On Jan 6, 12:31*pm, John B wrote:
1) the depot is at Cockfosters. Not sure what time this incident took
place, but if the train died in a place where it blocked or

MId to late afternoon.

significantly impaired depot access, then that's going to have an
obvious knock-on effect on the line

The knock on effects were caused (as usual), by terminators sitting at
arnos grove for ages before being sent back down the line. The concept
of stepping back seems to be a foreign one to the piccadilly line and
its staff.

2) normal line operation is based on turning some trains at Arnos

extremely expensive redundancy (in this case, adding signalling and
track work at Arnos so that it can be used as a full-capacity terminus

Its already got it. All it needs is for the staff to put some effort
into being ready for their train when it arrives and the signaller to
switch the route quickly. Arnos was a terminus anyway for a while
until the section to cockfosters was completed.

mirroring Cockfosters, or turning one of the stations north of Arnos
into an alternative reversing point - in either case, these will only

There is one - its called Oakwood.

B2003

On 6 Jan, 12:31, John B wrote:
1) the depot is at Cockfosters. Not sure what time this incident took
place, but if the train died in a place where it blocked or
significantly impaired depot access, then that's going to have an
obvious knock-on effect on the line

2) normal line operation is based on turning some trains at Arnos
(6ish) and some at Cockfosters (18ish). There are three reversing
platforms at Cockfosters; there is one at Arnos. If suddenly 24 trains
have to be reversed in the one platform, this is going to create some
fairly obvious bottlenecks.

Anyway... while junctions make life more complicated, they also
provide diversions and alternative routes. Whereas when you've got a
railway that's fundamentally two tracks with trains separated by only
a couple of minutes, any disruption is going to have instant and rapid
knock-ons - and the only way you can deal with that is to add
extremely expensive redundancy (in this case, adding signalling and
track work at Arnos so that it can be used as a full-capacity terminus
mirroring Cockfosters, or turning one of the stations north of Arnos
into an alternative reversing point - in either case, these will only
be useful for the few hours a month when access to Cockfosters has
completely failed).

I take your point about the depot and the smaller turning capacity of
Arnos Grove; I'd not considered these.

However, I really do think Boltar has a point here. It was mid-
afternoon, so most trains should have been out of the depot and
theoretically distributed evenly across the length of the line, thus
only a few would have been unavailable due to being trapped north of
Arnos Grove.

There is an additional turning point at Oakwood, apparently, so could
this not have been utilised? And although without having been there
at the time it's impossible to know for certain, but Boltar's claim of
trains sitting at Arnos Grove for AGES before turning around certainly
has the ring of truth to it, in my experience of the tube when things
go wrong. Why would they make trains wait for ages rather than
turning them around urgently? Does the signalling not permit it? If
not, why not? And how does it cope with the scheduled turnarounds at
Arnos Grove?

I'm really not a tube-basher and I think that on balance the tube does
pretty well at providing a service. However, I do think that they are
very unprepared for when things go wrong and seem incapable of dealing
with incidents quickly to stop them becoming major headaches.
Although of course it's easy to say that they could run things better
when we don't know all the constraints, but it does seem to me that
they could have done better in this instance, based of course on what
I've read here!

Patrick

On 6 Jan, 12:31, John B wrote:

(snips lots of worthwhile stuff)

Anyway... while junctions make life more complicated, they also
provide diversions and alternative routes. Whereas when you've got a
railway that's fundamentally two tracks with trains separated by only
a couple of minutes, any disruption is going to have instant and rapid
knock-ons - and the only way you can deal with that is to add
extremely expensive redundancy (in this case, adding signalling and
track work at Arnos so that it can be used as a full-capacity terminus
mirroring Cockfosters, or turning one of the stations north of Arnos
into an alternative reversing point - in either case, these will only
be useful for the few hours a month when access to Cockfosters has
completely failed).


One thing that never seems to get a lot of focus in such situations is
the failed train - what failed, why, and can better maintenance
prevent it or rather more realistically lessen the frequency of such a
failure occurring. It seems to be taken as a given, on both LU and the
mainline railway, that trains fail - of course some will, but I'm not
so sure this should be taken as a given as much as it seems to be.

