SRAWS (Signal Repeating/Southern Region AWS): why so little info

[AHBD]

....

To clarify, this is not actual cab signalling in the sense that lineside signals are not used, ...

Thanks, however my question was based on removal of lineside signals, as that saves cost, and in cab just seems better than small coloured dots zooming at one etc

 

[rmHawk765]

Thanks, however my question was based on removal of lineside signals, as that saves cost, and in cab just seems better than small coloured dots zooming at one etc

No doubt it's better, but the safety benefit of full in cab signalling versus a system that can still bring the train to a stand if necessary is rather small, whereas AWS/TPWS to KVB/PZB/ALSN/ETCS L1 LS is a fairly significant jump.

Cab signalling becomes significant really when you consider capacity. Even with fixed blocks it can allow trains to run closer and of course it is the stepping stone to a moving block system.

 

[takno]

Thanks, however my question was based on removal of lineside signals, as that saves cost, and in cab just seems better than small coloured dots zooming at one etc

The idea that you're going to save a lot of money by taking out the physical signals in such a setup is optimistic. The ramp/beacon is still fundamentally going to require the same power, data and interlocking to be installed in the area, so why not get some fairly cheap lights and repeat the information on them.

In any case, nobody would ever try to make the safety case when there's no backup for an AWS fault, and a real possibility of increased spads

 

[MarkyT]

No doubt it's better, but the safety benefit of full in cab signalling versus a system that can still bring the train to a stand if necessary is rather small, whereas AWS/TPWS to KVM/PZB/ALSN/ETCS L1 LS is a fairly significant jump.

Cab signalling becomes significant really when you consider capacity. Even with fixed blocks it can allow trains to run closer and of course it is the stepping stone to a moving block system.

How much headway benefit you get from cab signalling depends heavily on how continuous the better-aspect updates are. While a fully continuous implementation can update forward aspects before the driver even sees them, A cheap system with little infill will be much worse than visual interpretation in this respect if the associated protection features don't step up and allow reacceleration until a transponder at or near the signal is reached, even when the step-up had been seen by the Mk.1 eyeball many hundreds of metres before on approach.

Infill is expensive, requiring additional transponders, or track loop antennae. One way to tackle this in modern systems is by using data radio which is clearly what the higher levels of ETCS do on a wide scale. A localised radio system can also be used in conjunction with Level 1 Limited Supervision. Radio allows an infill distance range on approach, often corresponding quite closely with the visual range of a signal, to be covered entirely by transmission from a single point source.

Administrations' desire to remove lineside signals is not a new phenomenon.

The Pensylvannia RR experimented with no signals for their coded track circuit system but ultimately decided to retain them.

The Chicago & North Western RR removed many lineside signals in the 1920s after they implemented an inductive Automatic Train Control (ATC) system across their entire mainline between Chicago and Omaha. For nearly a century, C&NW trains traversed America's midwest relying entirely on cab indicators for plain track headway blocks, with optical lineside signals only at junctions and crossovers. They didn't even have any signage marking and labelling the fixed block extremities originally, but block markers were introduced subsequently, after rule changes came into force requiring crew to be able to report exactly where they were in radio communications.

As @takno says, with all that track circuit equipment out there, there was still a lot to maintain, but filament-based signals were a major headache, mainly due to the repeated lamp changing required, either following in service failure or often through a policy of routinely changing them before they're likely to fail, in very operationally critical areas particularly. Even the long-life 8000-hour version of the common dual filament SL35 lamp in the UK is rated for under one year of continuous illumination. They were under-run a little for longevity, but lamp changing was a significant proportion of the technician workload in filament days nonetheless.

 

[edwin_m]

The change to LED aspects does of course reduce ownership costs for signals, hence also the benefit of removing them.

 

[Kentish RYS]

The change to LED aspects does of course reduce ownership costs for signals, hence also the benefit of removing them.

However it is still not zero, and especially at re-signallings, they will be less and less common as the years go on on the GB Mainline network as ETCS becomes more common.

 

[takno]

However it is still not zero, and especially at re-signallings, they will be less and less common as the years go on on the GB Mainline network as ETCS becomes more common.

Signals away at this point with ETCS makes more sense. The sheer amount of development and testing, and the weight of international adoption makes it a clearly safe options. Moving-block or smaller-than-conventional-signalling block ETCS provide demonstrably better-than-signals level of safety and an increase in capacity, so it's bound to start appearing at scale.

Signals away based on an in-cab system that Southern Region had knocked up as a better-than-AWS signal support system would have been a different matter.

 

[Kentish RYS]

Signals away at this point with ETCS makes more sense. The sheer amount of development and testing, and the weight of international adoption makes it a clearly safe options. Moving-block or smaller-than-conventional-signalling block ETCS provide demonstrably better-than-signals level of safety and an increase in capacity, so it's bound to start appearing at scale.

Signals away based on an in-cab system that Southern Region had knocked up as a better-than-AWS signal support system would have been a different matter.

Yes very true the thread has broadened from just SRAWS but as the original system under discussion, the way a lot of systems of the time worked, it wouldn't make the same sense to remove the signals. ETCS works inna different way to transmit the signal to the train (especially when not level 1) so it is a no-brainer to go no-signals

 

[zwk500]

Cab signalling becomes significant really when you consider capacity. Even with fixed blocks it can allow trains to run closer and of course it is the stepping stone to a moving block system.

This is entirely dependent on how the specific system handles the braking curve in relation to block sections, and how it shows the aspects/limit of authority.

 

[rmHawk765]

This is entirely dependent on how the specific system handles the braking curve in relation to block sections, and how it shows the aspects/limit of authority.

Well yes, potentially I oversimplified. I mean a system with an actual target speed indication, not merely something which shows signal aspects in the cab.