RAIB Press Release - driver unaware of emergency speed restriction

[martin_wynne]

From RAIB today:

 

 https://www.gov.uk/government/news/overspeeding-incident-at-dauntsey-wiltshire

 

Martin.

[The Pilotman]

Its twenty years since I was on the railway but I always thought that an Emergency Speed Restriction became a Temporary Speed Restriction once enough time had elapsed for it to be published in the next Weekly Operating Notice. This ESR was over six weeks old; is that normal these days? 
From the report: “The 30/125 ESR had been allowed to remain in place for some time without being converted to a Temporary Speed Restriction (TSR)”

[phil-b259]

Not the first time this has happened in recent times....

 

See https://assets.publishing.service.gov.uk/media/5d4048b440f0b60a9f8021cc/R102019_190801_Sandy.pdf

[Trog]

I suppose that shooting the speed where a single detonator was placed at the warning board for every train until the speed was in the drivers notices would be much too old fashioned a way of drawing a drivers attention to an ESR in the modern age.

 

Usually this was a way of using up out of date detonators, some times two were placed on the rail next to each other just in case one did not go off. If a supervisor had a speed to shoot all his neighbours would contribute their old detonators usually in plastic sacks. I have several times spent an hour or two sitting on a sack of detonators beside a roaring fire, talking to the trackman shooting a speed adjacent to where we were working while awaiting the arrival of a ballast train.

Edited November 10, 2020 by Trog
Elucidation

[Dagworth]

16 minutes ago, Trog said:

shooting the speed

Please can you clarify what that means? As an ex train driver I have no idea.

 

Andi

[Rivercider]

30 minutes ago, Dagworth said:

Please can you clarify what that means? As an ex train driver I have no idea.

 

Andi

I see Trog has clarified the meaning now, but I knew what it meant, and assumed that it was a commonly used expression.

The first six months of my railway career was spent as a clerk in the permanent way clerical section of the civil engineers office in Bristol, and my dad spent much of his railway career as a clerk or supervisor in the civil engineers. I must have first heard the phrase back in the mid 1970s.

 

cheers

[SM42]

Shooting the road in my day was what was done before the boards were up.

 

I recall waving a yellow flag / lamp  from side to side came into it too.

 

Andy

[Oldddudders]

Baffled by all this. A suddenly imposed 20 mph speed on a 125 mph line? Why isn't every train being stopped and cautioned? And then hitting the dets as well? As for some freight operator complaining this aspect of safety was costing him £60k p.a., he should have been told it didn't matter if it was £60m! Do it or be out of railway operation!

[DY444]

Talk about setting up a driver to fail.  An ESR that had been there for ages which is changed to apply to passenger trains after he'd booked on, an ATP line and equipped train which was not configured with the newly changed ESR and total reliance on the driver noticing that one warning board had been changed.  He did notice the 125 marker had gone but assumed it had fallen off.  By the time he realised it hadn't it was too late to brake sufficiently. 

 

I can't help thinking there are shades of Nuneaton 1975 in this incident in the sense that both drivers were expecting to see something they didn't see and ran out of braking distance whilst trying to figure out why they didn't see it.  Thankfully the consequences this time were benign.

 

There are obvious dangers in applying a much lower speed to a pre-existing ESR as this incident shows.  It's a pity the ORR and others responsible for safety don't expend their energies in looking for these kinds of real dangers instead of wasting everybody's time and money worrying over what happens if a tall man on stilts wants to take a selfie on a newly electrified line.

Edited November 11, 2020 by DY444

[PaulRhB]

2 hours ago, Oldddudders said:

Why isn't every train being stopped and cautioned? And then hitting the dets as well?

There’s no requirement to caution once the boards are in place. A Dalek / Emergency Warning Indicator should only be in place until it’s published in the WON so until then the driver is expected to check the speed regardless. If the speed is assumed to be missing then the driver is expected to brake, to a stand if necessary, and contact the box by GSMR to report it. 

There are other issues you can probably guess with that . . .

Dets haven’t been used for this for years.

[caradoc]

The time required to convert an ESR to a TSR was a source of mystery and frustration to me too, as a Controller, and no doubt to the PW Staff as well, as they had to ensure the Emergency Indicator remained in working in order throughout. And of course, if it failed delays were incurred as trains were cautioned until it was rectified. Had this particular restriction become a TSR, the later reduction in permitted speed would have made it revert to an ESR and the appropriate additional equipment would then have had to be re-installed, with trains cautioned until this was done. 

 

What this incident does show is what an absolute minefield ESRs, TSRs, the equipment for and advising of has become, and although it was an unusual occurrence (I cannot recall having had to deal with a similar one), procedures should have prevented it. 