Connected to this - will the new 09TS for the Vic line and the new S
stock have some fancy but useful self-diagnostic systems on board, for
example? Maybe such things aren't that helpful but depot based systems
are - and in this context I mean system not just as in a computer but
a whole process. TBH I don't very little about the railway rolling
stock maintenance regimes that are in use and I'm sure that they have
advanced significantly over recent years (or at least I would hope
they have!) but I suspect more can be done.

On Jan 6, 1:52*pm, wrote:
1) the depot is at Cockfosters. Not sure what time this incident took
place, but if the train died in a place where it blocked or

MId to late afternoon.

OK, so depot isn't a major issue.

significantly impaired depot access, then that's going to have an
obvious knock-on effect on the line

The knock on effects were caused (as usual), by terminators sitting at
arnos grove for ages before being sent back down the line. The concept
of stepping back seems to be a foreign one to the piccadilly line and
its staff.

Indeed, it might be good to implement stepping back for emergency
reversals. And it's possible that it hasn't been put into place
because nobody cares about getting things going, or because of union
grumpery.

However, it's also possible that it hasn't been implemented because if
the line's in turmoil and all the trains (with drivers on board,
obviously) are stuck behind each other in tunnels along the route,
where exactly are you going to get hold of the extra spare drivers to
run the service? Do they sit around drinking tea at Arnos 29 days a
month just in case things go wrong? Presumably you need another chap
drinking tea at Northfields, and another at Hyde Park Corner, to cope
for failures at those locations...

2) normal line operation is based on turning some trains at Arnos
extremely expensive redundancy (in this case, adding signalling and
track work at Arnos so that it can be used as a full-capacity terminus

Its already got it. All it needs is for the staff to put some effort
into being ready for their train when it arrives and the signaller to
switch the route quickly.

I don't think that's true. The track layout at CULG implies that the
through platforms aren't reversible:
http://www.davros.org/rail/culg/piccadilly.html#layout
....giving you 1 reverser at a time instead of 3 at Cockfosters.

Arnos was a terminus anyway for a while
until the section to cockfosters was completed.

True, but I imagine they changed the signalling at the point when the
extension opened.

mirroring Cockfosters, or turning one of the stations north of Arnos
into an alternative reversing point - in either case, these will only

There is one - its called Oakwood.

No, that's a plain-track station with no reversing facilities. Trains
that are advertised as Oakwood terminators are going into the depot
(where they can be reversed, but complicatedly and slowly because
that's not what it's intended for).

--
John Band
john at johnband dot org
www.johnband.org

On Jan 6, 2:26*pm, Patrick Osborne wrote:
Why would they make trains wait for ages rather than
turning them around urgently? *Does the signalling not permit it? *If
not, why not? *And how does it cope with the scheduled turnarounds at
Arnos Grove?

Don't know; possibly not; possibly because it'd involve expensive
infrastructure upgrades that'd only occasionally be needed; and
because turning round a train in 10 minutes is a lot easier than
turning one around in 2m30s.

I'm really not a tube-basher and I think that on balance the tube does
pretty well at providing a service. *However, I do think that they are
very unprepared for when things go wrong and seem incapable of dealing
with incidents quickly to stop them becoming major headaches.

Information, when an incident is first starting to kick off, is the
worst bit. The number of times I've been on a service at a station
with multiple alternative options when things have gone wrong, and
have ended up picking the wrong one (despite having a better idea
about the system than most) because of inadequate/untimely/completely
false advice is definitely too high.

Although of course it's easy to say that they could run things better
when we don't know all the constraints, but it does seem to me that
they could have done better in this instance, based of course on what
I've read here!

It's difficult to say without knowing more about the problem (e.g. if
the broken train had broken coming out of the depot onto the
southbound=westbound between Cockfosters and Oakwood, then you
wouldn't be able to turn anything round north of Arnos and get it back
down south again), as you say.

I'm pretty certain there's someone at TfL charged with putting exactly
these kinds of questions to Picc and Tube Lines management, to ensure
that they do have the best contingencies possible given their
resources. It'd be nice to see exactly how that process works, and
what the problems were this time...

