Signalling Burbage

[The Stationmaster]

2 hours ago, RailWest said:

Note also that:-

 

a). 3 and 6 would lock each other, although that might be conditional on whether 9 was normal or reverse

b) 3 should lock 9 both ways

c) 10 should lock both 5 and 7 both ways.

Agree, although I doubt there would be any conditional locking on an old frame without tappet locking which is presumably what existed at Burbage.

[RailWest]

5 hours ago, The Stationmaster said:

Agree, although I doubt there would be any conditional locking on an old frame without tappet locking which is presumably what existed at Burbage.

I would agree. All the early boxes down the B&H lack frame type info, but I would guess DT ? and I doubt anyone would have gone to the trouble of adding the necessary extra tappet locking to such a simple location.

[richbrummitt]

11 hours ago, RailWest said:

Incidentally, out of curiosity I've looked now at the SRS GWR Box Register and they list Burbage as having 10 levers only. Hmm....I wonder if perhaps the frame was extended at any time......?

 

 Originally it was single track so would have had at least two fewer levers for the crossover. There's one lever that we know I haven't implemented though because we can see it in the photograph discussed above so that isn't an explanation. I cannot see 6 in that photo though. 

[RailWest]

11 minutes ago, richbrummitt said:

 

 I cannot see 6 in that photo though. 

Neither can I :-)  But 8 is certainly there.

 

I'm not sure about 4 either - there is 'something' in the distance that might be the back of it, but why 4 if no 6? Maybe there was only the siding exit shunts (8 and 9) - that would make some sense.

[richbrummitt]

12 hours ago, RailWest said:

Note also that:-

 

a). 3 and 6 would lock each other, although that might be conditional on whether 9 was normal or reverse

b) 3 should lock 9 both ways

c) 10 should lock both 5 and 7 both ways.

 

I've implemented this in the program I'm using and now I need 5 slots! I had previously made it a condition that 5 released 7 though and removing this would take me down to 4.

 

It seems by similar logic (to me anyway) as 3 that 8 should also lock 9 both ways?

 

 

[Grovenor]

1 hour ago, richbrummitt said:

 

I've implemented this in the program I'm using and now I need 5 slots! I had previously made it a condition that 5 released 7 though and removing this would take me down to 4.

 

It seems by similar logic (to me anyway) as 3 that 8 should also lock 9 both ways?

 

 

5 releases 7 is needed, removing it will need other locking to be added.

There is no reason to lock 9 both ways by 8 as a train passing 8 will not be occupying 9 during the move.

You could lock 9 normal by 8 (and vice versa) as back to back locking to avoid confusing a driver as to which way they should go.

[richbrummitt]

13 hours ago, RailWest said:

Neither can I :-)  But 8 is certainly there.

 

I'm not sure about 4 either - there is 'something' in the distance that might be the back of it, but why 4 if no 6? Maybe there was only the siding exit shunts (8 and 9) - that would make some sense.

 

Both 4 and 6 are on the Plan B of this location in Clarke, vol 1. In the beginning maybe only the exit shunts had signals and when the line was doubled the West exit was signalled properly, where it was crossing the up main, but it was not felt necessary for the East exit to change anything. 

[The Stationmaster]

18 hours ago, RailWest said:

I would agree. All the early boxes down the B&H lack frame type info, but I would guess DT ? and I doubt anyone would have gone to the trouble of adding the necessary extra tappet locking to such a simple location.

All I can find for the B&H is Colthrop which had stud locking.   There is little or nothing about original frames on the B&H Extension although Devizes had an early renewal to a tappet frame in 1911e and Pewsey received a 3 Bar VT in 1922,   According to SRS information Burbage Siding 'box closed in 1948 and judging by dates for frame renewals further east on the B&H it quite likely still had the original frame or did it get a new frame when the line was doubled?

15 hours ago, richbrummitt said:

 

 Originally it was single track so would have had at least two fewer levers for the crossover. There's one lever that we know I haven't implemented though because we can see it in the photograph discussed above so that isn't an explanation. I cannot see 6 in that photo though. 

 

But plus two FPL levers - so two point levers, 2 FPL levers either two or three running signal in each direction so if the latter that would give 10 levers and no independent ground signal levers at all.  If only two running signals in each direction then the total of 10 would allow two levers for ground signals. 

 

At the time of doubling the signalling would appear to have been considerably revised - definitely three running signals in each direction (if Clarke is to be believed), plus 3 levers working points so that takes us to nine levers in all before any ground signals are added.   I agree with your later comment about not making any alterations to the ground signals at the east end after doubling and I suspect that it possibly reflected an intention to work the siding by down train rather than anything (although things changed considerably after the signalbox was closed)

 

[The Stationmaster]

On 28/03/2019 at 12:02, richbrummitt said:

A little over a year on and I'm at a stage of committing to this. I'm not sure I have room for the additional interest of the down siding and looking at plans and pictures I cannot see evidence of a disc to at the RHS for access to the siding for an up train. 

