In article ,
Tom Anderson writes
However, (a) trains
would have to slow down for the station starter anyway (no idea what that
means!)

The station starter is the signal at the departure end of the platform;
while it is red, the train can't (legally) start away.

On LU the signalling is designed on the assumption that all trains will
be stopping at every station[*]. If a train runs through a station at
speed and passes a red signal, it is possible that the trip stops will
not stop it in time to prevent an accident [+]. Therefore drivers are
instructed to slow to 5mph (or in some places 10mph) when running
non-stop through a station.
[*] There are some specific exceptions, such as Turnham Green on the
Piccadilly, and the signalling in these places is altered accordingly.

[+] In general, a signal can only turn green if a train hitting the
train stop of the following red signal *at line speed* will be stopped
before the point of actual danger (e.g. another train). Designing for
the maximum possible speed of all trains would be unduly restrictive, so
designing to the speed limit of the line is a sensible compromise.

--
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:

Clive D. W. Feather wrote:

[*] There are some specific exceptions, such as Turnham Green
on the Piccadilly, and the signalling in these places is altered
accordingly.

In the case of Turnham Green, both lines do have home signals - the EB
line's homes are A631^A and A631^B, and the WB's are A630^A and A630^B.
Would the overlaps on these homes simply be longer than usual, to allow
a train to alternately stop or pass through at linespeed?


[+] In general, a signal can only turn green if a train hitting the
train stop of the following red signal *at line speed* will be
stopped before the point of actual danger (e.g. another train).
Designing for the maximum possible speed of all trains would be
unduly restrictive, so designing to the speed limit of the line is
a sensible compromise.


This doesn't make much sense. Are you saying that if a train passes a
red signal at linespeed or higher and gets tripped, the signal in rear
could change to green if the entire train manages to exit that signal's
overlap?

TheOneKEA wrote:
Clive D. W. Feather wrote:
snip
[+] In general, a signal can only turn green if a train hitting the
train stop of the following red signal *at line speed* will be
stopped before the point of actual danger (e.g. another train).
Designing for the maximum possible speed of all trains would be
unduly restrictive, so designing to the speed limit of the line is
a sensible compromise.


This doesn't make much sense. Are you saying that if a train passes a
red signal at linespeed or higher and gets tripped, the signal in rear
could change to green if the entire train manages to exit that
signal's overlap?

A signal is controlled by all the track circuits between it and the end of
the next signal's overlap. Therefore a signal will show red whilst there is
an occupied track circuit either between it and the next signal or in the
next signal's overlap.


--
Cheers for now,

John from Harrow, Middx

remove spamnocars to reply

In article . com,
TheOneKEA writes
[*] There are some specific exceptions, such as Turnham Green
on the Piccadilly, and the signalling in these places is altered
accordingly.
In the case of Turnham Green, both lines do have home signals - the EB
line's homes are A631^A and A631^B, and the WB's are A630^A and A630^B.
Would the overlaps on these homes simply be longer than usual, to allow
a train to alternately stop or pass through at linespeed?

Yes. Just as with a signal between stations. The overriding principle is
that a train stop hit at line speed should stop the train before the
point of danger.

[+] In general, a signal can only turn green if a train hitting the
train stop of the following red signal *at line speed* will be
stopped before the point of actual danger (e.g. another train).

This doesn't make much sense. Are you saying that if a train passes a
red signal at linespeed or higher and gets tripped, the signal in rear
could change to green if the entire train manages to exit that signal's
overlap?

Yes. But, in that case, the situation will still be protected. Um, let's
see:

|-O 1 |-O 2 |-O 3 |-O 4 |-O 5
-+---A---+---B---+---C---+---D---+---E---+---F---+---G---+---H---+---I-

Let's assume that braking distance from line speed is 1.4 times the
signal spacing. So:

1 is red if A, B, C, D, or E is occupied
2 is red if C, D, E, F, or G is occupied
3 is red if E, F, G, H, or I is occupied

etc. Suppose there's an obstruction at F. Signal 3 will be red because
it's within its block. Signal 2 will be red because a train tripped at
signal 3 won't stop until somewhere in G. Signal 1 can be green because
a train tripped at signal 2 from line speed will stop somewhere in E.

Now suppose a train runs past signal 1 at well over line speed and hits
the trip at signal 2. Signal 1 will be red at this point because the
train is occupying B and C. The train brakes but, because it was
speeding, it doesn't stop until somewhere in F. As the rear of the train
passes the E-F boundary signal 1 will revert to green *but* the
situation is still protected by signal 2 at red.

