|
|
Post by superteacher on Sept 2, 2019 19:14:32 GMT
The door open delay is because the system needs to confirm the train is stationary (and in the right place). Fair point, but this is an electronic system. It should be able to confirm this almost instantly. If the Central and Victoria Line systems can manage it, then it should be possible to programme CBTC to do the same. |
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Sept 2, 2019 19:21:46 GMT
The central and Victoria lines are not a Thales based system though so you can't directly compare them - in fact the Central and Victoria are not even comparable between themselves as even they are different from each other.
|
|
|
|
Post by Chris L on Sept 2, 2019 19:48:17 GMT
Distinct lack of Met trains at Moorgate in both directions around 7pm.
H&C and Circle trains jam packed.
|
|
|
|
Post by superteacher on Sept 2, 2019 20:17:17 GMT
The central and Victoria lines are not a Thales based system though so you can't directly compare them - in fact the Central and Victoria are not even comparable between themselves as even they are different from each other. The point I was trying to make is that it’s clearly possible to enable the door release in a short space of time. Maybe Thales should be looking at these systems and asking themselves why theirs can’t. Any additional time can thwart a high frequency service, and with the SSR timings will be key due to the many conflicting movements. |
|
|
|
Post by goldenarrow on Sept 2, 2019 20:27:59 GMT
Due to the planned commissioning of SMA 1 & 2, the latter encompassing the Metropolitan line between Finchley Road and Euston Square, platforms 1 or 4 at Baker Street are expected to be out of commission from Monday 2nd to Thursday 5th September. Platforms 1 and 4 have been in operation all day today. I'm glad I was proved wrong on that count, sounds like a very hectic day. Felt very sorry for staff on the ground today, it really shows how inept the current system of customer service information is given that it barley fulfils this role. |
|
|
|
Post by metrider on Sept 3, 2019 6:26:53 GMT
The central and Victoria lines are not a Thales based system though so you can't directly compare them - in fact the Central and Victoria are not even comparable between themselves as even they are different from each other. The point I was trying to make is that it’s clearly possible to enable the door release in a short space of time. Maybe Thales should be looking at these systems and asking themselves why theirs can’t. Any additional time can thwart a high frequency service, and with the SSR timings will be key due to the many conflicting movements. I'm with superteacher on this one. My background is technical, Electronics, Microprocessors, Hardware, Firmware, Software. I have a very good understanding in all of these areas. I also have a knack for actually fitting a solution to a need (rather than to an 'outsiders perception of a need' - for 'outsider - read 'engineer unconnected with the trade that the solution will be used in'). With proper analysis and design - there is no reason why a system cannot be designed without the pause before releasing the doors. The other systems prove that - whether the delay can be tweaked out of the current product (if there was the will to do so) is beyond my ability to guess as I have no inside knowledge of the actual architecture of the system. Whether the root cause in this case is insufficiently detailed analysis before design or some kind of conscious compromise reached during the design is also outside of my ability to analyse from the sidelines. Another relatively recent failing on the analysis front (well if several years can be classed as recent!) would be the PEA buttons on the 09 stock. The first time I was on one of the then new vic trains, the vulnerability if the PEA buttons stuck out to me like a sore thumb. It was immediately obvious to me that there would be a plethora of accidental activations. A year or two later (surprise, surprise) they all had shrouds fitted.... How something like that can be immediately obvious to an outsider to the railway trade after so many 'trade eyes' must have seen it before me is beyond understanding. Anyway - I've conspired to lead us off topic again (sorry!) - if there is more to say, maybe this is a discussion for another thread. I'm interested to know how the service improves over the next few days (so I can get back to my real desk!) |
|
|
|
Post by commuter on Sept 3, 2019 7:39:57 GMT
Platforms 1 and 4 have been in operation all day today. I'm glad I was proved wrong on that count, sounds like a very hectic day. Felt very sorry for staff on the ground today, it really shows how inept the current system of customer service information is given that it barley fulfils this role. I /think/ {and I stand to be corrected} is that the issue is that there is no timetable for it, it is effectively being made up on the day by the service control team and the train managers depending on who needs the training trips. In a situation like that it is very hard to provide detailed information on what trains will be cancelled and what trains will run, as it’s a fluid situation. |
|
DWS
every second count's
Posts: 2,487
|
Post by DWS on Sept 3, 2019 8:17:09 GMT
Tuesday 3rd September 2019, 0915, severe delays between Harrow on the Hill and Aldgate 😹🚂
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Sept 3, 2019 9:00:24 GMT
