2007 CTs on 2012 IRSE tables

[dorothy.pipet]

Comments and feedback please?

[PJW]

I have looked at the route and aspect tables; I'll return later re the points. In general I think they are very good, but I have included some feedback on specific items.

Notes sheet

Definitely worth keeping these on separate sht for the whole set of CTs, so recommend your approach to others.

I like the introductory notes and of course defining any $ note that is utilised later on the CTs.

Careful re the last one though, since there is a swinging overlap and therefore not quite that straightforward in the case of 231 in 113A(M)- there is specific column on the CT for “detect N or R at time of clearance but overridden whilst points can legitimately be swinging”. Also I think I am right that to the most recent standards, NR no longer detects facing points in the overlap, other than to condition out those tracks applicable to the potential overlaps which are not in use.

Also worth stating-
1. you are assuming SSI rather than RRI technology,
2. track circuits rather than axle counter,
3. policy relating to providing / not providing shunt route overlaps- this particular layout suggests provision at least beyond 127. Be aware that overlaps beyond a GPL as an exit signal are a lot more common (traditionally may not have been shown as overlaps but typically there was a 200yard exclusion zone beyond that GPL for passenger movements) than an overlap for a shunt route reading upto a main signal (something that has only really arisen in the last 10 years in standards- therefore really only about 7 yesr in actul commissioned sites). More recent IRSE layouts now typically provide shunt overlaps- however when a candidate is stating the principles they are adopting there can be a conflict between them and the given layout so it is worth stating explicitly- particularly as you have chosen to give an overlap on a goods line, then non provision of shunt overlaps seems rather inconsistent. The problem is that “UK mainline signalling practice” is continually evolving…..



113A(M).

I agree with note #1 but wonder whether worth including at such length. I do see sensible to explain in this case why you included in case examiner had a different view, but I think wording such as
#1: Route locking ensures maintained in circumstances in which 124A(S) is subsequently cancelled, but alternative option would have been to prevent 124’s approach locking releasing once a train en route to it.

Since 138A(M) holds 233R and the locking is held indefinitely even if the pre-set shunt 124 is replaced (beware those working on Western Region E10k relay rooms- this is NOT the case with the traditional “facing shunts”) then the opposing locking after 138A(M) is superfluous.

Not convinced that the “or 231R$37” condition on CJ is doing anything useful; unless 233N$37 then CJ will be replacing the aspect anyway by the entry above. As a tester I prefer the track to be listed only once with whatever the point conditions are, rather than having multiple interrelated entries as you have shown. Actually switch diamonds don’t really offer any flank protection and if a train waiting at 173 then eased back then I think it could come foul anyway regardless of the state of 233. Hence can be better to condition out CJ by a track sequence “CJ occ after CH, CJ occ” rather than via point detection (tend not to use route locking so that doesn’t give reversions during handsignalling / possessions etc)

Missed a foul track for the reverse overlap: (CK or 232R or 231N); don’t see there is any value in calling 232R though.

Aspect sequence does not name the signal ahead- the IRSE’s CT blank does not encourage you to show but I think I would.

No reference to signal replacement condition which I feel should be shown, particularly since you note the auto facility. The IRSE blank however does not provide any specific column so is easily forgotten.

No reference to signaller having different exit buttons for the M and W routes



113A(W)

No comments other than those duplicating the above; even if you didn’t feel need to state only the one ROL then still need the comment about route step-up. Actually I would have enhanced that, stating that step up will occur when a forward route is set and the relevant full overlap would then permit aspect clearance.



129A(M)

#3 I don’t see the need to call 253R. I would regard a 55m length into a sand drag as actually a better overlap than 75m onto the running line- it is not as if 253A are just trap points designed to through the train off the track. Of course at the time the layout was drawn, Goods lines would not have been given overlaps. However because you have written the note, the examiner knows where you are coming from and will mark accordingly- shows the value of stating your assumptions!

You originally showed 158B(S) as an opposing route but then deleted; I think that you should have explained why by stating an assumption, because I would have thought it was needed. Perhaps you are assuming that this route would set and lock 251N which would be sensible, but you have not made this clear.

Despite calling all the necessary flank points, the route is still vulnerable to a SPAD at 127, so aspect reversion once one is detected (or more traditionally just putting in the aspect the flank tracks SE, SF, SG) would be appropriate.

