Balise based ASDO - feasible to make this the standard?

[Bletchleyite]

The thread on Northern's guards deal made me wonder about this.

Balise based ASDO (automatic selective door operation) is clearly the gold standard. By having a balise at the platform start and end, the train can know precisely which doors are platformed, and so when the relevant member of staff presses the "open" button only safely platformed doors are released - or none if the whole train is off. It is also "sided" i.e. it can tell which side of the train is platformed, preventing wrong side release entirely.

This appears to be pretty "gold standard" to me compared to cheaper options like GPS which aren't sided and sometimes "misread" resulting in too few doors opening. I've had to run the length of a 10 car 730 due to a fault meaning only the front 4 released - the length of P6 which is never served by 730s!

What kind of costs would be involved in standardising on this obviously superior system? Is it viable for GBR to adopt one such system as a national standard?

 

[172007]

Deleted, changed my mind.

 

[Nym]

So...
RIS-2747-RST and RIS-2795-RST (Or Packet 44 data from ERTMS)?

 

[Dr Hoo]

So...
RIS-2747-RST and RIS-2795-RST (Or Packet 44 data from ERTMS)?

Could you expand this, please?

 

[zwk500]

So...
RIS-2747-RST and RIS-2795-RST (Or Packet 44 data from ERTMS)?

Packet 44 Data is absolutely the way ETCS schemes will be looking to do.

Could you expand this, please?

RIS-2747-RST is the RSSB standard for control of Exterior doors on Passenger Vehicles. It requires use of either track based or Global Navigation Satellite System (GNSS, because GPS is a specific system).

Packet 44 is the ETCS data message for miscellaneous 'national' messages. It is used for a range of applications, include ASDO.

RIS-2795-RST is the RSSB standard for Track to Train RFID Compatibility (i.e. for SDO Beacon Systems.

The thread on Northern's guards deal made me wonder about this.

Balise based ASDO (automatic selective door operation) is clearly the gold standard. By having a balise at the platform start and end, the train can know precisely which doors are platformed, and so when the relevant member of staff presses the "open" button only safely platformed doors are released - or none if the whole train is off. It is also "sided" i.e. it can tell which side of the train is platformed, preventing wrong side release entirely.

This appears to be pretty "gold standard" to me compared to cheaper options like GPS which aren't sided and sometimes "misread" resulting in too few doors opening. I've had to run the length of a 10 car 730 due to a fault meaning only the front 4 released - the length of P6 which is never served by 730s!

What kind of costs would be involved in standardising on this obviously superior system? Is it viable for GBR to adopt one such system as a national standard?

Fundamentally I agree - a track based system would allow both ASDO and CSDE automatically with very low margin for error.

The obvious way to do it is with ETCS-standard equipment sending Packet 44 but that requires the trackside equipment to be capable of sending ETCS messages and the train's on-board equipment to be capable of interpreting them. So very costly for the initial schemes, even if you don't have to fit the entire ETCS equipment at one go.

If you fit a non-ETCS standard system then you'll have to redo it all when ETCS does make its way round, and of course you'll have the inefficiency of multiple systems.

 

[Bletchleyite]

If you fit a non-ETCS standard system then you'll have to redo it all when ETCS does make its way round, and of course you'll have the inefficiency of multiple systems.

ETCS is a long way off on most of the network, so in reality by the time it does get there you will need to replace whatever is there anyway as it'll be at the end of life.

I know they did it on the Cambrian but that was a trial. I don't think ETCS is going to be on the West Highland Line in my lifetime, but when new units are built (which I would expect to be in my lifetime) a couple of cheap passive balises at Corrour would get rid of the "ceremony of the local doors" and all the confusion to tourists pretty easily.

 

[zwk500]

ETCS is a long way off on most of the network, so in reality by the time it does get there you will need to replace whatever is there anyway as it'll be at the end of life.

I know they did it on the Cambrian but that was a trial. I don't think ETCS is going to be on the West Highland Line in my lifetime, but when new units are built (which I would expect to be in my lifetime) a couple of cheap passive balises at Corrour would get rid of the "ceremony of the local doors" and all the confusion to tourists pretty easily.

On some lines like the West Highland as you say there's an argument for a separate, easier system. Although it's not just the passive balises - it's the train's management system to interpret the message. So for interim fitment on, say, the West Coast south (which is a decent while away from ETCS), where ASDO would be very welcome over the Unit de-select, you'll need to be amending the Desiros to permit ASDO.

Also, the critical thing holding up ETCS at the moment is fleet, and as ETCS rolls out (especially for Northern, with the WCML North and Transpennine upgrades proposing quite a lot of ETCS mileage in their area), the fleet will need ETCS fitting anyway. There'll be little point fitting a fleet for ETCS and a separate ASDO system, given ETCS can handle it.

And finally, don't forget the infrastructure side of the answer to platform lengths. Although it'll never be reasonable to extend every single platform for the longest possible train, so (A)SDO will be required no matter what.

 

[Bletchleyite]

And finally, don't forget the infrastructure side of the answer to platform lengths. Although it'll never be reasonable to extend every single platform for the longest possible train, so (A)SDO will be required no matter what.

And even where platform lengths are sufficient, balise based ASDO prevents wrong side release while allowing the slightly faster release provided by driver door release without the safety downside (or even where guard operation remained in full, there would be no need to open the local door first, just release everything at once because the balise did the safety check). On a local train with many stops, 10-15 seconds at each station really adds up.

 

[zwk500]

And even where platform lengths are sufficient, balise based ASDO prevents wrong side release while allowing the slightly faster release provided by driver door release without the safety downside (or even where guard operation remained in full, there would be no need to open the local door first, just release everything at once because the balise did the safety check). On a local train with many stops, 10-15 seconds at each station really adds up.

