Using LED Signals to Inform Operator to Set Correct Polarity in Next Track Section

[70021 Morning Star]

Apologies if this question is dealt with elsewhere, I've done a search but can't find anything.
 
I'm building some interesting features into my test track layout. These comprise a pair of turnouts and a Y point, forming a 'wye reversing turnaround' and, after a short section of single track, a much longer teardrop-shaped 'turnaround loop', for getting trains with long coach rakes turned around.
 
I'm using Peco 80 track, and to avoid conflicting electrical currents, have broken the layout into several electrically isolated sections. This all works fine, and is easily managed with just two power controllers.
 
Well, I say "easily managed", but the problem is keeping my brain up to speed with where the 'forward/reverse' settings should be on each controller! As I move from one electrically isolated section of track to the next, it's important that the two track sections are switched to the same polarity. Obviously the section that the locomotive is currently running over is ok - it's moving forwards - but what'll happen when it hits the next section? (By the way, good design of a multi section track usually uses switches linking the two power controllers to ensure you don't get opposing polarity: unfortunately, however, with 'turnarounds' it is impossible to avoid having adjoining sections with opposing polarity at certain times, hence this discussion.)
 
This is all simple DC Analogue. So, I'm thinking it should be easy enough to wire in a few of those 2 LED track level Signals. These should automatically tell the operator's tired brain if the two sections of track are set to the same polarity (green LED illuminated) or if a hurried flick of the forward/reverse switch is required (red LED illuminated).
 
In other words, when my locomotive is in track section 'a', I want the signal LEDs to tell me whether the polarity of the next section ('b') is the same polarity as section 'a'. Green for yes, Red for no.
 
The signals come with three connectors: black wire = common anode; green wire = cathode for green LED; red wire = cathode for red LED. Simples! But how do I wire this to provide the information I want?
 
Many thanks for any simple easy-to-follow explanation,
Rick

 

__________

Hopefully this picture will help. Suppose my locomotive is on the mainline, at bottom right of picture, approaching the isolator for the independently controlled section of the wye reversing turnaround, I need to know the polarity of this next section (powered by controller #2). Then, as the locomotive approaches the current isolator for the turnaround loop (powered by controller #1), I need to know the polarity of the left/right lines after the turnout. And as the train comes back around the loop, I need a visual confirmation that I've reversed Controller #2, to bring me back towards the mainline (else the locomotive [and operator] will go crazy, as the locomotive judders forwards and backwards, as it crosses between two sections with opposing current!). And, finally, I'd like a Signal-like confirmation that controller #1 has been set for the correct polarity (forward/reverse) for the intended direction of travel when emerging onto the mainline. -- I think this should be simple enough to achieve, but I'm no wiring expert. So HELP, please?

[70021 Morning Star]

After a lot of Googling (and head scratching) I think I have a circuit that'll work.

 

Probably not the most elegant solution, as it only tells me whether the next section of track is set to proceed forwards (from the operator's viewpoint).

 

I'll need to move some of my 'signals'. The two on the turnaround loop are ok where they are, but I'll need to position one on the wye reversing turnaround and one on the mainline (being careful it's not isolated by the Insulfrog pointwork). 

 

The circuit, illustrated below, simply tells me the polarity of the next section of track. It uses two LEDs, Green = GO and Red = STOP if the polarity isn't aligned to 'forwards'. So, as I'm approaching the wye reversing turnaround from the mainline, I need to check the signal is showing Green for proceeding 'up' the line. As I approach the turnaround loop, I need to verify that my pointwork is switched to the direction I wish to travel AND that I have a Green for GO in that direction, else throw the forward/reverse switch on the power controller so that the correct signal is shown before proceeding onto the loop. Then, as I return from the loop to the wye reversing turnaround section, I need to set it's power direction to show a Red signal for trains travelling 'up' the line, as I'll be coming back 'down' that line (notice that this Red signal will match the Red signal on the side of the turnaround loop I'm approaching from; so even my addled brain cell should be able to cope). Finally, I need reminding of the permitted direction of travel along the mainline, so I'll place the third signal where it's easily viewable along the left side of the board to indicate Green for 'up', or Red for 'down' operation from my viewpoint. 

 

Perhaps not a full prototypical signalling scheme, although it has some similarity to 'tokens' used on single track operations, but this is only a test track and it'll save me from burning out my remaining brain cell !