The central and Victoria lines are not a Thales based system though so you can't directly compare them - in fact the Central and Victoria are not even comparable between themselves as even they are different from each other. The point I was trying to make is that it’s clearly possible to enable the door release in a short space of time. Maybe Thales should be looking at these systems and asking themselves why theirs can’t. Any additional time can thwart a high frequency service, and with the SSR timings will be key due to the many conflicting movements. I'm with superteacher on this one. My background is technical, Electronics, Microprocessors, Hardware, Firmware, Software. I have a very good understanding in all of these areas. I also have a knack for actually fitting a solution to a need (rather than to an 'outsiders perception of a need' - for 'outsider - read 'engineer unconnected with the trade that the solution will be used in'). With proper analysis and design - there is no reason why a system cannot be designed without the pause before releasing the doors. The other systems prove that - whether the delay can be tweaked out of the current product (if there was the will to do so) is beyond my ability to guess as I have no inside knowledge of the actual architecture of the system. Whether the root cause in this case is insufficiently detailed analysis before design or some kind of conscious compromise reached during the design is also outside of my ability to analyse from the sidelines. Another relatively recent failing on the analysis front (well if several years can be classed as recent!) would be the PEA buttons on the 09 stock. The first time I was on one of the then new vic trains, the vulnerability if the PEA buttons stuck out to me like a sore thumb. It was immediately obvious to me that there would be a plethora of accidental activations. A year or two later (surprise, surprise) they all had shrouds fitted.... How something like that can be immediately obvious to an outsider to the railway trade after so many 'trade eyes' must have seen it before me is beyond understanding. Anyway - I've conspired to lead us off topic again (sorry!) - if there is more to say, maybe this is a discussion for another thread. is the thread split expert - I'm sure it'll happen soon with this one drifting...... What I was getting at is the way in which the correct side door enable system works is different. On the old legacy railway we use either loops of wire (literally set up as a loop under the platform) or tags attached to the platforms along with aerials on the trains that when lined up with each other, set up the correct side door enable system. Under TBTC (On the Jubilee & Northern lines) or CBTC (on the Circle, District, Hammersmith & City and Metropolitan lines) the correct side door enable system is managed by the signalling system. There will obviously be less delay in getting the door enable system set up if you only need to align something rather than prove what its doing first. Whilst I'm not particularly aware of the exact set up on the Central and Victoria lines, I wouldn't mind betting that at least in the case of the central line the correct side door enable system isn't controlled by the signalling system, hence they can open up quicker. |
|
|
|
Post by dm1 on Sept 3, 2019 10:15:20 GMT
There will obviously be less delay in getting the door enable system set up if you only need to align something rather than prove what its doing first. I think the point is that this should not be inherently true - at least not on a human scale. If the signalling system is designed with this in mind, there is no inherent reason why it cannot carry out the door enabling near instantaneously. Of course there might be some kind of compromises in the system that cause this - but the Victoria Line alone shows that this is not necessary. Objectively speaking if this proves to be a bottleneck (which on the Victoria line it definitely would have been) then I think there are perfectly legitimate questions about why the Thales system was designed in such a way that this delay occurs. |
|
|
|
Post by edb on Sept 3, 2019 11:00:13 GMT
It amazes me how some simple things like that are screwed up by supposedly better systems. Another example is that Oyster PAYG usually opens the barrier without a perceptible delay. Bank card contactless pauses for maybe half a second. And what is pushed in all of the announcements? contactless.... Given that all of these seconds (whether it be door opening or getting through the barriers) add up - surely they must have an impact on capacity. I've noticed that if I use my phone as a payment device rather than the card it's quicker, not a quick as oyster but also not card slow... |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2019 12:36:11 GMT
Correct Side Door Enable aka CSDE is currently not managed by CBTC this is a future software drop. Currently is uses the passive tags underneath the platforms still and there is no penned in date for the complete switchover to CBTC to control the CSDE
|
|
|
|
Post by goldenarrow on Sept 3, 2019 13:04:40 GMT
commuter , I understand what you are saying, but it doesn't hide away from the fact that information notbaly that communicated online was woeful. The information with regards to cancelled services was available, indeed some stations put up a list of cancelled services just after start of traffic. This is the board at Hillingdon, some service cancellations would only add ten minutes to a journey, others would push that up towards the 30min mark. You could’ve enabled alot more people to plan their journey and potentially save some grief if this was streamed online at the same time that this was chalked up. 
