|
Post by superteacher on Sept 12, 2018 12:49:16 GMT
There’s been a significant update to the CTBC software and given that it needs to be properly tested, the go live date for 0.5 has moved again..... New date is 17th/18th November 2018? 😂😂😂
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 12, 2018 13:43:19 GMT
Think it’s 3018
|
|
|
Post by goldenarrow on Sept 12, 2018 14:11:43 GMT
There’s been a significant update to the CTBC software and given that it needs to be properly tested, the go live date for 0.5 has moved again..... New date is 17th/18th November There’s been a significant update to the CTBC software and given that it needs to be properly tested, the go live date for 0.5 has moved again..... New date is 17th/18th November 2018? 😂😂😂 I think I feel a poem coming along... There was a place called OZ Due for replacement because Management had said earlier That 4LM should go to Bombardier But they ended up dropping it instead! So off it went to Thales So they thought they could toast a chalice But as the weather now turns grey And migrations zones still causing affray It looks like OZ may be on to stay! So is this to be our winter of discontent Where computers are the ones to lament As Westinghouse still reigns supreme With old signals still showing green Laughing at Hammersmith and the fabled digital dream...
|
|
|
Post by dm1 on Sept 12, 2018 15:02:46 GMT
There’s been a significant update to the CTBC software and given that it needs to be properly tested, the go live date for 0.5 has moved again..... New date is 17th/18th November Any details on exactly what has been updated/ what issues the update is meant to solve? In particular, is it an update that adds some kind of new functionality, or an update that is trying to make the functionaity that's already there err... function?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 12, 2018 15:04:37 GMT
Everything has been updated especially the code for the interlocking
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,317
|
Post by Colin on Sept 12, 2018 21:32:21 GMT
Any details on exactly what has been updated/ what issues the update is meant to solve? In particular, is it an update that adds some kind of new functionality, or an update that is trying to make the functionaity that's already there err... function? They're trying to make it work as intended rather than adding anything new - we are talking about implememting the system rather than modifying something that's already in use. Going back to June and the failed attempt to go live.... A number of technical issues cropped up during the "go live" weekend which passed the threshold of acceptability - mostly trains arriving at the CTBC boundary not able to correctly enter the CTBC area. It was also realised that availability of modified S stock trains would be tight for a full weekday service and that taking a step back and re-evaluating whether it was right to continue led management to the conclusion that they were not in fact ready to proceed. They decided to pull the plug and as I said at the time, it made sense to admit defeat. "They" decided to name a date in September (they did consider an August date but wanted to avoid the Notting Hill Carnival). Before the date arrived it was put back to October. October has now become November. Whilst on the face of it these constant date changes give the impression that the whole thing is in chaos; to be fair to the "the managment" they want to be certain that the next time they go live, it will actually go live. If they feel the system isn't ready, they'd much rather take a bit more time to test and then name a new date rather than go through the debacle of June all over again. In an email update that went out today, management are acutely aware that they're having to refresh Hammersmith & City/Circle line train opertaors as its been more than six months since they were originally trained and that District training is about to start (all District I/O's already done and are at risk of going over the six month threshold for refresher training). Refresher training of course incurs an additional cost to the 4LM project so the constant date changes cannot continue indefinately. Also, each time a new go live date is announced, depots have to release all their intructor operators as they are required to accompany drivers on their first few trips. Not only does this see changes to rest days and annual leave having to be planned, but a canceled go live means it all has to be undone again. So again the ever changing go live dates are not something that can keep happening. The go live delay does mean more and more S stock are being modified so at least in that regard there's less worry about having enough trains modified to run a full weekday service. Shadow running of a segregated fleet (ensuring only modified trains run on Hammersmith & City and Circle line workings) is currently being monitored to ensure the correct trains are being allocated on a daily basis. A number of software updates since June have been sorting out things like VOBC issues (Vehicle On Board Controller), CSDE (Correct Side Door Enable), emergency brake events, "NCT Lockups" (routes unable to be released once set), Trip cock mismatch at boundary points, spurious messages, dwell times, short target points, speed restrictions, platform hold not working and a few others I won't pretend to understand. The latest software version - the one they intend to go live with - is not being made available till mid October. Given that it needs to be tested and proved to be safe for certification to run a passenger service, you can see why the go live date has now moved to November. So the current plan is; 21st & 22nd October will now be the test weekend for the latest software (the go live version). 3rd & 4th November will be a trial operations weekend for SMA0.5 (extension of already planned SMA3 test weekend) 17th & 18th November SMA0.5 goes live 12th & 13th January SMA1 goes live (originally planned as SMA2 testing - SMA2 testing may be moved to another date) Hope that helps explain some of what's going on......
|
|
|
Post by MoreToJack on Sept 12, 2018 21:57:16 GMT
I don't feel the need or ability to expand on any of the above, but I will add...
