solidbond
Staff Emeritus
'Give me 118 reasons for an Audible Warning on a C Stock'
Posts: 1,215
|
Post by solidbond on Mar 29, 2005 20:24:12 GMT
Ah, now it makes sense ;D But why is WK1 cleared in that fashion? Now, at this point I would say 'Over to Tom!', but it seems he isn't too sure of the working in this area!! I would guess that it is due to there being no guarantee that the train will appear from Gunnersbury, until it actually approaches WK1. So rather than clear the signal HOPING the train will arrive, it waits until it KNOWS the train is going to arrive. |
|
DrJimi
Virtual District Line construction engineer and arborist
Posts: 365
|
Post by DrJimi on Mar 30, 2005 3:15:41 GMT
Pardon if I'm wrong or confused, but per my 1963 chart WK1 is on the EB slow from CP. Whereas WK11(ab) is on the RMD EB road, with the repeater (RWK11a) - which is I think the repeater SB is refering to - being just before the flyunder.
Which makes me wonder how/when WK11 does clear. Presumably the route has been set from GUN to TG and (per SB's observation) the train is approaching RWK11. This is the short section annotated "DD". Similarly, I'd expect WK1(ab) to clear if the route was set for a train from CP.
I do recall a video showing a train held at WK11b on the up-bank after the EB flyunder, so I'd assume the route had been set earlier for a CP EB train. Who gets the perferential treatment - a GUN train or a CP train? My gut would say WK11 would only clear if (a) a train was approaching RWK11a (i.e. imminent train wanting to merge onto the EB Slow) and route through WK1 was not already set. Also what's the sequence for WK11a and WK11b? I see there's enough overlap between then (397') to fit a train. Same question regarding Wk1a and WK1b.
Pardon the intrusion from a novice - I'm still trying to learn signals operation so I can model it realistically.
Best, Jimi
|
|
solidbond
Staff Emeritus
'Give me 118 reasons for an Audible Warning on a C Stock'
Posts: 1,215
|
Post by solidbond on Mar 30, 2005 9:14:53 GMT
You are of course correct about the signal numbers - I was relying on TheOneKEA to have the right signal numbers ;D ;D  (shame on me you ;D) WK11(ab) are indeed the signals coming up from the Richmond road, and WK1(ab) the ones from CP. It is indeed RWK11a which is the repeater I was referring to in my earlier answer. Again it is a guess, but the route would be called by the programme machine based on which train is due next. Because the PM can 'see' the Ealing road, it will 'know' where that train is, so will clear the signals immediately if that train is to be given the route. If the next train was due from the Richmond road, I guess the PM will allow a certain amount of time for that train to appear. If it doesn't appear in the allowed time, then it will select the route for the Ealing road instead. Because the route from Richmond, although set, will not have cleared, there will not be signals going back from clear to danger. As far as the clearing of the a&b for WK1 and WK11, if the route is clear they will both clear together, but if a train has just gone across the junction and the route is set for the other train, they will clear as the previous train moves out of the section, one at a time. Hope this makes sense, although I stress again, that I am guessing this is how it works as I am no signalling expert. Tom may be able to confirm if this is how PMs work. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 30, 2005 13:40:24 GMT
You are of course correct about the signal numbers - I was relying on TheOneKEA to have the right signal numbers ;D ;D (shame on me you ;D) Don't blame me, it was late at night ;D You are of course correct Dr. Jimi; WK11 is the EB converging home from Gunnersbury. WK11(ab) are indeed the signals coming up from the Richmond road, and WK1(ab) the ones from CP. It is indeed RWK11a which is the repeater I was referring to in my earlier answer. Again it is a guess, but the route would be called by the programme machine based on which train is due next. Because the PM can 'see' the Ealing road, it will 'know' where that train is, so will clear the signals immediately if that train is to be given the route. If the next train was due from the Richmond road, I guess the PM will allow a certain amount of time for that train to appear. If it doesn't appear in the allowed time, then it will select the route for the Ealing road instead. Because the route from Richmond, although set, will not have cleared, there will not be signals going back from clear to danger. As far as the clearing of the a&b for WK1 and WK11, if the route is clear they will both clear together, but if a train has just gone across the junction and the route is set for the other train, they will clear as the previous train moves out of the section, one at a time. Hope this makes sense, although I stress again, that I am guessing this is how it works as I am no signalling expert. Tom may be able to confirm if this is how PMs work. No, you've got it right. Both WK1 A and WK11 A are what I refer to in my head as the 'prime signal'; they are the first ones to be encountered by the driver when travelling towards the junction, and are thus the junction outer homes. WK1 B and WK11 B are the inner homes of the junction. The overlaps of the A and B signals will be such that a train passing A will occupy the track circuit right up to the very edge of the points; a train passing B will then occupy the track circuit over the points themselves and beyond to the next stick. The overlaps will be arranged such that A's overlap will include B's track circuit up to B's overlap at the next signal, thus accomplishing two things: 1. A train approaching the A signal will stop before hitting the points. 2. Trains can be compressed together before passing over the junction, thus improving headways. Tubeprune has a few images which explain something similar for the home signal arrangements used at stations. Hole picking time! |
|
DrJimi
Virtual District Line construction engineer and arborist
Posts: 365
|
Post by DrJimi on Mar 30, 2005 15:43:08 GMT
Got it! I had assumed that the "a,b" nomenclature was similar (if not the same) as for multi-home signals, and I'm familiar with how those are wired and operate (thanks to TubePrune et al). I was also checking other converging junctions to see if this was a common arrangement. Neither HLJ EB (DL/Picc merge) nor EB Slow at ACTN (DL/Picc flyunder EB slow merges) have a/b signals; however these examples are both within 20mph PSRs, so speeds will be low and the need for overlap protection is greatly reduced. Is it therefore the 'norm' to have a/b signals in convergence areas where speeds would be higher? Just curious.
Thanks, Jimi
|
|
Tom
Administrator
Signalfel?
Posts: 4,196 
|
Post by Tom on Mar 31, 2005 16:28:17 GMT
Tom may be able to confirm if this is how PMs work. It's been a few years but I'll try... P/Mc's have four main methods of control: Programme - machine signals trains in accordance with the timetable but also comparing the train it expects on the roll with the incoming train description. If there is a discrepancy it alerts the regulator at Earl's Court, allowing him or her one minute to intervene before the train is routed according to the TD (I'll check that last detail, I've got a 50-50 chance on this.). Programme Only - machine signals trains according to the Programme Roll only (hence the name) without reference to the train describer. First Come First Served - the machine works by the TD alone and whatever comes first will go through first. Push Button - the regulator pushes everything through manually. I would expect that the PM has indications of approaching trains from a similar distance on both lines, rather than being blind to the Richmond road. I think I'm going to have to read up on P/Mc's again though, as I haven't worked on them for three years or so. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 31, 2005 17:40:23 GMT
Just so people know; this is GB8! I went to richmond today!!  |
|
|
|
Post by igelkotten on Mar 31, 2005 18:24:30 GMT
It's been a few years but I'll try... P/Mc's have four main methods of control: Programme - machine signals trains in accordance with the timetable but also comparing the train it expects on the roll with the incoming train description. If there is a discrepancy it alerts the regulator at Earl's Court, allowing him or her one minute to intervene before the train is routed according to the TD (I'll check that last detail, I've got a 50-50 chance on this.). According to my papers from Coburg Street, you are entirely correct. /Igelkotten |
|
(shame on 