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2019 13:22:09 GMT
22 trains cancelled this morning
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Sept 3, 2019 16:54:35 GMT
There will obviously be less delay in getting the door enable system set up if you only need to align something rather than prove what its doing first. I think the point is that this should not be inherently true - at least not on a human scale. If the signalling system is designed with this in mind, there is no inherent reason why it cannot carry out the door enabling near instantaneously. I'm obviously not explaining this very clearly. On an S stock, on the current legacy railway, I get the correct side door enable indication just as the train comes to a stop - the momemnt it appears, the train is still physically (and noticeably) moving. On an S stock, on the CBTC railway, the signalling system needs to "see" that the train has physically stopped before it will offer the correct side door enable indication. All the technology in the world cannot possibly make it give that indication any quicker. Of course there might be some kind of compromises in the system that cause this - but the Victoria Line alone shows that this is not necessary. I haven't yet seen anyone confirm how the correct side door enable system works on the Victoria line. Is the signalling system controlling it or is it an entirely seperate platform mounted system? As it currently stands, the Victoria line proves nothing till we know! Correct Side Door Enable aka CSDE is currently not managed by CBTC this is a future software drop. Currently is uses the passive tags underneath the platforms still and there is no penned in date for the complete switchover to CBTC to control the CSDE Sorry but you are wrong. The tags are obviously still present as we could have reverted back to the legacy signalling system but we ain't using them. To prove it, when I did one of my Protected Manual driving trips on Saturday I apparently didn't stop accurately despite the indication in the cab being on 0 meters. I had to carry out an emergency door open. It was Edgware Road platform 2 BTW. I was pretty much bang on in terms of the chevron yet I had no correct side door enable indication. This was confirmed by the I/O on board (it was my second confidence trip) and the driver whose train we'd borrowed. Aside from the lack of indication, if we were using the tags I'd have either got a full door open pattern for the station (by virtue of being well within the chevron) or at the very least I'd have got overshoot mode (front doors only cut out) as I'd have passed the tags at a low enough speed. I had a test train operator on board too and he said it happened a few times during the day when drivers were in PM - he mentioned a P6 tag and explained the technicalities of how it worked (I'll be honest, I switched off momentarily at this point) and that he was going to suggest adjusting it's location. |
|
|
|
Post by MoreToJack on Sept 3, 2019 16:58:59 GMT
Don’t quote me at this stage but I believe that the Vic uses legacy CSDE and is not connected with the signalling system. DTG-R (and indeed Central ATP) is a very different system to T/CBTC.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2019 17:24:53 GMT
T6 tags (no such thing as a P6) are accurate stop tags they do not hold any other data apart it can stop the train within 48mm of its planned stopping position
T1 Crossover Tags are the basic tag used by the positioning system. They are installed every 25m
T2 Cross Border Tags are the same as a basic tag (T1) but include data that helps loop border detection
T3 Border Tags are used to determine borders between 2 virtual loops. They are also used at transition zones so that positioning can be established when entering the system from a Non - CBTC territory
T4 Not Used on 4LM
T5 Not Used on 4LM
T6 Accurate Stop Tags are installed around station stops in addition to crossover (T1) tags, they assist with stopping accuracy. As mentioned at the top
VT Virtual Tags are virtual and are usually placed around points where the train could pick up another tag at a diverging junction and stop the train from loosing position and EB'ing (Emergency Brake)
oh btw before you criticise again, the above is taken from the manual word for word
Before the question goes there is no physically loops so why have T2 or T3 tags in the first place the VCC still thinks there is loops outside hence the need for the virtual loop to fool the VCC as it can't see the TAG's physically
I can assure you CBTC does not run CSDE at present and won't do for a while, remember the train does a map of the virtual world within its VOBC on a compact flash card for the whole of SSL regardless if its a S7 or S8
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2019 17:26:19 GMT
Don’t quote me at this stage but I believe that the Vic uses legacy CSDE and is not connected with the signalling system. DTG-R (and indeed Central ATP) is a very different system to T/CBTC. No idea about CSDE on the Vic Central Line signalling is called Westrace |
|
|
|
Post by superteacher on Sept 3, 2019 18:10:26 GMT
Meanwhile . . ..