...none of us are holding our breath.
|
|
londoner
thinking on '73 stock
Posts: 480
|
Post by londoner on Sept 12, 2018 23:21:55 GMT
Any details on exactly what has been updated/ what issues the update is meant to solve? In particular, is it an update that adds some kind of new functionality, or an update that is trying to make the functionaity that's already there err... function? They're trying to make it work as intended rather than adding anything new - we are talking about implememting the system rather than modifying something that's already in use. Going back to June and the failed attempt to go live.... A number of technical issues cropped up during the "go live" weekend which passed the threshold of acceptability - mostly trains arriving at the CTBC boundary not able to correctly enter the CTBC area. It was also realised that availability of modified S stock trains would be tight for a full weekday service and that taking a step back and re-evaluating whether it was right to continue led management to the conclusion that they were not in fact ready to proceed. They decided to pull the plug and as I said at the time, it made sense to admit defeat. "They" decided to name a date in September (they did consider an August date but wanted to avoid the Notting Hill Carnival). Before the date arrived it was put back to October. October has now become November. Whilst on the face of it these constant date changes give the impression that the whole thing is in chaos; to be fair to the "the managment" they want to be certain that the next time they go live, it will actually go live. If they feel the system isn't ready, they'd much rather take a bit more time to test and then name a new date rather than go through the debacle of June all over again. In an email update that went out today, management are acutely aware that they're having to refresh Hammersmith & City/Circle line train opertaors as its been more than six months since they were originally trained and that District training is about to start (all District I/O's already done and are at risk of going over the six month threshold for refresher training). Refresher training of course incurs an additional cost to the 4LM project so the constant date changes cannot continue indefinately. Also, each time a new go live date is announced, depots have to release all their intructor operators as they are required to accompany drivers on their first few trips. Not only does this see changes to rest days and annual leave having to be planned, but a canceled go live means it all has to be undone again. So again the ever changing go live dates are not something that can keep happening. The go live delay does mean more and more S stock are being modified so at least in that regard there's less worry about having enough trains modified to run a full weekday service. Shadow running of a segregated fleet (ensuring only modified trains run on Hammersmith & City and Circle line workings) is currently being monitored to ensure the correct trains are being allocated on a daily basis. A number of software updates since June have been sorting out things like VOBC issues (Vehicle On Board Controller), CSDE (Correct Side Door Enable), emergency brake events, "NCT Lockups" (routes unable to be released once set), Trip cock mismatch at boundary points, spurious messages, dwell times, short target points, speed restrictions, platform hold not working and a few others I won't pretend to understand. The latest software version - the one they intend to go live with - is not being made available till mid October. Given that it needs to be tested and proved to be safe for certification to run a passenger service, you can see why the go live date has now moved to November. So the current plan is; 21st & 22nd October will now be the test weekend for the latest software (the go live version). 3rd & 4th November will be a trial operations weekend for SMA0.5 (extension of already planned SMA3 test weekend) 17th & 18th November SMA0.5 goes live 12th & 13th January SMA1 goes live (originally planned as SMA2 testing - SMA2 testing may be moved to another date) Hope that helps explain some of what's going on...... Thank you for this very informative post!
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,317
|
Post by Colin on Oct 30, 2018 20:14:47 GMT
The planned operational proving weekend coming up on November 3rd & 4th for SMA0.5 has been cancelled.
SMA0.5 therefore will not go live as intended on the weekend of November 17th & 18th.
As far as I'm aware, SMA3 testing is going ahead as planned this coming weekend (November 3rd & 4th).
Moving forward, all that is known at the momement is there will be a "software drop" sometime in January 2019 and new testing/implementation dates are yet to be determined......
|
|
|
Post by superteacher on Oct 30, 2018 21:20:56 GMT
The planned operational proving weekend coming up on November 3rd & 4th for SMA0.5 has been cancelled. SMA0.5 therefore will not go live as intended on the weekend of November 17th & 18th. As far as I'm aware, SMA3 testing is going ahead as planned this coming weekend (November 3rd & 4th). Moving forward, all that is known at the momement is there will be a "software drop" sometime in January 2019 and new testing/implementation dates are yet to be determined...... This is fast becoming a joke. They need to get this new system running soon because the near daily signal failures on the SSR are becoming a big issue,
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 31, 2018 0:53:48 GMT
If you knew what caused them then it would be a joke recently
|
|
class411
Operations: Normal
Posts: 2,724
|
Post by class411 on Nov 1, 2018 12:04:51 GMT
This is fast becoming a joke. They need to get this new system running soon because the near daily signal failures on the SSR are becoming a big issue, Not to mention forever shutting down the HC&C at weekends without providing a proper replacement bus service.
|
|
|
Post by ijmad on Nov 3, 2018 23:26:20 GMT
So if software drop is happening in January that means no chance of anything going in to passenger service until January now?