Similar comments as 113 re condition on foul track CJ, non statement of replacement condition etc

Approach release just on BH seems too restrictive- we want to keep the freight train rolling slowly but the driver that close to the signal will have the brakes fully applied and barely moving- it takes significant time to release the brakes on a long freight!. Hence should have included BF or, if we judge that the PLJI not readable from that distance, then BF occ for say 10seconds.

A/L release time not shown after you apparently changed your mind; I guess you intended to reduce to 60s given the route is MAR.


129A( C)

Some of the above comments also apply; you should also have made clear whether (as to modern standards) the signaller is given separate exit devices to select the route class and the relevant track circuit controls included at route level in each case. Again just because the IRSE’s blank CT did not lead you to this, shouldn’t mean that you overlook.

You have shown that “at least one of the tracks AB or AC have to be occupied for 60 sec” for the aspect to clear. If we ignore GK/RT0044 (Huddersfield) locking for a moment- I am not sure that I would provide just for trains into a Goods loop (but I have been out of mainline railways for a while!) then for a call-on we would definitely require the occupancy, but we also need approach release implemented on the entrance signal- in this case BH occupied does seem appropriate as I judge that this may be under 100m long (but worth stating an assumption or writing (BH for 7 secs) to make an allowance for it being a bit longer.
If we then consider GK/RT0044 locking then need to be a bit more explicit re what is a route level control and what is an aspect level, but clearly allowing the signaller to set the route once the first train in in the loop yet holding the aspect level until it is timed to a stand is sensible. Don’t get too hung up on this niceity and then forget the approach release!




146A(S)

Would have been worth stating an assumption / giving an explanation relating to opposing route locking imposed by routes up to 129. Clearly one would expect the overlap locking to be released after train timed to stand so that the junction could be utilised, but I agree there wouldn’t want to be able to set routes such as 146A(S), given the route box stipulation of having 124 off for aspect clearance. If the route setting were permitted one could get situation in which 124 is cleared for a train on BH, route 146A(S) then able to be set and as BH clears then 146 aspect would momentarily clear until 124 then returns back on. Therefore I would agree with your CT entries but it would be in your interest as a candidate to let the examiner know you had thought deeply about it, rather than merely forgotten to time out the locking!
In fact I might be tempted to include BH in the aspect of 146A(S) to avoid such timing problems if we are permitting the oversetting of this route whilst 124 could be off for a previous train.

Again you haven’t mentioned signal disengagement and of course for shunt signals the aspect should only replace once the berth track is clear (unless the train is now occupying more tracks than would be explained by the longest expected propelling move)- this is usually shown as $40 references on tracks on NR control tables with separate entries relating to the disengagement and hold-off conditions. Think you need to decide how to use the IRSE type blanks to depict one way or another.

The A/L should not be imposed if the berth track of a GPL is clear, but you ahve not shown this in the CT.



182C(M)

Should set, lock and detect 261N to give trapping from siding.
I interpret (although it isn’t completely clear) your control table as requiring the 138 route set control in the aspect level of 182 but not in the route level. You have only listed tracks clear up 138; you should have included some more tracks so that there is an aspect overlap (even though the layout plan doesn’t define a route level overlap since the relevant locking is being performed by the following route), so I would have added EA, CJ , BJ, (CK or 232R).

Comments and feedback please?

[PJW]

The points tables were also good.

For 233 however there is a typo as the locking in the overlap should of course be after 113A(M) not 133A(M).

The condition on that point call expression is stated as "w 231R$37" but I would not have expected detection to be relevant here, certainly not the way that SSI does swinging overlaps.
In general facing points are left where they happen to be lieing and any necessary facers beyond them set appropriately when the route is being set, it is only if not all the required overlap beyond them is available that the call is placed on the facers to move across to select a different overlap (since the route availability expression has already succeeded, one does know that at least one overlap must be possible).

Conversely the associated locking would need to be imposed unless the facing points can adopt their other position to swing to an alternative overlap. With RRI the circuits first check the other overlap is available, then call the facing points and then finally release the locking on those points no longer required in the former overlap only after the facers have achieved their new detection.
SSI actually just looks at the availability of the alternative overlap and simultaneously locks the new one, releases the old one and commands the facing points to move. It is actually non-compliant to standards in that respect; the locking of the overlap is still released even if the facing points never actually move.
So the actual $ ref used here does depend on the technology, but the locking should certainly be expressed as [113A(M) or 231N]; in spoken English it often is clearer to substitute the logical operator "or" by the word "except".

Also in that same expression should think about the possibility of the train coming to a stand in the platform but a fair way from the signal (perhaps it is a short train and the footbridge acces to the platforms is at the left hand end) so may need to time off the overlap with a train on BF but not BH; otherwise the junction area is locked up indefinitely.

Another thing is that the overlap locking on BJ, CJ isn't adding anything over and above the track occupancy that is stopping the points from moving by virtue of the dead locking anyway- the designer might perhaps decide to put in the data but it is logically superfluous and is certainly untestable (as opposed to route locking where there is "track bob protection" and the locking is held for 15s after a track section unexpectedly becomes clear). Just like route control tables only list as opposing route locking where actually needed and omits showing for those routes which are preventing by virtue of point availability whatever is actually there in the data, this is similarly superfluous and therefore should not be shown. If the designer does show then the tester must annotate, since cannot tick it and if left blank then looks as if not completed.

Hence I would have written for this entry:
[113A(M) or 231N (set & locked)]: BE [BF, BH ----or (BF or BH occ for 45s)]


235 (which you forgot to specify!)
This was good, but you sinned by utilising $39 without first defining its meaning.

Although you remembered the need for trapping on the vast majority of the relevant routes, you overlooked 129A,B.

You seem to have locked 235 after 109B(S) that I would not have done; there is a shunt ovbverlap shown beyond 127 but only including SE not SF. The locking is a bit restrictive re preventing what could otherwisehave been a simultaneous move and since the shunt overlap is actually shown then I don't feel that you should have locked; you'd have had a better excuse had the layout been silent on the subject. However given that you consider it necessary then in thi case it is correct that there is overlap locking on SE track (since does not deadlock points) but SF is superfluous (since it does).

A general item re saving valuable seconds in the exam relevant in several places but here in particular; where many routes have the same tracks for their maintained locking, then list the routes together and bracket so that you only hacve to write out the list the once for the several; pairing the routes from 146 & 148 would have been beneficial here for instance since they are so obviously the same.

All in all a very creditable set of Control Tables; the challenge is to do as many CTs so well within the time constraints of the exam; in 2007 candidates had 1 hour but within that time had also to draw a blank control table from memory and then arrange for these to be photocopied; if you could do both blanks within 12 minutes you were doing well.

You correctlly concentrated on the basic locking; it is important to do so in the exam. Don't worry about perfection; this is an exam not the day job.

If at work you only got 50 percent of the entries on a CT then you'd deserve to lose your license and your employment; in the exam you'd probably pass. Hence it is important to switch brain to "exam mode"; do attempt all CTs- whereas you might think that doing some well would be a better demonstration of your competence, in the exam it is necessary to accumulate marks. There is an allocation per route or per point; if you don't attempt all of them then you are throwing easy marks away whilst persuing others that are harder to get since the law of diminishing returns applies.

Comments and feedback please?

[dorothy.pipet]

You say "Aspect sequence does not name the signal ahead- the IRSE’s CT blank does not encourage you to show but I think I would."

I thought about this because Network Rail RRI tables list the exit signal twice, and SSI tables state the signal(s) ahead with the Aspect Sequence information. These tables have the exit signal on the LHS so I decided that ws sufficient.
Would you state the signal in the top row of the column?

[PJW]

Agreed- the blank does not encourage; the same is true for signal disengaging, last wheel replacement etc on the route & aspect CT as well as Time of Op locking and swinging overlap counter conditional on the points CTs yet I feel they are all needed.

As far as disengaging controls are concerned, I would follow the old Western Region practice of putting notes against the tracks in the tracks clear in aspect column. You wouldn't need to use the official WR #notes (predecessor to the national $ notes), but from distant memory:

#1 = stick track (the WR just used TC beyond signal without demanding berth TC occ)
#3 = stick track, but doesn't disengage when signal working auto

for forms of delayed replacement then in addition qualify by a 2nd note
#4 = non replacing whilst berth track occupied
#5 = non-replacing until stick track occupied and cleared.

