Collision and derailment at Neville Hill Depot (13/11/2019)

[AndrewE]

I also remember the Mark 3 coaches being extensively praised after the Ladbroke Grove and/or Southall accidents, obviously compared to to earlier generations of rolling stock.

And then criticised after Ufton, particularly with regard to window resilience in a rollover event.

So comparing Pendolinos and Voyagers, are we persuaded that the smaller windows of the P are safer than the much deeper V ones?
I think that accidents are so rare (and Greyrigg or Ladbroke Grove/ Southall - type accidents so much rarer still) that I have no worries about travelling in either - or any other current rolling stock for that matter. Even being in a train of preserved wooden bodied coaches on the main line (if it was to be allowed nowadays) would still be safer than walking, cycling or even going somewhere by car.

 

[swt_passenger]

Why not info from LNER seeing as this appears to be affecting their services.

Do any of the TOCs publish updates on damage repair to their rolling stock?

 

[ainsworth74]

The report into this accident has now been published and can be found here. A summary can be found below:

Summary​

At 21:41 hrs on 13 November 2019, an empty LNER Intercity Express Train, approaching the maintenance depot at Neville Hill in Leeds, caught up and collided with the rear of a LNER High Speed Train moving into the depot. The leading train was travelling at around 5 mph (8 km/h) and the colliding train at around 15 mph (24 km/h). No one was injured in the accident, but the trailing bogie of the second and third vehicles and the trailing wheelset of the fourth vehicle of the Intercity Express Train derailed to the right, by up to 1.25 metres.

The collision occurred because the driver of the Intercity Express Train was focused on reinstating an on-board system which he had recently isolated, instead of focusing on the driving task. This was exacerbated by him unintentionally commanding too much acceleration due to his lack of familiarity with the train.

The driver had isolated the on-board system at Leeds station because he had been unable to correctly set up the train management system. He had been unable to do this because ambiguous documentation from Hitachi, the train manufacturer, had led to LNER misunderstanding the required process for setting up the train management system when developing the content of its driver training programme.

The driver’s lack of adequate familiarity with the train probably arose because LNER had not recognised that his training needs were greater than for his peers.

The derailment occurred because the design of the Intercity Express Train is susceptible to derailment in low speed collisions. This susceptibility is related to the use of high-strength couplers with large freedoms of movement in pitch and yaw. These features were part of the train’s design. However, the impact of these features on the train’s resistance to derailment and lateral displacement in low speed collisions, was not considered by the train’s designers.

The crashworthiness standard used to design the Intercity Express Train did not specifically require consideration of the likelihood of derailment during collisions at lower than the 22.5 mph (36 km/h) specified design speed, nor did it include specific criteria for assessing the derailment performance. As such, the assessment and validation of the design did not identify any issues with these design features.

Recommendations​

RAIB has made five recommendations. Two recommendations are addressed to LNER and relate to correcting its understanding of the setup of the train management system and ensuring that the documentation provided by Hitachi has not led to any other safety issues. The other recommendations relate to:

• Hitachi to revisit the assessment of the design of the Intercity Express Train against the requirements of the crashworthiness standard
• LNER to assess the risk of a derailment of an Intercity Express Train involved in a low speed collision
• RSSB to consider whether it is appropriate for the crashworthiness standard to be modified.

 

[ainsworth74]

Mod Note: For updates on 800109 itself please see the existing thread here. I would ask that posts solely relating to getting 800109 back into service or it's future disposition continue in that thread and this one focus on the accident itself and fallout of that. There will obviously be some overlap but I'll just ask we do our best

 

[Grumpy Git]

Wow, what a major, major blunder by the designers. If you are going to have a collision make sure it is above 22.5 mph!

 

[Domh245]

Wow, what a major, major blunder by the designers. If you are going to have a collision make sure it is above 22.5 mph!

The fault for that lays more at the RSSB/standard body's door I'd have said. The designers and manufacturers design, optimise, and build it to the specification given to them, including compulsory standards for use on the network. That the RSSB/whoever wrote the standard managed to miss this from the standard does make me wonder about their appreciation of risk - it would seem to be all focused on severity and not on likelihood (similar patterns come to mind with regards to electrification). I would think that very low speed impacts are an order of magnitude more likely than even low and higher speed impacts.

 

[LucyP]

The real fault lies in the over-complicated TMS on the 800 if you read the full report. The report also shows how these new trains are developing faults, so so soon after manufacture.

 

[irish_rail]

On a minor note good to see RAIB referring to it as an Intercity Express Train and not the nauseating moniker of Azuma!

 

[Swimbar]

The real fault lies in the over-complicated TMS on the 800 if you read the full report. The report also shows how these new trains are developing faults, so so soon after manufacture.

The report makes it plain that the driver is the cause of the accident, albeit with very strong mitigating circumstances.
The major issue, to me, identified in the report is that it makes it plain that we should not be surprised if an Azuma derails in a very low speed collision.
Thankfully the train is likely to remain upright.

 

[Taunton]

The fault for that lays more at the RSSB/standard body's door I'd have said. The designers and manufacturers design, optimise, and build it to the specification given to them, including compulsory standards for use on the network.

I think it's a more general issue, one that has also been found in aircraft design. Designers are expected to be competent by those who buy their products. The standards, as issued by the authorities, are a backstop to this process, to make sure that a certain minimum set of standards are always incorporated.

Over time, however, the compliance with standards overtakes the production of an overall competent design, and the designers start to feel it is all they have to do, instead of regarding them just as a compulsory starting point. If it's not in the standard it doesn't matter. Hence, because there was no standard for the TMS interface, it ended up (at least initially) as a drivers' "aid" being more complex and confusing than the procedures it is meant to be protecting. Likewise there were diesel engines whose design required preheating before starting, while a separate designer put together an automated system that started them without doing so.

 

[Grumpy Git]

Basically a brand new train has suffered very significant damage due to a collision at 10 mph. Somebody somewhere should be having sleepless nights.

 

[LucyP]

Yes, but if the driver had been able to set the TMS properly, which he couldn't because of Hitachi's unclear instructions, leading to LNER and the simulator provider not understanding it properly, and not therefore training/simulating it accurately. The driver had a problem in the same week with it, where it transferred automatically from electric to diesel power, because it had not been set correctly, and he was worried about the report to Hitachi of a cold engine start. Once again, it would not set properly, so he turned it off, so it wouldn't switch over and report him to Hitachi again, and was then so concerned to turn it back on again, he chose the wrong moment to do so, and was more concerned with that, than watching where he was going.

It's the modern problem of computers that are supposed to help, getting in the way and being so complicated and taking so many prods of a finger on the screen to do something, that your attention gets diverted. You must have driven cars like that, where you are looking at the screen for too long, just to alter the heating or change the radio station, that in the past you could do without even taking your eyes off the road, because you knew where the dial or button was.

The fear of "big brother" reporting you to Hitachi for their own train allowing itself to be started automatically when cold, is taking over from the real task of driving the train. It's why Airbus airliners crash. The pilots are so fixated on all the 100s of warnings coming up on the screen because one major thing has gone wrong, many of which are not relevant to the emergency, that they forget to fly the plane.

 

[hwl]

The fault for that lays more at the RSSB/standard body's door I'd have said. The designers and manufacturers design, optimise, and build it to the specification given to them, including compulsory standards for use on the network. That the RSSB/whoever wrote the standard managed to miss this from the standard does make me wonder about their appreciation of risk - it would seem to be all focused on severity and not on likelihood (similar patterns come to mind with regards to electrification). I would think that very low speed impacts are an order of magnitude more likely than even low and higher speed impacts.

Nothing to do with RSSB directly, it is a European standard and the US requirements are very similar so this is some what a global issue.
The gap in the standards was effectively covered by recommendations in an older RSSB "T" project.

