(10-01-2011, 02:21 AM)ivanutd Wrote: Hello PJW,I am using your (and the usual modern NR nomenclature) for the routes, rather than that defined by the route boxes; if confronted with such in the exam then you need to give an initial sentence of explanation.
First let me wish you a Happy New Year.
I have uploaded below the CT's for all the signals and available routes on 1997 Layout and i'd like to ask, whenever you'll have time for this, please have a look and let me know of any issues you'll notice.
Thank you!
First looking at 201A(S) as an example:
Just list the route normals 122A(M) and 112B(M) without any point condition. Whichever the lie of 302, 201 can't set; there is no condition.
Note that 202A(S) and its tracks should not be listed; this route locks 302R until train has cleared BF and therefore the locking is purely via these points.
There should however be opposing locking from 203B(S) and indded there are also call-on routes as well as main routes from 122/112 that also need to be considered; you should be looking for:
a) any route with exit in rear of 107 (in this case none)
b) any route with exit at any opposite direction signal within route (in this case 108)
c) any route with exit at any opposite direction signal within overlaps (in this case none)
but then consider whether any are not necessary to list as opposing routes due to the presence of point locking.
Having decided (quite reasonably) that the opposing locking should time out by platform track occupied for a time
(note that you should have permitted the train to have released the locking whether it was occupying '(BH or BG) for 60' in the 5th column and thus also have included BH within the bracketted condition in the 4th column)
then it is nonsensical to demand BG, BH tracks clear within the aspect.
I agree it does seem sensible that 201A(S) is made permissive so that can add vehicles to a train already in the station (apart from anything else plan states the platform is permissive), so therefore the only track that can be proved clear is BF. You could though have stated that your railway's practices only permitted a shunting move into a clear platform and the IRSE would mark as such; HOWEVER YOU MUST BE CONSISTENT and in that case the opposing locking should NOT time out. However I'd actually advise you to treat shunts differently from main running signals so that you can demonstrate your familiarity with the difference between various route classes.
I am assuming the '4' in the disengagement entry is a typo.
Looking now at 107A(M)
In this case being definitely a non-permissive route, there should NOT be any entry in the 5th column- the opposing route locking should not time off and indeed those very tracks are correctly proved clear at aspect level.
Similar comment as before re point condition in the 3rd column.
You were right that the availability of 306N can be made conditional on 303N but you failed to include 305R w 303R. Actually I would also have included '305N w 303N' in order to provide flank and '304N w 303N'- though see later re how to express these conditions in the correct manner.
To be honest I'd have probably included the availability of 306N in either overlap (although would NOT have proved detection w 303R) in the 6th column).
Another small issue is that when writing the expression for point availability then it is conventional to express using 'or' conditions, so this would might have been expressed:
[303N, 304N, 305N, 306N or 303R, 305R]
on the assumption that we have decided not to care re 306 when the overlap is directed towards the branch. You might also care to define one lie of 303 as the preferred overlap if you felt this appropriate.
The opposing locking listed from 124A(M) is bogus; that route would be locking 303R until the train had cleared AE and therefore it is this which prevents the setting of 107A(M) for an overlap over 303N. Further it should NOT prevent the setting of 107A(M) since that route has the option of an overlap over 303R.
If we now turn attention to the 7th column, there is also some refinement needed to the tracks listed within he swinging overlap. Being a 'route relay man' at heart, I'd have worded as (DB or 303N) rather than (DB w 303R) but let that pass initially. BUT you do need to be careful re the logical correctness of your expression even if implemented in data:
'DB w 303R, BK w 303N' actually means the same as
'DB, 303R, BK, 303N' since both the 'when' and the ',' are logical ANDs.
If you evaluate your expression then you'll see that it is never TRUE since you claim to require 303R and 303N which obviously can never be satisified; you could have written: [(DB w 303R) or (BK w 303N)] to overcome this objection- however I'd still advise writing as: (DB or 303N), (BK or 303R).
Actually though it is not quite as simple as that, since even with the overlap set to the branch then you do need to worry about BK since that joint within the double junction is FOUL.
Therefore for that overlap you need to incorporate the expression (BK or 304R) so that the portion of BK on the Down Main is proved clear. Conversely it is certainly reasonable to assume that the BJ/DB joint is CLEAR, given the width of the platform and presence of the ROL at it.
So that would make the expression: (DB or 303N), (BK or 303R, 304R).
In the aspect sequence you have shown 107@YY up to 111@Y; but when will 111 show Y, since 113 has no R and 111A(M) is approach released?
In the comprehensive A/L lookback you have listed tracks as far back as BB. This means that the driver could already have passed 103@G and then the signaller cancels 107 which is then permitted to release its locking; the first the driver will know is when sighting 105@Y and then the train only has half the braking distance it needs to be able to come to a stand before being endangered by another train which perhaps has now been routed over 302R. You needed to take the A/L back at least as far as the sighting point of the furthest signal that changes aspect as a result of the route being cancelled- in this case 103 goes back to YY and so need to have included BA.
You should also consider 'overrun protection'- the reversion of the aspect if there is a SPAD at any of the signals protecting the route in question.
===========================================================================
Having looked at the first couple of routes, the impression I have got is that:
1. you have got your head around the Control Table presentation (and indeed demonstrated how efficient in terms of space and therefore to some respects also time),
2. you are broadly familiar with putting the required locking in the relevant boxes- actually you have got a lot of it right, but there are enough errors and inconsistencies to raise an alarm that although the veneer is quite impressive that there are problems below the surface- the examiners are on the lookout for such and merciless.
3. I have come to the conclusion that you don't actually always understand enough about the significance of what you are recording. Probably the activity is a bit too 'formulaic' rather than 'considered'; you do need to know the significance of the locking and WHY it is provided.
I suggest you probably ought to take a step back from attempting specific Control Table examples for a while and try to do more general reading / studying to increase the depth of comprehension of the subject. You have plenty of time prior to the 2011 exam to do this and then return to get some more Control Table practice. Anything that you can do to get a deeper appreciation. be it studying relay circuits, Computer Based Interlocking data, observing Principles testing or even a signaller setting routes etc for trains will assist. It is probably best as an iterative approach, so with a broad understanding of CTs then you should be able to pick up info relatively easily from other sources and then return to CTs with fresh eyes for a while before going back to build up understanding further.
PJW
