Signalling scheme help please!

[BluenGreyAnorak]

There you go. Slightly revised with the lie of the points altered as suggested:

 

 

Not quite sure how to illustrate a wide-to-gauge trap...

[TheSignalEngineer]

This document, although now probably superseded by a Network Rail edition, is good guidance plan symbols. Section 4.1.3 shows what you need, item 12 on page 12 of 30. Trains run from left to right on these, so you will need to reverse it for your case.

[WillCav]

This document, although now probably superseded by a Network Rail edition, is good guidance plan symbols. Section 4.1.3 shows what you need, item 12 on page 12 of 30. Trains run from left to right on these, so you will need to reverse it for your case.

 

The document has been superseded but the symbol remains the same.

 

WillCav

[BluenGreyAnorak]

This document, although now probably superseded by a Network Rail edition, is good guidance plan symbols. Section 4.1.3 shows what you need, item 12 on page 12 of 30. Trains run from left to right on these, so you will need to reverse it for your case.

 

That's a rather excellent link - thanks very much

 

Having thought about it a bit more, I'm not sure whether to include the trap on my diagram or not. Whilst I quite like the idea of it looking like a real signal box, in practice of course, it's a control panel for my model and cluttering it with non-operational details may be a bit unnecessary. I'm not about to start learning the black art of track building but I guess the compromise might be to use the Peco catch point as a working representation?

[The Stationmaster]

That's a rather excellent link - thanks very much

 

Having thought about it a bit more, I'm not sure whether to include the trap on my diagram or not. Whilst I quite like the idea of it looking like a real signal box, in practice of course, it's a control panel for my model and cluttering it with non-operational details may be a bit unnecessary. I'm not about to start learning the black art of track building but I guess the compromise might be to use the Peco catch point as a working representation?

 

I think your best course is to make a dummy representation of it as w-t-g traps are fairly distinctive in appearance so perhaps just use plain track with a couple of bits of filed rail closely alongside in the appropriate place? (BTW all being well signalling sketch tomorrow - assuming lunch today in honour of daughter's birthday is going to leave me rather knackered thus ruling out today. But because of the track layout the signalling will be very straightforward.)

[BluenGreyAnorak]

That sounds like a nice compromise too

 

I'll look forward to your sketch, Mike. It is very much appreciated. Whilst I'm some way away from planting signals, I have some free time over the upcoming half term week (children visiting grandparents ) so building the control panels is pretty imminent.

[BluenGreyAnorak]

Hi Mike,

 

Did you have chance to look any further at this, please? Control panel construction is on the cards this week

[WillCav]

BluenGreyAnorak,

 

I've made a start on a suggested signalling design. I have got a few more questions/assumptions to check before I can finish.

 

1) Which platforms do you want to allow platform sharing for arriving trains? I assume platform 4 will be as you may wish to stack DMUs there - The more platform sharing allowed, the more interesting the home signal becomes (and harder to build) due to more draw ahead arms (aka calling on arms).

 

2) The ideal situation would be to have an inner / outer home combination on the Down but there is not the space for that here - the next best would be one home with block acceptance point clear of the throat (better operationally - you can bring a train to the home signal whilst shunting the station throat). This does however make the home signal a fair distance from the station (400yds = c16ft on the real railway but this could be compressed to say 8ft to ensure a train stopped here does not foul the double junction in rear). Are you happy (visually) with a home signal this far away from the station?

 

3) I have assumed that you do not need a move from the Down road into the middle siding as this is used for running round predominantly - please advise if this is not the case.

 

4) I have assumed that all shunting is to take place on the Up road to keep the Down road clear for incoming services.

 

Regards

 

WillCav

[The Stationmaster]

Hi Mike,

 

Did you have chance to look any further at this, please? Control panel construction is on the cards this week

 

Sorry about the delay - certain 'domestic requirements' took priority then came Railex. i'll try to knock together a sketch this evening if I'm still awake as it's nice and straightforward (I've already said that several times I'm sure).

