Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Study guide 11.1
#1
Train detection systems.
Which signalling functions need to positively prove the presence of a train and which need to prove track clear?

Im struggling to answer/structure the answer and could do with some help in putting it into words.

Main safety functions I'm thinking are
  • To prevent trains taking conflicting routes
  • To maintain safe distance between trains
  • Protect trains against driver error
  • To ensure trains do not exceed their permitted speed
thanks in advance
Reply
#2
(06-03-2018, 05:19 PM)steak Wrote: Train detection systems.
Which signalling functions need to positively prove the presence of a train and which need to prove track clear?

Im struggling to answer/structure the answer and could do with some help in putting it into words.

Main safety functions I'm thinking are
  • To prevent trains taking conflicting routes
  • To maintain safe distance between trains
  • Protect trains against driver error
  • To ensure trains do not exceed their permitted speed
thanks in advance

Train detection systems:
  • Track circuits
  • Axle counters
  • Balises /radio tags/induction loop
Reply
#3
Have a look for Treadles (often used for Level Crossing strike-in)
Look up Rail circuits - I can't say I know much but I think they're an inverse of Track circuits.

Signal aspect sequences need to prove the absence of a train, similarly for moving points we prove there is no train on the points that might be derailed.

Level Crossing strike-in can probably be argued both ways - the sequence is initiated to lower barriers by a train approaching, but equally barriers are only maintained open by proof of no approaching train.

I'm probably thinking about it from a more Mod3 point-of-view though.
Reply
#4
(06-03-2018, 08:18 PM)dorothy.pipet Wrote: Have a look for Treadles (often used for Level Crossing strike-in)
Look up Rail circuits - I can't say I know much but I think they're an inverse of Track circuits.

Signal aspect sequences need to prove the absence of a train, similarly for moving points we prove there is no train on the points that might be derailed.

Level Crossing strike-in can probably be argued both ways - the sequence is initiated to lower barriers by a train approaching, but equally barriers are only maintained open by proof of no approaching train.

I'm probably thinking about it from a more Mod3 point-of-view though.

thanks
Reply
#5
This is an interesting subject. By default, detection is the absence of trains not the presence. That takes a little thinking about but the subtlety is very important.

Detection of trains is a comparatively rare thing.
1) As Dorothy states, certain LX treadles - they trigger an event based on a train striking one of them.
2) Tracks occupied for time (>=0s) - approach release for example.
3) Lime Street - checks for length of train but potentially this is also more checking for absence of trains on the other measuring tracks - a subjective argument could be had.

Am sure there are more examples but this isn't my domain... hopefully others can post some.
Le coureur
Reply
#6
I'm stumped.
Im looking at it from the point of track circuits and axle counters.

Axle counter is a discrete track vacancy detection system
Track circuit is a continous track occupancy detection system


I can understand a track circuit positively proving the absence of a train = power supply-track-relay-signal
If the train shunts the track and places signal to danger is this positively proving presence of a train?

I would say a track circuit needs to prove track clear and axle counter needs to positively prove the presence of a train.

Im going round in circles. Been reading as much info as i can but still cant come up with a coherent answer/explanation.

Thoughts would be most welcome!!!!!!
Reply
#7
I'd guess that this is not the way the question is intended.
Track circuits prove the absence of the train, if there is a power failure (example) the track section might be occupied or not.
Axle Counters also prove absence in that if the section is reset following a failure it indicates occupied util restored such that track clear can be proven.

Actually it looks like the question is focusing on Signalling functions rather than the type of train detection.
So for a signal to clear it needs the appropriate track sections proved clear (plus other things like points detection). So signalling function <Signal Off> is one that must prove track clear.

Approach release (Main Aspect Approach Released from Red for example) proves that a train has slowed down on the approach to a signal by testing Berth track occupied, or occupied for time, or (less common) by using a treadle.

To set a permissive move such as a call-on into a platform, various tracks need to be proven occupied or occupied for time. So <Set Call-on Route> must prove exit track occupied.
For the aspect <Signal Off for call-on move> must prove berth track occupied for time, permissive (exit) track occupied for time and the other tracks in route clear.

