Model railway help

Thread Starter

RobertB72

Joined Feb 6, 2021
37
Yes the number of diodes can get rapidly out of hand, an alternative is to use a microprocessor with plenty of pins as I did here signalling article but only if you are comfortable with programming.
As for the diode method imagine each point having an spco switch then wiring them in series exactly as your track diagram with the leftmost pole grounded, now each right hand tail wire will only be connected to ground (or - whatever you want to call it) when the points are set for that route. If you take each signal led pair, make them common anode with a single resistor to + then when you ground a signal wire the led will light. Now you arrange small diodes (eg 1N4148) with the anode to the signal wire and cathode to the point wire, best to make a cross hatched piece of paper to show each connection required. Where only a single signal is connected to a point wire no diode is required, the diodes prevent reverse feed conflicts when more than one point wire drives a signal.
I can sort of picture that in my head but I’m not good at drawing the diagrams for it I will ‘play’ about with it once I get some modelling time
I will attempt to draw one diagram and use different colour pens to represent the red green led feeds
I will put in where I think the diodes should go and also the direction
I have some signals that are wired the way you said
It has 2 inputs and only one out with a resistor attached to it
If I use diodes I think I can fathom out the wiring
 

eetech00

Joined Jun 8, 2013
3,859
Yeah the making one dependent bit is the hard bit
I can work out how to make that wotk for some aspects of it
Ie if I run the lead from the green led so that becomes the feed for the next switch but if that signal is red then the next signal is unlit on either red or green
I’m trying to devise a switching option so if one side of the turnout is set to red all subsequent turnouts show red at both sides of the turnout
I’m thinking of using the leds at these signals wired up to the previous red with diodes I can feed power into a string of them halfway along and only light up the following ones if that makes sense
It would be good, at this point, to create a truth table representing the behavior of each set of lights based on the positions of the turnouts. This will help us focus on the circuit requirements and keep things simple.
 

eetech00

Joined Jun 8, 2013
3,859
The track diagram reminds me of one (real) interlocking control system I designed for a transit railroad. If I remember correctly there were about 10 turnouts and a few crossovers and, of course, signals. Back in the day, it had what was called a "Line-o-Light" display. Display operation was based on entrance/exit requests. The route request would flash the route white from entrance to exit. When the controls were executed, and status returned, the route lights would change to steady white. As the train moved thru the route, each occupied track sections would change to red (as well as each passed signal), then change to dark, as the train cleared each track section.

Very cool to watch...
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
The track diagram reminds me of one (real) interlocking control system I designed for a transit railroad. If I remember correctly there were about 10 turnouts and a few crossovers and, of course, signals. Back in the day, it had what was called a "Line-o-Light" display. Display operation was based on entrance/exit requests. The route request would flash the route white from entrance to exit. When the controls were executed, and status returned, the route lights would change to steady white. As the train moved thru the route, each occupied track sections would change to red (as well as each passed signal), then change to dark, as the train cleared each track section.

Very cool to watch...
This is what I’m thinking will work
The two diodes at 4 a and 4 b would stop the green lighting up when the reds are being fed from switch 1 thus switch 4 won’t be getting any power from switch one if switch 1 is set to A
D429BADC-08F6-4AAE-A813-E4AEC9A59389.png
 

eetech00

Joined Jun 8, 2013
3,859
This is what I’m thinking will work
The two diodes at 4 a and 4 b would stop the green lighting up when the reds are being fed from switch 1 thus switch 4 won’t be getting any power from switch one if switch 1 is set to A
View attachment 230492
I need some clarification on how the turnout controls work.

The turnout control levers on the panel are "on-off-on", meaning, they are momentary when moved, and return to center off position when released. Correct? up/left=normal, dn/right=reverse?

Will each turnout position status be available from contacts located at the turnout switch points? 1NO and 1NC contact?
 

eetech00

Joined Jun 8, 2013
3,859
I need some clarification on how the turnout controls work.

The turnout control levers on the panel are "on-off-on", meaning, they are momentary when moved, and return to center off position when released. Correct? up/left=normal, dn/right=reverse?

Will each turnout position status be available from contacts located at the turnout switch points? 1NO and 1NC contact?
Also...There are no traffic signals involved, just turnouts. right?
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
I need some clarification on how the turnout controls work.

The turnout control levers on the panel are "on-off-on", meaning, they are momentary when moved, and return to center off position when released. Correct? up/left=normal, dn/right=reverse?

