SSR resignalling

Discussion in 'London Underground' started by Dstock7080, 22 Jun 2018.

  1. bluegoblin7

    bluegoblin7 Member

    Messages:
    795
    Joined:
    10 May 2011
    Location:
    OZ / OP
    If the tripcock did lift it wouldn't be necessary to go to the effort and expense of pegging down all the redundant trainstops!

    The tripcock relay 'switch' inhibits full speed trip operation, it doesn't affect the operation of the tripcock itself. The existing tripcock cut out switch can be operated in the event of a defective tripcock, with the train proceeding in RM to the nearest depot or sidings.

    (The above is slightly simplified to make it accessible to all audiences, before someone gets pedantic with me)
     
  2. philthetube

    philthetube Established Member

    Messages:
    1,666
    Joined:
    5 Jan 2016
    This is correct, the one difference however when tripped in an ato section is that after re-setting the trip the train is then immediately able to proceed at line speed with no time delay.
     
  3. bramling

    bramling Established Member

    Messages:
    6,209
    Joined:
    5 Mar 2012
    Location:
    South-east / Barnard Castle
    Makes sense. It would be quite a risk if a tripcock was inoperative under ATC and then for whatever reason failed to come back operative outside the ATC area. Safer and earlier all round just to avoid this.

    There will of course have to be a continued focus on keeping the infrastructure clear of obstructions. The Northern had issues with footballs around Brent Cross, which unfortunately coincided with a fast downhill stretch, so being tripped on a football would likely lead to a non-communicating train on a rainy day.
     
  4. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    Signal failure this morning at Paddington this morning making me wish for SMA1 sooner rather than later!
     
  5. londonboi198o5

    londonboi198o5 Member

    Messages:
    358
    Joined:
    28 Dec 2010
    Signal failures can happen any line no matter what system they use so CBTC is not going to get rid of problems. There are loads of things that can go wrong with CBTC that can cause just as much of a delay than your usual signal failure. Axel counter failure, point failure, Non communicating train, VCC failure, VOBC failure to name a few
     
  6. Malcolmffc

    Malcolmffc Member

    Messages:
    224
    Joined:
    19 Mar 2017
    But they will be rarer and/or have far less impact
     
  7. londonboi198o5

    londonboi198o5 Member

    Messages:
    358
    Joined:
    28 Dec 2010
    Less rare possibly can’t rule it out. Less impact nope. To make them less rare requires much more advanced equipment. What comes with more advanced equipment is more problematic when something goes wrong. So when it does go wrong it has a much larger impact. So you can expect rather longer delays when something goes wrong
     
  8. bramling

    bramling Established Member

    Messages:
    6,209
    Joined:
    5 Mar 2012
    Location:
    South-east / Barnard Castle
    If it’s anything like the lines resignalled to date, failures will be slightly rarer, but the centralised architecture of the Thales system means they can be more debilitating and far-reaching when they do happen. Likewise the design is such that the signalman’s workload can increase *very* considerably during even a quite small failure.

    Remember too that not all failures are down to the signalling itself, points are a major source of failure for example. At least on the sub-surface lines there’s no inductive loop cables, these have been a source of failure when damaged.

    Only a couple of weeks ago the Jubilee had a failure which lasted something like 12 hours. Naturally one thing led to another and when they tried to reverse a train somewhere else a points failure raised its head too. Sometimes bad luck just happens...

    Old signalling can be made to work reliably too. Tube Lines turned round the reliability of the Northern Line signalling in the late 2000s. SSR must surely have been affected by planning blight, namely that the railway has been about to be resignalled for the best part of a decade, thus making is hard to justify major improvements to the old signalling. The Piccadilly Line has also found itself in that position, although with no full line upgrade now in sight pretty major “interim” changes are now happening, like moving the control room and replacing the signal control system.
     
    Last edited: 21 Mar 2019
  9. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    Of course modern tech fails as well, but comparing a 2019 Thales installation to a decades old coloured light setup is like comparing a Casio wristwatch to a Cuckoo clock.
     
