Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Phantom Overlap
#1
Ok, so I'm not sitting the exam this year and I did actually pass it some time ago, but I need some information; standards etc. seem,,, well, short on detail.

My new project deals with phantom overlaps, and while I am familiar with the term I cannot remember the detail. My research so far has provided the following definition:

"A Signal Overlap that can be proved clear by the Signalling System, but which does not have a defined limit at the required distance from the signal."

Does anybody know why you would use a phantom overlap, what benefit it gives, and how it is achieved? Even better can you point to a GB Rail standard which contains any detail?
Reply
#2
(06-05-2010, 04:05 PM)Douglas Wrote: Ok, so I'm not sitting the exam this year and I did actually pass it some time ago, but I need some information; standards etc. seem,,, well, short on detail.

My new project deals with phantom overlaps, and while I am familiar with the term I cannot remember the detail. My research so far has provided the following definition:

"A Signal Overlap that can be proved clear by the Signalling System, but which does not have a defined limit at the required distance from the signal."

Does anybody know why you would use a phantom overlap, what benefit it gives, and how it is achieved? Even better can you point to a GB Rail standard which contains any detail?

A phantom overlap is defined when one does not want to incur the unnecessary cost / it is not physically feasible to put a track section limit e.g. in track circuiting impossibility of putting in IRJ, constraints re minimum lengths of TC, amount of physical stagger etc) but want a theoretical overlap there.

Most usual case may be where do not want to set and lock points beyond a signal but where there is enough length to have an adequate overlap (e.g. trailing points may be 150m beyond signal but want to define a 120m ROL but such a joint would be too close / foul of the points). Going to need a TC joint just beyond points anyway and putting one at 120m may not be practicable. Hence the overlap physically proves the entire track circuit clear (i.e. MORE than the phantom overlap position) but point set/lock/detection within that track circuit is omitted; if this point is a crossover then the ability to use the "other end" may be useful but if the crossover were locked then it wouldn't be. You generally wouldn't want to design a layoyut with POL or PROL; generally they result as a compromise when performing some alteration on an existing layout.

Another example may be two opposite direction signals that are say 360m apart. Want to be able to simultaneously signal trains up to each of them (perhaps one is a platform starting signal and that train is subsequently about to change direction). Could provide two separate track circuits so each had its own unique overlap, but since neither TC should become occupied in this scenario then save money and provide a single TC. We are not allowed to "share overlaps" but we can "share overlap track circuit" by defining a POL in the middle. The aspects of the two opposing direction routes up to the signals prove the entire 360m clear; however even if both trains then SPAD by 180m then they won't actually hit each other and it is just as safe (arguably more so as automatic reversion of a valid aspect if the other train SPADs) as if separate TCs provided.

Rather busy now to look out a standard (don't actually think there is much written down); I'l try next week if noone else has located in the meantime.

The IRSE does tend to feature them rather a lot in Layout 2's used for CTs and aspect sequence, so these should reveal some typical examples.

In real life I have very rarely actually encountered:
a) it is arguable that the Western Region Delayed Yellow (similar to Warning class route) is really a form of PROL but actually it is sometime a reverse PROL with more pointb locking than train detection!
b) In a sense where there are sidings immediately beyond an overlap with a trap point called to the trapping position despite being within the overlap and therefore "incorrect" for a normal trailing point in overlap, then one could argue that this is also a form of POL
c) St Pancras had one when the MML operated into what are now the Eurostar platforms of Eastern Interim Station
d) several instances of combined overlap track circuits for two opposing signals (actually I think it is most common in recent years with axle counters)
PJW
Reply
#3
P32 of GK/GN0802 gives a definition for both POL and PROL. I have seen several but the most memorable is the signal protecting the crossover on the up, country side of Ebbsfleet High Level.

It has a PROL as there are routes over the crossover bang road as occassional reversing moves or to the sidings. Hence, the platforms could both be full. The approaching train from the North Kent is signalled off that line to clear it for operational reasons. However, the approach is 1:40 until 20m in rear of the pin.

So, a ROL was introduced to allow the North Kent to EBB HL route to be set when a route from the station (down) across the crossover was set. However, from memory, the ROL is within the fouling point on the straight ahead route for the up line. It could not be on the diverging route due to the points being locked by the down route being set. Phantom, because it doesn't actually exist!

The approach to the pin was mitigated by a TSS and three OSSs mostly due to the gradient but part of the mitigation was the infrequency of the opposing move occuring despite the effects, due to the topology, being significant. It was the layout that forced a solution and the mitigations were complex.

Hope that helps. If you want more info, I will try to find the scheme and some of my notes.

Jerry
Le coureur
Reply
#4
Much appreciated chaps, thanks for answering.

There are a few in the vicinity of Vauxhall also, they appear to be PROL but the scheme plan shows only POL, trying to get my hands on the route tables...
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)