Semaphore signalling help

Having a problem making your model railway layout look real. Post questions and share the results of your model railway scenery here.
AJFE 2488
Posts: 94
Joined: Sat Jul 17, 2010 2:14 pm

Semaphore signalling help

Post by AJFE 2488 »

I'm in the process of sorting out multicored cabling for the electronics on my layout and I have been looking at incorporating working signals that are servo controlled. I've been trying to plan the signalling on my layout to work out the additional cabling I'll need for it and have been struggling to figure out how many signals I would need for my layout as shown below:
Image

Although I have spent the past 6 years helping out on my local heritage railway and have the basic idea of reading signals from left to right, I'm still not fully clued up on signalling layouts so I'd appreciate any suggestions on how best to signal mine so that it looks as realistic as possible. My layout will be making use of GWR lower quadrant and disc signals if that is of any use.
aleopardstail
Posts: 1526
Joined: Thu Apr 29, 2021 9:48 pm
Contact:

Re: Semaphore signalling help

Post by aleopardstail »

novice at it here, going through similar for my layout.

on the way into the station on the curve you want a signal to control the approach, like two arms for platform and the bay, and two smaller arms for the two goods roads (all red with white stripe, these probably below the two main arms on the posts as its a rural location by the look of it). that covers stuff coming in.

the end of the platform would have a pair of arms, again one for each road acting as starter signals (red ones, white stripe)

the goods road to the top in your drawing would maybe have its own starter covering departure to the main line, and potentially a shunt signal disc for the headshunt - this could also have been managed locally with a bloke and a flag using a local lever frame (the point interlocked by the signal box)

the lower headshunt could also be similar, and its sidings likely a shunt signal controlling entry.

not sure how the release crossover would be signalled, maybe a disc at each end

a lot comes down to "what looks right", basically any movement of a passenger train that would have warm self loading cargo onboard is signal controlled, ditto anything that can cross the path of such, pure goods areas likely not fully signalled and sidings very likely a local lever frame and bloke with flags
User avatar
Mountain
Posts: 6792
Joined: Mon Oct 24, 2016 3:43 pm
Location: UK.

Re: Semaphore signalling help

Post by Mountain »

Signals do two main things. Tell drivers if it is safe to proceed or not by controlling the passage of trains (And give advanced warning that they need to slow down ready to stop with distant signals), and tell drivers which routes have been set so they don't end up going the wrong way!
User avatar
Bufferstop
Posts: 13904
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: Semaphore signalling help

Post by Bufferstop »

Signalling is a huge subject and it's difficult to translate prototype practice into model terms. The only place I've seen home and distant signals working as the should is on Pete Waterman's making tracks layout where it's possible to stand at one end and see multiple four aspect signals changing in sequence, green, yellow or double yellow going back to red as the train passes, then changing back through the sequence as it passes each signal down the line. On a layout in a home you'd be lucky to get two blocks in succession. The best plan is for a signal to got to red when it's passed and change back through the sequence after a couple of timed delays.
Where should you put them, the end of platform roads, if it's a through platform it will be both home and starter. Before points in the main running line, if it's a light it gets a feather to indicate the diverging line is set, if it's a semaphore you get two arms main and subsidiary. The subsidiary is on a bracket on the appropriate side. On the approach to stations, to protect a train standing at the platform. There's a book by CJ Freezer which treets it in a pragmatic way, making it look right which is probably the best most of us can do. I think it's been reprinted recently.
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions
aleopardstail
Posts: 1526
Joined: Thu Apr 29, 2021 9:48 pm
Contact:

Re: Semaphore signalling help

Post by aleopardstail »

Bufferstop wrote: Tue Jun 25, 2024 10:06 pm Signalling is a huge subject and it's difficult to translate prototype practice into model terms. The only place I've seen home and distant signals working as the should is on Pete Waterman's making tracks layout where it's possible to stand at one end and see multiple four aspect signals changing in sequence, green, yellow or double yellow going back to red as the train passes, then changing back through the sequence as it passes each signal down the line. On a layout in a home you'd be lucky to get two blocks in succession. The best plan is for a signal to got to red when it's passed and change back through the sequence after a couple of timed delays.
Where should you put them, the end of platform roads, if it's a through platform it will be both home and starter. Before points in the main running line, if it's a light it gets a feather to indicate the diverging line is set, if it's a semaphore you get two arms main and subsidiary. The subsidiary is on a bracket on the appropriate side. On the approach to stations, to protect a train standing at the platform. There's a book by CJ Freezer which treets it in a pragmatic way, making it look right which is probably the best most of us can do. I think it's been reprinted recently.
there are really three problems here

