Is a new system needed to automatically stop trains which pass through a red signal?

[MarkyT]

Also, just to be clear, I do agree that TPWS is an improvement and has definitely improved safety on the railways. But it was and is only part of what was done.

Absolutely agree. When the decision was made to have a protection system, TPWS was the only one basically fully developed and ready to go immediately. Although I don't really like certain 'aspects' of the particular technology chosen there was nothing else that was at the same state of readiness for rapid nationwide deployment. Any other tech would have required yet more development and testing. The interfaces on track and trains were fairly easy which was a huge selling point for rapid implementation. ISTR most signals had already been assessed and ranked by a junction risk model process, taking into account factors like flank protection, traffic levels, even risks of different types of train (remembering that Mk1s were still around in quantity), so it was fairly simple to set a threshold for fitment.

Just a further note. The reason that ATP was not rolled out was cost. But a cut down ATP system was possible (at least in principle for the type used on the GWML). Not all signals had to be fitted and the operation of it could have been changed for lines that were only partially fitted. Indeed, not all signals were originally fitted for the GWML pilot ATP system, deliberate gaps were left (the gaps have since been removed by fitting the relevant signals with ATP).

Not ony cost, but a recognition from all railway administration by this time that all these (admittedly costly) proprietary systems, with supplier lock-in risks, were effectively obsolete already in the light of the upcoming interoperable open ETCS/ERTMS standards. For TPWS+, there was a limited supervision system proposed by Siemens using active switched Eurobalises, with a trial taking place near Slough and an HST equipped with the additional kit onboard. It had no chance of beating the rival Thales scheme to simply provide additional TPWS grids, as the trains were all already fully compatible with this. The Siemens concept could have had some more advanced functionality however such as preventing reacceleration for a time or distance beyond the speed trap, like PZB in Germany as it would have actually transmitted local speed information to a computer-based unit on the train. TPWS, with its simple spaced arm and trigger grids and onboard timer, has no state memory of a speed check thus encoded, so has no way of doing this in current incarnations.

 

[edwin_m]

Absolutely agree. When the decision was made to have a protection system, TPWS was the only one basically fully developed and ready to go immediately. Although I don't really like certain 'aspects' of the particular technology chosen there was nothing else that was at the same state of readiness for rapid nationwide deployment. Any other tech would have required yet more development and testing. The interfaces on track and trains were fairly easy which was a huge selling point for rapid implementation. ISTR most signals had already been assessed and ranked by a junction risk model process, taking into account factors like flank protection, traffic levels, even risks of different types of train (remembering that Mk1s were still around in quantity), so it was fairly simple to set a threshold for fitment.

TPWS was actually developed in response to a perceived need for a system that could be fitted quickly and relatively cheaply and captured most of the benefits of ATP, also avoiding the capacity downsides of that system. It didn't exist beforehand. The functional requirements were defined before the technology was chosen.

 

[MarkyT]

TPWS was actually developed in response to a perceived need for a system that could be fitted quickly and relatively cheaply and captured most of the benefits of ATP, also avoiding the capacity downsides of that system. It didn't exist beforehand. The functional requirements were defined before the technology was chosen.

I don't doubt that, but there had been a period of indecision on implementation that was only resolved by the Railway Safety Regulations 1999, requiring some form of trainstop system at certain high-risk locations by 2003, which forced Railtrack's hand. TPWS was the only rational option at that time, even though many professionals (and the suppliers obviously!) were still loudly calling for a more advanced full or part supervision ATP. The 'Joint Inquiry Into Train Protection Systems' report of 2001 stated that TPWS provided a 'cheap stop-gap' prior to the desirable widescale introduction of ATP and ERTMS, which I suppose, over a long enough timescale, is an absolute, inevitable truth.

 

[HSTEd]

This is true, but is a reason to go to L2 and take advantage of the functionality

There probably is not the industrial capability for significant resignalling acceleration. So the existing interlockings will have to remain in place for a long time yet.

My understanding is that the swiss systems had been migrating to align with ETCS anyway, so were close enough to have a conversion system. AWS/TPWS are nowhere near the ETCS architecture.

The Signum replacement balises installed did not require any interlocking side changes even when they moved beyond "EuroSignum" to full ETCS Level 1LS functionality.

Functionally they sat on the same relay contacts that were used for the Signum control signal and used them to control what (if any) datagram is sent to a train passing over them.

Such a functionality could be used as a direct replacement for TPWS, in that the system would transmit a speed enforcement datagram to the train as it passed over, but only if the interlocking would have done so by powering the TPWS system.

You can replace one pair of TPWS grids with one pair of balises with no interlocking changes at all. As each pair of TPWS grids is effectively capable of enforcing one set of speeds limits (with a passenger and freight value of a fixed ratio), a simple on-off switch is sufficient for the balise to replicate that functionality. The second balise in the pair would be used to provide different behaviour in the reverse direction.
The architecture of Level 1 is not really anything like the architecture of other levels, it needs no radio block centre or anything like that. Communication to the train is solely by euroloops or eurobalises.

The railway has been using different signalling systems for centuries. The idea that the railway would collapse because it can't transition to ETCS fast enough is laughable. We're still using Electric Token Block or Train Staff working on some line. We'd certainly be able to keep the AWS kit running on a handful of installations. RETB never got beyond 4 places.

It won't be the AWS and TPWS maintenance that kills it, it will be the life expiry of the legacy signalling infrastructure.

A major reason NEtwork Rail has finally decided to actually do something about ETCS is because it has become clear that the signalling supply chain cannot possibly sustain resignalling with conventional technology fast enough to avoid a reliability death spiral.

EDIT:

Here is a paper about the Signum replacement rollout that goes into some detail about the lineside equipment unit used