 

According to advice I've drawn the whole arrangement and plan to implement all the levers even though 1,2 and 12 will be off stage. This means an increase to 12 levels but I think it's going to be the beat thing. 9 remains as an indicator and operates from the same lever as the connection. I've looked at several similar arrangements and concluded the numbering thus:

 

 

Any comments on the numbering welcome before I start working on the interlocking, particularly 4 through 8. Also on the disc colours.  I think 4 and 6 are red and 8 and 9 white if I understand Mike correctly. 

Right locking using these numbers and keeping 9 as a non-independent ground signal (to avoid adding yet another lever) -

 

1.  Released by 2. 3.

2.  Locks 5.9     Releases 1.

3.  Locks 6. 7.    Locks 5. 9 both ways.    Releases 1.   (see notes A and D)

4.  Released by 5.  Locks 6. 8.    Locks 7 both ways.  (see Note B)

5.  Locks 2. 9. 11.  Releases 4. 6. 7.

6.  Released by 5.  Locks 3. 4. 7.

7.  Released by 5.  Locks 3. 6.    Releases 8.  (see Note D)

8.  Locks 4.

9.  Locks 2. 5.

10.  Locks 4,  Locks 5 both ways.  Releases 12.  (see Note D)

11.  Locks 5.   Releases 12.

12.  Released by 10. 11.

 

Note A.  Note that 3 has to lock 5 both ways to allow the crossover to be used during Single Line Working.

 

Note B. Locking 7 both ways ensures that 7 cannot move after 4 is cleared - and thus avoids having a Conditional release on 4.

 

Note C.  Ideally, and properly, 3 and 7 lock each other as they create a potentially conflicting movement although it is somewhat unlikely 

 

Note D. 10 should lock 9 with 5 reverse but to provide the lock without a condition would not leave an unworkable layout on the opposite line so it has to be omitted (unless somebody else has a bright idea?).

 

The above with the exception of Note C avoids any chance of any conflicting point/signal combinations particularly signals clearing oin both directions off the same line.

[richbrummitt]

On 01/04/2019 at 13:02, The Stationmaster said:

Right locking using these numbers and keeping 9 as a non-independent ground signal (to avoid adding yet another lever) -

 

1.  Released by 2. 3.

2.  Locks 5.9     Releases 1.

3.  Locks 6. 7.    Locks 5. 9 both ways.    Releases 1.   (see notes A and D)

4.  Released by 5.  Locks 6. 8.    Locks 7 both ways.  (see Note B)

5.  Locks 2. 9. 11.  Releases 4. 6. 7.

6.  Released by 5.  Locks 3. 4. 7.

7.  Released by 5.  Locks 3. 6.    Releases 8.  (see Note D)

8.  Locks 4.

9.  Locks 2. 5.

10.  Locks 4,  Locks 5 both ways.  Releases 12.  (see Note D)

11.  Locks 5.   Releases 12.

12.  Released by 10. 11.

 

Note A.  Note that 3 has to lock 5 both ways to allow the crossover to be used during Single Line Working.

 

Note B. Locking 7 both ways ensures that 7 cannot move after 4 is cleared - and thus avoids having a Conditional release on 4.

 

Note C.  Ideally, and properly, 3 and 7 lock each other as they create a potentially conflicting movement although it is somewhat unlikely 

 

Note D. 10 should lock 9 with 5 reverse but to provide the lock without a condition would not leave an unworkable layout on the opposite line so it has to be omitted (unless somebody else has a bright idea?).

 

The above with the exception of Note C avoids any chance of any conflicting point/signal combinations particularly signals clearing oin both directions off the same line.

 

Thank you ever so much for this. I've implemented the few things I didn't have. I was struggling with 9 being free and unsure what to do with it. I didn't have 3 locks 7 and vice versa either. Regarding note D 5 locks 9 so it is already locked when 10 is pulled with 5 reversed. Maybe I'm missing something or that is what the note means. 

[The Stationmaster]

1 hour ago, richbrummitt said:

 

Thank you ever so much for this. I've implemented the few things I didn't have. I was struggling with 9 being free and unsure what to do with it. I didn't have 3 locks 7 and vice versa either. Regarding note D 5 locks 9 so it is already locked when 10 is pulled with 5 reversed. Maybe I'm missing something or that is what the note means. 

You've found the deliberate(?) error - sorry.  It should read 10 locks 3 with 5 reverse (and the reciprocal) to avoid conflicting signal indications.  The rest is ok as I proof read it several times and corrected several 'wrong numbers' as I went. 

[bécasse]

21 hours ago, The Stationmaster said:

You've found the deliberate(?) error - sorry.  It should read 10 locks 3 with 5 reverse (and the reciprocal) to avoid conflicting signal indications.  The rest is ok as I proof read it several times and corrected several 'wrong numbers' as I went. 

