This article is in the "you can do it too" category. It's not as difficult as you can imagine for the average model railroader, especially with the information available on the Internet and with articles like this to guide you.
I have been looking for a way to construct my own detectors and signal circuits to keep the cost down. Also, there's a lot of satisfaction that comes from proving to yourself that you can do it.
I don't know a lot about electronics, but I can follow a schematic, I can read a meter, and I know how to solder. The rest falls into the category of "take it one step at a time and doublecheck what you're doing".
Sometimes this is referred to as a "smoke test". You don't want to see smoke!
Bill Payne, fellow member at our Nottawasaga Model Railroad Club (NMRC) knows more about electronics than I do so he took the lead on this project.
The NMR layout has been upgraded to DCC and we would also like to install signals on the layout. There's nothing like changing signals to keep people fascinated at train shows, even if the indications of the signals aren't quite right!
Before Christmas, 2009, Bill and I had already built and tested some signal circuits we had found on the Internet. The next step was to build some detector circuits that would work for DCC.
The ones we were working with were designed by R. Paisley. One of the block occupancy detectors is designed around a 555 timer and the other is based on a 339 comparator.
Both take power directly from one rail. The wire with the power passes through the hole in a small transformer. The transformer detects a change in voltage when a train occupies a block and passes this information to the signal controllers.
That's sort of a non-electronic way of describing what's going on. Here are the circuits by R. Paisley. We found that both work about the same. Bill is trying to see if they will also work for DC on his home layout that is not DCC.
The sensitivity of the transformers for detecting the current change can be changed by the number of loops in the wire through the center hole.
The second schematic for DCC detectors shown above is based around a 339 comparator. You can buy a single comparator chip, I beileve it's the 393. The 339, being a dual, allows more circuits to be built on one board.
Bill laid out four DCC detectors on one piece of stripboard. This is where you really have to take your time. He started by laying out the circuit on paper using a Stripboard Planning Sheet.
This paper layout shows where to insert components for the DCC detectors. The red dots on the plan indicate where to cut the traces on the copper side of the board.
Bill kept the top line on one side of the board for 12+ volts and the first line on the opposite side is 12- volts.
Once he was satisfied that the paper plan matched the schematic he turned the board over and carefully cut the traces by using a drill bit in a Dremel.
You can also use an X-Acto blade, but Bill feels the drill bit is a better guarantee that the trace is cut all the way through.
If you mess up, you can always add a jumper wire.
Bill laid out 4 blocks on one piece of stripboard. This is the final "beta-test" board we used to drive the signal logic.
We used a couple of alligator clips to live rails on the main layout (actually at Utopia East which is near the workbench).
The 12 volt power for the circuits is supplied by a simple DC power pack that came with a President's Choice train set.
We used a meter to adjust the voltage to close to 12 volts.
On a night when my other operators played with the trains on my Utopia Northern layout, Bill and I fired up the circuit. Well, not really "fired up". No smoke! One block wasn't working properly. We traced it to two pads for the chip that hadn't been soldered.
For our setup I used an old 3-foot piece of HO track. How old? You guess. Fiber ties, brass rails stapled to the fiber ties. I'd say probably 40 years old. Anyway, I took a Dremel with a cutoff disk and cut one rail in 3 places to make 4 blocks. We used a GE 44-tonner diesel for the test because it is short.
In the photo, the diesel engine is moving through block 2 that has turned red. The preceding block is yellow. The block ahead, block 3, is still green.
The way the circuit is designed it changes quickly when the locomotive enters the block, but there is about a 2 second delay as it leaves.
This is really a good thing because it allows for dirty track as the train moves through the block and looks good when the signals change. Pretty neat.
The GE 44-tonner has LED lights. It wouldn't activate the circuit until the engine was moving. This may be corrected by increasing the sensitivity. Another diesel, also with LED lights, triggered it OK.
A steam engine with 12 volt grain-of-wheat lights triggered the DCC detectors. We still need a little fine-tuning. We're almost there!
Currently, our signal circuit only gave the correct indications in one direction. The lights will change as the engine reverses through the blocks, but the indications for the block ahead and the block behind are wrong.
We were still looking for a signal circuit that allows bi-directional travel.
One of the sites that had interesting circuits, kits and parts was Hudson Telcom. Bill Hudson was helping us in 2010-2011. These were the early days of our experimentation.
After posting this page, I had a reply from Christopher, I believe from the United States. He wrote on my comments page "For the single track section, two strings of series-connected OR-gates can be used for each direction to provide the cascading tumble-down red aspect on opposing signals. Then at each end of the single track section, preferably before the passing siding islands, two signals can be wired so that the next signal will go yellow when the previous one is yellow to provide protection on either side of the single track section".
Thank you Christopher.
I managed to get the signals working on my Utopia Northern layout before dismantling it in 2014. I have since moved to a new home and am starting over again. In the meantime, fellow modeller John Houghton has redesigned some of Paisley's circuits and built detectors for his N gauge layout and has installed signals on our NMRC layout. We have also been learning a lot about using jmri for operations. All that on other pages.
Go to signal circuit experiment.
Go from "DCC-detectors" to the DCC overview page.
Go from "DCC-detectors" to signal construction.
Return from "DCC-detectors" to my Home Page.