Proposals for the Upgrade of the Sub-surface Lines
By District Dave
with comments and updates added by
District Dave writes:
I am sure that many of you will be aware that
there has, in the recent past, been a major restructuring of London Underground recently
under the much-heralded Public Private Partnership (PPP). I do not think
there is anyone who uses London Underground who would not agree that the system is in need
of major financial investment many of the trains are in need of replacement,
signalling systems are outdated and tracks require major renewal.
Much has been written about this and can be
found on the web, so I am not going to comment further, other than to say that, in its
simplest terms, the idea is that the infrastructure and assets of London Underground have
been leased off to private consortia for the next thirty years. It is the responsibility
of these consortia to upgrade and renew these assets and provide the means to an improved
service to the travelling public. The operational side of London Underground
and this effectively means those of us who are actually involved in the
provision of the service remains in public ownership. This is really
putting the situation at its simplest and, if you are not aware of more details, I would
urge you to do some research on the topic. Start on this site at the FAQs on the PPP page.
There is a wide divergence of opinion as to the
rights and wrongs of PPP, but it is now in place and a reality and what follows is an
overview of the plans as they affect the Sub-surface lines. But even though the
money is now available, there can be no magic fix building a whole new railway is
in itself a massive undertaking but renewing one whilst keeping it running (as far as
possible) at the same time is even more complex. There can be no overnight quick fix
and what I want to try to convey here is the complexity of the whole programme, the
timescales that will be involved and what will hopefully be the outcome at the end of it
Many of the details that follow in this article
are drawn from a document that outlines London Undergrounds vision for
the Sub-surface Lines. Ive used this as the basis of this article but I've
tried to explain areas which, in the main document, make a number of assumptions and
expand into more detail than is, in my opinion, relevant here whilst at the same time
trying to avoid internal jargon and corporate speak. So this is not
simply a copy of the document but my précis, with my own comments and expansions added as
I have felt appropriate in its preparation. Similarly I have not reproduced many of
the statistics or any of the graphical commentaries as these are going into too much of
the minutiae of the matter. Some of the matters addressed are also my own
interpretation of the content of the document.
So that you can appreciate correctly the parts
of London Underground that are being discussed here Ill define what makes up the
Sub-surface Lines. They are comprised of the Metropolitan, District,
Hammersmith & City and Circle Lines in essence all the lines that do not run
through deep tube tunnels and who operate (currently) the A, C and D Stock
trains. The observant amongst you may notice that there is no mention of the East
London Line. The ELL is not included within the scope of the review, I assume
because it is currently the subject of a whole programme of renewal and extension and
although, clearly, a Sub-surface line, falls outside the scope of this programme.
Programme for Peak Service Change
It is generally considered that the service
provision on the Subsurface railway (SSR) will change more than on any other line on
London Underground as a result of PPP.
Tubeprune comments that, curiously, the
Metronet and London Underground PPP teams refer to the sub surface network as
"SSL" or Sub Surface Lines. MetronetSSL is the official infrastructure
company name. SSR was the old term originally used in the days when it was offered
to Railtrack. They eventually refused it when they realised that they couldn't run
the Heathrow Express to the City and operate the existing frequencies as well. They
thought it was like Thameslink.
It is intended that SSR service frequencies will
improve and the services will also change from that offered at the present time
probably the most radical change will be that the Circle Line will, effectively,
disappear. The following will outline the phases of the upgrade.
This is interesting, since LU tried to get
rid of the Circle in the early 1990s. One Underground official suggested a
"panhandle" service. It wasn't really a panhandle, more of a crossed
loop. Trains would run from Hammersmith to Whitechapel via Shepherds Bush, Baker
Street, Liverpool Street, Tower Hill, S. Kensington, High Street, Baker Street, Liverpool
Street and Aldgate East. They would reverse at Whitechapel and then take the same
route back to Hammersmith via Liverpool Street, Edgware Road, S. Kensington, Tower Hill,
Liverpool Street, Baker Street and Shepherds Bush. It was tried on a number of
Notting Hill festival days but the conservative elements within LUL effectively killed
it. It need two extra trains to run but it put a turnround time into each Circle
trip. It was also disliked by the crews since each round trip was about 3˝ hours -
two of these, with a meal break, would make a complete duty within an 8-hour day.
