Mechanical Interlocking Museum Demonstartion

[Chen Melling]

Hello to all,

 

At the Israel Railway Museum we are planning a new exhibition dedicated to the basics and history of railway signalling. We have a good collection of artefacts to display, ranging from ground frames and block instruments to a complete set of miniature train staff apparatus and various semaphore posts (though most of the latter are much too tall to be included in the space available).
Below is a photo of the space, or rather wall, to be used for the new exhibition. You can also spot many of the relevant items, though we plan to redesign the entire space.

 

There are many challenges facing this project, besides the obvious limited funding. At the moment I am trying to design a demonstration of the principles of interlocking in general and mechanical interlocking in particular.  The main exhibit in this part will be the 2-lever ground frame in the photo below, which is interlocked such that the two levers can be together in either position, but you cannot reverse the red one until you reverse the blue one, and conversely, you cannot return the blue one to normal before returning the red one.

 

As I can't think of anything meaningful to connect it to inside the exhibition space, I plan on adding a simplified demonstration for the visitors to peruse, similar in nature the one presented in this FNRM video:

https://www.s-r-s.org.uk/videoPages/Tappet.php

but with the "tappet-lever"s connected to graphical representations of what they control. With the help of our archive manager I came up with the following simple layout of a junction/half a crossing station, and the corresponding dog chart (assuming no facing-point-locks, for simplicity):

I would be very thankful if the esteemed audience here could check whether the chart is correct, assuming one signal, at most, can be OFF at any one time, and the points position must correspond to that signal's aspect.

 

Chen

Edited February 1 by Chen Melling

[The Stationmaster]

What might be useful as an explanation to those looking at that is al ocking chart showing what each lever locks or releases in relation to other levers - which in any case is part of the design process.  Normally the locking chart is drawn up first and what some of know as the Dog Chart (which is what you have drawn) is worked out from the locking chart

 

It might also help understanding if you were to explain the basic principles of interlocking as a list.  You also need to explain the purpose of each signal arm on your sketch

[Grovenor]

Quite a bit of excess locking in that dog chart.

 

The bottom 2 bars are fine, 1 and 4 lock 3 and 2 and 5 released by 3.

So your double ended dogs for 1 locks 2 and 4 locks 5 are redundant as its already covered via 3..

The top two bars can be combined into one, giving 1 or 2 locks 4 and 5, and 4 or 5 locks 1 and 2.

 

just 3 bars in total.

[RailWest]

>>>...the points position must correspond to that signal's aspect.....

 

Not strictly true surely? For example, if a point is set 'normal' then the signal(s) for that route may be 'on' or it may be 'off'. 

[Chen Melling]

1 hour ago, The Stationmaster said:

What might be useful as an explanation to those looking at that is al ocking chart showing what each lever locks or releases in relation to other levers - which in any case is part of the design process.  Normally the locking chart is drawn up first and what some of know as the Dog Chart (which is what you have drawn) is worked out from the locking chart

 

It might also help understanding if you were to explain the basic principles of interlocking as a list.  You also need to explain the purpose of each signal arm on your sketch

 

You are of course correct. I attach my first locking chart ever, having already tried to integrate Grovenor's advice. Please don't hesitate to point out my mistakes:

[RailWest]

What about a FPL ?

[Chen Melling]

1 minute ago, RailWest said:

What about a FPL ?

 

As I wrote earlier, I omitted an FPL for simplicity.

This would definitely be included in a real Palestine Railways / Israel Railways design, but I feel it would be too complicated for our average visitor, who has not heard of a track turnout before. Many don't even know that trains don't have steering wheels!

[RailWest]

3 minutes ago, Chen Melling said:

 

As I wrote earlier, I omitted an FPL for simplicity.......

Sorry, missed that :-)

But it would give an ideal opportunity to demonstrate a lever which locked another one 'both ways', as well as showing that sometimes the release of one lever is reliant on the prior use of two (or more) other levers.

[Michael Hodgson]

1 hour ago, The Stationmaster said:

What might be useful as an explanation to those looking at that is al ocking chart showing what each lever locks or releases in relation to other levers - which in any case is part of the design process.  Normally the locking chart is drawn up first and what some of know as the Dog Chart (which is what you have drawn) is worked out from the locking chart

 

It might also help understanding if you were to explain the basic principles of interlocking as a list.  You also need to explain the purpose of each signal arm on your sketch

The words lock and release might seem intuitively obvious, but I think its probably also useful to explain what lock vs release mean, because I don't think the layman can read locking charts, although a chart with a column for locks and another for release is the easiest way of defining how if works, and helps pick up on Grovenor's comment about the redundant locks.

 

There is a logical sequence in which the locking is designed

1. First you lock points against conflicting routes, although it's trivial with this layout, as there's only one point.
2. Then you lock the signals so that they can only be cleared with their required point settings

which in this case means

• 1 and 4  lock 3 (in the normal position) while
• 3 (when reversed) releases 2 and 5

The converse of these is that

• if 3 is reverse, you can't get 2 or 5 and
• if 2 or 5 is off you can't move 3

Consequently if any of the signals is off it follows that the point can't be moved.

 

That just leaves signals for opposing movements  - which is why you would have

• 1 locks 4 (converses happen automatically)
• 2 locks 5

This is oversimplified because there are additional considerations not applicable here, such as distant signals being locked with the relevant running signals and shunt signals being locked against the running signal over the same route.

 

[Chen Melling]

1 hour ago, Michael Hodgson said:

The words lock and release might seem intuitively obvious, but I think its probably also useful to explain what lock vs release mean, because I don't think the layman can read locking charts, although a chart with a column for locks and another for release is the easiest way of defining how if works, and helps pick up on Grovenor's comment about the redundant locks.

 

The exhibition and the displays in it are aimed at the very general public, which is usually ignorant to most railway and technical matters and appears to have a very short span of attention. It must also be tri-lingual and accessible for the disabled.

Therefore the messages must be short and concise. I don't believe I will try to explain the intricacies of interlocking design, I am just aiming at getting the word "Interlocking" and its general meaning across.

Of course I also hope to arouse interest in those would-be enthusiasts who might want to learn more at our archives.

Edited February 1 by Chen Melling

[Chen Melling]

46 minutes ago, Michael Hodgson said:

 

This is oversimplified because there are additional considerations not applicable here, such as distant signals being locked with the relevant running signals and shunt signals being locked against the running signal over the same route.

 

 

Palestine Railways and Israel Railways never had distant signals in their mechanical installations, only inner and outer homes.

The German-inspired and supplied electric system does have distants, but these will be covered in the appropriate part of the new exhibition.

[Chen Melling]

Below is my attempt at improving the dog chart (partially) following Grovenor's advice. It is much smaller, but I feel like I've missed something. Any thoughts?

[Grovenor]

Well, that's not how I thought I described it, but it nearly does the job.

However, if you pull 3 and then pull 5, 3 should be held reverse but it isn't so can be moved back to normal.

[The Stationmaster]

20 hours ago, Grovenor said:

Quite a bit of excess locking in that dog chart.

 

The bottom 2 bars are fine, 1 and 4 lock 3 and 2 and 5 released by 3.

So your double ended dogs for 1 locks 2 and 4 locks 5 are redundant as its already covered via 3..

The top two bars can be combined into one, giving 1 or 2 locks 4 and 5, and 4 or 5 locks 1 and 2.

 

just 3 bars in total.

But don't forget he wishes to show and illustrate the way locking works so separating various functions in the locking could be a lot clearer to someone with no knowledge.  I think that makes a bit more sense for many viewers rather than than going in fora space saving layout of locking which might make it less easy to trace the various different functions.

[Grovenor]

Seperating is fine, but no need to duplicate, which would make demonstrating harder.

Do each lock seperately then you will have 6 bars rather than the 3 possible.

[Chen Melling]

1 hour ago, Grovenor said:

Well, that's not how I thought I described it, but it nearly does the job.

However, if you pull 3 and then pull 5, 3 should be held reverse but it isn't so can be moved back to normal.

 

Indeed I did not correctly understand or implement your initial notes. I hope this is better (though probably less explicit to the layman than my original, redundant version):

[Grovenor]

That looks OK now.

[Chen Melling]

Not contempt with an implementation where a double lock "floats" over a bar, I changed the chart thus, and am hopeful it still works correctly (I checked and double-checked, but would still appreciate another opinion):

[Grovenor]

Having a lock in the tray and not connected to any bar is a common technique, but your latest version is OK too.

