Colour light signals with DCC

[Dog Star]

So we have a single head, two aspect, colour light signal with red / green LEDs for illumination (design by @Pete Harvey and kit from C&L Finescale)... we are thinking that operating the signal from DCC ought to be possible. I have looked at the range of decoders on the DCC Supplies web site and the TCS two function decoder (without motor drive) appears to be a reasonable option.

Which WTers have experience of operating colour light signals under DCC and might care to share experience and / or offer guidelines?

thank you, Graham

 

[BrushType4]

Graham, I think DCC for a signal as you've described is overkill and needlessly complicated. Maybe others can put me right.
.

 

[Steph Dale]

Phil,
It needn't be if you've already got the DCC system installed...!

Graham,
Might be worth dropping a line to Mike Sheardown through the PM facility. He may not have colour-light experiance, but I'm sure will be able to advise on how to get points and signals working together effectively.

Steph

 

[Dog Star]

Graham, I think DCC for a signal as you've described is overkill and needlessly complicated.

Phil,

I have no experience in this area and I do not recall reading anything here or there about the subject... so in what way is this approach either overkill or complicated?

What do you suggest?

 

[BrushType4]

This is of course only my opinion but I'm thinking that a simple easy to use switch must be simpler, cheaper and easier than going to the expense of DCC and the awkwardness of it.

I mean, I like a track diagram with switches. Old school, I suppose but I've never found DCC quite as intuitive.

I'm as interested as you Graham for the answer.

 

[Dog Star]

This is of course only my opinion but I'm thinking that a simple easy to use switch must be simpler, cheaper and easier than going to the expense of DCC and the awkwardness of it.

As President of the Flat Earth Society I may agree with you. My Son may not given that (a) sound is everything and (b) running the layout from an I-pad is the way forward.

So, we our stock is DCC-fitted, we are progressing with how to manage coach lighting... and signal operation is the next frontier.

 

[Steph Dale]

Graham,
How are you controlling pointwork? That's about the point (groan, sorry ) I gave up on the simple apps, but I guess it's all solveable.
Steph

 

[Dog Star]

How are you controlling pointwork? That's about the point (groan, sorry ) I gave up...

Ah, that is the rub. We have turnouts operated by Fulgurex with the HCI through a mechanical lever frame (from the Shropshire and Herefordshire Area Group of the S4 Society) which has microswitches on the lever tails .

How to interlock turnouts and signals requires advice from Major Mandarin(?) .

 

[Steph Dale]

In that case I agree with Phil.

Steph

 

[Dog Star]

In that case I agree with Phil.

Fair enough, so what is your suggestion for control of signals where the power to the turnouts is managed by micro-switches which are operated by mechanical means?

Extending the lever frame is an acceptable answer .

 

[Steph Dale]

Hmm, yeah - perhaps 'if in doubt refer to the prototype'?
Any way you can do automatic interlocking so a combination of point positions/movements triggers the signal?

Steph

 

[SimonT]

Graham,
to bring two places for this question together. It is important to remember that DCC operation of signals and points can be done using the Lenz control system and by using Touchcab. However, you can only program the decoders using Lenz hand sets or more sophisticated programmers; Touchcab is only an operating handset. On Aberbeeg, the junction box signals and points are operated from an electrically interlocked row of switches that are laid out as a lever frame. The remote points and signals are operated using Touchcab. The use of Touchcab/DCC for the remotes saves on two sub switch boxes, two power supplies and two extra power leads; about 10 minutes in setting up and knocking down time.

The interlocking is achieved by a large number of relays. The relays are fed from SPDT switches for the signals and a point switch which combines a DPDT with a SPDT; the DPDT feeds the point motor (Tortoise on DC) and the SPDT feeds the relay drives. You could achieve the same with two micro-switches for point levers. Signals will only pull off when the route is correctly set by the sequence of relays that control that route. By placing all this circuitry on the return side, one can build in the back locks that are required for the junction. The start point for designing all this circuitry is a locking diagram. This is a matrix which has the points as the x-axis and the signals for the y-axis. The circuit diagram for Aber is a mitre square, but I can try to isolate part of it to illustrate all the above should it be of help.

