Drainage installation fault led to Carmont HST disaster
The Rail Accident Investigation Branch’s final report details the series of events that led to derailment of a ScotRail HST and the loss of three lives.
THE failure of a drainage system that had not been built to its design specification was the primary cause of the Carmont derailment on August 12, 2020, according to the report by the Rail Accident Investigation Branch.
Three people lost their lives in the accident – driver Brett McCullough, conductor Donald Dinnie, and passenger Christopher Stuchbury – and six other passengers were injured, including an off-duty Scotrail conductor.
The Scotrail HST was travelling at 73mph as it returned to Stonehaven after its intended path south was blocked by a landslip. It was derailed after hitting debris washed out of a drainage trench. The leading power car hit a bridge parapet before plunging down an embankment, with the trailer cars jack-knifing across both tracks.
Because the accident occurred during the Covid-19 pandemic, fewer people were travelling. RAIB estimated that the number of passengers on the train in normal times would have been between 25 and 50, and therefore there would have been a higher casualty figure.
Chain of events
The final report from RAIB about the accident some three miles south of Stonehaven says in the hours prior, the area suffered from near-continuous heavy rain, with nearly 52mm falling in three hours, which was close to the area’s average rainfall for the whole of August. This level of rain, RAIB says, is likely to occur every 100 to 144 years.
Train 1T08, the 06.38 Aberdeen to Glasgow
Queen Street, was a four-car HST (formed south-tonorth as power car No. 43030, Mk.3 coaches A/B/C/D Nos. 40622, 42007, 42564, 42145, and power car
No. 43140) but, because of the adverse weather, it was expected to terminate at Dundee.
Further south, there had been more than 20 other weatherrelated infrastructure problems, including the breach of a canal that flooded the EdinburghGlasgow main line. These incidents placed the operational control team under pressure. RAIB says “despite the severe nature of the disruption to Scotland’s railway infrastructure, no additional resource had been obtained for the control room, and the senior management ‘gold command’ structure had not been established to relieve the pressure on the controllers.”
The report further comments that controllers in Scotland, and elsewhere, had not been given sufficient guidance or training to enable them to effectively manage complex situations of the type encountered.
South of Carmont signal box, 1T08 was halted by an emergency call regarding a landslip ahead (reported by a northbound train), the HST halting 570 metres before the debris. After returning to Carmont wrong road and waiting for the Network Rail mobile operations manager to clip the points, 1T08 set off towards Stonehaven right road and was near line speed when it struck the debris washed out from a drain to a depth of between 135-170mm.
Faulty drainage
The RAIB report runs to 300 pages and goes into forensic detail regarding the ‘French drain’ that was installed in 2011/12 as part of a wider scheme to address a known problem with the stability of the earthworks in this locality. The drain comprised a 450mm diameter perforated pipe buried in a gravel-filled trench that ran for 306 metres along the edge of a field at the top of a slope running down to the railway. The drain then sloped down relatively steeply to track level. Catchpits (access chambers) were provided at intervals for inspection and maintenance. In its report, RAIB says: “The drainage system was not installed according to the design drawings and a low bund (artificial ridge) was constructed which was not part of the design. Consequently, on the morning of August 12, 2020, surface water flows were concentrated into a short length of the gravel-filled trench, which resulted in gravel and other stony material being washed out of the drainage trench and surrounding area. “The trench contained gravel (mainly between 20mm and 40mm in size) in accordance with normal practice for French drains. However, the use of this gravel in such a steeply sloping trench increased the likelihood of it being washed away should the water reach the drain as a concentrated flow.” The drain was constructed by Carillion, which failed to construct it to the designer’s requirements. Consequently, the drainage system was unable to perform as the designer had intended when it was exposed to particularly heavy rainfall. RAIB points to the construction of the bund as another factor in the derailment. The report says: “This bund, which was constructed outside Network Rail’s land, had the effect of diverting a large amount of water into a gully so that it all reached the drain at the same location, thereby increasing the propensity for washout of the gravel infill. RAIB found no evidence that the construction of the bund was notified to Network Rail or the designer.” RAIB also found other significant differences between the designed system and what was installed, nor that the changes were referred to the designer.
