Driverless Operation

[up1989]

Hello all,

ive been reading up about the ntfl, and I've seen the plan for driverless operation on the central and Picc by the 2030s but nothing mentioned about bakerloo driverless anyone know the reason why?

[Deleted]

Queen's Park to Harrow & Wealdstone and its National Rail signalling, its shared platforms with National Rail trains and the resulting compromise height platforms and difficulties with platform edge doors.

But with automation you can never say never, it's the way of the world.

[up1989]

I suppose they could stop the bakerloo north of Queen's Park?

[superteacher]

Oct 9, 2016 16:54:01 GMT up1989 said:

I suppose they could stop the bakerloo north of Queen's Park?

The main depot is at Stonebridge Park, so that would be a non starter.

[up1989]

Oct 9, 2016 17:08:10 GMT superteacher said:

Oct 9, 2016 16:54:01 GMT up1989 said:

I suppose they could stop the bakerloo north of Queen's Park?

The main depot is at Stonebridge Park, so that would be a non starter.

That is true,

would they still have someone on the front of the train or would dlr style be allowed?

[norbitonflyer]

Oct 9, 2016 17:08:10 GMT superteacher said:

Oct 9, 2016 16:54:01 GMT up1989 said:

I suppose they could stop the bakerloo north of Queen's Park?

The main depot is at Stonebridge Park, so that would be a non starter.

Unless a new depot is built at the other end of the line when/if it is extended to Lewisham/Hayes/Lewes/Dieppe.

[up1989]

If thats the case then its unlikely it will happen before i retire and I'm 27

[crusty54]

It may just take until the 2030s to replace the existing Bakerloo line trains.

[up1989]

Oct 10, 2016 2:11:21 GMT crusty54 said:

It may just take until the 2030s to replace the existing Bakerloo line trains.

Thats what I'm hoping! Ive just put in my nomination for the Bakerloo for when full time comes up (Currently NT Driver on part time) Hoping to have as much conventional driving before it all goes ATO.

[malcolmffc]

Even ATO is not currently planned for the Bakerloo as far as I know, for the reasons mentioned above. I suppose they could convert the section that isn't shared with NR (Elephant - Queen's Park) and have the remainder driven manually though.

[crusty54]

Oct 10, 2016 6:55:56 GMT malcolmffc said:

Even ATO is not currently planned for the Bakerloo as far as I know, for the reasons mentioned above. I suppose they could convert the section that isn't shared with NR (Elephant - Queen's Park) and have the remainder driven manually though.

whatever happens a signalling contract and new trains will be required at some point.

[domh245]

Oct 10, 2016 6:55:56 GMT malcolmffc said:

Even ATO is not currently planned for the Bakerloo as far as I know, for the reasons mentioned above. I suppose they could convert the section that isn't shared with NR (Elephant - Queen's Park) and have the remainder driven manually though.

Could they not use a similar system to the SSR on the Richmond and Wimbledon branches with an ATO overlay that talks to the NR system?

[up1989]

Oct 10, 2016 10:59:01 GMT domh245 said:

Oct 10, 2016 6:55:56 GMT malcolmffc said:

Even ATO is not currently planned for the Bakerloo as far as I know, for the reasons mentioned above. I suppose they could convert the section that isn't shared with NR (Elephant - Queen's Park) and have the remainder driven manually though.

Could they not use a similar system to the SSR on the Richmond and Wimbledon branches with an ATO overlay that talks to the NR system?

Thats what I was thinking as they are planning on having the Ato Overlaid on the NR Signalling

[rincew1nd]

Oct 10, 2016 7:33:18 GMT crusty54 said:

...a signalling contract...will be required at some point.

I beg to differ. If recent past is anything to go by, then at least two signalling contracts will be needed

[Deleted]

I can't see new trains being introduced and run with old signalling? It almost defeats the purpose of new trains. If the trains are going to be driverless, it's say to say it could most likely go ATO. I know Network Rail were once discussing the idea of trialling trains between Kings Cross and Kings Lynn or Cambridge where trains just run on axel counters with no signals whatsoever, so if that happens and it's successful, signals could definitely become a thing of the past! (Not in our lifetime I hope)

[phil]

Oct 15, 2016 23:01:41 GMT noddymac said:

I can't see new trains being introduced and run with old signalling? It almost defeats the purpose of new trains. If the trains are going to be driverless, it's say to say it could most likely go ATO. I know Network Rail were once discussing the idea of trialling trains between Kings Cross and Kings Lynn or Cambridge where trains just run on axel counters with no signals whatsoever, so if that happens and it's successful, signals could definitely become a thing of the past! (Not in our lifetime I hope)

