Jump to content
 

Decoding some mysterious signal diagram icons


Recommended Posts

51 minutes ago, 73c said:

http://www.bluebell-railway.co.uk/bluebell/hkwork_archive1.html

 

Click on the first picture on left to see the reverse. Shows lamp for the route stencil.

However, the route indicator that appears, distantly, in one of the photos on that link is the other (and much more common) sort of SR route indicator which worked rather differently. One (signal box) lever worked the appropriate route number/letter which remained hidden behind a shield until the actual signal was pulled off when the shield cleared revealing the route. (Thus the "semi-circular" route indicator needed only two levers for two routes, the other type needed three levers for two routes.)

 

I have a drawing for the more common type but not the semi-circular "Redhill" type even though I was well familiar with them.

SRrouteindicator(2).jpg.addd4856a8cdbd850df62233fdc20879.jpg

9 minutes ago, Nick C said:

 

Possibly a short train would be able to get lost if only a single bar was fitted? 

On Southern Electric routes, "spare" units were habitually stored in terminal platform roads between morning and evening (or Saturday midday) peaks.

Edited by bécasse
  • Informative/Useful 1
Link to post
Share on other sites

So as it pertains to model railway operation (and Caterham in particular) it would primarly be a process of the treadles locking points, which would lock the signals:

  • A would unlock platform starters 13 and 15
  • J & K would unlock 4 (and a host of other things in a chain, obviously)
  • G & H would unlock 3 (ditto)
  • B would unlock 15/13

 

C and L ( as well as 2 and 10) are a bit of a mystery to me. It would appear to me that there are two options, either 2 governs the entry to both down sidings and 10 governs a reversing movement into the up siding, or 2 governs the entry to No. 1 down siding, No. 10 governs the entry to No. 2 down siding?

Edited by Lacathedrale
Link to post
Share on other sites

  • RMweb Gold
6 minutes ago, Lacathedrale said:

So as it pertains to model railway operation (and Caterham in particular) it would primarly be a process of the treadles locking points, which would lock the signals:

  • A would unlock platform starters 13 and 15
  • J & K would unlock 4 (and a host of other things in a chain, obviously)
  • G & H would unlock 3 (ditto)

It doesn't seem like the up siding is locked by anything?

It seems that 2 controls access to both the No 1 and No 2 down sidings but doesn't indicate which route is about to be taken - so C and L presumably lock it but it feels a bit unclear?

It's usually the other way round, the treadles (or track circuits) back-lock the signals, which in turn lock the points. 

 

So for the example of a train approaching platform 1: The first thing would be that the signalman would pull points 20, and FPLS 7,17 & 19. He can then accept a train from Whyteleaf. When he gets 'train entering section' from Whyteleaf, he will clear signal 4 (and distant 1). Lever 4 is then electrically locked until the train hits treadle J - he can replace the signal to danger in an emergency, but the lever won't go fully back into the frame, and so the mechanical locking won't release. Once the train arrives, he can replace the lever and reset the road for the next move.

Link to post
Share on other sites

Thanks NIck, that's really clear in my mind.

 

I've been thinking about trying to work out a lever frame description/interlocking from the diagram, I've come up roughly with the below - no doubt there are efficiencies to be validated!

  1. Distant, unlocked by pulling off 3 or 4, locked by 2
  2. Ringed shunt signal to down sidings unlocked by 6
  3. Home Signal slotted (?) with route indication to No. 2 road, interlocked with 7, 17, 19, 16, back locked with G & H
  4. Home Signal slotted (?) with route indication to No. 1 road, interlocked with 7, 17, 19, 12, back locked with J & K
  5. Spare
  6. Crossover to No 1. down siding, interlocked with 7, 3, 4, and back-locked with L
  7. FPL for 6, interlocked with 9
  8. Shunt signal for No. 1 down siding to down main, interlocked with 6 - (how would one differentiate a right-away to the down main (to LoS) or up main (as departing train?)
  9. Crossover from No 1 down siding to up main, interlocked with 7
  10. Ringed shunt signal for a reversing move from the up main into the up siding, interlocked with  7, 9, 21, 22, 23
  11. Spare
  12. FPL for No. 20 crossover
  13. Platform starter for No 1. Road, interlocked with 12, 22, 21, 9 and back-locked by B
  14. Outer Home with white diamond back locked by A
  15. Platform starter for No. 2 road, interlocked with 16, 19, 21, 9 and back-locked by B
  16. FPL for No. 21
  17. FPL for No. 18
  18. Point to No. 1 down siding + catch point -  locked by 17 and back-locked by C
  19. FPL for No. 20 crossover
  20. Crossover down main to No 1 road, locked by 19 and 12 - maybe 22?
  21. Point No 2 road to up main
  22. Point No 1 road to up siding + catch point - maybe 20?
  23. Shunt signal interlocked with 22
  24. Shunt signal No 1. down siding locked by 18
  25. Spare
  26. Spare
  27. Spare