 

(All that remains is remembering to flip the pointwork. But, if I do the switching of points and power in sequence, this signalling scheme should also help me to avoid too many errors.)

 

Hopefully someone will come up with a more elegant solution. Ideas?

 

 

Rick

 

______________

The following (badly drawn) circuit diagram shows the two LEDs set in parallel, and pointing in opposite directions. The common anode goes to one track, while the two cathodes are combined and connected to the other track. I'll find out which track to connect to by trial and error.  (Don't forget to fit an appropriate resistor.)

 

Just hope the track level signals I've just ordered have the LEDs wired in opposite directions, else some fiddly rewiring may be needed. Why don't model railway LED suppliers provide essential information on their websites? Grrrr!!!

[70021 Morning Star]

Ok, it seems that the pre-wired dwarf signals I've purchased have both of the LEDs pointed in the same direction, so additional wiring is needed. They each have individual cathodes (inputs) and a common, shared, return (the anode). This means that I need to manage what I'll call the 'discriminator', which decides which cathode to power with a separate circuit, instead of utilizing the LED's own unidirectional diode functionality.

 

Again, after some head-scratching, I think I've come up with a solution, as follows:

 

Connect each of the LED cathodes to one rail. This means that one LED will receive a +ve voltage, while the other receives a -ve voltage, depending upon the current (forwards/reversing) polarity I've set for that section of track. As each LED is a 'polarized' unidirectional diode (LED = Light-Emitting Diode) only one LED will pass current. So far, so good.

 

Of course, I now need to provide the return. Problem is that my forwards/reversing polarity switching means that the return feed sometimes needs to be through the left rail and sometimes through the right rail. So I need to connect the common return (anode) wire to both rails; but, to avoid a short-circuit across the rails, I need to  ensure that current only flows into (or is it out of?) one of them (why do circuits flow from -ve to +ve?). This requires the fitting of two diodes as voltage regulators. One diode must be fitted on each side of the return circuit, with each diode oriented so that it only permits the return current and doesn't permit a short-circuit across the rails (that's the important bit).

 

If I've got this right, the circuit should be something like that shown in the following initial sketch. (Please note that convention says +ve and -ve values of rails are -ve on left for forward travel, so the sketch illustrates reversing. Also, I'm not certain that the 2 diode symbols are the correct way round; I'll ask a qualified electrician later today. AT PRESENT, I'M ONLY ILLUSTRATING THE THEORY FOR DISCUSSION PURPOSES. DO NOT ATTEMPT TO IMPLEMENT UNTIL I'VE HAD AN ELECTRICIAN CHECK IT OVER !! )

 

 

Fingers crossed,

Rick

 

______________

 

UPDATE

 

For correct wiring diagram, please see Suzie's first diagram, below.

 

Note: it is not necessary to bypass the pre-installed resistor, as this works just fine.

 

 

Rick

[Junctionmad]

There are two issues.

 

The simple one being , that at low dc voltages the leds wont light.

 

More importantly an led is a terrible diode , and has a typical max reverse current in the order of microamps

 

A better way is to connect a simple full diode bridge across track break between the sections , this will only output when the tracks are opposite polarity. Feed this through a voltage regulator circuit to produce a steady led power to a red led that then indicates " stop wrong polarity "

[70021 Morning Star]

Hello Junctionmad,
Thanks for your input. Much appreciated !

 

True, neither LED will light unless power is applied to the upcoming section of track. That's not a problem, as I'd expect both controllers to be set to the same power output to give me matched DC speeds as I cross the track break. Also, I'm thinking, if I stop my approaching train at a Red 'stop' signal, in order to flip the forwards/reverse polarity of the second controller to match the polarity of the current section, this will mean there is zero current in the current track section. Hmmm, that would scupper my original idea of using a circuit that determines whether the polarity changes across the track break, because only one section now has power. Therefore, if I want the signal to continue showing a Red or Green 'token', my circuit can only operate on measuring the polarity the upcoming track section, which does have power. (See my 'note', below.)  
 
I'm not to worried too much about LEDs being terrible diodes, as the pre-wired model dwarf signals I've purchased preclude using the LEDs as diodes to build the workings of the circuit. (Although, of course, your comment is most helpful for anyone who is thinking of going along that route for themself. Thanks.)
 