  10. bionic

    bionic Member

    Messages:
    653
    Joined:
    8 Nov 2013
    It was a modern Thales system that was responsible for the crash in Hong Kong the other day. I think that might have caused a bit more disruption to MTR than a track circuit failure at Paddington would to LU!
     
  11. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    I think we should wait to hear why that happened before jumping to conclusions. Besides, as I said earlier in the thread, automatic systems don't need to be perfect - they ideally just need to have a lower accident rate than humans in the same situation. Same with autonomous cars and automatic pilots on planes.
     
  12. bionic

    bionic Member

    Messages:
    653
    Joined:
    8 Nov 2013
    Wow! That's a very extreme view. So you think it's ok for automatic signalling to kill people as long as it kills less people than human error might? That is a hardcore line to take and I sincerely hope that you don't work in a position of responsibility on the railway with that attitude. Safety must always be the number one priority. I can't believe you are seriously advocating the use of automatic signalling systems that are not failsafe and could cause accidents.

    According to this article, MTR are blaming Thales...
    https://www.scmp.com/news/hong-kong...r-corp-blames-french-software-supplier-thales

    "Hong Kong’s rail giant has blamed the French supplier of its signalling system for failing to run a simulation that would have prevented a crash on Monday, according to a lawmaker and former rail boss."
     
  13. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    Of course, every effort should be made to make these systems tested in every conceivable scenario and every fail safe implemented. I didn't mean to imply that. However, you're never going to create perfect software - because software is only as good as its creators ability to envisage how it could go wrong.

    What's important is to judge safety critical systems rationally when the software does fail, because even the best systems will fail, and safety critical systems will eventually have a failure that results in injuries or deaths, it's almost inevitable.

    When this does happen, I think there is a tendency to knee-jerk and suggest the baby must be thrown out with the bath water and that no automated system will ever be good enough, and that's the narrow view, as it fails to take in to account the benefits they bring and the other accidents such systems prevent which aren't taken in to account, particularly as accidents of this nature are sensationally reported.

    Though perhaps saying only "as good" as a human is the wrong thing to say, and I could have worded that better.
     
  14. soz_p

    soz_p New Member

    Messages:
    1
    Joined:
    15 Oct 2018
    I don't think this is quite a fair comparison: it seems the MTR crash in Hong Kong was caused by someone not configuring the system correctly (thus allowing two conflicting routes), rather than a component degrading and breaking over time after heavy use. A more comparable case to the crash would be relays not being wired correctly in an interlocking - I wonder what kinds of disruption that can cause!
     
  15. bramling

    bramling Established Member

    Messages:
    6,209
    Joined:
    5 Mar 2012
    Location:
    South-east / Barnard Castle
    A modern system is only as good as the way it has been designed and tested, and even with an extensive testing regime undesirable or unexpected things can happen - look at the incident with the TSRs going missing on the Cambrian for a good example of this. Having said that, one can debate whether numbers on posts is any more robust - but naturally much depends on the organisation behind the scenes in respect of things like management of notice cases and how operational matters are communicated out to drivers.
     
  16. rebmcr

    rebmcr Established Member

    Messages:
    2,819
    Joined:
    15 Nov 2011
    Location:
    London
    You don't have to wonder, it happened at Waterloo!
     
  17. Class 170101

    Class 170101 Established Member

    Messages:
    4,906
    Joined:
    1 Mar 2014
    In terms of manual driving isn't it a requirement for drivers to keep up their competency in this respect? Isn't the Central Line timed for manual driving on Sundays due to fewer services in operation?
     
  18. Dstock7080

    Dstock7080 Established Member

    Messages:
    1,410
    Joined:
    17 Feb 2010
    Location:
    West London
    On the SSR CBTC when Protected Manual is selected the acceleration rate will be lower, the braking rate will be relaxed and the maximum speed will be lower, which means the optimum operation is in ATC.
     
  19. londonboi198o5

    londonboi198o5 Member

    Messages:
    358
    Joined:
    28 Dec 2010
    Yo clearly don’t work on the railways then do you.

    2 weeks ago the jubilee line had a “signal failure” this took out most of the line for 12hours!!
    And surprise surprise this line users a modern automatic signaling system.

    Don’t bet your Casio watch or cuckoo clock just yet.

