Boltar wrote:
[+] For example, on 1953-04-08 twelve passengers were killed in a
collision just in rear of signal A491, which had failed. The driver
of the rear train failed to control his speed after tripping past
A489.

And in other news hundreds died in Comet airliner crashes caused
by metal fatigue. Good thing we didn't keep on pressuring aircraft
and flying so high else who knows how many other people would
have died!

Are you seriously suggesting that 50 years later the controllers
still wouldn't know that there was a train on the section ahead of
a stuck signal and so to warn the driver behind?


Much of the railway works on Automatic signals. These are not repeated in
any signal cabin or control room. To do so would be wasteful of taxpayers
money.

The driver who passed A489 knew there was probably a train ahead of him
(it could have moved off). Nevertheless he failed to control his speed
and killed 12 people as a result.

If there was a train in the section ahead of him then why was he given
permission to move past the signal into that section in the first place
as
it seems to me the signal would have been red at that point even if
not broken. AFAICS all that was required was for control to tell the
driver to remain at said signal until the block ahead was clear, this
isn't rocket science.

B2003

Boltar wrote:
The driver who passed A489 knew there was probably a train ahead of
him (it could have moved off). Nevertheless he failed to control his
speed and killed 12 people as a result.

If there was a train in the section ahead of him then why was he given
permission to move past the signal into that section in the first
place as
it seems to me the signal would have been red at that point even if
not broken. AFAICS all that was required was for control to tell the
driver to remain at said signal until the block ahead was clear, this
isn't rocket science.


Because it's an automatic signal, (that means it works autimatically). It
normally shows green, proceed. Causes of it remaining at red, danger include
reasons additional to a train being on that particular section of track.
Such reasons include, a defective signalling circuit, a broken rail
(although that's not guaranteed to keep the signal at danger) or perhaps
flooding of the track. Being in an automatic area no signalbox or control
centre has control over it and so the signaller or controller cannot tell
the driver what the problem is. Therefore the driver cannot be "given
permission". S/He waits one minute and if the signal fails to clear there is
a standing instruction to proceed. This is the "stop and proceed" rule.

This rule relied on the driver travelling at such a speed that he can stop
before reaching an obstruction. Failure of a number of drivers to comply
with this simple rule resulted in a speed limiting device being fitted so
that after the tripcock is activated a maximum speed of 10mph is all that's
available for (IIRC) 5 minutes. Hence delays build up. Even if the whole
railway were controlled by signallers there would still be delay.

the driver what the problem is. Therefore the driver cannot be "given
permission". S/He waits one minute and if the signal fails to clear there is
a standing instruction to proceed. This is the "stop and proceed" rule.

So in other words the driver should only wait for 1 minute at any red
light
since the line controllers don't have a clue whats going on and
wouldn't
know if train A was in front of train B. In which case how come I've
been
in trains stuck at non broken red lights for up to 10 minutes at
various times?

Perhaps its about time LU moved into the 20th century , never mind the
21st and actually had electronic line maps in the controllers rooms so
they
had a bloody clue where the trains on the line they're controlling
actually are.

B2003

"Boltar" wrote in message
oups.com...
the driver what the problem is. Therefore the driver cannot be "given
permission". S/He waits one minute and if the signal fails to clear there
is
a standing instruction to proceed. This is the "stop and proceed" rule.

So in other words the driver should only wait for 1 minute at any red
light
since the line controllers don't have a clue whats going on and
wouldn't
know if train A was in front of train B. In which case how come I've
been
in trains stuck at non broken red lights for up to 10 minutes at
various times?

Perhaps its about time LU moved into the 20th century , never mind the
21st and actually had electronic line maps in the controllers rooms so
they
had a bloody clue where the trains on the line they're controlling
actually are.

B2003

Boltar...... The L/C and signal guys dont have a visual of whats on the
track. All they have is a line diagram that shows a light when a train is
occupying a section of track. That shows a train is there.
A signal failure manifests itself by.......thinking a train is in the
section ahead of the signal. So the signal goes red, thus stopping you and
everyone else. The diagram shows a train in the section, if it didnt the
signal could go green. So the first the L/C knows about it is when your
train calls up and says i've been here and its red. The L/C looks at the
diagram and sees a light in the section. He now thinks there is a train
there. Before he can authorise anyone to move he has to try and contact the
train by radio. If no-one responds he will assume we have a problem. Its not
actually a 'signal' failure. Its a track circuit down. Nothing wrong with
the signal, its reflecting what it thinks is a train ahead.
You may have been stopped for 10 minutes....perhaps you were the 5th train
in line.
If the T/Op cant contact L/C he then can carry out the appropriate
procedure. If its an auto, wait 2 minutes and go at a speed that he can
stop....etc etc. If its protecting points then he must contact L/C by any
means. If he cant get a direct contact, such as signl phone he nmust wait
for a visual/verbal direct communication.
They tried the system where trains set off at 2 minute gaps without the
hindereance of signals about 100 years ago.....too many crashes....You might
be a bit late for a meeting or dinner...but this is safe.

Mal

They tried the system where trains set off at 2 minute gaps without the
hindereance of signals about 100 years ago.....too many crashes....You might
be a bit late for a meeting or dinner...but this is safe.

I'm not suggesting ignoring signals, I'm just trying to figure out the
logic
behind the system they use now. Signal has failed so go slow across the
next 2 working signals. Don't get it. Signals don't fail to green
(supposedly)
so if the signal after the failed one is green why on earth go slow
past it,
why not just go normal line speed? It just seems OTT.

B2003

Boltar wrote:
They tried the system where trains set off at 2 minute gaps without
the hindereance of signals about 100 years ago.....too many
crashes....You might be a bit late for a meeting or dinner...but
this is safe.

