Mechanical Interlocking - How to learn the Lost Black Art?

[micknich2003]

Andy, here's a sketch I put together to illustrate a possible use of a "Butt". Obviously only one set of points can be Reversed at any one time, pulling any lever locks the other two normal.

[uax6]

That's novel. Some of these things are very difficult to picture from just the drawings, what does a butt actually look like?

(That article looks like it could be a good read...)

 

Andy G

[micknich2003]

7 minutes ago, uax6 said:

That's novel. Some of these things are very difficult to picture from just the drawings, what does a butt actually look like?

(That article looks like it could be a good read...)

 

Andy G

Chris, it's literally two Bridle Irons butted end on. As drawn pulling 1 pushes the right hand bridle out of No2 tappet and into No3 tappet, hence 1 locks 2 and 3. Mechanical locking being reciprocal if three is pulled both 1  and 2 become locked. Similarly, if 2 is pulled both locks are driven out and 1 and 2 becomes locked.

[uax6]

So simple when you explain it! The ingenuity of these designs is really quite brilliant.

 

Andy G

[LNERGE]

I will be cutting locking on the 5th,6th and 7th October. There will be one sliding tappet and a fair bit of mechanical sequential crammed in to an 11 lever tray. I'll post some pictures of the works.

 

So far we crashed a bit in just to cover a point and facing point lock..

 

 

 

 

[The Stationmaster]

One question that came up when Mick and I were talking yesterday was about mechanical sequential (aka rotation) locking and does any still exist anywhere on the national network?  I'm fairly certain that there is none left on any Western frames. (as in WR as it stood at the end of its life).    Although I know, according to drawings, there was still some around in the 1970s but in general full sequential locking seems to have been a later addition at many (most?) Western 'boxes so it had been done electrically.   In a few cases some tappet locking  had been added to older frames to provide sequential locking.

[Lacathedrale]

Thank you Mike @The Stationmaster - I've tried to put up a locking table based on what you said:

 

 

It seems strange to have such emptiness in the table, i.e. 1 doesn't do anything, it is rather just impacted by other levers. Same for 23 - it is released by A, and it is A which locks 1, 21, and B. The fact that B is locked means that 3 is also locked. I don't know how I could come up with the logical flow to end up with the same result (in this case, the dependence on A doing much of the locking)

[Adam88]

 

[The Stationmaster]

6 hours ago, Lacathedrale said:

Thank you Mike @The Stationmaster - I've tried to put up a locking table based on what you said:

 

 

It seems strange to have such emptiness in the table, i.e. 1 doesn't do anything, it is rather just impacted by other levers. Same for 23 - it is released by A, and it is A which locks 1, 21, and B. The fact that B is locked means that 3 is also locked. I don't know how I could come up with the logical flow to end up with the same result (in this case, the dependence on A doing much of the locking)

I didn't deal with Signal 1 in my comments but of course it locks, and is locked by points A & B.  

 

the Released By column complete is 3 released by B;  23 released by A (plus the relevant FPLs in the real world).

'Locks when normal' is the reciprocal of Released By, i.e. if lever B is standing normal signal lever 3 is locked normal3 .

Locks when reversed. - 1 locks A and B;  21 locks A;  A locks 1, 21 and B;  B locks 1 and A

Releases -  A releases 23; B releases 3

 

One of the keys to getting the most economical arrangement of locking it s tto first make sure the normal lie of any points is correct and also to remember that whenever possible points work as crossovers - e.g. in a 'ladder of single or double slips the two ends will be worked by different levers but those two levers will also be working any set of switches to make what amounts to a crossover.   One thing you can find in some older interlockings is that there was a lot of reliance on using FPL levers to help economise on locking.

 

for example looking back at my original sketch the two FPLs A1 and A2 would be worked by separate levers.  Assuming that the FPL is bolted when the lever is reversed  then in order to release signal lever 3 levers B and A2 would first have to be reversed.  But if B released A2 then there would only be a need for lever A2 to release Signal lever 3.  While it saves locking these sort of arrangements had their downsides when disconnections were taking place and tended to be frowned on on the WR (and probably elsewhere) in later years for that reason.

Edited September 23, 2020 by The Stationmaster

[micknich2003]

Marfleet was a small station on the Hull to Withernsea Branch, there's a bit of most things and the locking is reasonably easy to follow.kk

[The Stationmaster]

Now have a bit of fun with this one -  You can ignore the Limit of Shunt Board and backing arm 24.  And disc 28 would only read through 29 reverse.  (Image should enlarge if you click on it). No1 is the Down Distant signal, 38 is the Up Distant Signal; 7 is a Shunt Ahead below No.6 (the Down Starting Signal)

 

As far as the ground discs are concerned they are to WR standards so -   8 is a white light disc requiring either 9 or 29;   22 is a red light disc requiring  nil or 19;  28 is a red light disc requiring 29;  30 is a white light disc requiring 29;  31 is a red light disc requiring 32;   33 is a white light disc requiring 32.  White light discs could be passed at danger on the authority of the Signalman if the points to which they applied were standing normal.  All the discs in the sidings and bay would have red lights.

 

As ever a lot of simplification of the locking can achieved through the interlocking between levers working points - it comes back to the thing about conflicting routes.  Thus, for example, lever 9 would lock levers 19 and 29;  19 would lock 9, 17, and 29 and release 21;  29 would lock levers 9 and 19 (no need for it to lock 21 because 21 is released by 19.

 

 

[uax6]

On 23/09/2020 at 02:15, LNERGE said:

I will be cutting locking on the 5th,6th and 7th October. There will be one sliding tappet and a fair bit of mechanical sequential crammed in to an 11 lever tray. I'll post some pictures of the works.

 

So far we crashed a bit in just to cover a point and facing point lock..

 

 

 

 

Owww, shiny! Look forward to more reports on how this goes!

 

Andy G

[Lacathedrale]

I have recieved my IRSE green book - will review and reply to the thread - thank you all for the recommendations. The gent who runs the Plumpton Green channel also recommended it!

[The Stationmaster]

1 hour ago, Lacathedrale said:

I have recieved my IRSE green book - will review and reply to the thread - thank you all for the recommendations. The gent who runs the Plumpton Green channel also recommended it!

You will find it very useful I'm sure.  But one word of warning o it is very much built around an idealised way of doing things to the 'normal' standards and as such does not cover the individual foibles of the various Companies por different BR Regional Drawing Offices.

[47137]

I learnt a lot about the practicalities when I built one of the Modratec lever frames.

 

There are some notes near the beginning of this web page:

http://www.castellybwrdd.myzen.co.uk

 

This was about 8 years ago and it still works perfectly. An 'O' level or similar in metalwork is useful for making the bars and tappets.

 

- Richard.

[eastwestdivide]

Crikey, my head hurts, but in a good way. Any descendents of Babbage & Lovelace around?

[TheSignalEngineer]

On 23/09/2020 at 10:15, The Stationmaster said:

One question that came up when Mick and I were talking yesterday was about mechanical sequential (aka rotation) locking and does any still exist anywhere on the national network?  I'm fairly certain that there is none left on any Western frames. (as in WR as it stood at the end of its life).    Although I know, according to drawings, there was still some around in the 1970s but in general full sequential locking seems to have been a later addition at many (most?) Western 'boxes so it had been done electrically.   In a few cases some tappet locking  had been added to older frames to provide sequential locking.

About 50 years ago I did some work on a box with a Double Twist frame where there were  tappet trays added to provide sequential and I think also some conditional locking.

[micknich2003]

Here is an LNERly example of adding tappet locking to a non tappet frame, in this case  a Saxby "Rocker" at Howden East ex H&BRly.

[micknich2003]

Here is part of the locking from Springbank North, H&B Section, Hull a June 1924 installation and now long gone. As can be seen three routes and their respective Distant Signals read  to No3 the Down Main Starter. The BUTT under No2 tappet allows the Lock left of 3 to be pushed by No1 Distant or pulled by No6 or 9 Distants.

[Lacathedrale]

The IRSE booklets have been very helpful for me in understanding signalling and interlocking more fundamentally, thank you very much for the recommendation. I am still a bit fuzzy on the specifics, but I'm getting there!

 

My first draft of my own signalling diagram looks something like this (I know the levers need re-numbering!)

 

 

My very first draft for the interlocking is all over the place, though! For now I'm just trying to get the signals interlocking properly! I appreciate that a good deal of the interlocking further along will come from the turnouts, however.

 

 

1. P4 home - locks 2, 3, 4
2. P3 home - locks 1, 3, 4
3. P2 home - locks 1, 2, 4
4. P1 home - locks 1, 2, 3
5. Up main to P4 -
6. Up main connection -
7. Up main to P3 -
8. Up main to P2 -
9. P4 to down main -
10.  P2 to down main -
11.  P1/2 to down main -
12.  P3 to P2 crossover -
13. P1 to down main / loco shed area double slip -
14. P4 starter - locks 15, 16, 17
15. P3 starter - locks 14, 16, 17
16. P2 starter - locks 14, 15, 17
17. P1 starter - locks 14, 15, 16
18. Advanced starter - locked by 9 or 11
19. P3 crossover ground signal - released by 12, locks 20
20. P2 crossover ground signal - released by 12, locks 19
21. (spare)
22. Loco shed yellow ground signal  - released by 13, locks 24
23. Goods? shed ground signal  - released by 25
24. Headshunt to P1 ground signal released by 13, locks 22
25. Goods? headshunt -

 

After writing out the above, I wonder if 13 should be a barry slip rather than a double slip? I can't see a reason from P1 to engine/goods headshunt)

Might also be easier to route 'by default' the up into P3 and down into P2?  (rather than p3 and p4 respectively) ?

 

Edited September 28, 2020 by Lacathedrale

[The Stationmaster]

1 hour ago, Lacathedrale said:

The IRSE booklets have been very helpful for me in understanding signalling and interlocking more fundamentally, thank you very much for the recommendation. I am still a bit fuzzy on the specifics, but I'm getting there!

 

My first draft of my own signalling diagram looks something like this (I know the levers need re-numbering!)

 

 

My very first draft for the interlocking is all over the place, though!

A few points (sorry) -

1. Signal arms stand to the left of the post in British practice so your two platform starting signals have arms on the wrong side

2.  Single slips are usually arranged as ends of a crossover with a different lever working each end and as a crossover with another point end, And different ends of a crossover are worked by the same lever.  Thus in your diagram point ends 6 & 7 would be worked t by the same lever and the slip connect at 7 would be worked by lever 8.  9 and 12 are correct.   Look at how the numbering works in my sketch above.

3.  Points in sidings are worked by hand levers adjacent to the actual point and movements are controlled using handsignals - thus 22/23/24 are all superfluous.

4.  I assume 13 is a double slip - the end of it furthest from the running lines would be worked from the signal box as a trap point and disc 21 should be moved to cover any movements through it towards the running lines.  Depending on copmpany/Regional practice the other end wiould most likely be worked by a hand lever and not by the signal box.

5.  As mentioned previously practice varied so knowing which Company/BR Region/era being signalled can make a difference although mainly to signals however a ground sic at the toe of 11 and another at the Down Line toe of 9 are likely additions in most circumstances..

 

[Lacathedrale]

Thanks Mike, I really appreciate that! I guess by keeping 6 and 7 separate I figured that it would be better to ensure conflicting movements couldn't be signalled, but shows I was mistaken.

 

1) signal arms moved.

2)  6 and 7 amended as described, I think!

3) ground signals in sidings removed, points x and y converted to hand levers - unless X needs to be in the frame since technically it's a trap for the loco shed?

4) double slip changed to barry slip

 

 

Era is pre-group southern regions (specifically LCDR/SECR) in the edwardian period - so presumably shunt signals on the platform starters and calling-on arms on the homes?

Edited September 28, 2020 by Lacathedrale

[The Stationmaster]

You need to show the other end of No.5   Plus put a signal back at 13.

[Lacathedrale]

Ah! Yes, it's not very clear between 5, 10 and 11 what's going on - 5 and 10 are 'just' turnouts and 11 is a single slip - but I guess in terms of interlocking 5 and the opposide end of the slip would work effectively as a crossover.

 

 

1. shown 5 as the opposite end of the slip previously known only as 11
2. erased some of the tail of point 10 so it's a more clear arrangement
3. re-added ground signal 8 at point 13, if that's what you meant!

 

[Titanius Anglesmith]

If I may be so bold,

 

Point X should be double-ended with 13 to trap the siding, as you mentioned in a previous post. Y is still handworked. Ground signal 8 should be moved to the other side of X. 

 

May I also suggest a ground signal at the toe of 11 reading towards the platforms for running round etc?