 

[Robert Shrives]

Keeping ESR "Darleks" lit has always been a problem. A Darlek inside the Severn tunnel required batteries changing weekly and it meant closing tunnel to traffic while PW staff walked in and out of tunnel.  A dose of the common broke out and an empty stock train conveyed PW to the location. A bit of a COV 19 ballet inside train to maintain SD 2m rule and 15 later job done.  This was done twice!

The downside with the running of ESRs shown up on Cambrian coast where ERTMS speed supervision allows automatic remote supervision but where it is for a say a fishplate fault the speed supervision might be 250m before and after  thus a bit of over the top. Mind you recent system fault report has shown system not failsafe with system loosing defect data after a system  failure and restart...  

Not a perfect world .

Robert     

[Trog]

3 hours ago, caradoc said:

The time required to convert an ESR to a TSR was a source of mystery and frustration to me too, as a Controller, and no doubt to the PW Staff as well, as they had to ensure the Emergency Indicator remained in working in order throughout. ......

 

 

I suspect that some of these mysteries may be something to do with not spoiling the TSR or other, KPI figures.

 

KPI = Key Performance Indicators.

 

By the time I retired now some years ago keeping the figures looking good on the spreadsheets was getting to be way more important than wasting time buggering about with that on track stuff.

[Northmoor]

Why is the industry still relying on informing drivers of TSRs via pieces of paper (that they are expected to recall in detail hours later) and then seeing signage, rather than using the berth-dependent broadcast function in GSM-R?  That way you only inform drivers about to enter the area and don't bother anyone else.  GSM-R cost a fortune to implement, the rail industry should be exploiting every function it has.

[SM42]

As the report says, the speed was there to be seen. The human factor then came into play.

 

The challenge is how to overcome the erroneous conclusions that humans come to because of familiarity. We've all done it at some point in our lives.

 

The technology is in theory, available to help.

 

Andy

[caradoc]

16 minutes ago, Northmoor said:

Why is the industry still relying on informing drivers of TSRs via pieces of paper (that they are expected to recall in detail hours later) and then seeing signage, rather than using the berth-dependent broadcast function in GSM-R?

 

Apart from any other considerations there would still need to be equipment on the ground to show Drivers exactly where a TSR or ESR began and ended. And the signage should be positioned so that every Driver has time to reduce speed to the restriction, even if they were totally unaware of it before. However I do agree that in this particular, and as I mentioned above unusual, circumstance, a GSMR broadcast may well have been worth considering as a back-up to the other methods. 

 

[PaulRhB]

19 minutes ago, Northmoor said:

rather than using the berth-dependent broadcast function in GSM-R? 

Because it would need an acknowledgment, and therefore interaction with Signaller, for every train to confirm the broadcast was received so you may as well caution manually. Auto broadcasts also don’t work on first berths on a box area due to the disparity between TD step over being instant and the units GSMR switch to the new area being dependent on signal strength, not even digital technology can make that happen in a set place.

[PaulRhB]

28 minutes ago, SM42 said:

The challenge is how to overcome the erroneous conclusions that humans come to because of familiarity. We've all done it at some point in our lives.

 

The technology is in theory, available to help.

Hmmmm technology might not be the answer, I can think of simpler solutions to attract attention to a changed sign but that is hopefully what will come from this highlighting the actual problem with a RAIB report. This shows the advantage of these reports in identifying human behaviour traits. 

[iands]

Because it would need an acknowledgment, and therefore interaction with Signaller, for every train to confirm the broadcast was received ........

 

But by using the Berth-triggered broadcast calls function obviates the need to use the Cell-Broadcast option, thereby limiting the number of trains requiring to acknowledge the broadcast call, i.e. only the train(s) that occupy that specific berth whilst traversing the line concerned will receiver the call. To acknowledge the call the driver only has to depress the ST button once the broadcast call has ended (the driver is not required to set up a speech call with the signaller). This process allows trains to be cautioned without specially stopping at a red signal.

 

Auto broadcasts also don’t work on first berths on a box area due to the disparity between TD step over being instant and the units GSMR switch to the new area being dependent on signal strength, not even digital technology can make that happen in a set place.

 

All Berth-triggered broadcast call messages are pre-recorded by the signaller. The signaller will receive either a ‘failed’ or ‘not sent’ message if the call has not been received by the train radio. 

 

We should wait for the RAIB report for the full details (if the RAIB decide to go down that route) to see if the signal berth in this incident was in fact the first berth on the signalling area concerned.

[Northmoor]

Thanks @iands, you've confirmed my (what I accept is incomplete) understanding.

 

All, please understand I am in no way apportioning blame here (I am in no position to) or suggesting the GSM-R messaging is an alternative to existing methods; only that in the vast majority of scenarios, it should be an additional back-up to written notices and lineside signs.

