Good start- even though the bufferstop was obviously an after thought.
Note that London Underground does things slightly differently. A short distance prior to the buffers are a pair of Fixed Red Lights mounted on a vertical but pivoted thin post- the idea being that the movement authority is only as far as the FRL but if the train goes too far these will be knocked down and the train has a margin to continue sloing before it hits the buffers. The TCIs are positioned prior to the FRL, designed such that the front wheel of the rollingstock will hit them just prior to the FRL being knocked over by the front of the train- the assumption (which is very nearly true) is that on Metros there is only a single type of rollingstock utilising any particular line and hence can design the infrastructure to fit the trains like a glove.
I must admit that I wouldn't have thought to note that wide-to-gauge to have TCIs on the switch rails; however I can see that the train could leave the track as soon as the gauge widens and therefore the wheel may not get as far as the end of the stock rail, so that need to have one at first opportunity and no doubt that is on the outer edge of the switch rails. So well done for pointing that out (if the examiners are like me then they'd give particular credit when an answer actually teaches them something or revives a memory that had faded); however the "icing on the cake would have been to explain WHY the 4 are needed as I have described above).
In
part b) you were asked for
risks so you should have mentioned the potential loss (life, health, equipment damage, environmental damage, train service cancellation and disruption, public image) that would result from such an incident. Actually at this time there is a very obvious example in the news
from Canada re the Lac-Megantic accident . I bet Ed Burkhardt might be asking why there evidentally were no trap points guarding wherever the oil train had been stabled.....Be careful though - the question is about the mitigation that the TCI provides, not the trap points themselves.
You correctly talked about the associated
hazards (a train being foul of another train movement yet being invisible to the train detection leading to collision, but you should have briefly listed the risks associated with that collision (definitely equipment damage; a range of major and minor physical injuries and physiological effects; possibly further consequences resulting from outbreak of fire, a hazardous chemical spillage, water course contamination etc).
When explaining the TCI associated with the sliding bufferstops, you should have pointed out that if displaced by a minor collision then the ability safely to arrest another train on a later occasion will have been compromised; therefore the level of mitigation expected wouldn't be achieved and greater consequences would result- people in a train subject to rapid deceleration are likely to sustain injuries by being thrown around, objects coming loose etc. Hence the need to detect it has happened; if relatively minor initial collision then the driver may feel that has got away with it and not report it or there could be a subsequent failure to get around to rectifying if it did get reported. Having an alrm showing on the VDU screen (as per recent practice) reduces that chance.
There is also the more immediate hazard that if there is a collision that train may be partially derailed and, being constrained by one platform wall, is almost certainly going to become foul of the adjacent track in the same platform well. In this case it is not good enough to force the relevant TC to become occupied (though I note with pleasure that you wre careful in your wording earlier than you stated that operation would put the track circuit to be "not clear"- an undefined state so treated neither as clear not occupied) as permissive moves would still be given a PL aspect. You are also correct that the same applies for bay platforms adjacent to the through road and obviously the likelihood of collision and consequences of it would be greater in that scenario because of the speed of traffic likely to be higher.
I think what you should have mentioned in part a) and then worked into your explanation in part b) is whether the TCI would still be needed if the train detection were by axle counter- I do know that it is coming in part d) and you have partially addressed it, but feel not sufficiently for the various different circumstances.
I agree that having counted a train into a section that section will remain occupied even if the train has all its wheels subsequently removed from contact with the rails, so there is no hazard created by false clearance of the track section. However there is still the hazard, depending on the site layout, that the train has moved to be foul of a train movement that does not prove the section which contains the TCI as, provided the train is properly on that track, it is not foul. So it isn't just "because we always used to provide a TCI" that modern schemes still feature
some.
Even where this is a concern and there is a need to detect a derailment event and then include this as something that replaces other signal aspects, the functionality of a TCI can be provided in interloking data- if a track occupied without the associated trap points detected reverse, there is a definite likelihood that there has been a derailment; the interlocking then remembers that this has occurred until a technician confirms from site that there is really not a problem. This is what is known as a "virtual TCI"; hence the wording of the standard utilising the word "resetting".
Similarly there may be better ways of detecting that a sliding bufferstop is positioned where it should be than actually using mechanical movement to break a TCI, although this does have the advantage of being reletively cheap, simple, visual and effective.
Actually I think I would have used A3 paper in landscape and had 6 columns;
1. The headline as per part a)
2. The risks as per part b)
3. The hazards as per part b)
4. The mitigations as per part b)
5. The description of what happens as per parct c)
6. The applicability or otherwise for axle counter train detection as per part d).
Less writing prose, more clarity using the structure imposed by the tabular approach, more certainty that you were fully addressing all elements of the question.
Also more economical because don't need to repeat anything as one tends to have to if answer question sequentially.
In
part c) I think it would have been worth describing what happens should the TCI and its connection fall onto and make casual contact with the rail beneath it; an insulated TCI will be of opposite polarity and so shorts the track whereas the older non insulated TCI could by bad luck just repick the track circuit again. Obviously the modern practice of wiring the TCI to a repeat relay has no chance of getting the TC up again after it has been dislodged.
I think that modern SSI type schemes typically wire the TCI t a repeat relay and this then cuts the TFM input and then unsets a latch; this latch is used to prevent the clearance of relevant signals- I am not sure from memory whether it actually puts the track itself into a undefined state (my recollection is that it does not but is independent- however memory is clouded by the Ansaldo SEI implementation for Cambrian ETCS).
Certainly the technician is given a specificm failure message and it is modern practice to depict the "bow-tie" on the VDU screen (certainly Machynlleth for the Cambrian and TVSC for Reading station area)
Part d) was fine, apart from the implication that there is no requirement at all for TCI except to detect bufferstop movement- already discussed above.
If you could have done such an answer under exam conditions then it would have been a good Credit I think, not far from a Distinction.
I hope the above has showed some areas where you could have improved what you had written, but more imporrtantly made you think of whether you could have improved presentation methodology. There was nothing wrong with what you did; it was organised and legible, but I think could have got the same number of marks with less writing, thus saving time.
(09-07-2013, 08:29 AM)dorothy.pipet Wrote: An untimed attempt for comments please:
"a) Under what circumstances is it appropriate to provide track circuit interupters (TCIs)?
b) For the circumstances you have described in a), list the risks being addressed and describe how the provision of the TCI addresses these risks.
c) what happens when, and after, a TCI is operated?
d) What difference would it make if the area was fitted with axle counters rather than track circuit interupters?"
