Re: Signalling queries

[Eddie Oliver <eoliver@efs.mq.edu.au>]

Answers to a few queries as yet unanswered by other people in this
thread - apologies for having lost the attributions to said what:

 
> What is the protection (apart for extreme improbability on quiet lines) to
> ensure that a train can't enter the section travelling in the same direction
> under the same rules (on a ticket) but before the preceding train has exited
> the section?

In most such systems (e.g. the Victorian system described by someone
else) the first train has to report its arrival at the far end (or have
it reported if the location is attended), so absolute block is
maintained. However theoretically in some locations, time interval
working is available, i.e. the second train can leave n minutes after
the first (where the various values of "n" are tabulated for different
circumstances), and if the first train gets delayed along the way, it
has to protect against the following train. For obvious reasons, this
fell out of favour when brakevans and guards were abolished.

>> Yes, but the NSW version of train order working has many
>> software/hardware safeguards built in to minimise the likelihood of
>> such an incident - e.g. the location of a train is continually
>> monitored by GPS.

>What is the precise method of doing this?  Does the 'office' end of the
>system poll each train for it's location every x minutes?  Is this a
>function of the Train Radio system, or is there a seperate system that just
>uses the Train Radio as transport?

Basically "yes" to the question in the second sentence, and "it's a bit
of both" to the third sentence. The train radio system can directly
convey GPS information, but there is a separate computer system (called
a "watchdog") in the Train Control centre which deals with this
information separately from the main train order computer. The two
separate computer systems talk to each other, and any inconsistency
between their views of the situation is immediately highlighted to the
controller.

> So the difference between CTC and Train Order systems (both of which seem
> to be worked centrally) is that the movement of trains on CTC systems is
> governed by signal aspects monitored and controlled from a central
> location, while with Train Order working, train movements are governed by
> verbal/written confirmation direct from controllers in a central location?

Yes, with the understanding that the signal aspects in CTC are also
controlled by the state of the track occupancy, i.e. not just by the
actions of the controller - the hardware prevents any unsafe actions by
the controller.

> Is it possible to get some wire and short the rails so that the signal will
> go red even though there is no train,

yes

> or does it "audit" surrounding signal
> blocks to make sure that the circuit is actually being closed by a train
> passing through??

no, but it sometimes works the other way, in that "sequential" designs
can add extra confirmation that a train has actually cleared a track
circuit by checking that it has correctly occupied subsequent ones (so
as to reduce the risk that the track circuit "loses" a train because of
rust on the rails preventing proper contact, for instance). e.g. if a
train is going A-B-C, the sequential system would verify that circuit B
(and possibly C) has become occupied AS WELL AS that A has become
unoccupied. The sequential systems are used only where track conditions
are regarded as risky in this context.


> Okay, I've seen this [electric staff working] at Albion Park and Kiama. I think this is being
> replaced with CTC along with the electrification project now though....

The sections Kiama - Berry and Berry - Nowra will remain as electric
staff.