[The Stationmaster]

 

 

2) The ideal situation would be to have an inner / outer home combination on the Down but there is not the space for that here - the next best would be one home with block acceptance point clear of the throat (better operationally - you can bring a train to the home signal whilst shunting the station throat). This does however make the home signal a fair distance from the station (400yds = c16ft on the real railway but this could be compressed to say 8ft to ensure a train stopped here does not foul the double junction in rear). Are you happy (visually) with a home signal this far away from the station?

 

 

If you can find me an example of that anywhere on the GWR/Western Region I would be very interested to learn of it! (BTW it would be 440yds for a normal Clearing Point until such time as it was relaxed to 200yds for a c.l. distant). Nothing at all wrong with an ABR 5 acceptance in these circumstances.

[The Stationmaster]

That sounds like a nice compromise too

 

I'll look forward to your sketch, Mike. It is very much appreciated. Whilst I'm some way away from planting signals, I have some free time over the upcoming half term week (children visiting grandparents ) so building the control panels is pretty imminent.

 

Right - here it is (so you can work out how long it took when I got down to it)

A couple (well several probably) of items to note -

1. All disc signals are 'red', no yellows at all (but there are only 7 of 'em anyway).

2. The release crossovers are ground frame operated which they are most likely to have been on the real thing at this date - basically 2 levers in a ground for each crossover.

3. I have used a route indicator on the innermost Home Signal which - again - is the most likely prototype situation at this date otherwise it is a 5 doll bracket (which would no doubt look much more imposing, the choice is yours).

4. I have shown what on lesser railways would be called an Outer Home Signal as 'optional' because it provides some operational flexibility. here you are going to be into 'selective compression' and I would say a 'good train length plus a coach' is where to put it in rear of the next signal in advance (the one protecting the trailing crossover) but in view of the Advanced Starting Signal coming up from the junction you might start to look over-signalled. My advice then would be to do it whichever way looks best from an overall appearance viewpoint and adjust your operating practice and timetable to suit. An area you might wish to talk through further (there is a further option as well but you might think it too rrestrictive on how you want to run your trains so I'll ignore it for the moment.

5. Note also that whatever else you might have you have an Intermediate Home Signal protecting the trailing crossover from Down trains.

6. The Advanced Starting Signal (with lower arm distant for the junction) should be sufficiently in advance of the trailing point end in the Up line to allow a train to shunt without entering the advance block section.

 

 

[BluenGreyAnorak]

Thanks Mike, that looks excellent. As you no doubt guessed though, I do have questions...

 

3. I have used a route indicator on the innermost Home Signal which - again - is the most likely prototype situation at this date otherwise it is a 5 doll bracket (which would no doubt look much more imposing, the choice is yours).

Frankly, the thought of building an operational 5 doll bracket gives me the willys so I'm happy to stick with a route indicator What form would it take, though? Would it be one of those 'dot matrix' theatre boxes or something else? There also seems to be a small arm below the main arm and a ground signal: what are the purposes of those?

 

4. I have shown what on lesser railways would be called an Outer Home Signal as 'optional' because it provides some operational flexibility. here you are going to be into 'selective compression' and I would say a 'good train length plus a coach' is where to put it in rear of the next signal in advance (the one protecting the trailing crossover) but in view of the Advanced Starting Signal coming up from the junction you might start to look over-signalled. My advice then would be to do it whichever way looks best from an overall appearance viewpoint and adjust your operating practice and timetable to suit. An area you might wish to talk through further (there is a further option as well but you might think it too rrestrictive on how you want to run your trains so I'll ignore it for the moment.

I'm assuming that the function of both the Advanced Starter and the Outer Home are to enable trains to be effectively 'queued up' on the approach to the station? If so, then I'll forego the latter as, although it would increase capacity, the trains would be almost nose to tail.

 

There's one other detail that I'm not too sure about. There are three ground signals adjacent to points on the approaches to platforms 1, 2 and the central storage road. In each case, do they only serve the diverging road of each set of points or do they cover both routes with a straighforward 'proceed' or 'don't proceed'?

[The Stationmaster]

Been busy today so I'll come back tomorrow on this but the first item is easy - the route indicator would be the (G)WR mechanical version - sometimes referred to as 'the cash register' - a rather poor pic of one (below a centre pivot arm) is added below. The small arm is a 'Calling On' signal and is used to signal a movement into an occupied platform.

 

 

[BluenGreyAnorak]

Thanks for that Mike. I certainly won't be planning an operational one of those

 

I've been beavering away on the details of my control panel, which has included working out how the various combinations of points will be operated and which signals relate to them. I'm being fairly pragmatic here - whilst I want the flavour of a signal box, I don't want to ignore the short-cuts that being an electric model allow. Therefore, as I'm using tortoise motors, where two points are related they'll be operated via one switch and a link between the motors themselves.

 

I do want the points to be interlocked with the associated signals, even if only electrically, and as an operator aid I've decided to include LED repeaters for the signals on the panel. This is particularly relevant to the ground signals: I'm not sure if I'll ever make them as working models, but from an operational perspective each will have a lever (switch) and be interlocked with the associated pointwork. That's the plan anyway - I've been working on a wiring diagram for this lot and it's pretty fearsome!

 

It's not quite finished yet but attached is a copy of my diagram as it will appear on the panel itself showing both the station and junction areas. The image is about half the size of what the actual print will be.

 

 

I've tried to group the switches for points and signals in a logical manner and, as you can see, I've colour coded them as if it were a real signal box. Ground signals are indicated by their number being inside a circle on the coloured tab.

 

Seeing them both together like this does raise another question too. Assuming that each area were controlled by its' own box, how would the distant (S2) on the station departure line be controlled? It strikes me that its' setting is wholly dependant on the signal protecting the approach to the junction with the main line (S14 on the lower diagram) but in the real world they could easily be miles apart. Would it just rely on communication between the boxes or would it be electrically interlocked in some way?

 

In addition to all of this, I have actually constructed the carcass of the control panel itself and installed it on the layout:

 

 

 

 

I'm really pleased with the way it's come out. All of the wiring gubbins will be laid out inside the base, which should make for easy construction and maintenance, and there's even a handy place for my controller to sit

[The Stationmaster]

I've been beavering away on the details of my control panel, which has included working out how the various combinations of points will be operated and which signals relate to them. I'm being fairly pragmatic here - whilst I want the flavour of a signal box, I don't want to ignore the short-cuts that being an electric model allow. Therefore, as I'm using tortoise motors, where two points are related they'll be operated via one switch and a link between the motors themselves.

 

Good - all quite logical I think.

 

 

I do want the points to be interlocked with the associated signals, even if only electrically, and as an operator aid I've decided to include LED repeaters for the signals on the panel. This is particularly relevant to the ground signals: I'm not sure if I'll ever make them as working models, but from an operational perspective each will have a lever (switch) and be interlocked with the associated pointwork. That's the plan anyway - I've been working on a wiring diagram for this lot and it's pretty fearsome!

 

You do of course need to decide what you will use as your 'interlocking machine' - clearly the switches will not be interlocked (unless you have some cunning plan?) so it will be purely electrical interlocking between the switch and the function and the first thing you must decide is just how far you want that to go (draw a simple locking chart) and then how you will do it. Depending on the type of switches you use something might be possible with them if they have electrically separate poles and a lot of the locking you will need is fairly straightforward. Using relays is relatively simple - you can see what they are doing but they use power and have to be obtained, surplus GPO ones are a bit thin on the shelves nowadays. The next simplest alternative is probably using simple logic ICs with AND & NAND etc gates - can you still get them even? And finally you get to programmable processors - and pass beyond my knowledge area But whatever you use you should start by drawing up a simple locking chart because you will at least then know what you have got to do even if you haven't sussed a way to do it.

 

It's not quite finished yet but attached is a copy of my diagram as it will appear on the panel itself showing both the station and junction areas. The image is about half the size of what the actual print will be.

I've tried to group the switches for points and signals in a logical manner and, as you can see, I've colour coded them as if it were a real signal box. Ground signals are indicated by their number being inside a circle on the coloured tab.

 

Looks ok - one yellow switch too many tho' (S2 is not needed - the terminus has a Distant Fixed At Caution - the one below S13).