I'm not suggesting ignoring signals, I'm just trying to figure out the
logic
behind the system they use now. Signal has failed so go slow across
the
next 2 working signals. Don't get it. Signals don't fail to green
(supposedly)
so if the signal after the failed one is green why on earth go slow
past it,
why not just go normal line speed? It just seems OTT.


Because of something called an "overlap". In order to provide protection for
the next train ahead the signal is positioned before the start of the track
section it controls access to. The length of the overlap varies according to
the weight and speed of the trains plus an allowance for bad weather.

If driver's resumed normal speed after passing only one signal at clear or
caustion there might be another train or other problem withing the overlap.
Then no one gets home for tea.

If you want to learn about how the underground works, and why it's done the
way it is, try http://www.trainweb.org/tubeprune

On Tue, 31 May 2005, Brimstone wrote:

Boltar wrote:

They tried the system where trains set off at 2 minute gaps without
the hindereance of signals about 100 years ago.....too many
crashes....You might be a bit late for a meeting or dinner...but this
is safe.

I'm not suggesting ignoring signals, I'm just trying to figure out the
logic behind the system they use now. Signal has failed so go slow
across the next 2 working signals. Don't get it. Signals don't fail to
green (supposedly) so if the signal after the failed one is green why
on earth go slow past it, why not just go normal line speed? It just
seems OTT.

Because of something called an "overlap". In order to provide protection for
the next train ahead the signal is positioned before the start of the track
section it controls access to. The length of the overlap varies according to
the weight and speed of the trains plus an allowance for bad weather.

If driver's resumed normal speed after passing only one signal at clear or
caustion there might be another train or other problem withing the overlap.
Then no one gets home for tea.

According to http://www.trainweb.org/tubeprune/signalling1.htm, signals
are linked to the overlap between them and their section, as well as to
the section itself (the overlap, being part of the previous section, is
also linked to that section, of course). If that's true, there can never
be anything between a green signal and the end of its section, in the
overlap or not. Thus, it is always safe to go up to full speed at a green
signal. No?

tom

PS This is eerily reminiscent of concurrent computer programs. Things
queueing up on a critical section, regulated by wait and signal
operations. Such programs are notoriously and fiendishly difficult to
write correctly - or to read!

PPS Got as far as drawing this much of a highly entertaining diagram
before deciding text alone would do:

/----O /----O /----O
==== ==================== ==================== =======
^....^

Thought i'd include it in case anyone feels like drawing pictures and
wants somewhere to start!

tom

--
power to the people and the beats

"Tom Anderson" wrote in message
h.li...
On Tue, 31 May 2005, Brimstone wrote:

Boltar wrote:

They tried the system where trains set off at 2 minute gaps without the
hindereance of signals about 100 years ago.....too many crashes....You
might be a bit late for a meeting or dinner...but this is safe.

I'm not suggesting ignoring signals, I'm just trying to figure out the
logic behind the system they use now. Signal has failed so go slow
across the next 2 working signals. Don't get it. Signals don't fail to
green (supposedly) so if the signal after the failed one is green why on
earth go slow past it, why not just go normal line speed? It just seems
OTT.

Because of something called an "overlap". In order to provide protection
for
the next train ahead the signal is positioned before the start of the
track
section it controls access to. The length of the overlap varies according
to
the weight and speed of the trains plus an allowance for bad weather.

If driver's resumed normal speed after passing only one signal at clear
or
caustion there might be another train or other problem withing the
overlap.
Then no one gets home for tea.

According to http://www.trainweb.org/tubeprune/signalling1.htm, signals
are linked to the overlap between them and their section, as well as to
the section itself (the overlap, being part of the previous section, is
also linked to that section, of course). If that's true, there can never
be anything between a green signal and the end of its section, in the
overlap or not. Thus, it is always safe to go up to full speed at a green
signal. No?

tom

PS This is eerily reminiscent of concurrent computer programs. Things
queueing up on a critical section, regulated by wait and signal
operations. Such programs are notoriously and fiendishly difficult to
write correctly - or to read!

PPS Got as far as drawing this much of a highly entertaining diagram
before deciding text alone would do:

/----O /----O /----O
==== ==================== ==================== =======
^....^

Thought i'd include it in case anyone feels like drawing pictures and
wants somewhere to start!

tom

--
power to the people and the beats

My brain hurts now......

Mal

In article , Tom
Anderson writes
According to http://www.trainweb.org/tubeprune/signalling1.htm,
signals are linked to the overlap between them and their section, as
well as to the section itself (the overlap, being part of the previous
section, is also linked to that section, of course). If that's true,
there can never be anything between a green signal and the end of its
section, in the overlap or not. Thus, it is always safe to go up to
full speed at a green signal. No?

No.

In some situations it's as you described:

|-O 6 |-O 4 |-O 2
-------+---A---I-------B-------+---C---I-------D-------+---E---I--

6 green requires A, B, and C all clear.
4 green requires C, D, and E all clear.

But on plain line there's unlikely to be separate track circuits, so
it's more like:

|-O 6 |-O 4 |-O 2
---------------I------------X----------I------------Y----------I--

6 green requires X clear.
4 green requires Y clear.

I don't believe there are any photos on my web site showing it, but I've
certainly seen a signal stay green after a train has passed it until it
reaches the overlap block joint. On the Silverlink lines there are
places where, because of this, a single class 313 can disappear around a
bend before the station starting signal goes back to red.

--
Clive D.W. Feather | Home:
Tel: +44 20 8495 6138 (work) | Web: http://www.davros.org
Fax: +44 870 051 9937 | Work:
Please reply to the Reply-To address, which is: