Cambrian Line Radio Signalling failure - RAIB investigating

[big jim]

I’ve just spoken to my driver who informs me that....

 

A: the sighting speed restrictions over the various crossings are permanent speeds not temporary so appear in the planning area as reductions in speed but they still have the 100m approach and exit problem

 

B: crossings AOCL and ABCL as you approach the crossing the DMI will warm you that there is a crossing aproaching which you have to acknowledge or the brakes will come on, it will then bring you down to a speed to be able to stop in case the white light doesn’t flash but once in the crossing the speed doesn’t increase straight away it will ‘open up’ a good couple of vehicles past the crossing

 

C: arriva units can select 2,4,6,8 vehicles with degraded working for things like isolated brakes

[david.hill64]

The Rule Book instruction is that Drivers should accelerate their train as soon as it is on the level crossing.  Thus if a restriction of speed imposed by ERTMS does not allow that the system is not compliant with the Rule Book - are we coming across another instance here of people getting involved in operational safety matters who are not properly competent to do so?

 

 

Mike, I think it is going to be the other way round. ETCS/ERTMS fitment is a mandatory requirement within the EU, so the rule book will have to change to suit. It would be interesting to understand who has responsibility for doing this cross check and seeing what needs to change as a result of ETCS implementation. NR/ORR/RSSB? 

[The Stationmaster]

Mike, I think it is going to be the other way round. ETCS/ERTMS fitment is a mandatory requirement within the EU, so the rule book will have to change to suit. It would be interesting to understand who has responsibility for doing this cross check and seeing what needs to change as a result of ETCS implementation. NR/ORR/RSSB? 

 

That's a good point David.  as it happens I know the chap who did all the initial drafting of the ERTMS Rules for the Rule Book and I know that he did have somebody giving him technical input about how the system would, or wouldn't, do certain things.

 

But I think what we now come to is possible indication of the major disconnect within the industry because 'ownership' of the Rule Book lies with RSSB which has no operational responsibility for the railway.  In Railtrack days - particularly the privatised Railtrack - there was some sense in putting operational and safety documentation responsibility with an outside body and it could be, and was, initially staffed with some people who had past BR experience in various operational disciplines so could inject some commonsense among the nonsense 9and by golly there was some nonsense being put into RSSB back in the late 1990s.  

 

But now things are different and it really strikes me as far more sensible, for several reasons, to return operational and other standards responsibility to those who actually have to run the railway and enforce those standards and although much of the experience is lacking at least Nr has some elements of 'honest broker' in operational ands safety terms and it actually has to run the railway.  The other advantage of such a change would be to shorten (or in reality actually create) the feedback which is so critical with all sorts of safety standards and procedures.  

 

I know from various sources within the industry that getting the Rule Book amended, including getting nonsenses corrected (as was the case with an obviously incorrectly worded supplement last year) is so difficult it is high on impossible and is an over-bureaucratic and long drawn out process.  So, for example, there is little or no opportunity for informed peer review of Rule Book changes before they are published with no consultation at all at practical level.  In fact the disconnect was so great that until quite recently the official system of amendment of the Rule Book made it nigh impossible to physically amend it without the use of a photocopier and made it far more difficult for users to relate changes to their working situation.  The main way round this to date has been a sort of balkanisation via the medium of Rule Book Handbooks.

 

So coming back to your question while the preparation of the Rules lies in one place the matter of practically linking the effect on them of such as ERTMS might, even if recognised at ground level, not be a simple process.  From practical experience the shorter the feedback the more relevant any sort of Rule or procedure will be, and BR managed it reasonably well considering what a large organisation it was.

[Coryton]

Mike, I think it is going to be the other way round. ETCS/ERTMS fitment is a mandatory requirement within the EU, so the rule book will have to change to suit. It would be interesting to understand who has responsibility for doing this cross check and seeing what needs to change as a result of ETCS implementation. NR/ORR/RSSB? 

 

Is it as clear cut as that?

 

I presume the requirement is not that the UK moves to ETCS in exactly the form currently used on the Cambrian. So far as I know it's still under development so I don't think it's impossible that it could be modified to fit better the way the UK railway currently runs.

 

Out of "let's change the rules to make trains waste time at crossings because it's too much faff to reprogram ETCS" and "let's make ETCS behave sensibly" I know which I'd prefer. (And yes, I do appreciate that it's not just a case of bunging in a few lines of code, recompiling and off you go. I'm sure any changes to safety-related code aren't trivial to make. But still...)

 

(Of course maybe the UK won't in fact be required to implement it...but there are many reasons why it would make sense anyway).

[david.hill64]

Is it as clear cut as that?

Yes it is. In the same way that we can no longer mandate yellow fronts or set electrification clearances.

 

ETCS development has been going on a long time and will necessarily be a compromise between the rules, regulations and signalling principles from all EU railways.

[Junctionmad]

Yes it is. In the same way that we can no longer mandate yellow fronts or set electrification clearances.

 

ETCS development has been going on a long time and will necessarily be a compromise between the rules, regulations and signalling principles from all EU railways.

 

There isnt much choice I'm the matter, unless you wish to " roll" your own , like RETB, the only choice is one form of ERTMS. The likelihood is Level 3 will suit busy railways , even though the TMS of ERTMS is being overtaken by proprietary TMSs.  ERTMS Level 2 is where its hard to justify for medium to light lines ( primarily rolling stock fitment costs ) and RETB and other solutions may hold sway.

 

IN time stock will all come either ERTMS fitted or " ready " , rather like DCC.!!! and the costs of ERTMS will disappear into overall purchase costs 

[96701]

I have just spent half an hour going over the ETCS reference guides for designing ETCS schemes with a colleague due to this very incident.

 

In there, it says that the term "Temporary Speed Restriction" under ETCS includes Temporary, Emergency and Permanent Speed Restrictions. The TSR is applied through a TSR system, presumably by the technician on the technicians terminal next to the interlocking, which is then is included in the Movement Authority when that is transmitted to the train.

 

So I suspect that they aren't Temporary Speed Restrictions to protect the crossings, instead they could be Permanent Speed Restrictions, but the RAIB has used the ETCS definition.

 

Also, remember, this is not an ERTMS Line, I know that rulebooks etc. state that it is, but under the correct terminology, it is only an ETCS Level 2 fitted line, ERTMS is yet to be commissioned into use anywhere in the country.

 

Simon

Trust me, they are TSRs for the protection of level crossing users.

[martin_wynne]

RAIB have today published an interim report on the incident: https://assets.publishing.service.gov.uk/media/5bc871d5e5274a0956564a41/IR012018_181018_Cambrian_TSRs.pdf

 

Martin.

[david.hill64]

RAIB have today published an interim report on the incident: https://assets.publishing.service.gov.uk/media/5bc871d5e5274a0956564a41/IR012018_181018_Cambrian_TSRs.pdf

 

Martin.

It is very disturbing. Although the safety critical core of ETCS may be robust, the way it interfaces with other systems has resulted in a serious issue. Emergency line blocks - for example to protect urgent maintenance- are implemented with a zero speed TSR. If you cannot rely on this you have a serious problem. I look forward to the full investigation.

[Talltim]

What I don't get is why the driver of the 4th train, who reported the missing TSR, passed it doing 50mph? If they had passed it not realising it should be there, then they wouldn't have known to report it. If they had been expecting it, why were they going so fast?

[big jim]

It was probably one of those ‘hang on a minute’ moments where he thought to question it in hindsight

 

The thing with ertms is if it’s not on the screen there is nothing outside to warn you of a restriction coming up and if it’s not in the system then the DMI won’t be telling you to brake for it

Edited October 18, 2018 by big jim

[Coryton]

It was probably one of those ‘hang on a minute’ moments where he thought to question it in hindsight

 

The thing with ertms is if it’s not on the screen there is nothing outside to warn you of a restriction coming up and if it’s not in the system then the DMI won’t be telling you to brake for it

 

And presumably the driver wouldn't have had any cause to learn the point at which they needed to start to brake for the TSR because it should always just come up on the screen.

