Alsop-en-le-Dale Signalling Diagram: Help Now Needed with Interlocking

[Tortuga]

Thanks once again to everyone who helped me better understand Alsop-en-le-Dale’s signalling diagram. If you haven’t guessed, my eventual aim is to build and operate a model of this station as prototypically as I can, including fully interlocking the points and signals.

While construction of the layout is temporarily on hold (as explained in my layout thread)

Todays slack time on site has been spent trying to devise a locking table. With reference to what has already been said on this topic (particularly by  @The Stationmaster), here’s what I’ve come up with so far: am I on the right track?

(please refer to the linked signalling diagram

https://signalbox.org/~SBdiagram.php?id= 1057

as it seems to be the most complete)

 

Lever 1: released by 2 and 3 (both when reversed

Lever 2: locks 1 when normal, locks 10 and 23 and releases 1 when reversed.

Lever 3: locks 1 when normal, locks 10 and 24 and releases 1 when reversed and is released when 16 is reversed.

Lever 6: locks 7, 12 and 24 when reversed and is released when 8 is reversed.

Lever 7: locks 3, 8, 15, 16 and 24 when reversed.

Lever 8: locks 6 and 11 when normal and 24 when reversed.

Lever 9: locks 10.

Lever 10: locks 23 when normal and 2 when reversed. Is released when 9 is reversed.

Lever 11: locks 6 and 12 when normal, 8 when reversed and is released when 8 is reversed.

Lever 12: locks 6 when reversed and is released when 11 is reversed.

Lever 13: locks 3, 16 and 24 when reversed and is released when 14 is reversed.

Lever 14: locks 13 when normal, 15 when reversed and is released when 15 is reversed.

Lever 15: locks 14 when normal, 16 when reversed and is released when 17 is reversed.

Lever 16: locks 3 when normal, 13 and 24 when reversed and is released when 17 is reversed.

Lever 17: locks both 15 and 16 when normal.

Lever 23: locks 25 when normal, 8, 15 and 16 when reversed and is released when 10 is reversed.

Lever 24: locks 25 when normal and 8,15 and 16 when reversed.

Lever 25: is released when 23 and 24 are reversed.

 

Hope that’s clear enough! If not I’ll need some advice on how to share an Excel file...

Edited June 7, 2021 by Tortuga
Link to layout thread added

[Grovenor]

You have locks and releases muddled. The converse of 'released by' is 'releases'.

eg. If 1 is released by 2, then 2 releases 1.

If you attach your excel file we can chrck and amend where needed.

You just drag and drop the excel file in, see the paperclip below when you are writing a post.

[bécasse]

I suspect that the two home signals will be released by the levers for their adjacent FPLs standing normal, rather than the points themselves (although 24 will require 8N as well, on the assumption that 8R releases 11 and 15R releases 14). As I indicated above, the LNWR tended to be very parsimonious in its signalling equipment and chose the cheapest ways to implement locking. It was by no means unknown for LNWR distant signals to be locked by bars fixed to the levers which overlapped the home and starter levers, physically preventing them from being pulled unless both stop signal levers had already been pulled - however, given the date of the installation, it is probably unlikely here. One other issue that you have missed is that 2R will lock 24N and 24R will lock 2N so that trains can't enter the station from both directions at the same time.

 

I think that Keith has had a lot more experience with LNWR locking than I have and is probably in a better position to comment further.

[Tortuga]

Alsop Interlocking Table.xlsxRight. I’ve sorted an Excel version of the locking table. I’ve probably missed columns I should have and included columns I shouldn’t, but hopefully it’ll be clearer to those of you who are more used to this stuff than me!

Please feel free to poke it around and amend as necessary! I’ve not included any of the changes suggested so far in case my previous post inadvertently introduced confusion

Alsop Interlocking Table.xlsx

 

Edited June 7, 2021 by Tortuga
I’m an idiot (forgot to add the excel file)

[Grovenor]

Looks like you forgot to upload it.

[Tortuga]

3 hours ago, bécasse said:

I suspect that the two home signals will be released by the levers for their adjacent FPLs standing normal, rather than the points themselves

Wouldn’t that mean the homes could be cleared if the points were set for their diverging routes though? I thought an FPL was reversed to unlock the point, the point would then be reversed, then the FPL would be returned to normal in order to lock the point in its new position? Or have I got that completely backward?

 

3 hours ago, bécasse said:

One other issue that you have missed is that 2R will lock 24N and 24R will lock 2N so that trains can't enter the station from both directions at the same time.

I might’ve completely misunderstood your meaning here: do you mean at exactly the same time or that one train can’t enter in one direction while another is already in the station facing the other way?

I’m confused because Alsop was the crossover point for the daily trip workings and (allegedly) in the final years of the weekly mineral working from Alsop Moor Quarry, the length of the mineral train meant it had to “shuffle” forward once the passenger service had entered the down platform in order to let the passenger train depart!

[Michael Hodgson]

38 minutes ago, Tortuga said:

I thought an FPL was reversed to unlock the point, the point would then be reversed, then the FPL would be returned to normal in order to lock the point in its new position? Or have I got that completely backward?

 

There's two ways of doing it, and of course, sod's law applies, both are used (sometimes even at the same box, though I can't think of an example).  I have even seen cases where lever normal locks one point, lever reverse locks a different point.

1. Lever Reversed = Locked (the more usual situation)
2. Lever Normal = Locked  (as you described)

 

[Michael Hodgson]

48 minutes ago, Tortuga said:

I might’ve completely misunderstood your meaning here: do you mean at exactly the same time or that one train can’t enter in one direction while another is already in the station facing the other way?

 

On a single line if trains can enter the station at the same time there is a risk that one of them might overrun the loop exit points and collide with the other one as it arrives.  So they are not allowed to do so.  If trains have to cross, you keep both home signals at Danger,  When the first arrives, you lower the Home signal and let it into the loop.  Once it has safely come to a stand, clear of the points at both ends you let the second one in.  So the opposing home signals would be locked against each other. 

 

There's also a rule about restoring signals to danger where there are facing points - you must wait until the whole of the train is clear of the points.  That's a precaution against moving the points under the train.

 

There are a handful of exceptions - one survives on the West Somerset Railway, but only because the risk is lower because the station in question happens to be at the summit of a hill so there is less risk of a train being unable to stop.

[Tortuga]

Thanks @Michael Hodgson for explaining those two points; you’re explanation for locking the homes against each other makes perfect sense now - it hadn’t occurred to me (but should have) that a train might overrun the loop.

I can’t see that 17 would lock one point when normal and the other when reversed as both would need to be locked at the same time, so I assume the “more usual situation” of 17 being reversed to lock the points is more likely?

 

Out of curiosity, is there a particular reason for the ‘lever reversed = locked’ being more common? Was (is?) it to act as a visual reminder to the signalman that the FPL is ‘on’?

[micknich2003]

10 hours ago, Michael Hodgson said:

There's two ways of doing it, and of course, sod's law applies, both are used (sometimes even at the same box, though I can't think of an example).  I have even seen cases where lever normal locks one point, lever reverse locks a different point.

1. Lever Reversed = Locked (the more usual situation)
2. Lever Normal = Locked  (as you described)

 

Here is an example of a FPL lever locking "Both ways".

[beast66606]

Locks normal is the implicit understanding, locking reverse (when special circumstances need it) is stated to make things clear.

 

You don't need the locked when normal column.

 

Taking No.1 - the distant, it is released by (reversing) 2 and 3 - neither of those will be described as locking 1, even though they do, the fact that 1 is released by 2 and 3 is enough and adding that column can cause problems viz :

 

You have lever 9 normal locking lever 10 when normal, so how do you ever change points 10 ? - if lever 9 is reversed the points are mechanically locked "on the ground".

 

Personally I'd hang the locking off the points themselves and have the FPLs as mostly mechanical extras. so signal 2 locks lever 10 but is released by 9, which itself locks 10  (as it stands you can pull off 2 whenever you feel like it, same with 24)

 

Have another go at the locking as there are several mistakes and use the info above to help you - you need to focus on the locks (and remember locking will cascade so you don't need to list all levers all the time - eg 6 will lock 8, no need to mention 11 as 11 will be released by 8 so its implied and doesn't need adding to the locked list for 6) and released by columns and get rid of locks when normal.