 

I'm sure there are improvements to this - I'm re-drafting in Excel for an easier to read tabular view, and I'm happy to muse on it but if anyone has a specific correction that would be very helpful.

Edited by Lacathedrale
Link to post
Share on other sites

13 minutes ago, Lacathedrale said:

Thanks NIck, that's really clear in my mind.

 

I've been thinking about trying to work out a lever frame description/interlocking from the diagram, I've come up roughly with the below - no doubt there are efficiencies to be validated!

  1. Distant, unlocked by pulling off 3 or 4, locked by 2
  2. Ringed shunt signal to down sidings unlocked by 6 and probably also 7
  3. Home Signal slotted (?) with route indication to No. 2 road, interlocked with 7, 17, 19, 16, back locked with G & H
  4. Home Signal slotted (?) with route indication to No. 1 road, interlocked with 7, 17, 19, 12, back locked with J & K
  5. Spare
  6. Crossover to No 1. down siding, interlocked with 7, 3, 4, and back-locked with L
  7. FPL for 6, interlocked with 9
  8. Shunt signal for No. 1 down siding to down main, interlocked with 6 - (how would one differentiate a right-away to the down main (to LoS) or up main (as departing train?) It wouldn't :-)
  9. Crossover from No 1 down siding to up main, interlocked with 7 released by 6
  10. Ringed shunt signal for a reversing move from the up main into the up siding, interlocked with  7, 9, 21, 22, 23
  11. Spare
  12. FPL for No. 20 crossover
  13. Platform starter for No 1. Road, interlocked with 12, 22, 21, 9 and back-locked by B
  14. Outer Home Advanced Starting with white diamond released by Sykes indicator once 'line clear' given by Whyteleaf and then back locked by A
  15. Platform starter for No. 2 road, interlocked with 16, 19, 21, 9 and back-locked by B
  16. FPL for No. 21
  17. FPL for No. 18
  18. Point to No. 1 down siding + catch point -  locked by 17 and back-locked by C
  19. FPL for No. 20 crossover
  20. Crossover down main to No 1 road, locked by 19 and 12 - maybe 22?
  21. Point No 2 road to up main
  22. Point No 1 road to up siding + catch point - maybe 20?
  23. Shunt signal interlocked with 22
  24. Shunt signal No 1. down siding locked by 18
  25. Spare
  26. Spare
  27. Spare

 

I'm sure there are improvements to this - I'm re-drafting in Excel for an easier to read tabular view, and I'm happy to muse on it but if anyone has a specific correction that would be very helpful.

 

A few quick comments, but really the locking table needs to be drawn in the proper format with 'released by' and 'locks' columns

Link to post
Share on other sites

Thanks Chris, working on that now - hopefully my questions aren't improper but this is all very fascinating :)

 

Why would 2 be released by 6

What purpose do H and K fulfill if they're not used in the interlocking? All the pictures I have seen of Caterham in the 30's-40's show 3SUB/4SUBs stabled at the  bufferstops.

Similarly, why would 6 not be back-locked by L, but 18 would be back-locked by C?

 

Something like this?