--
John Band
john at johnband dot org
www.johnband.org

On Jan 6, 2:34*pm, Mizter T wrote:
On 6 Jan, 12:31, John B wrote:



(snips lots of worthwhile stuff)

Anyway... while junctions make life more complicated, they also
provide diversions and alternative routes. Whereas when you've got a
railway that's fundamentally two tracks with trains separated by only
a couple of minutes, any disruption is going to have instant and rapid
knock-ons - and the only way you can deal with that is to add
extremely expensive redundancy (in this case, adding signalling and
track work at Arnos so that it can be used as a full-capacity terminus
mirroring Cockfosters, or turning one of the stations north of Arnos
into an alternative reversing point - in either case, these will only
be useful for the few hours a month when access to Cockfosters has
completely failed).

One thing that never seems to get a lot of focus in such situations is
the failed train - what failed, why, and can better maintenance
prevent it or rather more realistically lessen the frequency of such a
failure occurring. It seems to be taken as a given, on both LU and the
mainline railway, that trains fail - of course some will, but I'm not
so sure this should be taken as a given as much as it seems to be.

Connected to this - will the new 09TS for the Vic line and the new S
stock have some fancy but useful self-diagnostic systems on board, for
example? Maybe such things aren't that helpful but depot based systems
are - and in this context I mean system not just as in a computer but
a whole process. TBH I don't very little about the railway rolling
stock maintenance regimes that are in use and I'm sure that they have
advanced significantly over recent years (or at least I would hope
they have!) but I suspect more can be done.

This rings a bell digs around piles of dusty stuff.

"[The trains] incorporate 'tell-tale' train equipment panels ..."

From the Piccadilly Line progress report, dated 1976.

Ah, and from Brian Hardy in 1976 "Improvements incorporated in this
stock are ... a train equipment fault-finding panel for the driver's
use ...".

Funny how things stick in one's mind. No great relevance really.

On Jan 6, 3:29*pm, John B wrote:
Indeed, it might be good to implement stepping back for emergency
reversals. And it's possible that it hasn't been put into place
because nobody cares about getting things going, or because of union
grumpery.

Wouldn't surprise me if that had something to do with it. However they
can do it if they want to - the victoria line is a good example.

I don't think that's true. The track layout at CULG implies that the
through platforms aren't reversible:http://www.davros.org/rail/culg/piccadilly.html#layout
...giving you 1 reverser at a time instead of 3 at Cockfosters.

No , they're all reversable - even platform 4 southbound. I've been on
trains terminating in all of them at some point or the other.
Admittedly only once for platform 4.

True, but I imagine they changed the signalling at the point when the
extension opened.

Given how people constantly remind us how old it is I'm not so
sure! )

There is one - its called Oakwood.

No, that's a plain-track station with no reversing facilities. Trains
that are advertised as Oakwood terminators are going into the depot
(where they can be reversed, but complicatedly and slowly because
that's not what it's intended for).

Nope, theres a crossover at oakwood just south of the station. Don't
take my word for it - check it out on google earth/maps , its clearly
visible. Whether its in use or not however is another matter I guess.

B2003

On Jan 6, 4:49*pm, wrote:
On Jan 6, 3:29*pm, John B wrote:

Indeed, it might be good to implement stepping back for emergency
reversals. And it's possible that it hasn't been put into place
because nobody cares about getting things going, or because of union
grumpery.

Wouldn't surprise me if that had something to do with it. However they
can do it if they want to - the victoria line is a good example.

I don't think that's true. The track layout at CULG implies that the
through platforms aren't reversible:http://www.davros.org/rail/culg/piccadilly.html#layout
...giving you 1 reverser at a time instead of 3 at Cockfosters.

No , they're all reversable - even platform 4 southbound. I've been on
trains terminating in all of them at some point or the other.
Admittedly only once for platform 4.

True, but I imagine they changed the signalling at the point when the
extension opened.

Given how people constantly remind us how old it is I'm not so
sure! )

There is one - its called Oakwood.

No, that's a plain-track station with no reversing facilities. Trains
that are advertised as Oakwood terminators are going into the depot
(where they can be reversed, but complicatedly and slowly because
that's not what it's intended for).

