the existing connections below, and then back down through
the PC board to its intended location. Care must also be taken when
soldering connections that are right beside each other, as it is easy to
unintentionally solder two separate connections together. There were also
several spots on the PC board where the soldered connections were shorting
out on each other because they were so close. To resolve this, I carefully
used a razor blade to chisel out a space in between each connection,
ensuring the connections were no longer touching.
You will probably notice when comparing my schematic to Rob’s schematics
that the LED traffic light portion of my circuit utilizes PNP transistors
to control the LED lights, whereas Rob’s examples show the LEDs connected
directly to the 20 outputs. The reason for this is because the 20 outputs
of the circuit are LOW (negative). To control the red, yellow, and green
LEDs of the traffic lights, the LEDs need to be supplied with a common
positive (+) current, connected to the anode of each LED. A separate
negative (-) ‘controller’ lead needs to be connected to the cathode of
each LED, and then to each output of the controller. However, I built my
traffic lights the opposite way, with a common negative and separate
positive controlling wires. Thus, my lights could only be controlled by
applying separate POSITVE current connections to each LED.
To get around this issue, I applied Rob’s wiring schematic where he
explains how to use the 20 output circuit to control high-current bulbs by
utilizing PNP transistors as switches. In other words, instead of directly
controlling the LED lights by having them directly connected to the 20
outputs of the controller, I used the 20 outputs to control the PNP
transistors. The transistors act like switches, either allowing or
stopping the separate positive, high-current 12V flowing through to the
bulbs. The transistor’s switching capability is controlled by its base
terminal, which is connected to the 20 outputs of the controller. I then
simply replaced the high current bulbs with resistors and my LED traffic
lights.
Once the circuit was completed, I needed to make the connections from the
controller to 2 – 10 position terminal strips, which I would then later
connect each traffic light to. To do this, I first cut a piece of tempered
hardboard, to which I attached the 2 terminal strips along the bottom edge.
I used ¾” brass flat-head machine bots applied from the back of the
hardboard to fasten the terminal strips, allowing the hardboard base to lay
flat. I also installed an additional 4 bolts through the hardboard base to
support the circuit board, allowing it to be secured without having its
bottom circuitry come in contact with the hardboard base.
The final step was to connect each positive output on the circuit board to
the screws on the terminal strips. I used high-quality phone cable to do
this, which is convenient because most telephone cable contains a red,
green, yellow, and black wire, making it easy to colour-code each terminal
based on what colour LED will be connected to it. I have a total of 6
traffic lights on my layout, so a total of 18 separate LEDs, 6 of each
colour, and one common negative, so a total of 19 connections. The first 9
connections will control 3 complete traffic lights in a north-south
direction, and the following 9 will control the other 3 complete traffic
lights in an east-west direction.
After I had attached everything to the hardboard base, I installed the
entire module unit under my layout. The circuit requires a 12 volt DC
power source, so I connected it to a 12 volt terminal on a previously
installed power terminal strip. I had already installed my actual
traffic lights on my layout’s intersections, running the wires to the
underside of my layout through 1/8” pilot holes. I utilized telephone
cord again here to connect the traffic lights to the terminals on the
controller module. After connecting each traffic light to its
corresponding power terminal on the control module, all that was left
was to test it. And just like that, I now had working traffic lights on
my layout!
Well, it wasn’t really “just like that.” I spent days and countless
hours pulling my hair out for over a week, testing and trying to
locate small short circuits that arose over and over again on both
the control module and in the traffic light wiring itself. With so
many small connections so close to one another, it’s pretty much
impossible to get it right 100% the first time. There were nights
were I literally had to walk away from the entire project in
frustration, but after sleeping it off, I always tackled it the next
day with a fresh and positive attitude. In the end, the final result
was a working, fully automated, synchronized traffic light system,
which I will definitely never take for granted. And to be honest, if
others don’t notice it, it’s only because it’s working how it
should.
Check out my YouTube video showing my hand-built traffic lights
and controller based on Rob Paisley’s 20 Output Sequencing Circuit
in action!