Edited June 8, 2021 by beast66606

[Tortuga]

I’d added the ‘locks when normal’ column to make it clear to me which lever locked which, but, as you say, duplicated this info in the ‘released by’ column and confused myself.

 

I think I’ve muddled the locking of the FPL levers with the mechanical locking at the points - from what has been said, if the FPL lever is reversed, the mechanical lock on the ground is locking the point; correct?

 

I think I’ve over complicated it by getting the impression that everything has to lock everything else. I realise I’ve muddled up 6 and 7 - I reread your earlier post - and I’ve got them the wrong way round.

 

I’ll give it another go, taking the guidance provided into account: are there any other columns needed? I don’t think I need a ‘locks both ways’ one?

[beast66606]

4 minutes ago, Tortuga said:

I’d added the ‘locks when normal’ column to make it clear to me which lever locked which, but, as you say, duplicated this info in the ‘released by’ column and confused myself.

 

I think I’ve muddled the locking of the FPL levers with the mechanical locking at the points - from what has been said, if the FPL lever is reversed, the mechanical lock on the ground is locking the point; correct?

 

I think I’ve over complicated it by getting the impression that everything has to lock everything else. I realise I’ve muddled up 6 and 7 - I reread your earlier post - and I’ve got them the wrong way round.

 

I’ll give it another go, taking the guidance provided into account: are there any other columns needed? I don’t think I need a ‘locks both ways’ one?

 

If you need the locks when normal info for model wiring etc. I'd add that at the end.

FPL = facing point lock = locks the facing point mechanically on the ground so the point can't be moved (there are other bits and bobs associated with this but in simple terms, it's locked)

I doubt you need a locks both ways column as you suggest.

 

Start with No.1 and think "What releases this ?" and then "When pulled what does it lock ?" and exclude anything in the "What releases this ?" column.

 

eg

 

No.1 - What releases this ? 2,3, When pulled what does it lock ? <2,3 rev> but they are already in the release column so nothing.

etc.

No.25 - What releases ....

 

Be methodical.

[beast66606]

Prototype example - Eccles Road, 1960

 

[Michael Hodgson]

4 hours ago, Tortuga said:

Out of curiosity, is there a particular reason for the ‘lever reversed = locked’ being more common? Was (is?) it to act as a visual reminder to the signalman that the FPL is ‘on’?

Probably just easier to implement - you'd have to ask a locking fitter.

Points governing entrance to the platform need to be locked Normal.  You wouldn't need to lock them reverse, because when they're set that way they are only used in the trailing direction and for shunt moves such as running round.  So the FPL would be one of the  releases on that Home.

[bécasse]

I have had a go at producing a locking table and, since it has been linked to several times already, I have included John Hinson's diagram to assist comprehension - please remember that this is copyright and shouldn't be reproduced elsewhere.

This locking table was only a quick draft so there may well be some errors in it and I am quite happy to amend it in response to comments.

Alsop-en-le-DaleLNWR.pdf

[Grovenor]

Well, beaten to it

 

Here's my take on it.

Alsop Interlocking Table-kn.xlsx

Edited June 8, 2021 by Grovenor
error

[Tortuga]

My thanks gentlemen. You both beat me to posting my own second attempt, so I’m going to spend any dead time on site tomorrow comparing your takes with mine to see how they compare!

[meil]

10 hours ago, bécasse said:

I have had a go at producing a locking table and, since it has been linked to several times already, I have included John Hinson's diagram to assist comprehension - please remember that this is copyright and shouldn't be reproduced elsewhere.

This locking table was only a quick draft so there may well be some errors in it and I am quite happy to amend it in response to comments.

Alsop-en-le-DaleLNWR.pdf 458.78 kB · 8 downloads

What stops 2 being reversed with 9 & 10 reversed?

 

Also, if I read this correctly, why should 9 be reversed to reverse 23? You wouldn't normally put a FPL in for a trailing move.

[Michael Hodgson]

2 hours ago, meil said:

What stops 2 being reversed with 9 & 10 reversed?

 

Also, if I read this correctly, why should 9 be reversed to reverse 23? You wouldn't normally put a FPL in for a trailing move.