[Edwin_m]

15 hours ago, Oldddudders said:

Baffled by all this. A suddenly imposed 20 mph speed on a 125 mph line? Why isn't every train being stopped and cautioned? And then hitting the dets as well? As for some freight operator complaining this aspect of safety was costing him £60k p.a., he should have been told it didn't matter if it was £60m! Do it or be out of railway operation!

According to the RAIB site it was a critical rail temperature issue in hot weather.  So probably a fairly remote chance of the rail buckling under the train, but if it had at 105mph it could have been very nasty.  

8 hours ago, caradoc said:

Had this particular restriction become a TSR, the later reduction in permitted speed would have made it revert to an ESR and the appropriate additional equipment would then have had to be re-installed, with trains cautioned until this was done.

They had to caution anyway until the boards were changed.  That might be quicker than putting the extra warning lights and magnet out to change it from TSR to ESR, but I suspect most of the time spent is getting someone out on the ground in the first place.  

[PaulRhB]

38 minutes ago, iands said:

But by using the Berth-triggered broadcast calls function obviates the need to use the Cell-Broadcast option, thereby limiting the number of trains requiring to acknowledge the broadcast call, i.e. only the train(s) that occupy that specific berth whilst traversing the line concerned will receiver the call. To acknowledge the call the driver only has to depress the ST button once the broadcast call has ended (the driver is not required to set up a speech call with the signaller). This process allows trains to be cautioned without specially stopping at a red signal.

It requires the interaction in the box to cancel the ST from the GSMR unit and then clear and replace the signal, so 3 interactions to the ST message

The boarded ESR requires no interactions and we are much busier with green zone line blocks these days so we don’t need distractions from those. 

 

38 minutes ago, iands said:

All Berth-triggered broadcast call messages are pre-recorded by the signaller.

I’m well aware as I do it regularly and it’s very useful for saving time as drivers call in straight away if they get it 

 

38 minutes ago, iands said:

The signaller will receive either a ‘failed’ or ‘not sent’ message if the call has not been received by the train radio. 

We sometimes receive that if they have got it! Also sometimes the driver gets a garbled message so has to call. 
 

38 minutes ago, iands said:

We should wait for the RAIB report for the full details (if the RAIB decide to go down that route) to see if the signal berth in this incident was in fact the first berth on the signalling area concerned.

I didn’t suggest that caused it but it does mean it cannot be used for all ESR’s, that was the point, it’s not a catch all solution. As I said in the next post the RAIB reports are good for identifying problems and I just expressed hope that the result will address it  

3 hours ago, PaulRhB said:

but that is hopefully what will come from this highlighting the actual problem with a RAIB report.

 

Edited November 11, 2020 by PaulRhB

[iands]

5 minutes ago, Northmoor said:

Thanks @iands, you've confirmed my (what I accept is incomplete) understanding.

 

All, please understand I am in no way apportioning blame here (I am in no position to) or suggesting the GSM-R messaging is an alternative to existing methods; only that in the vast majority of scenarios, it should be an additional back-up to written notices and lineside signs.

Having had 46+ years in the railway industry, I too have no desire to apportion any blame (I'll leave that to higher authorities if they deem it necessary). As @PaulRhB pointed out, GSMR is not 100% perfect, no system is, and it should be used to assist/complement other systems/procedures that are (or should be) in place.

 

The danger is, as others have alluded to, changing procedures that had been in place for years/decades (such as the use of detonators) for an alternative procedure that in theory should provide the same level of warning/safety, but in practice, for whatever reason(s), is found wanting and is less effective as the "procedure/system" it has replaced.

 

I understand the reasons for the authorities to keep staff safe and reduce/limit the need for them to go trackside and avoid potential contact with moving trains, and indeed to reduce as much as possible the use of detonators (dangerous things if you get on the wrong side of one!), but if there is a desire to replace a well practiced procedure/system on "health & safety" grounds, the new procedure/system must be proved to be as effective. It will be interesting to see how the industry (e.g. NR) demonstrate the effectiveness of any current procedures to the RAIB (and/or the ORR).

[Edwin_m]

50 minutes ago, iands said:

We should wait for the RAIB report for the full details (if the RAIB decide to go down that route) to see if the signal berth in this incident was in fact the first berth on the signalling area concerned.

There will be no report.  The website says they aren't taking it any further as any recommendations are covered by their report into the Sandy incident.  

[PaulRhB]

I like the reason given by the RAIB usually that their job is to investigate why not apportion blame. 
In cases like this I doubt there is blame as such but it has highlighted a human trait that can be looked at. 
Like iands says there’s a very difficult balance in safety of personnel on track and indeed their availability with the desire to minimise costs which is the politicians and users focus too. 

Edited November 11, 2020 by PaulRhB