Now this "full diode bridge and voltage regulator circuit" sounds interesting. Perhaps you could explain, in detail, how this works and give us a circuit diagram, please? (Be aware, I'm a software designer and model maker, with very shaky electronics skills. So I need all the help I can get !)

I presume you're suggesting I incorporate a 'full diode bridge rectifier', which looks like this (click on link) https://en.wikipedia.org/wiki/File:Diodebridge-eng.gif, somehow ? ...but, I'm not sure what the new circuit would look like ?? I'm thinking that my double diode circuit, for the anode return, is serving a similar purpose to half of your bridge rectifier, Junctionmad? In this case, we're only interested in the 'return', not both +ve and -ve. But, I'm no expert, hence the questions :-). Thanks. 

 

(Note: my original intent had been to fit something across the track break, to determine whether both sections of track had the same polarity. Then I realised that I already knew the polarity of the section on which the locomotive was running (as model railway standards say which polarity will move a locomotive forwards/backwards), so I only need to determine which polarity the next track section is set to and, apply the model railway standard to determine whether I should be displaying a Red or Green 'token' signal for the next section of track. This simpler task is what my currently proposed circuitry is attempting to do, if that makes sense? Also, as stated above, if a DC locomotive is controlled to a stop when encountering a Red 'stop' token, there's no longer any power going to the track where the locomotive is sitting.)
 
Many thanks,
Rick

[Suzie]

It is possible to use one of these signals as you wish. It is a shame that they keep being fitted with resistors.

 

The first diagram shows how to use the signal as is with the addition of a couple of diodes, but you have to be a bit careful because you will exceed the reverse voltage of the LEDs if track voltage is above 6V

 

 

The second diagram is a better way to do it if you can bypass the built in resistor. This will get round the low reverse voltage of the LEDs.

 

 

[Crosland]

More importantly an led is a terrible diode , and has a typical max reverse current in the order of microamps

 

With two LEDs wired anode-cathode and cathode-anode, the LED that i silluminated will protect the other from excessive reverse voltage.

 

Andrew

[CourthsVeil]

… (why do circuits flow from -ve to +ve?)…

 

 

… that's because Benjamin Franklin "…for a reason that was not recorded, … identified the term "positive" with vitreous electricity and "negative" with resinous electricity." [Wikipedia, "Electric charge / History"].

He imagined a flow from glass with plenty of electric fluid to other matter (e.g. silk) and a flow from other stuff to amber, the latter (in his view!) lacking electric fluid.

During the 19^th century electric engineers became acquainted with this view: current flows from +pos. to -neg.

 

Unfortunately, however, J.J.Thomson detected in 1897 that the opposite is true!!!

Thus the term "… circuits flow from -ve to +ve" expresses the view of physicists.

 

Whereas the +/-labels on batteries etc. follow the engineer's convention.

 

Therefore Wikipedia quite correctly states: "This is a source of confusion for beginners." [same article]

 

HTH

   Armin

[big jim]

Ive wired led's to the frogs of my hidden points and made a panel so when they are set in the correct direction the leds on the panel are green, when set to a diverging route they are red, a simple idea and only 3 wires, the led strips i use (12v) share the negative feed and the positive is fed from the outside if the points

 

don't know ifs what you have in mind but you may be able to do something with that idea

 

Have a look here

http://www.rmweb.co.uk/community/index.php?/topic/69543-big-jims-roundy-roundy-shed-layout-scenics-underway/?p=1100325

[RJS1977]

I'm just wondering whether the LEDs are necessary at all or whether you could use the point accessory switches to set the polarity correctly.

[Junctionmad]

Ive wired led's to the frogs of my hidden points and made a panel so when they are set in the correct direction the leds on the panel are green, when set to a diverging route they are red, a simple idea and only 3 wires, the led strips i use (12v) share the negative feed and the positive is fed from the outside if the points

 

don't know ifs what you have in mind but you may be able to do something with that idea

 

Have a look here

http://www.rmweb.co.uk/community/index.php?/topic/69543-big-jims-roundy-roundy-shed-layout-scenics-underway/?p=1100325

Yours is s DCC layout. Hence slightly different ( ie constant voltage etc )

[big jim]

Good point!!

[Junctionmad]

With two LEDs wired anode-cathode and cathode-anode, the LED that i silluminated will protect the other from excessive reverse voltage.