Simon

PS This post is intended to meld a short e-mail conversation between Graham and I with this thread.

 

[Dog Star]

On Aberbeeg, the junction box signals and points are operated from an electrically interlocked row of switches that are laid out as a lever frame.

The interlocking is achieved by a large number of relays. The relays are fed from SPDT switches for the signals and a point switch which combines a DPDT with a SPDT; the DPDT feeds the point motor (Tortoise on DC) and the SPDT feeds the relay drives. You could achieve the same with two micro-switches for point levers. Signals will only pull off when the route is correctly set by the sequence of relays that control that route. By placing all this circuitry on the return side, one can build in the back locks that are required for the junction. The start point for designing all this circuitry is a locking diagram. This is a matrix which has the points as the x-axis and the signals for the y-axis. The circuit diagram for Aber is a mitre square, but I can try to isolate part of it to illustrate all the above should it be of help.

Interesting that you use DPDT for operating the switch motors (Tortoise) and we have SPDT for the same purpose (Fulgurex). Maybe the difference is that we are using 18VAC for the power... with a pair of diodes in the input feed to the switch bank for rectification. The SPDT takes the input feed and connects that power source to one or other of two wires to an individual switch motor. All switch motors share a common return.

I am familiar with locking tables having been taught same by signalmen in ex-GW&GC boxes (late 1960s, mid afternoon when traffic was light and the rest of the class was doing "games"). I shall be pleased to see what you have done Simon so isolate away and publish either here or on the Big Hill thread.

regards, Graham

 

[SimonT]

Graham,
here goes. The signals were built by Peter Kibble and he designed the signalling system; I just did the soldering. Power is supplied to the signal servo and goes to earth through the interlock relays and finally the switch controlling the signal.

Firstly a shot of the complete wiring diagram (it misses out the supply side and only features the flow from the signal through the interlock system).

The relays are the red blocks and some turnouts require up to three relays; all are worked by the SPDT side of the switch. The diagram became this... (there is more!)

The blue and white wires are the feeds to the Tortoise motors, the grey the feed and earth to the relay coils and the black is the signal wiring. As you can see turnout 23 requires 3 relays working in parallel.

Now to study one particular interlock. This is lever 31 which controls the up branch crossover. There are three affected signals, the Down Home, the Up Starter and the Up Crossover Starter, required for the up departure of the Ebbw DMU from the Down platform; yes they really did depart wrong line.


The circuit diagram is a straight copy of the appropriate part of the circuit diagram. If 31 is normal (switch open) signals 1 and 50 can go to earth. If 31 is pulled (closed) only 49 can go to earth when pulled. Things might become a little more clear in the next post as I can't upload the required file at the moment.

Simon

 

[SimonT]

Attempt 2

Each relay has one of these sheets, the top schematic represents the contacts in the relay. Contacts 9 to 12 are switched output and 1 to 4 normal input and 5 to 8 are pulled input. Contacts 13 and 14 are the coil contacts. The rectangle with the numbers in circles is a schematic of the relay mounting base. This is a copy of the final master sheet, there was a scribbled version that I filled up as I did the work. The feed from signal comes from the signal, through switch 1 and into the relay at contact 4. At Normal it exits the relay at contact 12 to go to earth so the signal can be pulled off. With the relay at Normal the feed from switch 49 cannot go to earth at contact 12 and thus the signal cannot be pulled; if the relay is pulled, switch 1 can no longer go to earth and now switch 49 can go to earth. This basic functionality can be built up to allow more complex interlocking.

And then there are the back locks.......

Simon

 

[Dog Star]

Thank you Simon,

All is clear now - I see how the principle can be extended for more complex formations.

Back locks? The floor is yours.