Carmont (near Stonehaven) derailment - 12 August 2020

[Taunton]

If you read the RAIB report, paragraph 183 makes it clear that there is no record of any check being done after the work was complete.

That's correct, reading between the lines it seems that NR had contracted out their client side surveying function, to Arup, who would normally have done the post-installation check but didn't get an instruction so didn't do it. That in itself is pretty poor, as their instruction for the project should include the whole thing, not little bits given out separately and subject to being overlooked or budget cut. The same paragraph also states that Carillion, the contractor, could have done it, but that is worse, not only because they had no instruction to do so but you just don't get people to "check" their own work. All up, I do wonder who finally signed off the work so that Carillion got paid for it.

 

[Master29]

Can’t say I’m hugely surprised. Sounds like a lot of mouth music and hot air by various parties involved. It’s time for these parties to get serious and start doing what they’ve said they would.

Can’t find the link but also the ORR publicly said last week that they’ve found no tangible evidence of any remedial action being taken following the RAIB recommendations into the crash.

Sounds like typical modus operandi where quangos are concerned.

 

[najaB]

All up, I do wonder who finally signed off the work so that Carillion got paid for it

That's a key question, as answering it could lead to rather uncomfortable conclusion that contractors were being paid for work that may or may not have actually been done.

 

[Davester50]

but you just don't get people to "check" their own work.

You'd think that, but Boeing were self-certifying 787s until recently!

 

[najaB]

You'd think that, but Boeing were self-certifying 787s until recently!

To be fair, there is a bit of difference between certifying units off a production line and one-off engineering/construction projects.

The former should be well within the competency of an organisation like Boeing - and was for several decades.

 

[Davester50]

To be fair, there is a bit of difference between certifying units off a production line and one-off engineering/construction projects.

The former should be well within the competency of an organisation like Boeing - and was for several decades.

There's a big engineering difference between a drain at, err, Zero feet, and an aircraft at 40,000.

 

[najaB]

There's a big engineering difference between a drain at, err, Zero feet, and an aircraft at 40,000.

True. But when you're making hundreds of the same thing there should be less scope for mistakes to creep in (eg. measurements on plans read wrong) and more chance to catch them if they do - "Why doesn't this one look like the others?"

 

[Davester50]

True. But when you're making hundreds of the same thing there should be less scope for mistakes to creep in (eg. measurements on plans read wrong) and more chance to catch them if they do - "Why doesn't this one look like the others?"

You're comparing a complex, custom designed (each lessor or airline specifying a near unique fit out) aircraft with a drain?
Okay then.

 

[najaB]

You're comparing a complex, custom designed (each lessor or airline specifying a near unique fit out) aircraft with a drain?
Okay then.

Boeing isn't responsible for cerifying interiors.

And this "just a drain" failed and killed six people. If not for COVID travel restrictions it could easily have been dozens.

 

[Taunton]

Furthermore, my experience with Carillion (who we can mention openly as they are no longer around), at the time I think the No 2 construction/civils contractor in the UK by turnover, was that they actually did not employ the operatives in many instances, but subcontracted it out. It is seen in the report where they subcontracted the design of the works out to Aecom, having taken responsibility for the design but not having a design team, I wonder if they subcontracted the actual works out.

Regarding the aircraft manufacture, an airline ordering one sends a full time engineering representative to the factory, who goes in every day and checks it's all being done in accordance with the agreed build, the right parts, the right checks, etc. Construction works of the Carmont scale may not have this (HS2 does), but there are the periodic visits, checks, approvals, and sign-offs. And the prime documentation trail for this is with the client (Network Rail), not the contractor. Just like you having your car MOT'd - it's you responsible to keep the paperwork, not the garage.

 

[Davester50]

Boeing isn't responsible for cerifying interiors.

And this "just a drain" failed and killed six people. If not for COVID travel restrictions it could easily have been dozens.

There's more to custom manufacturing than just interiors, (even the retrofit of an interior caused the loss of an aircraft).
The point, before going further off topic is an agreement with @Taunton that complex work shouldn't be signed off by the same person doing the work.
The Boeing example I gave was to agree with him, as self-assurance has become near worthless in some industries.

 

[najaB]

The point, before going further off topic is an agreement with @Taunton that complex work shouldn't be signed off by the same person doing the work. The Boeing example I gave was to agree with him, as self-assurance hase become so

I get that. The point I was making was in agreement that civil engineering work should definitely not be self-certified, but when it comes to mass-produced items it isn't feasible, nor should it be necessary, for a third-party to assess each item off the production line.

The design, yes. The first few items, definitely. But each and every one isn't going to work. They were aiming to roll a 787 off the line every two days - an engineering team big enough to quality check that many planes (plus a 737 every day) was going to be almost as big as Boeing's assembly team.

 

[Davester50]

The design, yes. The first few items, definitely. But each and every one isn't going to work. They were aiming to roll a 787 off the line every two days - an engineering team big enough to quality check that many planes (plus a 737 every day) was going to be almost as big as Boeing's assembly team.

Well, better call the FAA then, and tell them they're wrong to take back control then.

 

[Falcon1200]

And this "just a drain" failed and killed six people.

Not to diminish in any way what was a dreadful accident, but in the interests of accuracy it was actually three people killed at Carmont.

 

[Taunton]

The point, before going further off topic is an agreement with @Taunton that complex work shouldn't be signed off by the same person doing the work.
The Boeing example I gave was to agree with him, as self-assurance has become near worthless in some industries.

There are very well developed procedures in civils and building works to sign off completion, the ultimate sanction of course being that you don't get paid. Experience of self-signoff is that it depends very much on the ability of the organisation, but it certainly doesn't go well with a contractor selected on the basis of being bottom bidder, or one which is inherently focused on maximising revenue and minimising costs, passing control of these from the top management down through various levels of staff whose only assessment is these financial goals on a spreadsheet.

Which is where Carillion, and the comments of their Receiver and various others more recently involved, comes in ...

 

[najaB]

Not to diminish in any way what was a dreadful accident, but in the interests of accuracy it was actually three people killed at Carmont.

Sorry, yes. I was thinking of six people on the train.

Well, better call the FAA then, and tell them they're wrong to take back control then.

The FAA only took back control because they had to. They said the reason that they trusted Boeing to self-certify is specifically because Boeing had the necessary resources to do it and they didn't.

The fact that FAA is in control is a major factor in why Boeing's projected production rates are so much lower now than they were.

 

[ainsworth74]

Another update regarding this accident:

Network Rail is to face court action after an Aberdeenshire train crash which claimed three lives.
Driver Brett McCullough, 45, conductor Donald Dinnie, 58, and passenger Christopher Stuchbury, 62, died when the train derailed at Carmont on 12 August 2020.
The train hit a landslide after heavy rain.
Network Rail is due to face criminal action at the High Court in Aberdeen on 7 September.
The court roll, which is published by the Scottish Courts and Tribunals Service, states the prosecution will be under a section 76 indictment.
This procedure suggests that a guilty plea may be offered.

'Fundamental lessons'​

Network Rail said: "The Carmont derailment and the tragic loss of Christopher Stuchbury, Donald Dinnie and Brett McCullough was a terrible day for our railway and our thoughts remain with their families and all those affected by the accident.
"While we cannot comment on the ongoing legal process, the Rail Accident Investigation Branch report into Carmont made clear that there were fundamental lessons to be learnt by Network Rail and we have supported the investigation process."
Six other people were injured when the 06:38 Aberdeen to Glasgow service hit washed-out debris at Carmont, south of Stonehaven.
The train was returning to Aberdeen due to the railway being blocked further down the line.

Stonehaven crash: Network Rail to face fatal derailment charges

Three people died when a train hit a landslide after heavy rain near Stonehaven in 2020.

www.bbc.co.uk

Hat tip to @brad465

 

[brad465]

By coincidence I passed through the area to and from Aberdeen this weekend, wondering now if I'm cursed.

 

[BRX]

Another update regarding this accident:

Stonehaven crash: Network Rail to face fatal derailment charges

Three people died when a train hit a landslide after heavy rain near Stonehaven in 2020.

www.bbc.co.uk

Hat tip to @brad465

I wonder if any legal consequences will come about for anyone who was involved at Carillion. Of course, Network Rail had a duty to make sure work carried out was done properly but so did Carillion. Do the people involved there escape responsibilty simply because the company has disappeared as an entity (whereas NR hasn't)?