Will each turnout position status be available from contacts located at the turnout switch points? 1NO and 1NC contact?
yes they are momentary so there won’t be any way of knowing from the switch alone what way the turnout is set
But I plan to get round this by taking a feed from the last green leds and running them to a control panel to a white or blue led

theses will only be lit if the last green light is lit

so at a glance I will see why one of the 5 possible ends is live and I will know that gives me access to and from turnout 1
 

eetech00

Joined Jun 8, 2013
3,859
yes they are momentary so there won’t be any way of knowing from the switch alone what way the turnout is set
But I plan to get round this by taking a feed from the last green leds and running them to a control panel to a white or blue led

theses will only be lit if the last green light is lit

so at a glance I will see why one of the 5 possible ends is live and I will know that gives me access to and from turnout 1
What if we were to design a circuit to remember which way the control lever was thrown?
We could do that with a small set/reset latch circuit using a relay, or IC, or couple of transistors.

The lever would drive the circuit. The circuit would send a pulse of the correct duration to the switch machine to move it normal or reverse.
We could then use the state of the "last command" to represent the switch position, and use it for the logic to light the proper lights.
 
Last edited:

Thread Starter

RobertB72

Joined Feb 6, 2021
37
What if we were to design a circuit to remember which way the control lever was thrown?
We could do that with a small set/reset latch circuit using a relay, or IC, or couple of transistors.

The lever would drive the circuit. The circuit would send a pulse of the correct duration to the switch machine to move it normal or reverse.
Would that be very involved or is it something a novice could do also would it be very costly
 

eetech00

Joined Jun 8, 2013
3,859
Would that be very involved or is it something a novice could do also would it be very costly
An NE556 (dual 555 timers) cost about $0.50 USD. Same for a CD4013 dual FF.
Should be simple to do on a perfboard. Have you ever built a 555 timing circuit on a perfboard? or any circuit using an IC?

But this brings up another question:
I believe the switch machine requires a pulse to change position. How did you intend to "pulse" the switch machine?
 
Last edited:

Thread Starter

RobertB72

Joined Feb 6, 2021
37
Should be simple to do on a perfboard. Have you ever built a 555 timing circuit on a perfboard?

But this brings up another question:
I believe the switch machine requires a pulse to change position. How did you intend to "pulse" the switch machine?
I have no experience at all with electronics apart from using a couple of leds in a signal

the power will be supplied from an AC. Accessory outlet on my trains controller I think it 15/20 v ac
 

eetech00

Joined Jun 8, 2013
3,859
I have no experience at all with electronics apart from using a couple of leds in a signal

the power will be supplied from an AC. Accessory outlet on my trains controller I think it 15/20 v ac
OK....probably more than you want to take on at this time.

So...how do you anticipate the finished control panel will operate from a users (you) perspective?
(assume the switch machine will operate according to the position selected by the lever)
Can you give a step by step "operator interacting with control panel" example?

I know it seems like alot of questions but please bare with me...
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
OK....probably more than you want to take on at this time.

So...how do you anticipate the finished control panel will operate from a users (you) perspective?
(assume the switch machine will operate according to the position selected by the lever)
Can you give a step by step "operator interacting with control panel" example?
I will hopefully draw a miniature map of the railway and at every turnout position I will drill where the miniature toggle switch will fit
I will drill two holes one at A and B referring to the earlier diagrams and Will fit an led here probably blue
These will be wired in series to the next led and the last led at each siding wil be connected to the last green signal
So where as the signals will be fed from the throat end of the sidings these will be fed from the sidings end back towards the throat
So say we have turnout 1 set to turn to the right ie towards turnout 4 and 4 set to B
The green signal at B will be ‘live’
This will feed an led on the mimic board at 4b which will run to say 2 leds before turnout 1 where another led will be at 1B
So a visible route will be seen
Because the leds only allow one way travel I think I can put these converging as long as only one input is fed
 

djsfantasi

Joined Apr 11, 2010
9,156
OK....probably more than you want to take on at this time.

So...how do you anticipate the finished control panel will operate from a users (you) perspective?
(assume the switch machine will operate according to the position selected by the lever)
Can you give a step by step "operator interacting with control panel" example?

I know it seems like alot of questions but please bare with me...
The TS is using the Peco brand of model railroad turnouts. They use a dual coil motor, and activated by a short pulse. The Peco control unit is a SPDT toggle switch that requires pushing the toggle past its limit to operate a momentary contact. The toggle then remains in position to indicate the turnout track position.