It doesn't matter in any of these cases whether track sections are TC, AxC or something else. Have a think about these cases and others, what happens if the track has failed [clear when occupied] or [occupied when clear] and what will happen. Is it a wrong-side or right-side failure?
Reply
#8
Ah, right, I sort of see where the question is leading now. It’s the wording that’s throwing me a bit. 
Cheers for that.
Reply
#9
Reading the questions is easy. Interpreting the question more challenging. In the OP, the word function was used.

One method of understanding a question is to pick out the key words and use those to ensure are being covered in your answer.
- "Draw a diagram... "
- "State three reasons for... "
- "Give the pros and cons of... "
Le coureur
Reply
#10
(08-03-2018, 01:01 PM)steak Wrote: I'm stumped.
Im looking at it from the point of track circuits and axle counters.

Axle counter is a discrete track vacancy detection system
Track circuit is a continous track occupancy detection system


I can understand a track circuit positively proving the absence of a train = power supply-track-relay-signal
If the train shunts the track and places signal to danger is this positively proving presence of a train?

I would say a track circuit needs to prove track clear and axle counter needs to positively prove the presence of a train.

Im going round in circles. Been reading as much info as i can but still cant come up with a coherent answer/explanation.

Thoughts would be most welcome!!!!!!

A track circuit attempts to prove the absence of a train; the TR should be energised unless a train 's axle diverts current.  Its failure mode is the same as track occupancy, so we cannot really be sure that there is a train there when the TR is down.  Have to be careful when using for signal approach release, for keeping a signal off for Last Wheel Replacement, for a permissive move, for releasing an overlap after the signal berth track has been occupied for a time etc.

Whereas we use the axle counter's output similarly, the axle counter does not directly prove absence.  The heads detect the presence of a wheel; because they are in pairs then the direction of that wheel's motion can be determined. Hence the evaluator can keep count of the net changes.  It has to be told when the section is initially clear and by adding and subtracting it knows when it gets back to the same state again.  If it loses power for an instant, then it hasn't got a clue.

A single axle counter head can be used as an "electronic treadle" or "position detector"; indeed London Underground often does.  The PDs are used to determine where the front of a train has reached as a form of "belt and braces" before a set of points is regarded as free to be thrown after the dead locking track circuit has cleared.  Previously high frequency "overlays" were used instead- the train axle was detected by the relay picking as it completed the circuit in the vicinity of the rail connections (the impedance of the rail prevented a train more than around 10m away being "seen".  In the 1980s British Rail also used for implementing approach release  to avoid needing to divide the actual track circuit into multiple sections.

Rather unusual but can use a "rail circuit" to detect presence of train by completing electric circuit through the axles. For example the RH&DR used (and probably still does although more recently they use a form of axle counter instead) for initiating level crossings- with rails only 15" apart and alongside the sea with salt water contamination, track circuits not practicable.  Relay picked when train within 10m length of track; used much like a treadle contact.

As Dorothy has said, treadles are heavily used for level crossings on NR as  presence detectors- for AHBC for example they generally are used to be the back-up means of ensuring that the Strike-in track is dropped by an approaching train.  However also used either side of the road in order to determine the direction in which the train has traversed the crossing and get the barriers up again afterwards.

When a train reports its own position relative to the last balise / RFID tag / Norming Point that it encountered, this is certainly a "presence" detection.  One of the problems of eliminating lineside track detection is what to do when a train goes non-communicating and how to get the railway working properly again in a safe manner. Even worse if power failure etc. and when trying to resume operations first have to establish how many trains were actually in the area and account for them all.

However as Dorothy says, the question was about SIGNALLING FUNCTIONS, not the form of train detection per se.  Actually seems to be a bit more mod 3 than mod 5 in that respect.  However I think the things is that for the application you should be aware of the requirement and be aware of the failure modes of the form of train detection selected.
PJW
Reply


Forum Jump:


Users browsing this thread: 7 Guest(s)