Are you talking about ERTMS? www.networkrail.co.uk/aspx/12275.aspx

Current plans are for it to be fitted as an overlay on the GWML while retaining lineside signals, but on the ECML south of Peterborough it was planned to do away with lineside signals altogether next time the signalling kit gets renewed

[piccboy]

Oct 15, 2016 23:01:41 GMT noddymac said:

I can't see new trains being introduced and run with old signalling? It almost defeats the purpose of new trains. If the trains are going to be driverless, it's say to say it could most likely go ATO. I know Network Rail were once discussing the idea of trialling trains between Kings Cross and Kings Lynn or Cambridge where trains just run on axel counters with no signals whatsoever, so if that happens and it's successful, signals could definitely become a thing of the past! (Not in our lifetime I hope)

Traditionally, the underground has always introduced new trains on old signalling, but of course there are 4 ways new trains could be introduced on new signalling.

1) Upgrade old trains to run on new signalling, install new signalling, introduce new trains. Pros New trains would not have to have any legacy signalling system installed, so cheaper to manufacture. Cons Expense of designing and fitting old trains with new signalling equipment including issue of lack of space for said equipment on the 72 stock.

2) Shut down the entire line, upgrade signalling system, remove old trains and introduce new ones. Pros Cheapest option in terms of actual costs of new signalling and trains as no need to upgrade old trains, install legacy equipment in new trains, run dual signalling systems, quickest way to have new system completed as can be worked on 24/7. Cons Loss of service for passengers during upgrade, loss of revenue from passengers using other services to complete journeys, impact of increase passenger load on other Underground lines, compensation to passengers effected?

3) Introduce new trains onto old signalling system. Remove old trains, Upgrade signalling system. Pros quickest way to get new trains in service, less impact on passenger service as upgrade can be spread over longer period of time, i.e. worked on at nights when service closed. Cons New trains would have to be fitted with legacy signalling equipment so cost.

4) Install new signalling system to run parallel with old, introduce new driverless trains, remove old trains, removed old legacy signalling system. Pros new trains cheaper as no legacy equipment needs to be installed. Cons cost of designing and installing a new signally system that can run parallel with old in a fail safe manner.

IMHO the current method of introducing new trains able to run on old signalling system is probably the cheapest and has less impact on customers.

[metman]

Agreed. It's been happening for the last 22 years!!

[rincew1nd]

You missed out one major con of option two, the need to store all the new trains somewhere once built, and somewhere for the old trains to go before they get chopped up.

Would only work with the W&C.

[Chris M]

Isn't option 4 essentially what was done with the Victoria Line (minus the driverless bit)

[piccboy]

Oct 16, 2016 10:06:35 GMT rincew1nd said:

You missed out one major con of option two, the need to store all the new trains somewhere once built, and somewhere for the old trains to go before they get chopped up.

Would only work with the W&C.

The new trains would be stored at the existing depots / sidings as they are built, old trains would removed to scrappers, and space at scrappers would be determined by how many used, and where. In the case of the Bakerloo line, its direct connection to the National rail network would make it logical to move old and new stock by rail. That could also open up the possibility of storing old trains in unused but functional sidings on National Rail.

I appreciate that the signalling upgrade would more than likely be completed before all the new trains are built and delivered, and I would suggest that this time period be used for a thorough testing and bedding in of the new signalling system and trains. At some point there would be enough trains to run a limit passenger service.

Oct 16, 2016 10:41:21 GMT Chris M said:

Isn't option 4 essentially what was done with the Victoria Line (minus the driverless bit)

That's correct. In my opinion going from one ATO system to another would be relatively easier than going from a traditional fixed block, tripcock / trainstop, coloured signal, manually driven system if only for the simple reasons that, in the example of the Victoria line, it had a relatively precise train identification system, overlaid with a completely electronic signalling system. The traditional system has a number of mechanical systems that would need to have a reliable method of obtaining signalling information to pass to the computerised signalling system.

I would like to say that although I used the term "relatively easier" above to describe the Victoria Line upgrade, I appreciate and give immense respect to the teams that successfully developed, built and installed that system.

With regards to the driverless bit, from noddymac's original post he mentioned driverless as going to ATO, I assumed he meant ATO as implemented on existing Underground lines i.e. Central, Jubilee, Northern, Victoria lines with a Train Operator / Instructor Operator at the front able to override the system and bring the train to a safe stop in the event of an Emergency, open and close doors, etc. As opposed to a system like UTO (Unattended Train Operation) where there is no member of staff on board and the train does everything.

