Re: [General] Signal failures

[signal_spotter@my-deja.com]

In article <399C7B7F.DB85087@yahoo.com>,
  Sam Eades <eadess@yahoo.com> wrote:
>
>
> Tony Gatt wrote:
>
> > Sam Eades <eadess@yahoo.com> wrote in message
> > 399B3A20.CD6D392C@yahoo.com">news:399B3A20.CD6D392C@yahoo.com...
> > > I have been working at Thomastown station (on the single line
between Keon
> > Park
> > > and Lalor on the Epping line) when all Automatic signals between
Lalaor
> > and Keon
> > > Park have failed. This has required the driver to stop at each
automatic
> > signal
> > > (they are only about a train length apart) for 30 secs, then
proceed with
> > > caution to the next. It has added about 15 to 20 mins to the
journey time.
> >
> > Yes, and by observing the rules correctly, he made it to his
destination
> > without incident :)
>
> I'm not saying that he shouldn't have done what he did, I think that
he was
> totally in the right. I'm just commenting on how 1 computer failing
can cause 20
> min+ delays for every train.
>
> An idea I have for alleviating problems such as these is perhaps a
switch on the
> signal so the Automatic Signal can be "Switched Out". This switch
would
> extingush the signal lamps and lower the trip arm, as well as
extending the
> track circuit to the next fixed signal. So instead of having 2 short
sections
> with 1 defective and 1 operational signal, you have 1 long section
with 1
> operational signal. Could it work?
>
> Sam
>
>
But by the time the driver waits a prescibed time (to determine that
the signal is at stop because of a track circuit/train stop failure and
not a train ahead), then gets out of the cab, unlocks the door, flicks
the switch, locks the door, gets back in the cab and choofs off, it
will take about the same amount of time. Then you have a scenario of a
switch bypassing the signalling system (unless you make it one-shot in
which case every train will have to go through the same rigmarole) and
then when the fault clears who is going to flick the switch off to
resume normal signalling. What you propose has some merits in theory
but won't work in real life. Also what you have done by
introducing 'block' working is limit the block to 1 train in the entire
section. This would have a more disastrous effect on train running than
having many trains following each other albeit at a slower speed. Also
I would be surprised that an auto section is computer based unless it
was an extension of a computer based interlocking. It would not be
economical to install a computer based system for an auto system alone.
So generally the reason an auto signal is at red is because there is a
train in section, the track circuit's fuse has blown, the rail is
broken or the trainstop has failed to return to normal (generally due
to the previous train damaging the trip arm). All these factors would
occur regardless of the type of system in place.

regards Ian
Signal Design Engineer



Sent via Deja.com http://www.deja.com/
Before you buy.