    A modern signal system took a line out for over 12 hours yet a decade old signaling system can be up and running again very quickly sometimes in a matter of minutes.
     
  20. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    Actually I thought it was a pretty good analogy. If a Casio wristwatch stops working, you can't do much. Maybe you can replace the battery and reset it, but it that's about it. It's probably gonna be useless until you do.

    With a Cuckoo clock there's more to go wrong, but if the cuckoo doesn't pop out any more, maybe you can get your hands dirty and fiddle with the gears until it does again. Plus, a problem with the bird probably won't stop the clock ticking...
     
  21. bramling

    bramling Established Member

    Messages:
    6,209
    Joined:
    5 Mar 2012
    Location:
    South-east / Barnard Castle
    A myth went round on the JNUP lines suggesting that manual operation only allows 90% of the performance ATO allows. This was swiftly debunked! In fact I’ve even seen a download of a train doing a trip in ATO and then a trip in PM over the same line with a proficient driver - apart from the PM profile looking like it had been drawn by someone with shaking hands, when laid on top of each other the speed profile was identical. Indeed, for those drivers with good nerves it’s quite possible to hit a platform manually faster than ATO would with the speed held for virtually the whole platform length in the small window between target speed and maximum safe speed, especially if using trackside features as a marker to give a certain braking point (just watch out if the software gets changed without warning!).

    If things are different on SSR it would be interesting to know why.
     
  22. londonboi198o5

    londonboi198o5 Member

    Messages:
    358
    Joined:
    28 Dec 2010
    Protected manual is much slower on the new system on the SSR at present.
     
  23. bramling

    bramling Established Member

    Messages:
    6,209
    Joined:
    5 Mar 2012
    Location:
    South-east / Barnard Castle
    So does it actually display a different target speed in PM compared to what it does in ATO? If so, that’s very different to JNUP.
     
  24. swt_passenger

    swt_passenger Veteran Member

    Messages:
    19,847
    Joined:
    7 Apr 2010
    London Reconnections have just published an article giving their latest view of the SSR resignalling:
     
  25. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    I know they're speculating a little bit, but exciting to have the entire Circle operating under the new signals by the end of the year.

    A post from the other place says there are now only 5 S8s (out of 58) and 21 S7s (out of 134) that have yet to have ATO gear fitting. They will likely be able to finish the S8s by the time SMA1/2 is commissioned so no problem there running any of the Met trains in/out of Aldgate. The S7s will take longer to finish but the ones not fitted can just be allocated to District Line diagrams until September (and those still not fitted can run the Wimblewares until the end of the year). <-- incorrect, see below (oops!)
     
    Last edited: 23 Apr 2019
  26. Dstock7080

    Dstock7080 Established Member

    Messages:
    1,410
    Joined:
    17 Feb 2010
    Location:
    West London
    Currently the last S8 is at Derby to be converted, all others have returned converted.
    4 S7s are currently away being converted, 8 S7s remain in London unconverted.

    Paddington-Edgware Road is due to go CBTC at the end of July, so District S7s heading east from High Street Kensington must be CBTC fitted by then.
     
    Last edited: 26 Apr 2019
  27. cactustwirly

    cactustwirly Established Member

    Messages:
    4,647
    Joined:
    10 Apr 2013
    Location:
    Sunny Tuscany
    Why would District line trains need to be CBTC fitted, as surely it's only the Paddington(H&C) - Edgware Rd section being fitted, not the Paddington (Praed St) - Edgware Road section?
     
  28. Mojo

    Mojo Administrator Staff Member Administrator

    Messages:
    16,912
    Joined:
    7 Aug 2005
    Location:
    0035
    It's the same track.
     
  29. ijmad

    ijmad Established Member

    Messages:
    1,100
    Joined:
    7 Jan 2016
    Location:
    UK
    My mistake, eyes went squiffy looking at diagrams, evidently.
     
  30. swt_passenger

    swt_passenger Veteran Member

    Messages:
    19,847
    Joined:
    7 Apr 2010
    Yes, the same point was made in the comments under the London Reconnections article. Might have aided the explanation to highlight Praed St Junction as a key location where 4 becomes 2...
     

Share This Page