The Situation Today
As I mentioned above, the SSR is made up of four
services but, to put more detail on this, the following list shows the make up of the
total service during the peaks:
The Metropolitan Line utilises 44 A stock
trains each having a total capacity of 1,048 passengers.
The District Line utilises 67 D Stock trains
each having a capacity of 968 plus 10 C Stock trains each having a capacity of 1,014.
The Hammersmith & City line utilises 15 C
Stock trains The Circle Line utilises 14 C Stock trains.
These provide a total of 150 trains in service
in the central area of London providing 28 trains per hour (tph). The total daily
kilometrage approaches 62,000 km.
A word of warning on the train capacity
figures shown here. The passenger capacity of a train is calculated by adding the
number of passengers who could fit into the standing area to the number of seats. In
the Underground's calculation, they use a figure of 7 passengers per square metre of
available standing area. Now, imagine a square metre and then try to get 7 people,
with briefcases and handbags, to stand it it. No? Can't imagine it?
Quite right too. People just won't do it under normal conditions. It's
much more like 4 per square metre.
Another thing to bear in mind is that the
train is not evenly loaded along its full length. A good example can be seen on the
Victoria Line every day, where many of the station exits are at the south end of the
platforms and the trains are packed at that end but with room for more at the north end.
So, any attempt to relate LU's capacity
figures with what actually happens on the train needs to be looked at with a serious
reality check. Thus, in the case of the District, the D Stock capacity is really not
more than 673 or 70% of the planner's figure, assuming it is equally loaded throughout,
which it isn't.
The Proposals to be Implemented by 2011
Through improvements to signalling systems
(though not replacement at this stage), many of which are aimed at relieving various
locations and areas that, at the moment, cause delays to services, it is intended that the
peak services will be increased to 30 tph from the existing 28 tph. It is also intended
that the Circle and Hammersmith & City Line services will have been replaced by what
is being called a T Cup service, the yellow line shown in Fig. 1 below.
Fig. 1: Diagram showing the proposed "T-Cup"
service, in yellow, which will replace the Circle service.
The T-Cup service will run from Hammersmith to
Edgware Road via Tower Hill, that is, trains will start at Hammersmith, go via Edgware
Road and Liverpool Street round what is currently known as the Outer Rail and
then, on arrival back at Edgware Road, the trains will reverse and follow the reverse path
back to Hammersmith. The intention is to create a recovery period (i.e. a layover
point) in the trip, which the Circle line does not currently have and which causes many of
its problems. At the present time trains operating the Circle Line simply go
round and round, so once running late there is nowhere that lost time can be
caught up, unless the controller decides to reform (renumber) all the trains,
thus putting them back on time.
The Hammersmith & City line service between
Liverpool Street and Barking will no longer exist. The H & C service will
reverse at Aldgate. The Metropolitan's Barking service will be covered by extending
the Metropolitan Line services beyond Liverpool Street to Barking.
In Tubeprune's view, the new service
proposals present a very risky strategy. First, the number of trains passing though
Praed Street Junction will increase from the present 22.5 to 30, in each direction in each
hour. With delays at Edgware Road caused by crew changes, where the allowance is at
least 60 seconds for each train stopping, the effect will be to lock up the junction at
regular intervals. It happens now, so what the result will be will be after a 25%
increase in the number of trains is anyone's guess. With new "safety"
restrictions on signalling design and higher train speeds, it is likely that eastbound
trains will have to be held at Praed Street home signal rather than the station home
signal just outside Edgware Road station. This will lose a train berth and further
In addition, the plan is to reverse 15
trains an hour at Edgware Road (the "T-Cup and Wimbledon services) in two platforms.
With crossover clearance times, each train will get less than 6 minutes standing in
the platform. This will require split second timing and very close supervision,
something the Underground doesn't do well at the moment.
The second issue is the plan to extend
Metropolitan services to Barking. This was predicted by Tubeprune and described here. It was tried before
in 1939 and failed then because another flat junction is put into the route of a long
service (over an hour in each direction from Watford or Uxbridge) and this will cause
delays over the whole of the sub surface network. Good metro planning requires that
services are kept short, not made longer and less manageable. It will also require
the route to Barking to be gauged for A Stock running, since the stock is currently barred
east of Aldgate because of infringements at St Marys, Whitechapel, near Bow Road, Barking
On top of all this, LU proposes to introduce
the T-Cup service in 2009, before any C Stock replacement trains are delivered.
