Scherb signal logic circuit

Continuing this series of posts on signals let’s have a little more info on Jeff Scherb’s signal logic circuit. When I first started looking for a signal system for the Piedmont Southern I first dug out the info on Bruce Chubb’s CMRI and also Dick Bronson’s RRCir-Kits designs. Both of these use detectors with current sensing transformers so there is no direct electrical connection between them and the DCC power bus. Also, both are compatible with JMRI so computer layout control is possible and you can develop a virtual CTC panel for operations. I wanted to avoid the complications of designs that use diode based detectors because the DCC power bus feeds through them and they decrease the voltage on the track. At the time I had not run across the Azatrax designs so they didn’t factor into my search and evaluations but they appear to be a good option (I need to dig into them a bit more).

Another factor was the Southern Railway did not start using CTC in the area I model until the mid-1960s and I am stuck in the 1950’s. At that time they were still using ABS signals which means no computer connections or virtual CTC control panel was necessary or even desirable. I wanted a logic circuit that would independently control three aspect signals for the double track mainline without the need for the complications of computer interfaces and direct turnout control by the dispatcher. Had I desired computer assisted or controlled operations then the CMRI or RRCir-Kits would have been the way to go.

Looking at the CMRI and RRCir-Kits material I decided they were overkill for my needs. Then I dug out the March 2001 MR article by Jeff Scherb that I had set aside. It seemed to offer everything I needed. The circuit was designed to work with the CMRI detectors which was a plus in that it also meant that the BD20 detectors would be compatible too. The good thing about the BD20s is they are small, reasonably priced, and do not have to be built. At the time the CMRI boards were only available in kit form.

So, as a test I cobbled together three of the Scherb logic circuits and wired them up as described in the article for a test. As you can see from the photo there are only 9 components so this was a very easy job. I also added screw terminals to make the connections easier. In the article Jeff described how to test the operation of the circuit by simply manipulating the inputs so you can easily familiarize yourself with their operation right on your workbench before you commit to purchasing a bunch of components and block detectors.

I don’t think I’ve ever seen an electronics circuit article in the hobby magazines without at least one error and this one was no exception. The table that gives the various resistor values was screwed up although the circuit diagrams are correct. In this case the values for R1 and R6 were transposed. Also, once I assembled several boards and hooked them up, I had issues with the green LED not completely going off as designed. This only occurred in the constant on and not the approach lighting configuration. By simply increasing the value of R1 from 470 to 750 ohms I was able to correct that. I am not sure if it is a difference in the LED current requirement or a difference in the transistors. At any rate the change worked for me.

There is another circuit described by Jeff in the April 2001 MR that provides for searchlight signals instead of 3-aspect signals. It too had errors so make sure to look for the corrections in the June 2001 issue Railway Post Office column. I also have a pdf of a design change Jeff made at a later date that simplifies the searchlight circuit and fixes an overheating resistor issue. For the searchlight signal update: Click here


  • Thanks Larry for bringing this article to light with improvements. I have a couple of questions. First, would you please add a picture of the backside of your perfboard to show how you connected all of the components together? I have not worked with perfboard before. Also, do you have a source for the screw connectors you added to the perfboards (part number, website, etc)?

    • Thank you Dwaine for asking for the photo and screw connector source to use on the perf boards.
      OK Larry, the balls in your court–joel

    • I’ll get something up for you today or tomorrow. It’s been a busy week dealing with proofs for my May column and finishing my June column along with addressing requests for more photos for my new book. It always seems to hit at once. I got the screw terminals and most if not all components from All Electronics. Perfboard is getting hard to find and this came either from Jameco or Mouser. I got large sheets and cut them down into smaller rectangles. For the connections on the back of the perfboard I literally used the long cutoffs left over from the resistors and soldered them all together. If you plan it right you can just run the full leads through the holes in the perfboard and bend them into place. Cut them so they overlap and do a quick solder job to finish it. Them use the cutoffs to connect the components with short leads. If you are just making a few boards to experiment with you can solder in the LEDs and wires between boards instead of buying the screw terminals. If you can’t lay your hands on the specific MR issues a one month subscription to the MRArchive is available for about $5. For 30 days you can download every old article you ever wanted.

  • Jeff refers to pg 91 of “your book”. What is the title of that book?

  • 0n the scherb signal logic info, what is used for detection?

    • As I said in the post it was designed to be used with the CMRI detector but will work with just about any detector that uses an open collector transistor circuit such as the BD20 which I also discussed.

  • Thank Larry, I’ll take a copy of the PDF as well

  • This has been an interesting thread that may get me started on my long desired signal system. Is anyone aware of PC boards for the Scherb logic system? I would be interested in purchasing some for evaluation. I am not fond of bread boarding duplicates.

    • Unfortunately there is no circuit board. The original article included the board images so you could get some boards made or make them yourself. For experimenting it is easy to just wire them up on a piece of perfboard.

  • Larry does the logic circuit on page 91 of your book have the correct component values?

  • Where do we find the pdf of a design change Jeff made at a later date that simplifies the searchlight circuit and fixes an overheating resistor issue.