[big jim]

Yes, and I believe that caused an issue early on in testing as every train was braking at exactly the same point for the same restrictions causing the track to wear quickly in the same place rather than evenly over a wider area based on drivers ‘ball size’ some break early, some brake late!

 

I think the speed curves were relaxed slightly to allow for it in the end

[leopardml2341]

The TSR has been in place since 2014?

 

When does a T become a P? Or does it need to?

[The Stationmaster]

The TSR has been in place since 2014?

 

When does a T become a P? Or does it need to?

 

Only 4 years then - I knew of some in BR days which had been there for a decade or more (on very secondary routes).   Simple difference between the two in BR operational terms is that permanent restrictions of speed are taken into account in the point-to-point running times a but the time cost of temporary restrictions of speed is not (that was how it was in BR days, might be different now).  In civil engineering terms I think it would be mainly down to other depts being prepared to accept the permanent time cost and the engineer not having the budget to restore normal line speed

[Talltim]

I read somewhere that in ERTSM all restrictions are the same as they can be entered, edited or deleted and pushed out to the trains almost immediately (or not at all as the case may be...) As David Hill says above, even a total line block is a speed restriction (of 0)

[big jim]

Ive had a speed restriction Come up for a broken fishplate outside Newtown and by the time I’d come back it from Machynlleth had been repaired/replaced and the restriction removed

[Coryton]

The TSR has been in place since 2014?

 

When does a T become a P? Or does it need to?

 

Wasn't this actually to control speed approaching an open level crossing, implemented as a TSR?

[big jim]

Wasn't this actually to control speed approaching an open level crossing, implemented as a TSR?

There were many imposed but not on open crossings, they were put on foot crossings deemed not to have enough sighting for pedestrians

[ess1uk]

When is the full report out?

On 18/10/2018 at 13:30, martin_wynne said:

RAIB have today published an interim report on the incident: https://assets.publishing.service.gov.uk/media/5bc871d5e5274a0956564a41/IR012018_181018_Cambrian_TSRs.pdf

 

Martin.

 

[martin_wynne]

RAIB full report published today: https://assets.publishing.service.gov.uk/media/5df8fa1be5274a08de86827d/R172019_191219_Cambrian_Coast_line.pdf

 

Martin.

[The Stationmaster]

I am slightly surprised (in some respects) that the Recommendations do not wholly address NR's misuse of TROSs on a longer than 6 months basis. Or the fact that while the NR's TROS  data was stored in a volatile manner this could also have been a consequence of a system developed in France/Italy being used in British operating circumstances where there is a very different attitude to TROSs and the manner in which they are advised to Drivers.

[big jim]

That’s a very techy report!

 

my IT expert ERTMS signing fellow driver will he wetting himself with excitement over some of the jargon, terminology and diagrams in there

 

as I put a few pages back when this first came to light I too experienced a missing TSR one night while down there with a 97 on a ballast job, at first I thought it had been lifted/removed but when it reappeared the next night I reported it, turned out it wasn’t anything to do with this issue but still an oddity

 

ive recently been asked to refresh ERTMS and the Cambrian, the route, lovely, the signalling, joy of joys, can’t wait to get back to that!

[johnofwessex]

Do I detect a lack of enthusiasm for ETRMS?

[Robert Shrives]

It has to be said that having read most but not all of the report it is one of the best I have seen for a long time and getting back towards reports of a useful nature compared  - in my mind  almost child like language in some reports a few years back. So well done to the authors.

 

It will be interesting to see how this develops for other ETRMS projects in UK and elsewhere in Europe.   

 

Robert