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,346
|
Post by Colin on Sept 3, 2019 18:18:04 GMT
oh btw before you criticise again, the above is taken from the manual word for word I wasn't criticising in any way. Blimey, its far too easy to be misundertood on this forum with just written text! What I did say was all the evidence, that is: - My training
- The "confidence" I/O accompanying me
- The 4LM people out & about the railway on Saturday
- The test train operator on board my train
- The way the train behaved
All point to CBTC controlling the CSDE It's only you that has said different. So it was on that baisis that I said you were incorrect. No offence was or is intended. T6 tags (no such thing as a P6) are accurate stop tags they do not hold any other data apart it can stop the train within 48mm of its planned stopping position T1 Crossover Tags are the basic tag used by the positioning system. They are installed every 25m T2 Cross Border Tags are the same as a basic tag (T1) but include data that helps loop border detection T3 Border Tags are used to determine borders between 2 virtual loops. They are also used at transition zones so that positioning can be established when entering the system from a Non - CBTC territory T4 Not Used on 4LM T5 Not Used on 4LM T6 Accurate Stop Tags are installed around station stops in addition to crossover (T1) tags, they assist with stopping accuracy. As mentioned at the top VT Virtual Tags are virtual and are usually placed around points where the train could pick up another tag at a diverging junction and stop the train from loosing position and EB'ing (Emergency Brake) Thank you for the detailed explanation - I wasn't far out considering I was only half listening! |
|
|
|
Post by gjs on Sept 3, 2019 18:30:28 GMT
22 trains cancelled this morning Southbound from Wembley Park seems to have essentially stopped at 7pm this evening. Passengers need to be given a better explanation by now....day 2 of chaos. |
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,758
|
Post by Chris M on Sept 3, 2019 18:56:57 GMT
On an S stock, on the current legacy railway, I get the correct side door enable indication just as the train comes to a stop - the momemnt it appears, the train is still physically (and noticeably) moving. On an S stock, on the CBTC railway, the signalling system needs to "see" that the train has physically stopped before it will offer the correct side door enable indication. All the technology in the world cannot possibly make it give that indication any quicker. Regardless of what specifically controls it or why, it is a serious design flaw in the overall 4LM system(s) to require the train be proved completely stopped before correct door side enable indications can be given because of the significant dwell time increase it causes. If it is not possible for a CBTC signalling system to give CSDE indications before the train has completely stopped then responsibility for providing such indications needs to be given to a different system that can. |
|
|
|
Post by danwoodhouse on Sept 3, 2019 20:06:41 GMT
pardon me for not understanding this to the full, but is all this to do with the eventual automatic train operation of the sub-surface lines?
|
|
|
|
Post by superteacher on Sept 3, 2019 20:13:30 GMT
pardon me for not understanding this to the full, but is all this to do with the eventual automatic train operation of the sub-surface lines? Yes, in part. The two pages of this thread will give you the details. |
|
|
|
Post by danwoodhouse on Sept 3, 2019 20:25:10 GMT
pardon me for not understanding this to the full, but is all this to do with the eventual automatic train operation of the sub-surface lines? Yes, in part. The two pages of this thread will give you the details. So what's a C.B.T.C then? |
|
|
|
Post by superteacher on Sept 3, 2019 20:40:28 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2019 20:41:21 GMT
Communication Based Train Control
|
|
|
|
Post by jimbo on Sept 4, 2019 4:42:59 GMT
Some of these comments seem to suggest that the Met was lax on training its drivers, but I think that is a misunderstanding of the situation. Those closer to the action could explain, but I believe that Met drivers had completed their classroom and simulator training. What they needed was practical experience on their train using the new signalling. This is defined as at least six trips, two manually driving, with an Instructor Operator. They need to switch over the train for CBTC, proceed through the new signalling, and then switch the train back for conventional signalling. The Instructor will then document all this and leave them, crossing over to accompany another driver back. Most of the Instructors received their experience during test running on trial weekends, but to include all Met line drivers for at least six trips needs a full service running for a couple of days at least, and there will still be those on rest days, etc to catch up with. The service can carry passengers but can never be as efficient as a regular day. Hard to justify closing to passengers on a regular weekday.
This was based on experience resignalling the Jubilee and Northern lines. On the Jubilee the first implementation was between two country stations to provide all drivers with experience, but the full implementation took so long that they were out of practice by then. A similar plan for the Barnet branch was dropped for that reason, I believe, only delaying the real implementation. On the Met such an introduction would have been restricted by the branches splitting north of Harrow. Of course, the two earlier Sub-surface resignallings planned to start out at the extremities to reduce delays to main services, but it was decided to start in the centre this time to bring advantages to the most passengers as soon as possible. Hopefully once this week is over, things will go as planned.
|
|
rincew1nd
Administrator
Junior Under-wizzard of quiz
Posts: 10,286 
|
Post by rincew1nd on Sept 4, 2019 5:56:13 GMT
Admin comment
Can we keep this thread to the specific problems on the met line this week please, we have the more general thread for more general questions.
|
|
|
|
Post by metrider on Sept 4, 2019 6:13:20 GMT
I think the posters at Harrow advertised a "slightly reduced" service 2nd-5th Sept. Is it likely that this will actually be sufficient to complete what is required? (I find it hard to comprehend that it could), or will this issue drag on a number of days longer then taper off?
|
|