 

image.png.798267decb33fccd68bb6a5ff9bfbf31.png

26Spare----

27Spare----

Link to post
Share on other sites

5 minutes ago, Lacathedrale said:

Thanks Chris, working on that now - hopefully my questions aren't improper but this is all very fascinating :)

 

Why would 2 be released by 6

What purpose do H and K fulfill if they're not used in the interlocking? All the pictures I have seen of Caterham in the 30's-40's show 3SUB/4SUBs stabled at the  bufferstops.

Similarly, why would 6 not be back-locked by L, but 18 would be back-locked by C?

 

Something like this?

 

image.png.798267decb33fccd68bb6a5ff9bfbf31.png

26Spare----

27Spare----

 

>>>>Why would 2 be released by 6..

Because 2 reads across 6 into the sidings.

 

>>>What purpose do H and K fulfill if they're not used in the interlocking?

They probably locked 3/4, but that would be front locks not back locks.

 

>>>Similarly, why would 6 not be back-locked by L, but 18 would be back-locked by C?

Missed that one :-( Generally speaking points were not locked/back-locked by treadles, only signals - there were exceptions but not relevant here. Treadles provided releases, not locks.

 

  • Informative/Useful 1
Link to post
Share on other sites

  • RMweb Premium
11 hours ago, Lacathedrale said:

The gantry itself seems very strange - it looks like there are two poles ahead of the gantry on the up and down lines, then the main arch whose horizontal bar extends wider than the legs, and then yet another vertical leg on the far left. None of that makes much sense to me, any advice would be gladly appreciated.

I think what you are seeing is stay-poles for bracing the gantry, one either side on both sides of the line. The whole thing is freshly painted, and the stay-pole on the right of the line, this side of the gantry, is almost hidden from view by the end of a building.

  • Like 1
Link to post
Share on other sites

  • RMweb Premium

It looks as though the gantry originally spanned three or four tracks, but by the time of the photograph the additional line on the up side had been taken out- see  here

 

The post on the left could have been for a signal applying to that line.

 

Although the SRS diagram is dated 1945, the interlocking changed quite soon afterwards as a result of the accident described here (sorry about the poor quality of the report printing - a wartime economy measure). To try and make sure that if a train left the terminus against a signal at danger, it would be routed onto the up line rather than into head-on collision with an arriving train, the Southern rearranged the way the pointwork in the station throat was controlled. The technique became known as "Caterham locking", and is still sometimes referred to as such today- see here (paragraph 11) for how it didn't quite work as planned at Bognor Regis in 2008.

Edited by 4069
Link to post
Share on other sites

That old plan looks very impressive!

 

I had wondered too about there having once been an extra line on the left. Also, the post on the far left almost in the trees must have been another 'doll' (otherwise why would the gantry stretch that far?), so there must have been more signals on the gantry than in the later SRS diagram.

 

The point-work in the throat appears to have been altered and re-numbered again after the Down Sidings were abolished.

Link to post
Share on other sites

I think you're right @4069 - this is a snapshot from the 1912 OS-grid map - to the right of 'S.P' there are two dots - maybe the two gantry masts? Both outer tracks are headshunts for the sets of sidings so unlikely to have signals? Maybe however due to the clearances the masts were oriented as such?

 

image.png.19d775c1acb9a0baf98ed50f71f48392.png

 

@RailWest - The old plan is very impressive, and the most complicated of all. Certainly the runarounds and turntable would make 'unusual' passenger services possible, but the vast majority unless I'm mistaken were just simple push-pull commuter trains, twice an hour!

Edited by Lacathedrale
Link to post
Share on other sites

  • RMweb Gold
14 hours ago, 4069 said:

It looks as though the gantry originally spanned three or four tracks, but by the time of the photograph the additional line on the up side had been taken out- see  here

 

The post on the left could have been for a signal applying to that line.

 

Although the SRS diagram is dated 1945, the interlocking changed quite soon afterwards as a result of the accident described here (sorry about the poor quality of the report printing - a wartime economy measure). To try and make sure that if a train left the terminus against a signal at danger, it would be routed onto the up line rather than into head-on collision with an arriving train, the Southern rearranged the way the pointwork in the station throat was controlled. The technique became known as "Caterham locking", and is still sometimes referred to as such today- see here (paragraph 11) for how it didn't quite work as planned at Bognor Regis in 2008.