Another one for the file of: Complies with TSI not all UK inc. best practice

 

[GRALISTAIR]

The major issue, to me, identified in the report is that it makes it plain that we should not be surprised if an Azuma derails in a very low speed collision.
Thankfully the train is likely to remain upright.

Basically a brand new train has suffered very significant damage due to a collision at 10 mph. Somebody somewhere should be having sleepless nights.

very worrying to me. Perhaps not as bad as the 737Max situation but I think an Informed Sources law may apply here.

 

[Bungle]

It's why Airbus airliners crash. The pilots are so fixated on all the 100s of warnings coming up on the screen because one major thing has gone wrong, many of which are not relevant to the emergency, that they forget to fly the plane.

I take it from this comment that you’ve never flown an Airbus...

The wider point about distraction from critical tasks is entirely valid and appears to have been a major factor here, but that applies to any form of transport.

 

[Domh245]

Nothing to do with RSSB directly, it is a European standard and the US requirements are very similar so this is some what a global issue.
The gap in the standards was effectively covered by recommendations in an older RSSB "T" project.

Another one for the file of: Complies with TSI not all UK inc. best practice

Is it not within the power of the RSSB to issue "supplementary" standards to these though, that could have required a fifth test for a low speed collision like this?

 

[GRALISTAIR]

The wider point about distraction from critical tasks is entirely valid and appears to have been a major factor here, but that applies to any form of transport.

I do agree. This is a fair point. I get very distracted driving a modern car which should not be the case. I don’t think it fair to 100% blame the driver.

 

[ainsworth74]

It almost reads to me like the driver had been set up to fail to be honest.

 

[Crossover]

It almost reads to me like the driver had been set up to fail to be honest.

I feel this is a good summarisation to be honest

 

[Grumpy Git]

As someone who installs and trains people on computer based systems via touch-screen interfaces, I suggest the biggest failing of all systems is that the programmers have never been in the shoes of the people who have to use the interface on a day to day basis.

The last company I worked at, brought in two generations of the same equipment over a five year period. At no point did they ever consult us (the guys installing the kit and training operators) on what we might suggest would be an improvement.

 

[falcon]

It almost reads to me like the driver had been set up to fail to be honest.

The driver is supposed to look wher he is going and control the speed of the train.

 

[ainsworth74]

The driver is supposed to look wher he is going and control the speed of the train.

Not sure anyone is disputing that? But that's a touch simplistic considering all the contextual information as to why they ended up not doing that.

 

[Royston Vasey]

The report makes it plain that the driver is the cause of the accident, albeit with very strong mitigating circumstances.
The major issue, to me, identified in the report is that it makes it plain that we should not be surprised if an Azuma derails in a very low speed collision.
Thankfully the train is likely to remain upright.

Basically a brand new train has suffered very significant damage due to a collision at 10 mph. Somebody somewhere should be having sleepless nights.

Yes the damage to the train is less relevant compared to the likelihood of it exceeding the normal kinematic envelope, fouling an adjacent line and a secondary collision being much worse. Imagine this had happened on the ECML slows immediately prior to another train approaching on the adjacent fast line, for example.

 

[BrettSy96]

Sorry if I have missed this information or if I’m asking in the wrong thread but what happened to the HST powercar? Has it been Scrapped?

 

[Rhinojerry]

As someone who installs and trains people on computer based systems via touch-screen interfaces, I suggest the biggest failing of all systems is that the programmers have never been in the shoes of the people who have to use the interface on a day to day basis.

The last company I worked at, brought in two generations of the same equipment over a five year period. At no point did they ever consult us (the guys installing the kit and training operators) on what we might suggest would be an improvement.

Nail on the head..

 

[Swimbar]

Sorry if I have missed this information or if I’m asking in the wrong thread but what happened to the HST powercar? Has it been Scrapped?