|
|
|
Post by goldenarrow on Nov 3, 2018 23:47:26 GMT
So if software drop is happening in January that means no chance of anything going in to passenger service until January now? Not necessarily, software drops also known more usefully as code deployments can and have taken place on other Thales signalling products that are already in active service. Given the number of years that the 4LM’s migration of signalling will take place over, I’m sure that further modifications to software will need to take place during which passenger carrying trains will be in use. As SMA 0.5 will be the first of many so I can understand the impetus to get it right, but it really hinges on the severity of the changes that need to be made, the Jubilee line is still receiving software mods six years from its conversion so it is something that can and often does just go on in the background although this projects insatiable capacity to dither over things means that I too share your scepticism.
|
|
Tom
Administrator
Signalfel?
Posts: 4,101
|
Post by Tom on Nov 4, 2018 6:50:41 GMT
One of the factors relating to dates is the requirement for a possession to change the old signalling over to the new; you can't just change over on any night as there isn't quite enough time.
There are, naturally, other factors influencing things which I'm not in a position to go into publicly, but the rumour mill amongst staff is pretty active and not entirely innaccurate.
|
|
|
Post by MoreToJack on Nov 4, 2018 9:56:38 GMT
So if software drop is happening in January that means no chance of anything going in to passenger service until January now? Not necessarily, software drops also known more usefully as code deployments can and have taken place on other Thales signalling products that are already in active service. Given the number of years that the 4LM’s migration of signalling will take place over, I’m sure that further modifications to software will need to take place during which passenger carrying trains will be in use. As SMA 0.5 will be the first of many so I can understand the impetus to get it right, but it really hinges on the severity of the changes that need to be made, the Jubilee line is still receiving software mods six years from its conversion so it is something that can and often does just go on in the background although this projects insatiable capacity to dither over things means that I too share your scepticism. CBTC will not be commissioned until at least January. The need to also undertake 'Operational Proving' once that software update has happened will dictate things to an extent. Certainly the rumour mill within service control seems somewhat accurate - the rumour mill on CBTC in other departments (excepting signals, naturally)... less so.
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,317
|
Post by Colin on Nov 4, 2018 13:31:44 GMT
Certainly the rumour mill within service control seems somewhat accurate - the rumour mill on CBTC in other departments (excepting signals, naturally)... less so. Service control staff are closer to the action so to speak, so will inevitably be more aware of whats really going on. On the train side we are heavily reliant on senior management keeping us in the loop, otherwise the rumour mill takes over with ever more fanciful rubbish being peddled as fact. Some of us do our best to keep abrest of what's really going on but we are at the bottom of the information chain so any truthful facts we do pick up are often a week out of date by the time its filtered through from the top level. Sadly there's far more nonsense rumours going round than truthful facts at the moment.....
|
|
Tom
Administrator
Signalfel?
Posts: 4,101
|
Post by Tom on Nov 4, 2018 17:10:36 GMT
The software update has happened - this was the reason for the last test closure. What is now needed is the weekend of operational proving with that software (assuming another update isn't forthcoming).
|
|
|
Post by cudsn15 on Nov 4, 2018 23:38:59 GMT
I can't help but feel it's like they are trying to re-invent the wheel. I kinda get it's tricky to amalgamate 4 lines into 1 system but the fact it's not even getting past the first stage (0.5) after all this time does make me wonder about the competence of those involved. Yes it has to work and be safe and carry out everything that's being asked of it - but the time its taking to write, test and lock software really is stretching the credibility of the professionalism of these people. As the software can be updated why not just concentrate on the basics and get the signalling sorted and all the ancillary stuff like Correct Side Door Enabling and the rest be done by the trained driver in attendance on each train until such time that the computer is callable of making the very expensive driver even more redundant?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 4, 2018 23:57:21 GMT
Speaking as a software developer / engineer - it is highly likely that by the time stage 0.5 launches the signalling code will be 98-99% complete even for things that are not used at that stage. I'd like to dream that it will be a modern agile and modular system - but knowing how it usually goes in the industry, most likely it is a giant blob that contains everything possible for the entire system. So after the initial launch, the follow-ups should be much quicker.
Think of it as bowel movements: if you are having issues (with not having any), and then finally squeeze out the first turd - the rest tends to follow much quicker.
|
|
rincew1nd
Administrator
Junior Under-wizzard of quiz
Posts: 10,241
|
Post by rincew1nd on Nov 5, 2018 5:41:54 GMT
Unless you suffer an anal fissure in the process in which case it just prolongs the agony! Morning all
|
|
|
Post by whistlekiller2000 on Nov 5, 2018 7:20:24 GMT
Unless you suffer an anal fissure in the process in which case it just prolongs the agony! Morning all Anyone else reading this while taking their errrr 'morning constitutional'? SSR Resignalling has never been so visceral.......
|
|
Tom
Administrator
Signalfel?