[Harsig]

I have to say that we may very well see option 4 used more often in the future when ATO lines are next upgraded. This is because ATO signalled lines have rather more trainborne signalling equipment than trains operating on a conventionally signalled line. As such considerably more work would be required to convert a train from one system to the other, and it would probably not be practicable for a train to be fitted with both systems at the same time. This makes dual fitted signalling systems a more attractive alternative.

[up1989]

Oct 15, 2016 23:01:41 GMT noddymac said:

I can't see new trains being introduced and run with old signalling? It almost defeats the purpose of new trains. If the trains are going to be driverless, it's say to say it could most likely go ATO. I know Network Rail were once discussing the idea of trialling trains between Kings Cross and Kings Lynn or Cambridge where trains just run on axel counters with no signals whatsoever, so if that happens and it's successful, signals could definitely become a thing of the past! (Not in our lifetime I hope)

We will see new trains running on existing signalling whilst the signals are upgraded to seltrac (or other type) of ato. Similar to whats happening on the SSR network, as the other option will be to close the whole line during the upgrades. 

[rincew1nd]

Oct 16, 2016 12:46:43 GMT piccboy said:

Oct 16, 2016 10:06:35 GMT rincew1nd said:

You missed out one major con of option two, the need to store all the new trains somewhere once built, and somewhere for the old trains to go before they get chopped up.

Would only work with the W&C.

The new trains would be stored at the existing depots / sidings as they are built, old trains would removed to scrappers, and space at scrappers would be determined by how many used, and where. In the case of the Bakerloo line, its direct connection to the National rail network would make it logical to move old and new stock by rail. That could also open up the possibility of storing old trains in unused but functional sidings on National Rail.

There simply isn't space to store a whole new fleet of trains anywhere on LU metals, let alone the Bakerloo line. Just look at the roll out of the S Stock, where one of the major factors that have held things up is the delay in completing depot enabling works.

Bringing things back to the Bakerloo, it wouldn't surprise me if we see the option of new trains on old signalling, as the "lights on sticks" north of Queens Park may well be around longer than the 72ts!

[Deleted]

Oct 16, 2016 21:07:35 GMT rincew1nd said:

Oct 16, 2016 12:46:43 GMT piccboy said:

The new trains would be stored at the existing depots / sidings as they are built, old trains would removed to scrappers, and space at scrappers would be determined by how many used, and where. In the case of the Bakerloo line, its direct connection to the National rail network would make it logical to move old and new stock by rail. That could also open up the possibility of storing old trains in unused but functional sidings on National Rail.

There simply isn't space to store a whole new fleet of trains anywhere on LU metals, let alone the Bakerloo line. Just look at the roll out of the S Stock, where one of the major factors that have held things up is the delay in completing depot enabling works.

Bringing things back to the Bakerloo, it wouldn't surprise me if we see the option of new trains on old signalling, as the "lights on sticks" north of Queens Park may well be around longer than the 72ts!

In fairness, with option 2, which is shutting down the entire line, upgrading the signalling, removing the old trains and introducing the new ones, there's no need to find additional space for the new trains, because you've shut the line down, precisely in order that you don't have to operate the old trains and the new trains simultaneously, to avoid signalling incompatibility. So you wouldn't need to hold onto any old trains (which you did in the case of the S Stock rollout), so as long as you've got rid of an outgoing train before an incoming one arrives, there's no problem. And, of course, you can start mass-dispatch of your old rolling stock the moment you've shut the line down.

So the problem is really more the fact that passengers have no service until the upgrade is finished.

Well, except, of course, where you're ordering more new trains than you had old trains, which you almost certainly will be. But there's plenty of time to construct the extra space while you're also doing the signalling fit-out. You would, of course, have been constructing that space in any case, and where you've shut the line down, you can get rid of your old stock very quickly, whereas the new stock will take a while to be built and delivered, so there'll be plenty of time before all the old stock is used up.

What is more of a problem, in that respect, is of course that an entire batch of rolling stock takes an age to construct, so if you're not going to run any trains side-by-side, it'll be a long long time before you have enough trains to restart a service. piccboy suggests using this time to thoroughly test the system, but I think more likely what you'll see is your rolling stock lying unused for a long time, which won't do it any good. Of course, you could reopen your line in sections, but there you would probably have the opposite problem. I imagine a fair amount of rolling stock will be ready before the signalling is ready to go on the first section, which will leave it to gather dust.