Clear? Or have I answered the wrong question?

--
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:

Clive D. W. Feather wrote:
In article . com,
TheOneKEA writes
In the case of Turnham Green, both lines do have home signals - the
EB line's homes are A631^A and A631^B, and the WB's are A630^A and
A630^B. Would the overlaps on these homes simply be longer than
usual, to allow a train to alternately stop or pass through at
linespeed?

Yes. Just as with a signal between stations. The overriding
principle is that a train stop hit at line speed should stop the
train before the point of danger.

Thanks, that was what I thought.

This doesn't make much sense. Are you saying that if a train passes
a red signal at linespeed or higher and gets tripped, the signal in
rear could change to green if the entire train manages to exit that
signal's overlap?

Yes. But, in that case, the situation will still be protected. Um,
let's see:

snip

Clear? Or have I answered the wrong question?

I understand now. But it doesn't seem sensible to place the signals
that close together, or only hold one signal in rear of the signal
protecting an obstruction at danger.

In article . com,
TheOneKEA writes
I understand now. But it doesn't seem sensible to place the signals
that close together,

Signal spacing is chosen to meet various requirements, including maximum
throughput of trains. For example, on LU there are usually several
signals approaching a station, because this allows a train to draw up
close as the previous train departs. This is *better* than spacing the
signals further apart, but could mean that there are four or five red
signals behind a train under some circumstances.

or only hold one signal in rear of the signal
protecting an obstruction at danger.

Why? If one red signal can protect the obstruction, what's the need for
more?

--
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:

Clive D. W. Feather wrote:

Signal spacing is chosen to meet various requirements, including
maximum throughput of trains. For example, on LU there are usually
several signals approaching a station, because this allows a train
to draw up close as the previous train departs. This is *better*
than spacing the signals further apart, but could mean that there
are four or five red signals behind a train under some
circumstances.

I've seen numerous examples of this around the system and guessed that
the rationale was something similar to what you've just stated.


or only hold one signal in rear of the signal
protecting an obstruction at danger.

Why? If one red signal can protect the obstruction, what's the need
for more?

You just showed that under certain circumstances, one signal is not
enough to protect an obstruction (or at least I think you did...)

TheOneKEA wrote:
Clive D. W. Feather wrote:

Signal spacing is chosen to meet various requirements, including
maximum throughput of trains. For example, on LU there are usually
several signals approaching a station, because this allows a train
to draw up close as the previous train departs. This is *better*
than spacing the signals further apart, but could mean that there
are four or five red signals behind a train under some
circumstances.

I've seen numerous examples of this around the system and guessed that
the rationale was something similar to what you've just stated.


or only hold one signal in rear of the signal
protecting an obstruction at danger.

Why? If one red signal can protect the obstruction, what's the need
for more?

You just showed that under certain circumstances, one signal is not
enough to protect an obstruction (or at least I think you did...)

You only need one signal to proect an obstruction. Additional signals mean
that a following train can enter a platform more closely behind the one
departing whilst still maintaining a safe distance between them.

In article .com,
TheOneKEA writes
Why? If one red signal can protect the obstruction, what's the need
for more?
You just showed that under certain circumstances, one signal is not
enough to protect an obstruction (or at least I think you did...)

Um, no.

You have to make *some* assumptions when designing a signalling system.
The ones LU make a
(1) train stops will stop trains in the design distance;
(2) trains won't be exceeding the speed limit at the point they pass a
red signal in the worst situation.

Within those assumptions, one signal is all that's needed. There may be
two or more, but only one is doing the protecting.

--
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:

In message . com,
TheOneKEA writes

[*] There are some specific exceptions, such as Turnham Green
on the Piccadilly, and the signalling in these places is altered
accordingly.

In the case of Turnham Green, both lines do have home signals - the EB
line's homes are A631^A and A631^B, and the WB's are A630^A and A630^B.
Would the overlaps on these homes simply be longer than usual, to allow
a train to alternately stop or pass through at linespeed?

Yes. A Non-stopping Picc train doesn't have any speed restriction on
the fast lines through Turnham Green.

The normal 'rule' is 5 mph at a station starter. However, if we (the
Picc) are run down the local (District) line from Acton to Hammersmith,
we can pass the starters at 25 mph., so there must be some adjustments
of the overlaps there too.
--
Steve Fitzgerald has now left the building.
You will find him in London's Docklands, E16, UK
(please use the reply to address for email)