non-reciprocal locking

[PJW]

Hi Signalling professional:

Currently, I am reading the railway signalling book (by O.S.Nock) and on page 54 it explained the term ''non-reciprocal locking'' and ''delayed yellow condition''. But I still don't get what it is trying to explain.

Can anyone explain it plainly?

sorry for my poor comprehension skill on the book


thanks

First I'll attempt 'non-reciprocal' ; then the actual description of the "delayed yellow".

Generally the majority of interlocking is reciprocal (i.e. symmetric converse).
In mechanical signalbox terms:
if lever 1 locks lever 2 then lever 2 locks lever 1.
i.e. we are wanting to prevent both levers being able to be reverse simultaneously, so whereas either can be pulled initially, once the first one has been pulled, then the other one cannot.

In a mechanical locking arrangement, this reciprocity arises naturally from the operation of the locking nibs in the tappet blades, however in an electrical installation (or indeed a mechanical signalbox in which the locking is provided by electric lever locks) then the reciprocal locking has to be explicitly provided (i.e. one circuit is the electric lever lock for lever 1 which includes the proving of the position of lever 2 and a totally separate circuit is the electric lever lock for lever 2 which includes the proving of the position of lever 1).

However in some less common circumstances it is desired to have "non-reciprocal locking"; i.e whereas the locking of lever 1 is dependent upon the position of lever 2, the converse is NOT desired- lever 2 is to be free from any dependency on lever 1.
In a mechanical signalbox the classic example of this would be requiring the signaller to pull the levers in a specific order, a semaphore can only be pulled if the one to which it reads is initially at danger, but then that signal can itself immediately be cleared (the reason for this is that the logic is part of the means of enforcing the overall requirement of ensuring that a new "line clear" is obtained from the signalbox in advance for each separate train).
It is possible to implement this locking entirely mechanically by using nibs with springs etc, but generally it was difficult and was easier to achieve electrically so this became the more common means even on what was otherwise a mechanically interlocked installation.
======================================================
So now coming to the specific text on page 54:
[and I remember struggling with this wording myself when studying this book some 30 years ago- I really don't think that this book is written in a good way to explain to the novice.......]
what it is trying to say is-
a) whenever practicable, signal 10 to sigal 12 should be established with a full overlap which means that sigal 8 would be Green, signal 10 Yellow and signal 10 Red
b) there are circumstances where this is too restrictive as it prevents other movements over 112 points and therefore the sigaller is given the option of selecting a "restricted overlap" [i] [this is the current nomenclature although you'll find that it is called "reduced overlap" within this book as that was the nomenclature used back then]
c) if this shorter overlap is chosen then the risk of overrun is mitigated by deliberately holding signal 10 at Red (called here a "Delayed Yellow") until a train arrives at it (and thus would have received a Yellow at signal 8), thus ensuring a slow speed approach and thus lowering the likelihood of an overrun at signal 12. Point 112 remains free and thus for example signal 52 can be used for a different train movement.
d) however should 112 then be placed normal and the restricted overlap beyond signal 10 permitted to become a full overlap (e.g. by Forward Route Setting from signal 12), then at that time points 112 must become and remain locked (because once signal 10 has been allowed to clear then "the cat is out of the bag" and the train may now be approaching faster). In more modern terminology it is the equivalent of a Warning route "stepping-up" to the Main route.
e) Hence whereas signal 10 does not always require 112 normal, 112 is locked to be locked normal once signal 10 has cleared with a full overlap.

To be honest I myself wouldn't have used the nomenclature "non-reciprocal locking" in this context; I see what he is trying to say, but doesn't seem particularly appropriate and certainly not helpful to the explanation.

Does the above help?