It's been stripped for spares at Neville Hill

 

[Scotrail314209]

To me I find it concerning that the Azuma came off and sustained quite severe damage in that low speed collision.

It makes me worry about how an IET would hold up in a derailment similar to Carmont..

 

[ainsworth74]

To me I find it concerning that the Azuma came off and sustained quite severe damage in that low speed collision.

It makes me worry about how an IET would hold up in a derailment similar to Carmont..

Considerably better than the HST did. The IET has been rendered extremely hard to restore to operational condition but there's nothing in the report that suggests it wasn't crashworthy (it doesn't sound like the driver even had to get medical attention and he was literally at the sharp end).

Indeed it rather reads like this was the worst case scenario. Due to design decisions made to meet the requirements of the both the IEP specification and the standards around crashworthiness it appears that we ended up with couplers that have a higher than usual strength to resist collapsing and they allow for more pitch and yaw due to, again, the requirements of the IEP specification. So in this scenario where the couplers do not deform because not enough energy has been generated by the collision and so they can move to a greater degree than might be expected you end up with a more significant chance of derailment. The RAIB said (paragraph 112):

Overall, the risk of derailment appeared to be greatest at low speeds: very low speeds do not generate enough energy to lead to a derailment and high speeds create enough energy to collapse the inter- vehicle couplers which appear to reduce the risk of derailment.

Whilst the front of thing might look spectacular it doesn't follow that the trains are unsafe in collisions (again, driver did not appear to need medical attention). But they are more prone to derailment in just the right conditions (fast enough to have enough energy to cause a derailment, too slow to have enough energy to collapse the couplers and help keep the train on the rails). I also didn't spot the RAIB offering any comment on the damage to the train itself which I would have expected them to do so if it hadn't have stood up to the damage as well as expected. At Carmont, however, the speed of the train would have been sufficient for all of the various safety features in the design to kick in and work as intended. Certainly having read the report I would, if I had to be in accident as catastrophic as Carmont, rather be in an IET than a HST.

 

[43096]

Not sure anyone is disputing that? But that's a touch simplistic considering all the contextual information as to why they ended up not doing that.

But from the report it states the driver stopped - briefly - and then started moving again, but for whatever reason thought that was a good time to fiddle with the APCO. Given that he knows he has the HST ahead of him, that’s pretty dumb. And even dumber knowing that he’s on performance management as a result of two SPADs.

So fundamentally, it’s driver error (bordering on stupidity).

 

[ainsworth74]

But from the report it states the driver stopped - briefly - and then started moving again, but for whatever reason thought that was a good time to fiddle with the APCO. Given that he knows he has the HST ahead of him, that’s pretty dumb. And even dumber knowing that he’s on performance management as a result of two SPADs.

So fundamentally, it’s driver error (bordering on stupidity).

Yes, I don't think anyone has disputed that? I'm simply trying to point out that there are then a lot of things behind that! Of course he should have been paying attention to where he was going but the report spends a lot of time looking at all the things that may well have played a part in why they didn't do that. Which is surely worthwhile rather than just saying (in perhaps not so many words) "drivers an idiot". Surely a lot of accident reports could boil down to "someone was an idiot" but if we don't look at the factors that lead to them to make a stupid mistake we'd end up having a lot more stupid mistakes!

For instance, it strikes me that LNERs training and competency management made some pretty serious oversights regarding this driver. Should have been given the same training as his colleagues considering what had happened to him over the previous two years? Or should he actually have had a more significant training course perhaps with more supervised driving? If had then the accident may not happen as he'd be more familiar with the performance of the IET and the foibles of its TMS. If Hitachi had made it clearer in their training materials to LNER that you need to press "Check Stops" to get a headcode to be accepted by the TMS not just "Home" he wouldn't have been worrying about the TMS at all.

Should the driver have been paying attention the road ahead? Yes, 100% of course he should that lapse of judgement is incredible. But there's an awful lot of things going on behind the scenes that played a part in how he ended up having that awful lapse.