Nope, theres a crossover at oakwood just south of the station. Don't
take my word for it - check it out on google earth/maps , its clearly
visible. Whether its in use or not however is another matter I guess.

There is, and the latest Quail shows platform 2 as being
bidirectional.

Also, the crossovers at Arnos Grove are laid out for the sidings* to
be accessed from all platorms (and vice versa), so it's probably
intended that a train could go into any platform and then head south
(either into the sidings or down the line having left the sidings
previously). No point in having the crossover otherwise.

*South of the station.

On 6 Jan, 16:38, MIG wrote:

On Jan 6, 2:34*pm, Mizter T wrote:

(snip)

One thing that never seems to get a lot of focus in such situations is
the failed train - what failed, why, and can better maintenance
prevent it or rather more realistically lessen the frequency of such a
failure occurring. It seems to be taken as a given, on both LU and the
mainline railway, that trains fail - of course some will, but I'm not
so sure this should be taken as a given as much as it seems to be.

Connected to this - will the new 09TS for the Vic line and the new S
stock have some fancy but useful self-diagnostic systems on board, for
example? Maybe such things aren't that helpful but depot based systems
are - and in this context I mean system not just as in a computer but
a whole process. TBH I don't very little about the railway rolling
stock maintenance regimes that are in use and I'm sure that they have
advanced significantly over recent years (or at least I would hope
they have!) but I suspect more can be done.

This rings a bell digs around piles of dusty stuff.

"[The trains] incorporate 'tell-tale' train equipment panels ..."

From the Piccadilly Line progress report, dated 1976.

Ah, and from Brian Hardy in 1976 "Improvements incorporated in this
stock are ... a train equipment fault-finding panel for the driver's
use ...".

Funny how things stick in one's mind. *No great relevance really.


Well, yes, if only to treat with a (un)healthy dose of salty
scepticism any claims that a new train is a self-diagnosing healing
machine.

I do like the optimism of a Piccadilly line progress report from 1976,
given that it opened in 1906! Not saying that that fact invalidates
the notion that progress is possible on such a line of course, as it
obviously is. Do LU still do yearly progress reports and call them
that, or was the report not part of an annual series and instead more
of a one-off?

On Jan 6, 5:48*pm, Mizter T wrote:
On 6 Jan, 16:38, MIG wrote:





On Jan 6, 2:34*pm, Mizter T wrote:

(snip)

One thing that never seems to get a lot of focus in such situations is
the failed train - what failed, why, and can better maintenance
prevent it or rather more realistically lessen the frequency of such a
failure occurring. It seems to be taken as a given, on both LU and the
mainline railway, that trains fail - of course some will, but I'm not
so sure this should be taken as a given as much as it seems to be.

Connected to this - will the new 09TS for the Vic line and the new S
stock have some fancy but useful self-diagnostic systems on board, for
example? Maybe such things aren't that helpful but depot based systems
are - and in this context I mean system not just as in a computer but
a whole process. TBH I don't very little about the railway rolling
stock maintenance regimes that are in use and I'm sure that they have
advanced significantly over recent years (or at least I would hope
they have!) but I suspect more can be done.

This rings a bell digs around piles of dusty stuff.

"[The trains] incorporate 'tell-tale' train equipment panels ..."

From the Piccadilly Line progress report, dated 1976.

Ah, and from Brian Hardy in 1976 "Improvements incorporated in this
stock are ... a train equipment fault-finding panel for the driver's
use ...".

Funny how things stick in one's mind. *No great relevance really.

Well, yes, if only to treat with a (un)healthy dose of salty
scepticism any claims that a new train is a self-diagnosing healing
machine.

I do like the optimism of a Piccadilly line progress report from 1976,
given that it opened in 1906! Not saying that that fact invalidates
the notion that progress is possible on such a line of course, as it
obviously is. Do LU still do yearly progress reports and call them
that, or was the report not part of an annual series and instead more
of a one-off?

Ah sorry, I think it was more specifically a Heathrow Extension
progress report. I don't know how many there might have been, because
it isn't numbered, so I'd cynically guess only one ever.