He is using a set of add-on contacts to activate additional circuitry. The add on contacts provide a DPDT switch. One pole is often (but not necessarily) used to power the track frog* The other pole is used for signaling. So, inputs to his track signals are independent from positioning the track position.


The switching will be done via accessory switches that clip on to the point motors so as I change the way the point is set at the lights will change if that point is actually changing the route if it’s behind a point that is already set the other way the signal won’t change until that point is set
If that makes sense
I have the option of using these

one is a single pole on on switch the other uses 2 single pole on on switches

I have had success with the single switch and 2 signals
By wiring the red led from one signal to the same pole as the green from the other signal And vice versa

so the signal at turnout 1A is green and point 1B is red
And then when the turnout is changed 1A is red and 1B is green

But if I use that method on all signals
If turnout 1 is set straight
There will still be a green signal showing at turnout 4 either A or B depending which way that turnout is set

View attachment 230445

View attachment 230446
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
The TS is using the Peco brand of model railroad turnouts. They use a dual coil motor, and activated by a short pulse. The Peco control unit is a SPDT toggle switch that requires pushing the toggle past its limit to operate a momentary contact. The toggle then remains in position to indicate the turnout track position.

He is using a set of add-on contacts to activate additional circuitry. The add on contacts provide a DPDT switch. One pole is often (but not necessarily) used to power the track frog* The other pole is used for signaling. So, inputs to his track signals are independent from positioning the track position.
I imagine my panel looking like what I’ve attached but with switches drilled into the board with just the toggles showing

so assuming
Turnout switches 1 and 2 are set straight and 3 is set to left
That would mean that the signal at 3a is green and I would run this lead to point V under the mimic board
This woukd light up the led at
3a 2b and 1a and also the led before 1 at the far left of the diagram


1943A900-441F-4917-9C3D-BA949007F54A.jpeg
 

eetech00

Joined Jun 8, 2013
3,859
The TS is using the Peco brand of model railroad turnouts. They use a dual coil motor, and activated by a short pulse. The Peco control unit is a SPDT toggle switch that requires pushing the toggle past its limit to operate a momentary contact. The toggle then remains in position to indicate the turnout track position.

He is using a set of add-on contacts to activate additional circuitry. The add on contacts provide a DPDT switch. One pole is often (but not necessarily) used to power the track frog* The other pole is used for signaling. So, inputs to his track signals are independent from positioning the track position.
Thanks for the clarification djsfantasi. It really helped.
But now I have another question. :)
Does the TS have to provide a pulse generator circuit to control the switch machine? Or does it come with the control unit and/or turnout?
 

djsfantasi

Joined Apr 11, 2010
9,156
Thanks for the clarification djsfantasi. It really helped.
But now I have another question. :)
Does the TS have to provide a pulse generator circuit to control the switch machine? Or does it come with the control unit and/or turnout?
The “pulse” is created by the momentary contacts in the turnout controller. Purely mechanical. There is no pulse generator involved.
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
Thanks for the clarification djsfantasi. It really helped.
But now I have another question. :)
Does the TS have to provide a pulse generator circuit to control the switch machine? Or does it come with the control unit and/or turnout?
The current will come from the normal power supply for the train controller through a toggle switch then to a point motor which is two coils that are magnetised for a brief moment

this ishow I envisage wiring the leds from the last signal on each track

3CA9CD59-F75F-46A0-90E1-821D50A29201.jpeg3CA9CD59-F75F-46A0-90E1-821D50A29201.jpeg
 

Thread Starter

RobertB72

Joined Feb 6, 2021
37
The current will come from the normal power supply for the train controller through a toggle switch then to a point motor which is two coils that are magnetised for a brief moment

this ishow I envisage wiring the leds from the last signal on each track

View attachment 230525View attachment 230525
I’ve been doing some reading ad I think a not gate chip might come in handy
Ie the green led for each signal would take its feed from the preceding switch

and the chip could be connected to this feed so if the green led isn’t getting power the red led will automatically light

This seems plausible and relatively cheap solution.
I’d use one chip per turnout just to cut down on wiring

im still a bit unsure on the wiring of these chips but il have a read up on them
 

eetech00

Joined Jun 8, 2013
3,859
But if the lever generates a pulse, a latch circuit will be needed to remember the lever position (or turnout position) and keep the led lit.

or am I overlooking something?
 
Top