The investigation also discovered another failure in the management process: “Network Rail’s project team were probably unaware that the 2011/12 drain was significantly different from that intended by the designer and therefore did not take action. Had they been aware of this, it is possible that the consequent risk would have been recognised and remedial actions taken.
“Although Network Rail had a project team, they were not required by Network Rail business processes to check that the drain was being installed in accordance with the design. They therefore relied on a contractual assurance process that required Carillion to refer proposed changes to the designer, Arup, for approval.”
Previous warning
The report points out that in December 2012, after heavy rain, the landowner took a photograph showing water flowing from a side channel and slight erosion to the gravel surface. This image was passed to Carillion or Network Rail, but no evidence has been found as to receipt or any action. Further erosion was likely evident during a 2013 inspection, with RAIB commenting: “It is very unlikely that the slight erosion of the gravel surface would have been immediately recognised as a precursor to a sudden washout affecting railway safety.” It concludes: “This was clear evidence of a problem requiring action such as repair, monitoring, and/or reference to the drain designer. This
“RAIB considers it more likely than not that the outcome would have been better if the train had been compliant with modern crashworthiness standards”
was a missed opportunity to recognise the effect of the bund on water flows.”
After the train returned to Carmont, there was a conversation between the driver of 1T08 and the signaller at Carmont, the signaller stating that the line was ‘fine’ and that the driver could proceed at normal speed – no one being aware that the line was obstructed.
At the time, there was no written process that required train 1T08 to be instructed to run at a lower speed on its journey between Carmont and Stonehaven following an intense rainfall event, and no such instruction was given by route control or the signaller.
HST safety issues
RAIB undertook a detailed study on the crashworthiness of the train, which dates from the mid-1970s. It determined the fatal injuries were caused by secondary impact of the driver with the cab windscreen; loss of survival space in the vestibule of the first carriage (the report covers age-related corrosion of the trailer cars and how the carriages deformed); and ejection of the passenger through an open gangway. The first two vehicles underwent extreme movements and rolled over onto their roofs before coming to rest, subjecting the occupants to violent vertical, lateral, and longitudinal movements. Of the 61 windows in the train, 22 were broken or failed to the subsequent fire that occurred. RAIB notes there were many large shards of glass in the leading coach that caused lacerations.
When the carriages were refurbished in 2019, contractor Wabtec noted localised corrosion and carried out authorised reports. However, RAIB says there was no formal criteria for judging the tolerability of the corrosion and that repairs were based on engineering judagment, nor were there photographic records of the work actually done.
In the severe impact, the driving cab became detached, RAIB noting that the “impact conditions were significantly beyond those in which even modern cabs are designed to provide.”The leading power car’s fuel tank ruptured, but RAIB did not investigate what caused the fire.
Trains built since the adaptation of a revised group standard in 1994 would have had a number of design features intended to provide better protection for occupants and keep vehicles in line should they collide with an obstacle or derail. As the HST predates these standards, the report adds: “While it is not possible to be certain about what would have happened in the hypothetical situation with different rolling stock in the same accident, RAIB considers it more likely than not that the outcome would have been better if the train had been compliant with modern crashworthiness standards.”
The report has resulted in 20 recommendations that have implications across the rail industry (see panel).
These recommendations are passed over to the Office of Rail and Road to progress and report back to RAIB.
Industry reactions
Scotland’s minister for transport Jenny Gilruth said: “While the RAIB report is very clear there was nothing about the way the train was driven which caused the accident, the primary cause – that Network Rail’s contractor did not construct the drainage system correctly – will add to families’ pain.
“We must seek to learn the lessons from this incident to improve further the safety of all who work and travel on the railways of Scotland.”
Network Rail chief executive Andrew Haines admitted there are “fundamental lessons to be learnt by Network Rail and the wider industry”. He added that similar locations around the country have been inspected since and, post-accident, the company set up a taskforce to understand the impact of rainfall, but also to better manage at risk infrastructure.
RMT general secretary Mick Lynch said: “It is shocking that although the failure of the sub-contractor Carillion was one of the key causes of this tragedy, the fact that the company is now defunct means those responsible may not be properly held to account.”
“The RAIB report goes into forensic detail regarding the ‘French drain’ that was installed in 2011/12 as part of a wider scheme to address a known problem with the stability of the earthworks in this locality”