 

Seeing them both together like this does raise another question too. Assuming that each area were controlled by its' own box, how would the distant (S2) on the station departure line be controlled? It strikes me that its' setting is wholly dependant on the signal protecting the approach to the junction with the main line but in the real world they could easily be miles apart. Would it just rely on communication between the boxes or would it be electrically interlocked in some way?

 

S2 would be worked by the 'box at the junction - by yellow switch (=lever No 15). This signal could only be cleared when 14, 9, and 3 have been cleared. You could of course not bother with switch 15 and make the distant automatic depending on the three stop signals being off, not very prototypical but it saves a switch. Also the distant and the stop signal (S1) will be 'slotted' to ensure that the Distant will only clear to off when the stop signal above it is 'off' and for modelling purpose electrically slotting the controls is very simple.

Hope all that has helped a bit but I think your control panel is very definitely heading the right way and while not prototypical for numbering arrangement it is, I reckon, very well laid out for operational purposes - almost like some OCS panels in the real world with the switches disposed geographically.

[BluenGreyAnorak]

You do of course need to decide what you will use as your 'interlocking machine' - clearly the switches will not be interlocked (unless you have some cunning plan?) so it will be purely electrical interlocking between the switch and the function and the first thing you must decide is just how far you want that to go (draw a simple locking chart) and then how you will do it.

No, it will just be electrical interlocking I'm, afriad. A proper mechanical solution is just too complicated, so my operators will just have to learn to watch the panel

 

S2 would be worked by the 'box at the junction - by yellow switch (=lever No 15). This signal could only be cleared when 14, 9, and 3 have been cleared. You could of course not bother with switch 15 and make the distant automatic depending on the three stop signals being off, not very prototypical but it saves a switch. Also the distant and the stop signal (S1) will be 'slotted' to ensure that the Distant will only clear to off when the stop signal above it is 'off' and for modelling purpose electrically slotting the controls is very simple.

Ah ha, I thought so! I'll lose S2 and arrange for it to operated automatically as you suggest. Does it really need to be interlocked with all three following stop signals though? I'm particularly thinking of light engine moves from the station to the depot that would never pass S3 and wouldn't require S9 to be 'off' either?

 

Perhaps I'm mis-undertanding how S14 should be used though? I'd assumed it was there to sub-divide the section between the station and junction to increase capacity. I imagined that whilst a train was being held at the junction, a second would be allowed to draw up to S14 in order to clear the station. Conversely, with only one train in the section I thought it would be able to depart the station and have S14 'off' even if it is to be held at S9?

 

Hope all that has helped a bit but I think your control panel is very definitely heading the right way and while not prototypical for numbering arrangement it is, I reckon, very well laid out for operational purposes - almost like some OCS panels in the real world with the switches disposed geographically.

Oh yes, enormously. Thanks very much

[The Stationmaster]

Ah ha, I thought so! I'll lose S2 and arrange for it to operated automatically as you suggest. Does it really need to be interlocked with all three following stop signals though? I'm particularly thinking of light engine moves from the station to the depot that would never pass S3 and wouldn't require S9 to be 'off' either?

 

So those engines would pass the Distant Signal at caution (i.e. it would not be 'off'). A Distant Signal can only be cleared when all the stop signals which apply to the same line in the same direction at a signal box have been cleared - hence me listing all of the stop signals in that direction at the junction which are applicable from the terminus towards the principal route. (It can sometimes be more complicated than that but at this point you only need to know the basic situation for your layout and the relevant WR practice - we've n need to get involved with complexities such as splitting distants fr junctions or 'hurry-up' distants on the GC mainline and so on.)

 

 

Perhaps I'm mis-undertanding how S14 should be used though? I'd assumed it was there to sub-divide the section between the station and junction to increase capacity. I imagined that whilst a train was being held at the junction, a second would be allowed to draw up to S14 in order to clear the station. Conversely, with only one train in the section I thought it would be able to depart the station and have S14 'off' even if it is to be held at S9?