Typically an entry could be:
AE#1#4#5, AF, AG, AH
so very economical with space and no need for an extra box on the CT and can get into the habit of always qualifying the first track in the tracks clear entry- so less chance of forgetting.


To return now to the actual question here; yes the NR tables do show the exit signal in two columns for good reason. Although usually the same entry in each,

1. a move to bufferstops does not prove the bufferstop alight and there is no sequence dependent upon it

2. a call-on move is set to exit signal button, but again not proved alight and no sequence depending on it

3. in isolated 3 aspect signalling, the route is set to the next red / green signal as the exit, but the spect sequence is derived from the intermediate yellow / green signal and the entrance signal proves BOTH the exit signal and its repeater alight

4. some sequences involve banner signals that are proved "alight or controls off" and which also affect aspect sequence, holding the signal in rear to yellow at best if not alight but off; "geen banners" are more complicated, as are co-actors and splitting distants.

Hence in preparing for the exam you need to consider how you would use the IRSE CT blank to be able to depict any of these possibilities- some might need the use of the last column to add as note, or
a cross reference from a relevant other column to a specific free-form entry on your initial CT Notes sheet; I thinkthis would certainly be the only option for a junction signal having a flashing aspect sequence in rear.

The reason why I would (in the simple case you are thinking about) have put the signal number at the top of the column is that to me
it is an important principle that on NR, for main routes at least, there is an end of movement authority proved alight and, where necessary to protect a junction or other high risk conflict, a TPWS loop proved active.
Agreed it may be better alternative to state this as an explicit assumption in your initial notes sheet. You still need to think about what to do in the less simple cases mentioned above.

However the important thing is, one way or the other, to make sure that the examiner knows you know the principle and knows that you have thought about it- that will be what gets you the mark; don't leave them wondering......

You say "Aspect sequence does not name the signal ahead- the IRSE’s CT blank does not encourage you to show but I think I would."

I thought about this because Network Rail RRI tables list the exit signal twice, and SSI tables state the signal(s) ahead with the Aspect Sequence information. These tables have the exit signal on the LHS so I decided that was sufficient.
Would you state the signal in the top row of the column?

[dorothy.pipet]

On the Aspect sequence, agreed that there are good reasons for being able to state the Lamp proved/exit signals and the next signal separately. This was one of my first queries over the SSI-style tables since they try to combine the two - not ideal for the more complex situations.
So an general note along the lines "Aspect Sequence Signal Ahead refers to the Exit Signal in Column 3 unless otherwise stated".
I need to compile a list of these which I can learn and scrible out fast.

[PJW]

Yes I would recommend that; you will note that there are some such attempts posted on this website but that they tend to predate the IRSE CTs and therefore whereas they may give inspiration would need to be updated. I imagine that I won't be the only one looking forward to seeing your revised list.

I didn't say that I won all by battles when the group was compling the new 11202 standard.....

On the Aspect sequence, agreed that there are good reasons for being able to state the Lamp proved/exit signals and the next signal separately. This was one of my first queries over the SSI-style tables since they try to combine the two - not ideal for the more complex situations.
So an general note along the lines "Aspect Sequence Signal Ahead refers to the Exit Signal in Column 3 unless otherwise stated".
I need to compile a list of these which I can learn and scrible out fast.

[dorothy.pipet]

Then can you point me towards these previous attempts please? I've had a little rumage and it doesn't seem obvious.

Who wins all their battles? Our use of the SSI-style tables is evolving as we go, having used them for several large resignalling projcts now.

[PJW]

I agree that isn't easy to find!
I definitely remember "AlexGoei" from Singapore posting several items on this topic 3 or 4 years back, but can't recall if in attachments or as a straight post. Didn't locate- just possible that was in private email to me. Thought there were some in other CT offerings but again haven't found readily.

I have however found two of my own by using the Search facility

http://irseexam.co.uk/thread-967.html?highlight=notes

http://irseexam.co.uk/thread-275.html?highlight=notes

If I chance upon any others I will come back and add them here.

Then can you point me towards these previous attempts please? I've had a little rumage and it doesn't seem obvious.