Where they will get the additional 8 trains they require is a mystery because the 8
additional trains ordered from Metronet are 8-car trains ordered for the Barking extension
service. The plan is that all the A Stock will be replaced before the C Stock (after
all, it is older) and the signalling immunisation required for the new AC traction drive
system is planned on that basis.
By this time, the D Stock fleet will have been
refurbished to extend their service life and to improve the ambience of the
type. Delivery will have commenced of a new fleet of trains for the whole of the SSR.
The plan is for the A Stock to be replaced first,
It is intended that there will be a
uniform stock for all the SSR lines that will offer improved top speed and
comfort. The new stock will also be equipped with Automatic Train Operation (ATO)
and Automatic train Protection (ATP) equipment, although signalling will need to be
upgraded before these capabilities can be used. The provision of air-conditioning is
also a possibility.
There are two major factors that will need to be
taken into account as far as the new trains are concerned: At present the length of the
three stocks varies considerably. The A Stock is comprised of 8 cars, each approximately
51ft in length. The C Stock trains are made up of 6 cars, also approximately 51ft
and the D Stock trains comprise 6 cars, each approximately 60ft in length. The
seating configuration of each of the three stocks also differs greatly the A Stock
has a relatively high number of seats, the C Stock relatively few. The D Stock falls
between the two.
It is intended that the first of these factors
will be overcome by making all car lengths 51ft (approx.). Trains for the
Metropolitan Line will be 8 cars, as at present. The trains for the T
Cup service (C Stock replacement) will be made up of 6 cars again as at
present. The D Stock replacement will be 7 cars, which will roughly equate to the
length of the existing trains.
How the second factor is to be addressed though
is less clear. Whether there will be a standard interior layout irrespective of the route
on which the train is being operated or perhaps there could be an airliner-style facility
to reconfigure the interiors according to the route operated by being able to slot in or
out seating as deemed appropriate for the route. Personally I think this is
unlikely, the plans to date seem to suggest a compromise seating arrangement. I am sure
this will become clearer once the plans for the trains are more advanced. (It
has been confirmed to Tubeprune that a uniform seating layout is the plan at present).
One matter that is clear though is that each car
will have three sets of double doors a major problem with the D Stock is the single
leaf door configuration which are slow to open and close and which result in extended
dwell times at stations. By 2011, the total number of trains in service
will have risen to 173 and as implied above will be a mix of the old and new
fleets. The projection is that the District line will require two more trains, the
new T Cup service a further five and the Metropolitan Line a further sixteen
to allow for the extended service to Barking. (Current stock delivery plans do not
quite line up with this).
Proposals to be Implemented by 2014
It is anticipated that, by 2014, further
improvements will have allowed a further increase in the number of trains operating
between Uxbridge and Barking by 4tph. This should result in a train frequency on
what is today the north part of the Circle Line to 34tph. By the end of this phase
the A and C Stocks will have been withdrawn from service. If the D Stock is not
replaced yet, where will the two extra trains required for the 2011 services described
above come from?
It is also expected that the ATO and ATP will be
able to be utilised over the parts of the lines solely operated by the new trains. The
total number of trains in service is envisaged to be 180 and only new trains would operate
on the northern parts of the lines. This should allow improved running times that
should then reduce the number of trains required to be in service by about ten. At
this point and, as implied by the introduction of ATO/ATP, new signalling systems will
have been installed and become operational.
It is intended that a new Service Control Centre
to be located at Neasden will by now be operational, at least for the Northern services.
Ultimately this centre will be the control point for all the SSR lines, meaning that Earls
Court, Baker Street and all the existing small signal cabins will be decommissioned.
Tubeprune is told by informed sources that
the PPP contract calls for two control centres. Don't put all your eggs into one
basket, seems to be the word here.
Proposals to be Implemented by 2017
The introduction of the final part of the plan
will allow an increase of about 4tph between Wimbledon and Barking, increasing the South
Circle capacity to 34tph to match that on the Northern Circle. It also increases the
service from the City to Barking to 34tph, which is an increase of 10tph over the
The control and signalling systems, the track
and the fleet will, by now, all have been upgraded. All sections will now be
controlled from Neasden (doubtful) and all trains running under ATO/ATP.
