Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
2014 Q2 Release of Route
#1
Forum has been very quiet this year, but actually I have had a lot of attempts from a range of individuals, particularly recently; everyone seems very shy in sharing their work......

However I have permission to put some attempts on which I have commented on the Forum, so here is one of them attached with some comments written on and others below.

2014 Q2:

This was probably your least good attempt; it wasn’t bad, but nothing like as good as the others in the batch.

Question asked about release of route locking, which yes certainly includes the release of Approach Locking but of course should also talk about the progressive release of route as the train proceeds through it. 
To an extent TORR is also relevant, but all this is doing is initiating the process; the equivalent of the signaller pulling the button.  If you had introduced it in that light, it may have been reasonable to have started with this, but you majored on it too much.  
The fact you actually started with the stick disengagement (which yes does turn out to be one element of TORR) definitely started the answer off on the wrong foot; the signal stick is an ASPECT level control to stop the signal re-clearing and (apart from its inclusion as a condition in TORR) is not relevant to route releasing. You actually seemed to treat it separate to TORR as you next considered A/L Release.



In part b) you didn’t make clear these were alternative releases (but you did later on); nor that lookback is generally not provided as an option.
Then you talk about TORR (see later) but say nothing of the sectional route release, track bob protection 15sec timing if next section not occupied and this was a serious omission.
 
Also you were weak re the REASONS.  I think there is a similar but not identical) question on IRSE Forum that should be helpful to you to improve this area.

This question would have been a Pass, but only just I think.
 
As far as TORR is concerned, here is some history that might help with your query.

  1. Originally it was novel and incorporated on some large Southern panels from the mid / late 1970s.  there was concern re putting signals to danger too early; also automating the releasing of the route which had always been signaller responsibility.  Therefore many controls included to minimise risks.

  2. In those days there was (almost) no auto route setting.  Lots of signals were given Auto buttons.  Made sense to make use of the comprehensive A/L lookback circuitry that was generally provided.  The proof used was GSR down (i.e. train had passed signal when showing proceed aspect), plus the TASR (sequence of 1st and 2nd track occupancy (normal sequential track release for A/L) plus “something extra”. 
    Generally this something was the “lookback free” (i.e. no train approaching- so very much belt and braces- train really must have entered route because two things were telling us that it was beyond signal plus confirmed by something else saying that also no longer approaching).  
    If we didn’t have that then could use 3rd track occupied (and this soon changed to a sequence of 2nd and 3rd) or perhaps a treadle (that we really, really tried to avoid). Slight problem was that didn’t always have 3 tracks; also it could mean that TORR didn’t happen as quickly as might want so slightly restrictive.

  3. With increasing ARS, needed more TORR. 
    Also less need for Auto buttons (since ARS only sets routes for train at last minute - well to achieve two greens ahead of the train- ; signaller may well do much earlier or could delay train if he gets distracted later, so more chance of wanting to change decision / ability t do so). 
    Lookback A/L is time consuming to test particularly if SSI (although ironically far quicker if MAP file is also used for TORR).  
    Also if points in rear happen to be moving at instant train passes signal, then TORR failed- wasn’t obvious to signallers why.  Hence fell out of favour; life settled down to 1st and 2nd for TISP, then 2nd & 3rd for TORR. 

  4. Decided that better to make it occur earlier and so then TISP used berth & 1st, allowing TORR to use 2nd & 3rd.  Slight reduction is safety one could argue, as some of the independence lost since same tracks as GSR. 
    Also for a station platform where could be occupied by multiple train, then A/L release would fail so would need to time out.  Less of an issue now that platform sharing is strongly discouraged by ORR (although of course some places still common and hence likely to use different tracks to avoid issue). 
    From what you say, it seems that in recent years people have decided to swap the TORR to use berth and first with the TISP; that at least means that if berth track stay occupied then it is only TORR that fails (so signaller pulls button) rather than A/L (so no great delay)- seems sensible.  Actually I was unaware of this, so I have learnt something.  Have asked around when this formally changed and seems that formally it hasn't, only in practice!

  5. TORR has been around for years now; everyone much more comfortable with it.  Indeed it is now being done in SIL2 Control System rather than interlocking, particularly where an existing RRI being transferred to ROC control.

  6. It certainly was a surprise when you said Derby had unconditional TORR.  I have checked; this is the same for Port Talbot.  I asked around- no standard acknowledged this yet at all, but projects specifying contractually…….
    We wait for the Christmas (in more than one sense) for of the new NR interlocking standard Principles which hopefully will resolve all this.

  7. Regarding the order; actually it has ALWAYS been possible for the TORR condition to be achieved before the A/L release; it is just stored waiting for it to happen and once A/L has actually released, then the entrance button is pulled by TORR and that is when the signaller sees TORR happen.  In fact in the old days, the best way to test the MAP file was
    a) drop berth TC
    b) leaving that occupied, do sequential release of A/L (watch SSI memory to make sure had occurred)

    c) walk the train backwards from the signal under test, first on 1 then on 2 then on 1 track etc at a time until get behind signal in rear- see the route TORR as expected
    d) leave that track occupied, clear and replace that signal to approach lock, then go through steps a and b again, lift that signal's berth track.  See still does not TORR this time, then gradually walk back as before to the next signal in rear and repeat.
    i.e. “preselect the A/L release, then exercise TORR”.
    However if you had 3rd track TORR, then could satisfy the TORR condition first (i.e. pre-select it), then operate the TISP tracks in the incorrect sequence (to see not release), followed by correct sequence (and see the TORR)

.
 
PJW
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)