 

[Horizon22]

I wonder if any legal consequences will come about for anyone who was involved at Carillion. Of course, Network Rail had a duty to make sure work carried out was done properly but so did Carillion. Do the people involved there escape responsibilty simply because the company has disappeared as an entity (whereas NR hasn't)?

This is going to be a classic "responsibility / accountability issue". Whilst NR may have contracted it out, they still remain accountable for the work, even if Carillion were responsible.

 

[yoyothehobo]

This is going to be a classic "responsibility / accountability issue". Whilst NR may have contracted it out, they still remain accountable for the work, even if Carillion were responsible.

Its a strange one as Carillion in not building what they were supposed to build and implementing something else, essentially became designers of the work under CDM regulations.

 

[najaB]

Its a strange one as Carillion in not building what they were supposed to build and implementing something else, essentially became designers of the work under CDM regulations.

As I understand it, they were required to inform NR of the changes to design. The fact that they didn't would at least partially absolve NR of liability. Especially if the changes weren't obvious once the work was completed.

 

[Peter Mugridge]

As I understand it, they were required to inform NR of the changes to design. The fact that they didn't would at least partially absolve NR of liability. Especially if the changes weren't obvious once the work was completed.

Carillion used a third party back office / administrative subcontractor for a lot of their internal functions.

That company still exists - would it have any potential liability for failing to complete paperwork and communications correctly?

 

[matchmaker]

Apparently the court hearing is a so-called "Section 76" Indictment, which means that a guilty plea is to be tendered. It is quite likely that the form of the plea will have agreed between NR and Crown Office.

 

[Taunton]

I've made my comments previously in this thread, no need to repeat, but feel any prosecution, like the RAIB report, is heading in the wrong direction.

If Carillion did it wrong that was for professional Network Rail, and their professional engineers, to identify. If my plumber put in a new water tank in my loft in 2012 and it fell through the ceiling today, in 2023, that is by now my responsibility, not the original plumbers.

I still have no answer to why the train was allowed to return at full line speed, to be wrecked at 75mph, when it was known that in the conditions there had been multiple landslips all around and multiple trains were stranded up and down the line by this.

The HST power car did absorb by far the brunt of the smash, how fortunate it was not a dmu front car with passengers that was tipped right down the embankment.

The HST power car which struck head-on the almost-new 165 at Ladbroke Grove came out of it far less destroyed than that more modern train, which was shredded down to the underframe. Look at the accident pictures. Why was the railway not prosecuted for the 165 design afterwards?

 

[ainsworth74]

I still have no answer to why the train was allowed to return at full line speed, to be wrecked at 75mph, when it was known that in the conditions there had been multiple landslips all around and multiple trains were stranded up and down the line by this.

As per the RAIB:

230. Although aware of these obstructions on either side of train 1T08, route control staff did not take any action to assess the risk to the operation of 1T08 on its return to Stonehaven and took no action to run the train at a reduced speed. With the exception of Network Rail’s stranded train risk assessment process (see paragraph 232), there was no process requiring route control staff to assess risks to specific trains such as 1T08 beyond general situation awareness possessed by individuals within the control function. The resources available to control (see paragraph 236) and the volume of incidents which were being managed meant that such reliance on individual situational awareness did not assure the safety of individual train operations.

231. The apparent lack of awareness about weather-related risk to train 1T08 contrasts with action taken when train 1A43 was reversed at Laurencekirk station because flooding at Newtonhill meant it could not continue its scheduled northward journey. Train 1A43 had arrived at Laurencekirk station at about 07:16 hrs and remained there until, at 08:28 hrs, route control staff instructed the Laurencekirk signaller to return the train to Dundee as train 1Z43 (paragraph 52). When giving this instruction, an informal exchange between route control and the Laurencekirk signaller identified that the train had been at Laurencekirk for over an hour, and led control to instruct the signaller that he should advise the driver of 1Z43 to run at a reduced speed. The instruction was given, but there was no opportunity for the driver to implement it as he stopped the train shortly after leaving the station because he encountered a landslip (paragraph 53).

236. Route control is routinely resourced (paragraph 196) to deal with the problems likely to arise as a normal consequence of railway operation. Typically, these will include equipment failures, staffing issues, and the problems associated with adverse weather that is normally encountered. This resource can be insufficient to deal with exceptional events. Evidence that route control staff were overloaded on the morning of 12 August 2020 is provided by witness evidence, the number of events listed in appendix D, the severity of some of these events (for example, the canal breach at Polmont) and omission of the stranded risk assessment for train 1T08.