 

Not so much 'misunderstanding' as 'not fully understanding' S14 is the Outer Home Signal (in non-Western terminology, on the Western it would be called the Home Signal) for the junction and what it allows you to do is have a train leave the terminus towards the junction at the same time as a train has been accepted towards/is approaching the junction on the Main Line. It does this because in the real world it provides an extra safety margin (a quarter of a mile in old money) between the first stop signal and the junction (or to put it another way - the spot where two trains could collide). In Absolute Block semaphore signalling - which is basically what you are replicating in your layout of signals - there is something called a 'Clearing Point' which runs for 440 yards = quarter mile) in advance of the rearmost stop signal at a signal box (and before anyone shouts I am ignoring colour light Distant Signals here).

In order for a Signalman to accept a train that 440yds must be clear with no conflicting movements being made or signalled or accepted on the block. By inserting S14 we have moved that 440yds - selectively compressed for the 4mm scale world - back clear of the junction. So a train can leave the terminus for the junction provided the line is clear as far as S9. Any following train would have to be held at S1 until the first train has passed in advance of S9 and the signalbox at the junction. Further - any train held at S14 is only allowed to approach S9 once certain conditions have been met - the junction is set for S9 (i.e. P20 is reversed) and no train has been accepted or is approaching on the Main Line unless it has been brought to a stand at S11/12. Now re-read this paragraph - it's fairly simple once you get the hang of it and might become an important factor if you start involving yourself in timetable operation and the like, on the other hand you might not want to work the signals prototypically but at least they will look right. If you're at Members' Day we can have a full run through of the way you would work your signalling if you want to mimic the real world (or I can pop over and take you through it as I'm not too far away).

[Grovenor]

Just one little point on your diagram, point P9 needs its normal lie reversed so it leads into platform one. It then works as a crossover with P10, as you have done with the actual switches.

Regards

Keith

[BluenGreyAnorak]

So those engines would pass the Distant Signal at caution (i.e. it would not be 'off'). A Distant Signal can only be cleared when all the stop signals which apply to the same line in the same direction at a signal box have been cleared - hence me listing all of the stop signals in that direction at the junction which are applicable from the terminus towards the principal route.

Ah, I see. I understand now.

 

Not so much 'misunderstanding' as 'not fully understanding' S14 is the Outer Home Signal (in non-Western terminology, on the Western it would be called the Home Signal) for the junction and what it allows you to do is have a train leave the terminus towards the junction at the same time as a train has been accepted towards/is approaching the junction on the Main Line. It does this because in the real world it provides an extra safety margin (a quarter of a mile in old money) between the first stop signal and the junction (or to put it another way - the spot where two trains could collide). In Absolute Block semaphore signalling - which is basically what you are replicating in your layout of signals - there is something called a 'Clearing Point' which runs for 440 yards = quarter mile) in advance of the rearmost stop signal at a signal box (and before anyone shouts I am ignoring colour light Distant Signals here).

In order for a Signalman to accept a train that 440yds must be clear with no conflicting movements being made or signalled or accepted on the block. By inserting S14 we have moved that 440yds - selectively compressed for the 4mm scale world - back clear of the junction. So a train can leave the terminus for the junction provided the line is clear as far as S9. Any following train would have to be held at S1 until the first train has passed in advance of S9 and the signalbox at the junction. Further - any train held at S14 is only allowed to approach S9 once certain conditions have been met - the junction is set for S9 (i.e. P20 is reversed) and no train has been accepted or is approaching on the Main Line unless it has been brought to a stand at S11/12. Now re-read this paragraph - it's fairly simple once you get the hang of it and might become an important factor if you start involving yourself in timetable operation and the like, on the other hand you might not want to work the signals prototypically but at least they will look right.

Ah, ha. That makes more sense too I do want to run a timetabled operation so this sort of detail is excellent.

 

If you're at Members' Day we can have a full run through of the way you would work your signalling if you want to mimic the real world (or I can pop over and take you through it as I'm not too far away).

Thanks Mike. I can't make Members Day unfortunately, but I'd love to take you up on your offer. I want to work on buildng the control panel over the next few weeks and perhaps then we can sort something out.

[BluenGreyAnorak]

Just one little point on your diagram, point P9 needs its normal lie reversed so it leads into platform one. It then works as a crossover with P10, as you have done with the actual switches.

Regards

Keith

I can't believe I missed that. I must have been staring at it too long