 

I don't quite understand how a change to the locking would have prevented that particular accident though - the report says that the collision occurred on the single slip, and that no points were run through, meaning that the outbound train had not passed points 20a - so even if points 21a had been set for the up line, the front of that train would have been in almost exactly the same place - would the slight difference in angle have made enough difference? Having the guard and platform staff check the signal before giving the ready to start would have prevented it...

Link to post
Share on other sites

I've been looking at this a bit more, and just wanted to be sure, that:

  • Signal 2  governs all shunting movements from the down main into the whole station (i.e. all possible routes from that point -  No. 2 road (via 18 and 21), No. 2 siding (via 18) and No 1 siding (via 6)
  • Signal 10 similarly govern shunting movements from the up main into both the up siding and No. 1 road (via 21 and 22) , the No. 2 road (via 21), and the No. 1 down siding (via 9 and 6)

 

Link to post
Share on other sites

In some ways Caterham was a most unusual Southern London outskirts terminal.

 

Firstly, there was the nearby Guards' Depot (the first "modern" military barracks in Britain) which meant that the station had to have provision for occasional but potentially substantial military traffic.

 

Secondly, after electrification in 1928 the normal service pattern was for off-peak services to/from Charing Cross but peak-hour services which went only as far as London Bridge. Furthermore this service was combined between London and Purley with trains to/from Tattenham Corner, the norm, in the days of the original "3-SUB" units, being 3-cars to/from Caterham and 3-cars off-peak/6-cars on-peak to/from Tattenham Corner. When the "3-SUB" units were made up to 4-SUBs post WWII, the norm became 4-cars to/from Caterham and 4-cars to/from Tattenham Corner but this led to complaints of overcrowding on the Tattenham line where peak-hour provision was reduced by 2-cars per train. Consequently, once 2-EPB units became available in the mid-1950s, the peak-hour trains were worked by 4- and 2-EPB units instead of SUBs with 4-EPB going to Caterham and 6-EPB (2+4) going to Tattenham Corner, and this pattern remained unchanged for several decades. Although services were more frequent at peak hours, the fact that they worked a shorter route (turning back at London Bridge) meant that the only extra stock required was the 2-EPB strengtheners added at Tattenham Corner (and stabled there at other times). As a result of this there was never any need to stable units at Caterham other than overnight, probably the only London area Southern terminal which normally had no units stabled in the platforms during the day.

 

Incidentally on Epsom race days complete 10-EPB trains ran all day between Charing Cross and Tattenham Corner with Caterham served by a 4-EPB shuttle to/from Purley.

  • Like 1
Link to post
Share on other sites

  • RMweb Gold
2 hours ago, Nick C said:

 

I don't quite understand how a change to the locking would have prevented that particular accident though - the report says that the collision occurred on the single slip, and that no points were run through, meaning that the outbound train had not passed points 20a - so even if points 21a had been set for the up line, the front of that train would have been in almost exactly the same place - would the slight difference in angle have made enough difference? Having the guard and platform staff check the signal before giving the ready to start would have prevented it...

A change to the locking or the normal lie of the points at Caterham would have made no difference at all to the way in which the collision occurred on the single slip and the Inspecting Office made no recommendation to that effect referring only to the use of of a 'Catch' (sic) point and then dismissing it as impracticable.  While it is in many respects far more logical to arrange the lie of points so that departing trains from a terminus would runaway in the right direction in the event of a SPAD that is all it would achieve and it would obviously not mitigate against a collision in the event of the paths of trains crossing each other following a SPAD all that might mitigate against that is the use of TPWS or full ATP of some sort.

 

Similarly having the Guard responsible for checking signal aspects before giving the ;ready to start' signal is no guaranteed defence and the result can lead to what is known as a 'ding, ding & away' situation where the Guard etc doesn't check the signal aspect and neither does the Driver because he thinks the Guard has checked lt.  Hence the Bellgrove collision in 1989 and Hyde in 1990 with Bellgrove in particular having far more similarities to the Caterham collision than to that at Eastbourne.