237. It was possible for further resources to be brought into control to deal with abnormal circumstances. One option was an additional member of staff brought on duty to operate a specific ‘weather desk’. This individual would provide additional support to other control staff managing weather-related incidents. This relied on a competent individual being available and willing to work overtime. A ‘weather desk’ had been introduced in response to adverse weather in Scotland in June 2020 (paragraph 215). However, no ‘weather desk’ was operated on the night of 11/12 August 2020 and there is no evidence suggesting that serious consideration was given to seeking volunteers to staff this.

238. An alternative source of additional resource is the implementation of senior management incident control, commonly referred to as ‘gold command’, and described in NR/L2/OPS/250, ‘Network Rail National Emergency Plan’. This deploys a cadre of senior managers who can be tasked with managing a specific problem or incident (such as an adverse weather event). Although Network Rail procedures describe this implementation as being decided by an EWAT meeting, the RCMs within the Scotland route control sometimes did so without an EWAT meeting.

239. During the night of 11/12 August, the night shift RCM considered implementing the gold command structure to better manage the weather issues. However, despite the level of disruption already known about and the forecast of further extreme weather received at 02:51 hrs, he did not do so because he judged that, by the time the necessary staff had been mobilised, the need for the additional support would have receded. Had a gold command structure been implemented earlier, it is possible that the railway’s ability to respond more effectively would have increased.

240. The following morning, before the accident at Carmont occurred, the day shift RCM, in conjunction with the head of integrated control, decided to implement ‘gold command’ to manage the recovery from the weather issues. However, the time taken to implement this decision meant that this did not become operational until 10:12 hrs, shortly after the accident happened. A second separate command structure was then created to manage the issues related to the accident.

241. It is possible that, had additional resources been available at route control earlier on August 12, consideration of damage elsewhere in Scotland, including the four events in the Laurencekirk/Newtonhill area, would have led to recognition of the potential threat to train 1T08 from the extreme weather in the vicinity of the train. It is possible that this would have led to actions being taken to mitigate that threat.

253. The Carmont signaller had no indication that the line was obstructed between his location and Stonehaven when authorising the movement of train 1T08 to Stonehaven. After passing his signal box, train 2B13 had passed the location of the washout at 07:07 hrs and reached Stonehaven station at 07:13 hrs, with its driver subsequently confirming that he had seen no indication of a problem (paragraph 50). The Carmont signaller was told that this train had reached Stonehaven during a phone call with the Stonehaven signaller at 07:19 hrs, during which they concluded there was no known obstruction of the up line, so this could be used for the northbound movement of train 1T08 if necessary (the movement was actually made on the down line). Neither signaller had received any indication of an obstruction at the washout location when, at and after 09:08 hrs, they exchanged the messages needed to authorise the northward movement of train 1T08 to Stonehaven.

254. By the time the Carmont signaller authorised this movement at 09:28 hrs, the weather had improved significantly (paragraph 57). Evidence that the driver of train 2B13 had not seen a problem when passing the washout location at 07:07 hrs indicates that the driver of train 1T08 would have seen no evidence of a problem when passing the site in the southbound direction eight minutes earlier at 06:59 hrs. RAIB has concluded that, when train 1T08 headed north towards the washout, there was no rule in place that required the train to proceed any slower than the normal maximum permitted speed of 75 mph (121 km/h) at that location.

255. When the signaller gave the driver of train 1T08 authority to travel ‘wrong direction’ (in the down direction along the up line) as far as the crossover, he also advised the driver that, due to the power failure affecting his signal box, he might have difficulty in clearing the signal permitting train 1T08 to travel along the down line to Stonehaven. In the event, he was able to clear the signal normally. During this conversation, the driver queried whether there was any speed restriction to Stonehaven; the signaller replied that the line was fine between Carmont and Stonehaven, and that the driver could proceed at normal speed. In response the driver said that he would be in no rush to get there.

It would seem that the RAIB have addressed why and the answer appears to a route control overloaded by issues that morning and something slipping past them and a driver and signaller who were not required to proceed at caution and, with the information available to them locally, not aware that there may be any reason to suspect a problem.