What you might do, therefore, is try to have your cake and eat it, by making a headstart on the rolling stock building, getting a fair amount of the order built, without using it. Then shut the line down, then install the signalling (while more trains are still being built), then open the line back up. And that's where you'd really have a storage space problem.

[rsdworker]

well in Paris - the line 1 was changed from Manual to Automatic required having new trains and old trains mixed till the signaling system was done then moved old trains to another line where no automatic trains operation exists

[Chris M]

You might be able to store rolling stock on parts of the running lines that signalling installers/deinstallers don't need access to. e.g. on the Bakerloo line, if you were starting to replace the signalling from the south end, you could store all your old rolling stock nose to tail on the northbound line from Queens Park southwards while the installers work south of their, leaving the southbound line free for rolling stock deliveries into London Road Depot. As you'll be removing old trains from the north end, the south end of the line of trains will be moving northwards as the signalling works do.

[rincew1nd]

Oct 16, 2016 21:19:09 GMT tut said:

Oct 16, 2016 21:07:35 GMT rincew1nd said:

There simply isn't space to store a whole new fleet of trains anywhere on LU metals, let alone the Bakerloo line. Just look at the roll out of the S Stock, where one of the major factors that have held things up is the delay in completing depot enabling works.

Bringing things back to the Bakerloo, it wouldn't surprise me if we see the option of new trains on old signalling, as the "lights on sticks" north of Queens Park may well be around longer than the 72ts!

In fairness, with option 2, which is shutting down the entire line, upgrading the signalling, removing the old trains and introducing the new ones, there's no need to find additional space for the new trains, because you've shut the line down, precisely in order that you don't have to operate the old trains and the new trains simultaneously, to avoid signalling incompatibility. So you wouldn't need to hold onto any old trains (which you did in the case of the S Stock rollout), so as long as you've got rid of an outgoing train before an incoming one arrives, there's no problem. And, of course, you can start mass-dispatch of your old rolling stock the moment you've shut the line down.

There is still an assumption that there is space at the manufacturer for all this new stock.

If I was a procurement manager I'd want new assets in service ASAP, not sitting around waiting!

[Deleted]

Oct 16, 2016 22:07:56 GMT rincew1nd said:

Oct 16, 2016 21:19:09 GMT tut said:

In fairness, with option 2, which is shutting down the entire line, upgrading the signalling, removing the old trains and introducing the new ones, there's no need to find additional space for the new trains, because you've shut the line down, precisely in order that you don't have to operate the old trains and the new trains simultaneously, to avoid signalling incompatibility. So you wouldn't need to hold onto any old trains (which you did in the case of the S Stock rollout), so as long as you've got rid of an outgoing train before an incoming one arrives, there's no problem. And, of course, you can start mass-dispatch of your old rolling stock the moment you've shut the line down.

There is still an assumption that there is space at the manufacturer for all this new stock.

If I was a procurement manager I'd want new assets in service ASAP, not sitting around waiting!

Oh no doubt about it, it's a ridiculous strategy, it would never be adopted.

But it's certainly possible. There's no need for any space at the manufacturer, as soon as it's built, it goes to LU and onto the line. Of course there's space on the line to receive the new train, because one of the old trains will already have been dispatched to make way for it. How can we do this so quickly? We don't need to provide any service at all, so we can start getting rid of the old trains immediately, so we don't need to hold off deliveries of the new trains from the manufacturer at all.

The problem is, as you say, apart from wiping out an entire line for however many months and months, that those new trains are gonna be more rust and dust than train!

[philthetube]

Oct 16, 2016 21:07:35 GMT rincew1nd said:

Oct 16, 2016 12:46:43 GMT piccboy said:

The new trains would be stored at the existing depots / sidings as they are built, old trains would removed to scrappers, and space at scrappers would be determined by how many used, and where. In the case of the Bakerloo line, its direct connection to the National rail network would make it logical to move old and new stock by rail. That could also open up the possibility of storing old trains in unused but functional sidings on National Rail.

There simply isn't space to store a whole new fleet of trains anywhere on LU metals, let alone the Bakerloo line. Just look at the roll out of the S Stock, where one of the major factors that have held things up is the delay in completing depot enabling works.

Bringing things back to the Bakerloo, it wouldn't surprise me if we see the option of new trains on old signalling, as the "lights on sticks" north of Queens Park may well be around longer than the 72ts!

I think there might be lights on sticks when the 72 replacements depart