Link to post
Share on other sites

23 hours ago, Lacathedrale said:

Thanks Chris, working on that now - hopefully my questions aren't improper but this is all very fascinating :)

 

Why would 2 be released by 6

What purpose do H and K fulfill if they're not used in the interlocking? All the pictures I have seen of Caterham in the 30's-40's show 3SUB/4SUBs stabled at the  bufferstops.

Similarly, why would 6 not be back-locked by L, but 18 would be back-locked by C?

Something like this?

 

This would be my take on the interlocking

Caterham.pdf

Link to post
Share on other sites

Worked semaphore distants were quite rare because the Southern Railway fixed many of them as an economy measure (on the grounds that it made little difference to the speed of approach to a terminal station), so the fact that this one was retained, and that the two available routes shared the same home (with an indicator, of course), suggests that it could be cleared for either route.

Link to post
Share on other sites

Keith @Grovenor thank you very much.

 

I have started going through your take and hopefully I can ask a few questions when it's a bit more settled in my mind? While doing research for the model side of this station, I came across a couple of shots of the accident described earlier:

 

image.png.13b16da6e69ef7a0fbab5ac65f78a289.png

 

image.png.001fc84aa4f78fef88da6b44333c0175.png

 

As this occured in 1945 it's roughly concurrent with our signal plan. The west of the station (side against which the signal box is located) is wooded as shown above, and so presuambly the disc signal shown is No. 24, with the approximate location of the photographer being between the down sidings, indicated by the pink arrow on this cropped aerial photo:

 

image.png.f8219b5621aa0f54e4ccd509b312646e.png

 

 

Link to post
Share on other sites

Until the frequency reductions implemented in September 1958 the Charing Cross-Tattenham Corner/Caterham trains ran every 20 minutes. The approximate journey time to/from Tattenham Corner was 54 minutes and to/from Caterham  40 minutes, so 6 units were required to work the Tattenham Corner trains and 5 to work the Caterham ones. However, the Tattenham Corner unit lead between Charing Cross and Purley in both directions, so a unit that worked to Tattenham Corner on one round trip worked to Caterham on the next and the Tattenham Corner headcode [01 on EPB stock] was displayed throughout between Charing Cross and Tattenham Corner in both directions, with the Caterham headcode [93 on EPB stock] only normally seen between Purley and Caterham (again in both direction). More modern 4-SUB stock displayed the same numerical headcodes as EPB stock, albeit usually using stencils rather than roller blinds, but the older 4-SUBs, and all "3-SUBs", displayed the letter stencil S with two dots over the letter for Tattenham Corner and one dot for Caterham.

Edited by bécasse
Link to post
Share on other sites

@bécasse - http://www.semgonline.com/headcodes/eheadcodes/eheadcodes08.html  and http://www.semgonline.com/headcodes/eheadcodes/eheadcodes05.html show some different headcodes for the earlier periods:

 

Alpha 1939:

H-two dots: London Bridge - Caterham via Forest Hill

L-bar: Caterham - Purley empties, Victoria - Caterham via Streatham Common and Purley

P: Purley - Caterham

 

Numeric 1939:

80: Vic-Purley,

80-bar: Purley-Caterham (ECS)
81: London Bridge-Caterham

3: Purley-Caterham

 

The post-1960's page shows what you have described for numeric numbers, albeit with the addition that 3 became 43 for Purley-Caterham (seemingly not used if operated as you described), but nothing on the 'S' alpha-code - presumably because it's outside the timeline?

 

Also (strangely) I've found conflicting evidence of the 3sub joining/splitting at Purley - some sources show 6 to Tattenham and 3 to caterham (going as far as to describe a riot at Purley during a blizzard when one Caterham service was cancelled and the following redirected to Tattenha Corner) and the others showing the inverse. I'd be inclined (based on your evidence also) to assume the former.

 

With regard to your comments around the use of a 2EPB strengthener for Tattenham Corner trains, and no need whatsoever to stable trains at Caterham, I am somewhat confused that the central section would continue to reserve a headcode exclusively for the use of ECS workings between Purley and Caterham - do you have any insight on that?