 

Andrew

Not where the series resistor is so wired in a common anode 2 aspect signal

The post by Suzie shows it correctly

[70021 Morning Star]

.
SUCCESS - I NOW HAVE WHAT LOOKS LIKE A SINGLE-TRACK SIGNALLING SYSTEM THAT INDICATES THE TRACK'S OPERATING DIRECTION !
 
Suzie takes the prize - thank you - as she corrected the orientation of the diodes (which I had reversed in my sketch). No need to bypass the factory-fitted resistor, as it works just fine.
 
So, go ahead and use Suzie's first circuit if you want to be a cheapskate and use just two power supplies to build a reversing loop and wye reversing track, as well as powering the mainline, without confusing yourself in operating it.
 
[ I'll edit this post with some photographs tomorrow. BTW, the dual-aspect Dwarf Signals are rated for 12V with their factory-fitted resistor. My expert tested on his workbench and said they'd be fine for normal operating voltages <12V.

 

Phone camera won't take a decent photograph. Either daylight too bright or LED dazzling camera. I'll get my proper camera out and upload some pictures in the next day or so. Apologies for the delay.]
 
Thanks, everyone, for your contributions,
Rick

[Il Grifone]

 

… that's because Benjamin Franklin "…for a reason that was not recorded, … identified the term "positive" with vitreous electricity and "negative" with resinous electricity." [Wikipedia, "Electric charge / History"].

He imagined a flow from glass with plenty of electric fluid to other matter (e.g. silk) and a flow from other stuff to amber, the latter (in his view!) lacking electric fluid.

During the 19^th century electric engineers became acquainted with this view: current flows from +pos. to -neg.

 

Unfortunately, however, J.J.Thomson detected in 1897 that the opposite is true!!!

Thus the term "… circuits flow from -ve to +ve" expresses the view of physicists.

 

Whereas the +/-labels on batteries etc. follow the engineer's convention.

 

Therefore Wikipedia quite correctly states: "This is a source of confusion for beginners." [same article]

 

HTH

   Armin

 

 

 

The problem is that, having a negative charge, electrons effectively flow backwards...

 

Thus, having muddied the waters even further, I'll shut up!

 

I was thinking along the lines of  something to detect the difference in potential between the sections (ideally you need the two controllers set not only to the same polarity but the same 'speed', but couldn't think of anything simple, not least because the output voltage is load dependent

 

Personally I would have arranged the circuitry for 'cab control' and used a single controller per train.

[28XX]

At 3 x 10^8 metres per second it doesn't really matter which way current flows. It's just easier to analyse a circuit if you choose one.

[CourthsVeil]

The problem is that, having a negative charge, electrons effectively flow backwards...

…

 

 

Why should they run "backwards", just because we name their charge 'negative' ??     

 

   Armin

[70021 Morning Star]

.

NEXT QUESTION - GETTING THE SIGNAL TO WORK WITH INSULFROG POINTS

 

Thanks for solving my first question, now let's see if you can figure out how to use these signals to ensure that Peco Insulfrog turnouts (a.k.a. 'points) are correctly set.

 

Remember, this is an Analogue DC layout and the 2 aspect (red and green) LED Dwarf Signals are pre-wired with single cathodes from each LED  and a common anode with a suitable resistor for 12V operation.

 

Assuming our locomotive is moving forwards, towards the turnout, it'd like to use two sets of signals - one on each spur - to tell me which way the turnout is set, showing Green for the set direction and Red for the other. 

 

So, as I approach the turnout, the signals will show Green on the branch the Insulfrog will direct the train, and Red on the other spur. So far, so good. Also, of course, when this section of track is running in the opposite direction, BOTH signals should show Red.

 

Note: this whole section of track is powered by a single power supply, so we don't have a problem of moving across a join between two sections of track with independent power supplies with opposing polarity; it is guaranteed that both sides of the turnout have the same +ve or -ve polarity. 
The difficulty arises because there's no power getting past the Insulfrog turnout onto the line that should be showing a Red light.

 

 

I'm thinking the answer might be to use the same wiring diagram that Suzie supplied, but to connect the Red LED cathodes to the other branch ?

 

 

Any suggestions?

 

Thanks,

Rick

[Il Grifone]

Why should they run "backwards", just because we name their charge 'negative' ??     