There are differences between the two tables and I find the second easier to understand ...

 

Using Grovenor's locking, 2 is released by 9, so that has to be reversed.

10 must be Normal as it is locked in that position by pulling 9.

So the sequence is

1. ensure 10 lever Normal
2. pull 9, locking 10 in that position, releasing 2
3. pull 2

Again using Grovenor's version, 9 isn't reversed for 23.  10 must be reverse and that locks 9 Normal.

 

The pull plates list he added at the bottom helps to understand it from the signalman's point of view, and these plates would be attached to or fixed besides the levers concerned.  It is always easier to understand when you start with everything normal in the frame and think about the sequence in which things have to happen :

1. set all the points needed for the movement
2. lock them
3. clear the stop signals in the order in which they will be encountered
4. pull the distant if applicable
    

 

 

[beast66606]

Moving forward with Keith's (Grosvenor) version

 

Pull plate for 1 would not need 9 and 16, these are handled by 2 and 3 respectively.

 

11 - there's an argument that 11 could be released by 8 and 16, which would then lock out 15 and 17 so the pull plate would become 16, 8

13 - pull plate would probably only show 14

15 - the route into the siding (which would be flagged) could use the FPL so 15 would not lock 17, local instructions would state points 15 need to be locked if used in a facing direction into the yard - or similar wording.

 

25 - same comment as 1 for the pull plate, no need for 10 and 17.

 

 

[Grovenor]

40 minutes ago, beast66606 said:

Moving forward with Keith's (Grosvenor) version

Pull plate for 1 would not need 9 and 16, these are handled by 2 and 3 respectively.

11 - there's an argument that 11 could be released by 8 and 16, which would then lock out 15 and 17

15 - the route into the siding (which would be flagged) could use the FPL so 15 would not lock 17, local instructions would state points 15 need to be locked if used in a facing direction into the yard - or similar wording.

Pull plates, matter of opinion, I prefer to show all the levers that need to be pulled.

11 released by 16, I wouldn't, what do you see as the advantage?

17 locks 15 both ways, why add that complication, there are not going to be any passenger moves into the yard?

[meil]

1 hour ago, Michael Hodgson said:

There are differences between the two tables and I find the second easier to understand ...

 

Using Grovenor's locking, 2 is released by 9, so that has to be reversed.

10 must be Normal as it is locked in that position by pulling 9.

So the sequence is

1. ensure 10 lever Normal
2. pull 9, locking 10 in that position, releasing 2
3. pull 2

 

 

Yes, that's the normal routing but what stops me pulling 10 reversed, 9 reversed then 2 reversed?

[beast66606]

21 minutes ago, Grovenor said:

Pull plates, matter of opinion, I prefer to show all the levers that need to be pulled.

11 released by 16, I wouldn't, what do you see as the advantage?

17 locks 15 both ways, why add that complication, there are not going to be any passenger moves into the yard?

 

You may - but the LNWR didnt

 

Flank protection but it's the usual devil and deep blue sea, setting the points to the wrong loop protects a shunt move but may divert a train into a head on.

 

We don't know if the points into the yard were locked when reversed or not so have offered an alternative, there's an argument the points could have been locked as the mechanical side would be set up anyway - without knowing the details we cannot be certain.

[Michael Hodgson]

2 hours ago, meil said:

Yes, that's the normal routing but what stops me pulling 10 reversed, 9 reversed then 2 reversed?

After you have pulled 10, you can't pull 9 because 10 locks it.  As you can't pull 9, you still can't pull 2.

 

It would be different if 9 locked 10 both ways. 

If there were another signal allowing you to run into the wrong platform, 9 would indeed need to lock 10 both ways.  It would also be necessary for both the levers for the signals into both platforms to look additional at the position of 10, since the FPL on its own would not be enough.  In addition to adding another signal and a lever, this would make the locking more complicated - good reason not to provide a signal for the "wrong" platform unless you have a genuine traffic reason for doing so.  

 

It's also one of the reasons you don't lock trailing points.  There is another reason, in that if a train runs through a trailing point which is incorrectly set more damage is caused if that point is bolted by the FPL.