I suspect that there are actually another pair of "typos" as I am reasonably certain that the references to note D against levers 3 and 7 should read note C. 

Apart from that I am not at all sure that 3 and 6 would have mutually locked each other on an "ancient" tumbler-locked frame. Undoubtedly that would have been the practice from at least the mid-1920s and probably since tappet-locking became the norm rather earlier*, but the limitations of tumbler-locking meant (as The Stationmaster has demonstrated in note D) that mutual "direction" locking was sometimes not possible and my suspicions are that it would have been considered bad practice (both at the time and, indeed, later) to provide it on some signals and not others, and, obviously, every bit of extra interlocking cost money and had to be justified. Leaving 3 and 6 mutually unlocked would not have allowed any conflicting moves, and the avoidance of conflicting moves (and settings) was the original purpose of the provision of interlocking. Furthermore the continued presence of the (quite adequate) point indicator guarding the exit from the "loop" over 9 points suggests to me that this would have been a location where the original interlocking was never renewed (other than the like for like replacement of any worn out parts).

* For pairs of running signals it undoubtedly would have been earlier but here 6 is a dolly. Although not immediately relevant here, a LSWR (tappet-locked) locking diagram dating from new works in 1914 which is in my possession shows running signals with mutual "direction" locking but dollys without, which might suggest that that was the IRSE recommended practice at the time.

Edited April 4, 2019 by bécasse
typo

[richbrummitt]

38 minutes ago, bécasse said:

For pairs of running signals it undoubtedly would have been earlier but here 6 is a dolly. Although not immediately relevant here, a LSWR (tappet-locked) locking diagram dating from new works in 1914 which is in my possession shows running signals with mutual "direction" locking but dollys without, which might suggest that that was the IRSE recommended practice at the time.

 

If 6 and 3 are unlikely to have been locked is it also the case that 4 and 10 need not be? What about 4 and 6, and 8?

 

Not sure it will save me any slots...

[The Stationmaster]

14 hours ago, bécasse said:

I suspect that there are actually another pair of "typos" as I am reasonably certain that the references to note D against levers 3 and 7 should read note C. 

Apart from that I am not at all sure that 3 and 6 would have mutually locked each other on an "ancient" tumbler-locked frame. Undoubtedly that would have been the practice from at least the mid-1920s and probably since tappet-locking became the norm rather earlier*, but the limitations of tumbler-locking meant (as The Stationmaster has demonstrated in note D) that mutual "direction" locking was sometimes not possible and my suspicions are that it would have been considered bad practice (both at the time and, indeed, later) to provide it on some signals and not others, and, obviously, every bit of extra interlocking cost money and had to be justified. Leaving 3 and 6 mutually unlocked would not have allowed any conflicting moves, and the avoidance of conflicting moves (and settings) was the original purpose of the provision of interlocking. Furthermore the continued presence of the (quite adequate) point indicator guarding the exit from the "loop" over 9 points suggests to me that this would have been a location where the original interlocking was never renewed (other than the like for like replacement of any worn out parts).

* For pairs of running signals it undoubtedly would have been earlier but here 6 is a dolly. Although not immediately relevant here, a LSWR (tappet-locked) locking diagram dating from new works in 1914 which is in my possession shows running signals with mutual "direction" locking but dollys without, which might suggest that that was the IRSE recommended practice at the time.

 

It wouldn't have been 'tumbler' locking but most likely double twist or possibly stud.  And looking at locking charts for GW frames with those sorts of locking you'll find that by the end of the 19th century that sort of locking existed as amendment notes don't record any change.  The only change you normally find - although not necessarily at a wayside place such as Burbage is where some tappet lockimng had been added to create conditional locks or more easily enforce sequential locking of running signals.

 

What happened originally at the siding end toe of number 9 points is open to conjecture but later photos of Burbage clearly show (the back of)  an independent ground disc signal there - on the 'wrong' side of the line (for a very good reason).  This might well have, and probably did, date from the time when the line was doubled and inevitably at the time the lever frame would have been extended and the locking would have been renewed or very extensively altered to reflect the additional point work and added signalled movements as well as, obviously, removing the interlocking between signals applying to opposite direction through running.  Records from the usual online sources are absent so it is not known exacly what took place but it would have involved substantial changes.  Incidentally the interlocking of a near contemporaneous lever frame (to that of the time of doubling at Burbage) on the Berks & Hants Extension has running signals locking ground signals applying to movements in the opposite direction from the same signal section and it is only omitted where conditional locking would be needed to enforce it.

 

The GWR was still going very much its own way on some long established interlocking practices to the end of it existence and some Reading Drawing Office locking design practices continued for a very long time into BR days (in the mid 1980s I was even advised by one of the WR's best locking experts that I could use one of them in the locking I was designing for use on a preservation site as it would avoid a conditional release).