 

   Armin

 

'Forward' is positive to negative (conventional current). Electrons actually travel from negative to positive, which is thus 'backwards'. (All down to the wrong polarity being chosen as 'positive' in the first place.

[Il Grifone]

.

NEXT QUESTION - GETTING THE SIGNAL TO WORK WITH INSULFROG POINTS

 

Thanks for solving my first question, now let's see if you can figure out how to use these signals to ensure that Peco Insulfrog turnouts (a.k.a. 'points) are correctly set.

 

Remember, this is an Analogue DC layout and the 2 aspect (red and green) LED Dwarf Signals are pre-wired with single cathodes from each LED  and a common anode with a suitable resistor for 12V operation.

 

Assuming our locomotive is moving forwards, towards the turnout, it'd like to use two sets of signals - one on each spur - to tell me which way the turnout is set, showing Green for the set direction and Red for the other. 

 

So, as I approach the turnout, the signals will show Green on the branch the Insulfrog will direct the train, and Red on the other spur. So far, so good. Also, of course, when this section of track is running in the opposite direction, BOTH signals should show Red.

 

Note: this whole section of track is powered by a single power supply, so we don't have a problem of moving across a join between two sections of track with independent power supplies with opposing polarity; it is guaranteed that both sides of the turnout have the same +ve or -ve polarity. 

The difficulty arises because there's no power getting past the Insulfrog turnout onto the line that should be showing a Red light.

 

 

I'm thinking the answer might be to use the same wiring diagram that Suzie supplied, but to connect the Red LED cathodes to the other branch ?

 

 

Any suggestions?

 

Thanks,

Rick

 

It's beginning to get complicated! The red LEDs will light as you suggest but only for one direction of travel. Perhaps the answer is to use 'proper' semaphore signals connected to the point tiebars. They would only indicate the point setting of course, but anything more 'correct' would require complicated switching and dedicated solenoids for the signals.

[Suzie]

There are two problems here that complicate things - DC and dead frog. I think that the problem is insurmountable unless you use a high frequency lighting unit and then use DCC methods of point position indication. This will require rewiring your points as if they were live frog with frog switching (and because you are DC power routing).

[70021 Morning Star]

I'm pretty sure that cross-wiring the Red/Green lamps to the different tracks should work. I'll run a few practical tests and post the results.

 

Rick

[AndyID]

At 3 x 10^8 metres per second it doesn't really matter which way current flows. It's just easier to analyse a circuit if you choose one.

 

 Electron current is quite a bit slower than the speed of light. Might be around one third.

[70021 Morning Star]

Ok Folks,

I've been playing around with some LEDs, Diodes, & Resistors, and have come up with a circuit that works.

 

The result is two pairs of two aspect Green-Red signals that sit to the left of each branch of the turnout. With the power supply set to forwards, the line that the Insulfrog points are set to will show a Green signal, while the other track shows a red signal. Throw the points switch, the selected line changes and the signals change accordingly. Also, if the power control is flipped to reverse, then both signals shine Red.

 

This is a very simple circuit that tells the operator how the points (and power) are set, averting embarrassing derailments and conflicting power settings on the layout. The Diodes provide the circuit 'logic' for determining whether each Red LED illuminates when the opposite branch line is live forwards (locomotive moving up the line) or whether the power is reversed (locomotive coming back down the line) as well as preventing short-circuits.

 

Maybe someone will find it useful,

Rick

 

 

CIRCUIT DIAGRAM

For clarity, I've only drawn half of the circuit. Both sides are the same. Essentially, when moving forwards, the left tracks drive the left Green LED and the right Red LED, and visa versa. (The diagram shows the left Red LED and the right Green LED.) If the power is in reverse, the two outmost rails (which are always powered on Insulfrog points), power the double Red signal. As you see, the Red LED is wired the two outermost rails for this purpose. The Diodes provide the 'logic' and prevent short circuiting. Track connections should be made after the plastic frog on each of the branch lines.

 

Note: I think I missed excess back current protection for the Green LED in my original post. I've added what I think is the correct additional protective circuitry - circled in red - but this has not been tested and is missing from the photograph of the breadboard. It may cause a short-circuit ! Maybe it needs another diode, in series, between the LED and the outside rail? Will test and update.

 

(I've also added a snapshot of the test circuit on a 'breadboard' for those who like concrete examples. The power rails of the breadboard represent the two pairs of rails on the track.)