Total trains in service will now be in the region of 180, all new stock. District
Line services will use all the final increase in the number of trains.
At this point Ill briefly summarise what
all this has brought about, in terms of trains in service and the kilometrage achieved.
Today the SSR has 150 trains in service, covering 62,000km each day. At the end of
all this there will be 180 trains in service, a 20% increase, covering a total of 73,700km
each day, an increase of 18.9%. So, those are the plans as to what the SSR services
will look like at the end of the programme but there are some pretty radical changes
The question of the refurbishment, upgrading and
renewal of stations is outside the terms of reference of this article, but, of course, the
changes will have an impact on stations. It is well known that commitments have been
given that there will be improvements to stations and there are several major projects
already underway and a number of other projects have been announced. But many
stations have been neglected for many years.
I assume that there is a similar plan in place
for a programme of station refurbishment in the same way that the report around which I
base this article. However, anyone who uses London Underground on a regular basis
and particularly during the peaks will be only too well aware that many
stations are currently extremely congested, to the point that crowd control measures have
to be put into effect on more or less a daily basis at some key interchange points and
trains instructed to non-stop for safety reasons. A lack of station capacity could
seriously affect the benefits that the improved train service will offer. Steps are
being taken to ensure that station capacity does not detract from the introduction of the
train service upgrade. This will include forecasts for future demand at stations,
assessments of where it is anticipated crowding will increase and the effect this will
have on journey times and to consider and amend (if necessary) the station decongestion
programme that is already in place. Other day-to-day station activities will need to
be re-examined if the upgraded service is to be fully realised.
Staff Management, Deployment & Training
So, what of the people at the sharp
end of the train service the Train Operators? What will the effect of
all this be on them? It appears that there could well be quite a few changes.
But it doesnt appear to be all bad news. At the moment there are a total of
906 Train Operators required for the lines making up the SSR at the end of the
whole programme this is expected to have increased to 1,006 as would be suggested by the
increased number of trains that the plan envisages actually putting on the railway.
The statistics apparently show that the
effective driving time per running duty has dropped over the last seven years by about 6%.
I presume this loss of efficiency has come about through the reduction in the
working week and the time spent not driving when on meal break (this is 30 minutes plus
walking time but it is not unusual that the time spent off a train could be up
to an hour therefore the excess is considered as lost time).
There is also a concern that this could also be affected with the introduction of
family friendly rostering which is intended to complement LUs
work-life balance policy a matter I may comment on elsewhere at some
Further matters are also being considered.
The introduction of a uniform rolling stock could offer the possibility of greater
flexibility of routes covered, subject of course to appropriate road training having been
given. There is also the matter of Train Operator depots. Currently there are
seven on the SSR but it is considered that they are not ideally located in view of the
changes outlined above. The plan at the moment is actually for the number to
increase and for there to be some relocation to the various termini the idea of
this being that it would get away from the problems of a train occupying a platform on the
running line whilst waiting for a crew relief and thus preventing trains
behind it running through in the event of a late relief or non-availability of a driver
and so on. Of course, other train and signal related grades will be
affected too and during the implementation there will be implications as far as
training staff in new equipment too.
In conclusion, I emphasise that all that
Ive outlined is very much that its been written from a blueprint, which
Im sure will be amended and fine tuned as time progresses but I hope that it has
been of interest and at least gives an idea of how the future of the SSR at least is
expected to evolve. As, when and if further news, changes or amendments come to
hand, I will try to keep this up to date as far as possible.
Further reading is available from Metronet. Images on the Metronet site give an
impression of the interior and exterior views of the proposed new rolling stock.
These trains will be built by Bombardier. The interior view suggests that the
trains will have walk through cars, whether this proves to be a reality or not
only time will tell. There are clearly advantages to this configuration but, in the
light of the need to make trains up in different lengths dependant on route, can this be
achieved whilst retaining flexibility? Yes, if you plug in your laptop after
reforming the train in the depot so that you can tell it how long it is and make the ATP
will work properly.
For more information go to District Dave's