Why was the railway not prosecuted for the 165 design afterwards?

Presumably because the 165 design met the applicable standards and requirements for a railway vehicle and had been properly constructed to those requirements. Unlike the drainage arrangements at Carmont which were not completed to the applicable standard and Network Rail failed to identify as having failed to do so over many years subsequent. If it had turned out that BREL had failed to build the 165 to the agreed design then it may well have been the case that prosecutions followed.

If it turned out that when the 165 had been

 

[yoyothehobo]

If Carillion did it wrong that was for professional Network Rail, and their professional engineers, to identify. If my plumber put in a new water tank in my loft in 2012 and it fell through the ceiling today, in 2023, that is by now my responsibility, not the original plumbers.

I dont agree with this. Carillion were contracted to build what had been designed are were provided with the specification for which to build it (a design that incidentally would not have caused the washout that caused the crash). Carillion neglected their contractual obligations to build what they had been contracted to, for reasons that are unknown. In changing the design they become defacto designers on the project.

The design should have been checked following construction however, and when found not to be compliant, Carillion should have been forced to rebuild it all at their own expense.

 

[35B]

I dont agree with this. Carillion were contracted to build what had been designed are were provided with the specification for which to build it (a design that incidentally would not have caused the washout that caused the crash). Carillion neglected their contractual obligations to build what they had been contracted to, for reasons that are unknown. In changing the design they become defacto designers on the project.

The design should have been checked following construction however, and when found not to be compliant, Carillion should have been forced to rebuild it all at their own expense.

And in the failure to do that, we see the accountability come back to the client - Network Rail.

At present, as we don't know whether the charges relate to the drainage works or to the operational decisions on the day, it's premature to draw conclusions about whether this prosecution is or is not reasonable. A point that goes double when we consider that the legislation involved is Scottish, and therefore may differ in important particulars from the English legal concepts that we are more familiar with.

The HST power car which struck head-on the almost-new 165 at Ladbroke Grove came out of it far less destroyed than that more modern train, which was shredded down to the underframe. Look at the accident pictures. Why was the railway not prosecuted for the 165 design afterwards?

If the 165 design met the applicable legal requirements, and the unit was in good physical order, what would there have been to prosecute for?

Again, we need to come back to the question of what charge(s) Network Rail are facing before leaping to conclusions about whether the prosecution is or is not appropriate.

 

[Dieseldriver]

I still have no answer to why the train was allowed to return at full line speed, to be wrecked at 75mph, when it was known that in the conditions there had been multiple landslips all around and multiple trains were stranded up and down the line by this.

Hindsight much? In the time 1T08 waited to crossover to return north the weather cleared substantially. 1T08 had already traversed that section of line in the southbound direction with nothing amiss and a train had travelled north through the incident site prior to the derailment with nothing amiss. It’s a cruel situation but as a Driver myself, prior to the Stonehaven incident I am sure that I also would have felt comfortable to reach linespeed.
If I were in that exact scenario now, post Stonehaven, there would definitely be a major red flag that would influence my decision making. This is knowledge that no one on the day was privy to.

 

[bramling]

And in the failure to do that, we see the accountability come back to the client - Network Rail.

At present, as we don't know whether the charges relate to the drainage works or to the operational decisions on the day, it's premature to draw conclusions about whether this prosecution is or is not reasonable. A point that goes double when we consider that the legislation involved is Scottish, and therefore may differ in important particulars from the English legal concepts that we are more familiar with.

If the 165 design met the applicable legal requirements, and the unit was in good physical order, what would there have been to prosecute for?

Again, we need to come back to the question of what charge(s) Network Rail are facing before leaping to conclusions about whether the prosecution is or is not appropriate.

I don’t think comparisons to Ladbroke Grove are that useful. The 165 bodyshell was subject to absolutely massive forces, not helped by the HST power car being of traditional heavy underframe construction. The 165 was never going to come out well. Indeed it is probably to its credit that cars 2 and 3 came out of it generally intact apart from the leading end of car 2. ISTR it was highlighted that the welds were of comparatively poor quality though. Was it not the case that the leading DVT at Great Heck was essentially destroyed as well? That is the only comparable recent collision, thankfully. (I am ignoring Cowden in this as the forces there were several orders of magnitude less, and in that accident both trains were underframe stock).