You are not exactly short of space in this example.

[Chen Melling]

So now I have a correctly functioning version which is reasonably easy to understand and implement as well as being interesting and economical in space.
I want to thank all those who responded and helped me achieve this, and I will let you know when we actually create the display.

[bécasse]

There is one other issue, if there are two signals on the same post (ie not bracketed) ahead of a facing point, the normal convention is that upper and lower relate to left and right in that order (and likewise if there are more than two routes/arms). You have them the other way round. Bracketing signal 2 to the left, even marginally, might be the obvious answer.

 

Signals mounted one above another relating to different routes were once commonplace but started to become rarer during, say, the 1880s, at least for running roads. 

[The Stationmaster]

36 minutes ago, bécasse said:

There is one other issue, if there are two signals on the same post (ie not bracketed) ahead of a facing point, the normal convention is that upper and lower relate to left and right in that order (and likewise if there are more than two routes/arms). You have them the other way round. Bracketing signal 2 to the left, even marginally, might be the obvious answer.

 

Signals mounted one above another relating to different routes were once commonplace but started to become rarer during, say, the 1880s, at least for running roads. 

However for what amounts to a subsidiary movement with a smaller arm the usual cinvention is that irrespective of the 'left-to-right rule' the principa running arm is mounted at the top irrespective of teh direction of divergence signalled by the smaller arm.

 

There is one example I know of ( on the ER, exGN section in Lincolnshire) where the subsidiary arm was mounted above the running arm - the only time I've ever seen a photo of such an arrangement

 

Going back to the start I think a locking chart would still be a useful addition because it will show the locking which is required  leading then to the diagram which shows how the requirement will be met.

 

The example below is slightly different in that for space reasons the arm for the subsidiary routes has a route indicator but both of the routes it leads to are to the right of the running signalled route

 

https://www.flickr.com/photos/68861278@N03/14786921808/

 

[Grovenor]

Since this museum exhibit relates to the Israel/Palestine railways then the UK conventions on signal arm placement may not be relevant anyway. Railways with predominently single line and crossing loops tended to do what suited them.

[The Stationmaster]

1 hour ago, Grovenor said:

Since this museum exhibit relates to the Israel/Palestine railways then the UK conventions on signal arm placement may not be relevant anyway. Railways with predominently single line and crossing loops tended to do what suited them.

The railways in. palestine and Israel havea very complex history with regaugings (from narrow to stanradrd on some routes, new construction to stnadard fa gauge and some of teh narrow gauge left.  Originally French built and controlled the railways were run by the British during the Mandate period with lots of updating going on during that time plus they had been under British military controls at various stages during the two World Wars.

 

After 1948 and the effective partition of not only some of Palestines borders with neighbouring states but also within the country as Israeli and Arab dominated areas split and changed with routes vanishing or becoming divided.  Then to the modern IR  situation with massive modernisation and infrastructure improvement.

 

So there would have been early French operating practice succeeded by British Military Railway practice in places and then British management of. the network for over 20 years working to British practice but with a varied loco fleet.  It must be very interesting putting together museum exhibits reflecting that but from what we've seen in the is thread the interlocking seems more represeentative of British practice than any other

 

 

 

 

[Chen Melling]

4 hours ago, bécasse said:

There is one other issue, if there are two signals on the same post (ie not bracketed) ahead of a facing point, the normal convention is that upper and lower relate to left and right in that order (and likewise if there are more than two routes/arms). You have them the other way round. Bracketing signal 2 to the left, even marginally, might be the obvious answer.

 

Signals mounted one above another relating to different routes were once commonplace but started to become rarer during, say, the 1880s, at least for running roads. 

You are correct, in principal, and this was also how the Palestine Railways regulations describe the situation, though The Stationmaster's comments below also apply in our case, and we have life-size examples of both types in our museum's collection. Nevertheless, I intend to amend the final graphic to show a bracket signal, and thus hopefully make it clearer and more intuitive to the general public.

 

3 hours ago, The Stationmaster said:

However for what amounts to a subsidiary movement with a smaller arm the usual cinvention is that irrespective of the 'left-to-right rule' the principa running arm is mounted at the top irrespective of teh direction of divergence signalled by the smaller arm.