 

Lastly, I've read that from 1939 to '49 there was only one train an hour on the Caterham branch, presumably a shuttle service - but which is directly contradicted by the headcodes provisioned in 1939 above?

 

EDIT: I do thank you all very much - I think (if it's OK with everyone who is part of this discussion) - further conversation around the Caterham branch and it's operation I'll post to my main planning thread here:

 

And keep this thread for discussion of the signalling and interlocking?

 

Cheers!

Edited by Lacathedrale
Link to post
Share on other sites

On the recommendation earlier I picked up Pryer's OPC 'A Pictoral Record of Southern Signals' book and I was a little dissapointed to see how little there was of Southern and SECR infrastructure compared to the LSWR - a third of the book dedicated to their pneumatic signals!  Anyway, it did provide some clarification that the ringed signals are 'siding' signals, not 'shunting' signals - and help clarify some other points, i.e. that the limit of shunt board at Caterham exists to limit the movement of trains in the wrong direction on the down line (i.e. authorised by the yellow armed ground signal no. 6 or ground signal no. 24)  from entering into the box in rear. I knew this in theory, but reading the official explanation was helpful.

 

I'm still not 100% about the use of Sykes Lock and Block  but have sketched what I believe the process would be between Whyteleaf and Caterham using the SR-issue Sykes L&B device (rather than one with a plunger and tablets), i.e. with  two indicators showing the line status from the previous box, and his own section - either 'Normal' or 'Line Clear' or 'Train on Line'.

  1. WHY calls CAT, requests Line Clear via bell codes
  2. CAT sets commutator to Line Clear for his lower, WHY's upper indicator
  3. WHY sends 'train entering section' to CAT, CAT acknowledges 
  4. CAT sets commutator to 'Train on Line' for his lower, WHY's upper indicator.
  5. WHY sends 'train out of section' to CAT, CAT acknowledges
  6. Activation of treadle at CAT, permits Commutator to be set from Train On Line back to normal
  7. CAT sends 'train out of section' to WHY, WHY acknowledges

I think this maps pretty directly to a model railway situation, albeit with a detector circuit rather than a treadle.

 

I think this would also be the basis for a secondary mode of operation, where two people are purely signalmen, and the trains run themselves based on the signal aspects and occupancy...

 

Link to post
Share on other sites

24 minutes ago, Lacathedrale said:

On the recommendation earlier I picked up Pryer's OPC 'A Pictoral Record of Southern Signals' book and I was a little dissapointed to see how little there was of Southern and SECR infrastructure compared to the LSWR - a third of the book dedicated to their pneumatic signals!  Anyway, it did provide some clarification that the ringed signals are 'siding' signals, not 'shunting' signals - and help clarify some other points, i.e. that the limit of shunt board at Caterham exists to limit the movement of trains in the wrong direction on the down line (i.e. authorised by the yellow armed ground signal no. 6 or ground signal no. 24)  from entering into the box in rear. I knew this in theory, but reading the official explanation was helpful.

 

I'm still not 100% about the use of Sykes Lock and Block  but have sketched what I believe the process would be between Whyteleaf and Caterham using the SR-issue Sykes L&B device (rather than one with a plunger and tablets), i.e. with  two indicators showing the line status from the previous box, and his own section - either 'Normal' or 'Line Clear' or 'Train on Line'.

  1. WHY calls CAT, requests Line Clear via bell codes
  2. CAT sets commutator to Line Clear for his lower, WHY's upper indicator
  3. WHY sends 'train entering section' to CAT, CAT acknowledges 
  4. CAT sets commutator to 'Train on Line' for his lower, WHY's upper indicator.
  5. WHY sends 'train out of section' to CAT, CAT acknowledges
  6. Activation of treadle at CAT, permits Commutator to be set from Train On Line back to normal
  7. CAT sends 'train out of section' to WHY, WHY acknowledges

I think this maps pretty directly to a model railway situation, albeit with a detector circuit rather than a treadle.

 

I think this would also be the basis for a secondary mode of operation, where two people are purely signalmen, and the trains run themselves based on the signal aspects and occupancy...

 

Delete item 5 :-)

 

  • Agree 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...