Circuit for signal lights

This forum is dedicated to Riding Scale Railroading with propulsion using other than steam (Hydraulics, diesel engines, gas engines, electric motors, hybrid etc.)

Moderators: Harold_V, WJH

Post Reply
User avatar
Posts: 263
Joined: Thu Nov 20, 2008 1:32 am
Location: Oxnard, CA

Circuit for signal lights

Post by pwcphoto » Mon Jan 11, 2010 1:59 am

Curious to see what various clubs are using to control their signal lights. How do you overcome the problem with moisture falsing the signal etc? We have a number of lights around the track but they seem to be very sensitive to water in the ground.

Any diagrams of circuits that work and are in operation would be appreciated.

"I used to be a rocket scientist, now i am just a space cadet"

RMI 3.75" scale Prairie 2-6-2 under construction.
View Progress

User avatar
Posts: 655
Joined: Mon Aug 18, 2008 5:35 pm
Location: San Diego

Re: Circuit for signal lights

Post by tburzio » Sat Jan 16, 2010 10:03 am

pwcphoto wrote:Curious to see what various clubs are using to control their signal lights.
The two biggest installations which are wet some of the time are at Train Mountain and Redlake Junction. Friends of Train Mountain is having teething problems with their system from what I can make out.

We're also looking at signals for our railroad, which gets deluged with sprinkler geysers which have fertilizer water (conductive) for a golf course. Initial readings trying to get a voltage difference between rails on our track aren't promising, since the electrolyte laden water makes a conductive path between the rails. The best I can figure to try next is to energize one or both rails and then get a reading vs a known ground (metal spike pounded into the ground) across a big resistor (high voltage, low current). I really like Redlake's plastic insulated joiners.

If we can get electronics working in a harsh environment, we can reduce energy consumption by only putting out a pulse down the rails every second or so. This should be fine for a railroad, since even our fastest guys can't get across a whole block in under a second! :-) PIC chips can monitor the line for the absence of pulses and determine if there is a train present.

Did you notice that single 1/2 watt outdoor water resistant (!) Red/ Green/ Blue LEDs are now available?

User avatar
Posts: 291
Joined: Thu Jan 09, 2003 11:16 pm
Location: Canton, Ga

Post by RichD » Mon Mar 22, 2010 10:39 pm

I was responsible for the NGLS signal system in Atlanta, GA. (Duluth)
This worked extremely well wet or dry.
What you decribe is a common misunderstanding in view of the result.
Full size RR circuits are similar and have the same problem from
false current paths. Using LEDs and such only makes the problem worse.
On a miniature track a simple track circuit detector comprised of a relay
in series with the track as a switch and battery voltage works well.
However, the devil is in the details as they say.
The leakage can be quite heavy depending on block length and moisture.
The correct design (or cure) is to require the circuit components to pass
a much higher current to be insensitive to the leakage, especially on release.
175 mA leakage was not unheard of.
Since the Duluth track system was built there are now relays available to
help the situation from the standpoint of low coil resistance and contact
current capacity. Modern autos are loaded down with electronics and thus
you will find on the surplus market small 50 ohm relays that are perfect
for signal designs. Parallel the Red lamp (incandescent) for more current.
A second relay in each detector unit controls the green and yellow aspects.
This relay coil is connected to the next blocks red lamp as an advance or
distant signal warning.
Except for power, there are only two wires following the track. Both are the
distant relay controls. This system is fully automatic bi-directional.
No solid state electronics are used making for a weather resistant system.
If you have any further questions I can be reached at

Post Reply