Yes, good point about Correct-Side Door Enable (CSDE) which is a separate but related part of the doors control system. Beacon-based CSDE and ASDO may even allow the 'priming' of doors before the train stops (as other railways globally do or some UK light rail systems), which Satellite-based systems probably don't have the necessary accuracy for adjacent tracks.

Saving 10 seconds on a 40 second door cycle across an all-stopper can add up quite quickly.

 

[Bletchleyite]

Yes, good point about Correct-Side Door Enable (CSDE) which is a separate but related part of the doors control system. Beacon-based CSDE and ASDO may even allow the 'priming' of doors before the train stops (as other railways globally do or some UK light rail systems), which Satellite-based systems probably don't have the necessary accuracy for adjacent tracks.

Saving 10 seconds on a 40 second door cycle across an all-stopper can add up quite quickly.

Indeed.

Sadly Merseyrail had to drop this feature due to too many wrong-side releases, but if CSDE was part of it it could come back. Certainly all Stadler kit can just have this turned on in software. In Switzerland it's often used as a means of requesting a request stop, too!

== Doublepost prevention - post automatically merged: Yesterday at 11:28 ==

Saving 10 seconds on a 40 second door cycle across an all-stopper can add up quite quickly.

With regard to this I often point out that the start of the rot on Ormskirk-Liverpool journey times (once 25 minutes, now about 35*) was when hustle alarms were added to the 50x units. Just under 10 seconds at each stop (more if they were reopened) caused a load of delay and eventually running times were extended to 27 minutes to take account of it. No doubt the effect on the Southport line was even greater.

* Though I think they are deliberately slowed at present, because with both ends being single platform termini you can't run a bit faster unless the terminus is unoccupied. Headbolt mercifully avoids that by being two platforms.

 

[MarkyT]

Packet 44 Data is absolutely the way ETCS schemes will be looking to do.

Agreed for early pilots, though it's likely standard functionality for this will be adopted eventually within the core ETCS protocols in some future baseline.

RIS-2747-RST is the RSSB standard for control of Exterior doors on Passenger Vehicles. It requires use of either track based or Global Navigation Satellite System (GNSS, because GPS is a specific system).

Packet 44 is the ETCS data message for miscellaneous 'national' messages. It is used for a range of applications, include ASDO.

RIS-2795-RST is the RSSB standard for Track to Train RFID Compatibility (i.e. for SDO Beacon Systems.


Fundamentally I agree - a track based system would allow both ASDO and CSDE automatically with very low margin for error.

The obvious way to do it is with ETCS-standard equipment sending Packet 44 but that requires the trackside equipment to be capable of sending ETCS messages and the train's on-board equipment to be capable of interpreting them. So very costly for the initial schemes, even if you don't have to fit the entire ETCS equipment at one go.

A bare-bones reader and basic onboard processing unit could plausibly interpret packet 44 passive balise messages for various purposes without requiring a full ETCS onboard computer and DMI screen set up.

This would be a little like the custom Eurosignum modules for introducing L1 limited supervision ETCS on SBB's legacy fleet, and conceptually similar to TPWS onboard fitments on many older UK trains with a new combined TPWS/AWS unit in the same space as the old AWS equipment, using the same traction and braking interfaces.

More modern trains with active onboard ETCS computers and screens could emulate the same functionality, reading the same balises.
At some future full ETCS scheme, there will likely still be passive balises in the same places for marking the extremities of the platform, just as the Thameslink core ATO employs additional positioning balises, so it might be 'just' a case of reprogramming the originals to fit in with the new scheme, but in any case they're passive devices with no electrical connection, so a simple swap out if necessary.

 

[Bletchleyite]

That they're passive is certainly a key aspect of it given that many stations are in the middle of nowhere, particularly those with short platforms.

 

[swt_passenger]

Isn’t there still a ‘stop short’ issue with the Desiro version of ASDO used by SWT/SWR, because only the leading vehicle reads the balise? The ’gold standard’ presumably requires every individual carriage to make its own checks, AIUI this was the case with the 458‘s setup?

 

[Bletchleyite]

Isn’t there still a ‘stop short’ issue with the Desiro version of ASDO used by SWT/SWR, because only the leading vehicle reads the balise? The ’gold standard’ presumably requires every individual carriage to make its own checks, AIUI this was the case with the 458‘s setup?

Either that or the train measures how far it has travelled since the cab passed the balise.

 

[Starmill]

It's basically impossible to accidentally open the doors on the wrong side on a 701 already. And they don't need to stop anywhere that doesn't have beacons. Arguably this problem has been technologically solved for decades...

 

[HSTEd]

Isn’t there still a ‘stop short’ issue with the Desiro version of ASDO used by SWT/SWR, because only the leading vehicle reads the balise? The ’gold standard’ presumably requires every individual carriage to make its own checks, AIUI this was the case with the 458‘s setup?

Conceptually you could also fit multiple balises in the platform, and the opening length changes as the balise reader on the leading vehicle passes over each in turn.

This would allow a cut down balise reader to be fitted to trains without ETCS rated odometers.

 

[zwk500]

Conceptually you could also fit multiple balises in the platform, and the opening length changes as the balise reader on the leading vehicle passes over each in turn.

This would allow a cut down balise reader to be fitted to trains without ETCS rated odometers.

Or fit multiple balise readers on each vehicle, which I think is similar to at least one ASDO system in use (TPE?).

 

[D365]

Or fit multiple balise readers on each vehicle, which I think is similar to at least one ASDO system in use (TPE?).

Vehicle underframes are already crowded; I dread to think how multiple balise readers could be feasibly accommodated.