1. with signalling, especially in the earlier eras, for every rule there are many exceptions so there is no one "correct" way but many
2. on a model the trains tend to drive the signals not the other way - it can be done but its a pain, indeed in a lot of cases the signals never move but are purely cosmetic (nothing wrong with this)
3. scale compression...

the final point being the most troublesome, e.g. a distant signal is ~1,000 yards from the closest home one (see above about exceptions..) this is the best part of 40' in OO, even if you compress the scale noting your trains are say a quarter of the length of real ones thats still 10' - so actually this sort of branch station is probably the only time you could use one - and then assuming you have a decent length scenic section between the station and the yard.

of course "rule of cool" applies, but in general terms, if the route could be taken by a passenger train, or by a train that could cross such a route - stick a signal there, if it looks decent, if not don't worry too much
User avatar
Wolseley
Posts: 176
Joined: Sun May 22, 2016 8:19 am
Location: Hills District, Sydney, Australia

Re: Semaphore signalling help

Post by Wolseley »

Here's another vote for making sure this book is in your library:

Image
AJFE 2488
Posts: 94
Joined: Sat Jul 17, 2010 2:14 pm

Re: Semaphore signalling help

Post by AJFE 2488 »

I've had a crack at figuring out the signals + positions on the layout and came up with this design. Have based it on all of your suggestions along with how we've got the signals arranged at Kidderminster. Apologies for the handwriting but hopefully you should be able to get the gist of it.

Image

No'1 Bracket took a while to figure out as originally I had separate semaphores but changed it to have 1 with a route indicator to the bay and engine line, 2 more for the main platforms and a disc to the goods yard. Groundframe positions should be obvious on the goods yard points and within the MDP area. No'7 Bracket would most likely be platform mounted to cover the bay platform and the engine line and removes the need for the bay signal on No'4 hence the scribbling out.
User avatar
fourtytwo
Posts: 277
Joined: Mon Oct 31, 2016 7:41 pm
Location: North Lincs UK

Re: Semaphore signalling help

Post by fourtytwo »

AJFE 2488 wrote: Fri Jun 21, 2024 11:32 pm I'm in the process of sorting out multicored cabling for the electronics on my layout and I have been looking at incorporating working signals that are servo controlled.
Beware multi-core & servo's, there fussy little beasts at the best of times and don't much like there short cables extending, far less being bundled in with a load of others! In my case I drive the servo's from as near as practicable under the baseboard. Adding 1-2' of servo extension lead is generally ok but beyond that is often troublesome without "special measures" :wink:
A fresh start in OO, DC Steam
User avatar
Mountain
Posts: 6792
Joined: Mon Oct 24, 2016 3:43 pm
Location: UK.

Re: Semaphore signalling help

Post by Mountain »

AJFE 2488 wrote: Mon Feb 17, 2025 12:35 am I've had a crack at figuring out the signals + positions on the layout and came up with this design. Have based it on all of your suggestions along with how we've got the signals arranged at Kidderminster. Apologies for the handwriting but hopefully you should be able to get the gist of it.

Image

No'1 Bracket took a while to figure out as originally I had separate semaphores but changed it to have 1 with a route indicator to the bay and engine line, 2 more for the main platforms and a disc to the goods yard. Groundframe positions should be obvious on the goods yard points and within the MDP area. No'7 Bracket would most likely be platform mounted to cover the bay platform and the engine line and removes the need for the bay signal on No'4 hence the scribbling out.
I was going to say "Should signal 4 be a single signal and another single signal should be on the left of the track near it inbetween the two tracks..."
Bigmet
Posts: 11004
Joined: Tue Aug 14, 2007 2:19 pm

Re: Semaphore signalling help

Post by Bigmet »

I agree about no 4, two independent starting signals, probably on the platform end. There should also be a ground signal for the double slip, to control movement off the bay platform into the yard.
ChrisGreaves
Posts: 394
Joined: Sun Feb 21, 2021 12:32 pm
Contact:

Re: Semaphore signalling help

Post by ChrisGreaves »

aleopardstail wrote: Wed Jun 26, 2024 9:04 am 2. on a model the trains tend to drive the signals not the other way - it can be done but its a pain, ...
From a logical point of view this is debatable (... so I shall debate it! :lol: )

In Real Life a human standing in a signal box controls the signals (and points) and so governs train movements.
In Modelling Life a human standing near the baseboard controls the trains (by turning dials on a power unit).

I concede that Scale can be a problem, but in the matter of controlling signal and train movements that need not be so. Especially with the move towards electronics and DCC, and cheap laptops available to emit electrical/electronic signals.

Suppose a Real Railway with four signal boxes, one at each station, each signal box housing a human who pulls levers to control train movements on their four miles of track. Those four signalmen between them control sixteen miles of track that serve the four railway stations. This is real Life in the 1940s.
Now replace the mechanical levers with electric push-buttons so that even a 5'6" geriatric weakling like me could cause signal arms and point blades to move. In the automobile world this is called "power steering".

We need a set of buttons or dials that cause a signal arm to move, and that same mechanism can be used to activate a track circuit causing a train to spring into life and move forwards.

Where there is a distant (Yellow) signal urging caution, the DCC current that powers the train can be reduced, so that the train slows down. That Distant" signal is linked to the governing signal further down the track.

I see no logical reason why the locomotive(s) can not be controlled via circuits that control signal arms.

I agree that setting this up might/would be a pain, but that said, regardless of the size or scope of the layout, once the first circuit is developed (a lot of trial, error and, no doubt, muttered imprecations), it serves as a template for all the controls on the rest of the layout.

The costliest part of this plan is effecting that FIRST control; the rest should follow easily.

Indeed, I suppose that a 200-foot long layout could be converted using trained volunteers to implement a template procedure set up by the designers. That is, construction of the component parts, station by station, block by block, signal by signal, could be accomplished by anyone without arthritic hands.

Of course, we would love to have an historic diary of the trials and tribulations for posterity.

Cheers, Chris
aleopardstail
Posts: 1526
Joined: Thu Apr 29, 2021 9:48 pm
Contact:

Re: Semaphore signalling help

Post by aleopardstail »

Well for DCC you have, essentially, described he automation system I am working towards

the MQTT broker "knows" the state of a signal, the software driving the train can query that and take appropriate action - hence if something else changes that signal, they train will still obey it.

it also means signals default to the danger aspect, and clear when a route is set, not as is often seen in model form (and in reality an longer sections) being purely automatic "next block is clear"

personally I love it, its overly complicated and frustrating but love it
AJFE 2488
Posts: 94
Joined: Sat Jul 17, 2010 2:14 pm

Re: Semaphore signalling help

Post by AJFE 2488 »

Wolseley wrote: Sat Jun 29, 2024 12:32 pm Here's another vote for making sure this book is in your library:

Image
Haven't made any further progress with the signalling planning since my last post but I have been able to get hold of a 2nd hand copy of that book earlier today for a couple of quid from a sales wagon after doing a bit of lighting up this morning. Hopefully that'll give me a lot more guidance on this as you guys have recommended.

As for powering the signals, I do intend on having servos near the signals with a control board sited nearby for each cluster. That'll eliminate the multicore on servo issue that was previously mentioned but I would still want to be able to control all signals from the layout mimic board.

Contrary to the discussion that has gone on about automating trains with signals, as interesting as it is, that is not something I plan on implementing with my layout as I would prefer to just stick with a controller for the locomotives with separate buttons/ levers for accessories. If I was doing another layout or a large club layout then yes maybe but not for this one.
User avatar
Mountain
Posts: 6792
Joined: Mon Oct 24, 2016 3:43 pm
Location: UK.

Re: Semaphore signalling help

Post by Mountain »

I didn't know Mr Freezer has written a signalling book. I know he wrote a lot of books!
Mike Parkes
Posts: 858
Joined: Sat Sep 01, 2007 8:25 pm

Re: Semaphore signalling help

Post by Mike Parkes »

I would suggest
1 should have a distant equivalent on the approach to it somehere before passing the turntable
Around the same location going the other way there would be an "outer" home signal marking the limit of shunting movements and only trains cleared to the section alllowed to pass it
2 would only be ground signal
4 would solely relate to the main platform and ought to be sited to the left of tracks if possible
I do not see a need for a signal equivalent to 4 for the line from the double slip as any movements would be cleared by the signal for the double slip.
7 should relate soley to the bay paltform and a separate ground signal provided for the siding below that
The station loco release crossover should have appropriate ground signals on all approaches to it - only one set is shown
Post Reply