Posts: 4,101
|
Post by Tom on Nov 5, 2018 7:29:23 GMT
@cudsn - The problem is that you seem to be thinking of this as a computer -based signalling system. It's not. It's an all-inclusive train control system, which incorporates ATO, ATP, Signalling, CSDE, etc all into one system.
I'm in a position where I'm aware of the issues and what the delays are. I'm not able to explain them publicly however, but I can assure you there is no issue with the competence of people - it's that this is something which needs to be right first time as the access on SSR isn't forgiving enough to go back later and fix things.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 5, 2018 13:30:43 GMT
@cudsn - The problem is that you seem to be thinking of this as a computer -based signalling system. It's not. It's an all-inclusive train control system, which incorporates ATO, ATP, Signalling, CSDE, etc all into one system. I'm in a position where I'm aware of the issues and what the delays are. I'm not able to explain them publicly however, but I can assure you there is no issue with the competence of people - it's that this is something which needs to be right first time as the access on SSR isn't forgiving enough to go back later and fix things. You sure about that I could name a few
|
|
|
Post by roman80 on Nov 5, 2018 20:49:03 GMT
Software aside, how is the trackside hardware installation progressing?
|
|
|
Post by Dstock7080 on Nov 5, 2018 21:31:40 GMT
Software aside, how is the trackside hardware installation progressing? Lineside antennae poles appearing around Hammersmith (District) and as far east as Dagenham Heathway. Signalling equipment rooms being constructed at Barking, East Putney and Wimbledon. Despite the rescheduling of commissioning it hasn’t slowed the installation teams.
|
|
Tom
Administrator
Signalfel?
Posts: 4,101
|
Post by Tom on Nov 5, 2018 21:59:35 GMT
In theory, installation can continue regardless of any delay to commissioning.
|
|
|
Post by philthetube on Nov 6, 2018 3:51:46 GMT
These problems make me wonder if the initial option to start with outer areas rather than the circle would not have been the better choice,
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,317
|
Post by Colin on Nov 6, 2018 8:28:31 GMT
These problems make me wonder if the initial option to start with outer areas rather than the circle would not have been the better choice, It's very easy to assume its a case of intsall the equipment, turn it on and away we go but you've got to bear in mind there's four seperate streams that have to come together at the right time. They are: 1) Cable runs, cabling, lineside equipment and lineside equipment rooms need to be installed. 2) Hammersmith SCC needs to have the correct number of signallers trained to operate the VCC's (whilst also maintaining normal signalling coverage on the legacy system). 3) S7 and S8 fleets need to be converted ready to work with the new CBTC system - each train has a week long visit to Derby. 4) Drivers need to be trained - each driver has a week long course (again whilst also maintaining the published timetable). Refresher training is carried out for both service control and drivers if they go more than six months over the date of original training, so there is a cost and training logistic's problem each time the implementation date is delayed. Believe me, 4LM do not want to keep delaying the start date as the further we get into training staff, the more staff need retraining. So, the Hammersmith to Latimer Road section was chosen as its a relatively self contained part of the railway. It's also the closest part of the railway to the new SCC. Compared to the District or Metropolitan lines, less drivers initially need to be trained and less S7 stock are required to be ATO converted. Each driver needs to be "floor walked" by an instructor operator for their first few trips so having a smaller initial area makes the whole thing much easier to manage. As they move along the North side of the Circle its much easier to then introduce the West end District line drivers (Paddington to Edgware Road) with a few more S7's converted, then the Met drivers with the S8's fully converted (Paddington/Finchley Road to Euston Square), then finally the East end District drivers (Euston Square to Stepney Green/Monument) by which time the project will have gotten all of the S7 fleet fully converted. If the District had been chosen, say the Upminster to Becontree section, you'd need a lot more S7's to be ATO converted much earlier in the project timeline, and twice as many drivers would need to be trained [and re-trained if there is any initial slippage]. Same with the Metropolitan - indeed the whole S8 fleet would have to be fully converted before anything else could happen. By doing it in the way they are, it makes each element of the project easier to manage and integrate.
|
|
Tom
Administrator
Signalfel?
Posts: 4,101
|
Post by Tom on Nov 6, 2018 8:55:20 GMT
Colin's spot on, and all I can say is he has only just scratched the surface of the many parts that need to come together.
There's also the need for maintainers to be trained (both fleet and signals), and taking his first point, the interfaces between the old and new need to be joined up a long way ahead of any commissioning. For example, the point controls and detection need to be interfaced before any test operations can begin, and there needs to be a suite of tests undertaken before considering trial operations and a commissioning date.
To interface the points, you need the wayside installation to be reasonably progressed, which needs the equipment room(s) to be finished before that can commence. And so on...
|
|