Should Network Rail undertake regular overnight inspections of 'high risk' track sections

[RobShipway]

Moderator note - split from:

ASLEF circular to members on 'legacy traction' and its safety.

Doing the rounds on other forums, a circular to members regards the safety of legacy rolling stock - HSTs get the main mention, but also 'legacy traction' - i.e. BR-era with reference to of course Carmont and the lack of recommendations of the report being implemented. I'm not looking for a...

www.railforums.co.uk

"My own view is that the campaigners about Grenfell Tower can get stuffed. People have lived in those flats without issue for years, there's only been one fatal accident involving spread between flats in the last 15 years..."

Sometimes it takes a tragedy to point out that something needs to be done. Had it not been for COVID, that train would have been full and we'd have been looking at a death toll possibly reaching 100.

It is, with trains, in my view valid to say "we'll stop them hitting stuff" (TPWS having been the big step in that direction in recent years), but that means something will need to be done to seriously reduce the risk of landslips, which have caused a number of derailments (e.g. that LM Class 350) in recent years.

Precisely! It’s such a non argument to come out with. Climate change conditions in recent years with freak weather events have most likely increased the chances of another Carmont happening. In the last 2 years I’ve seen landslip events lineside which have never had any previous history and such areas weren’t on the high risk register. It seems anything is possible these days.

But that is where Network Rail, should be checking the routes where landslip is possible prior to any trains travelling on those routes each morning. I know when there was landslip on the Uckfield line, between Uckfield and Crowborough, that I believe go reported to Network Rail or the line was checked by network Rail, before any trains left either London or Uckfield, so that the line could be closed. This is something that should have happened at Carmont. Yes, Aslef have the right to make sure that their members are safe. Whilst possibly better protection may have saved the driver of the HST at Carmont, an inspection of the route checking for possible areas of landslip should have been made by Network Rail, prior to any train travelling the route. If a check was made, then I would like to know how Network Rail determined that a landfall would not occur?

I know with leaving near cliffs that get bashed with water by the sea, that there is quite often more recently due to global warming boulders falling from those cliffs, especially over the seven sisters cliff area. But when inspecting the cliffs, you can see where cracks are showing, which is where a boulder is likely going to be breaking off the cliff to the crowd. I would imagine the same would have been the case on the hills above the line at Carmont, if it has been checked before any travel had travelled. This to my mind is what Aslef should be pushing for, rather than saying the issue is totally down to the ex - BR stock and pushing for it's replacement. Yes, the BR stock needs replacing, but it is better to make the lines safe first or whilst new stock is being sorted out, rather than just expecting new stock to just resolve the issue.

 

[takno]

But that is where Network Rail, should be checking the routes where landslip is possible prior to any trains travelling on those routes each morning. I know when there was landslip on the Uckfield line, between Uckfield and Crowborough, that I believe go reported to Network Rail or the line was checked by network Rail, before any trains left either London or Uckfield, so that the line could be closed. This is something that should have happened at Carmont. Yes, Aslef have the right to make sure that their members are safe. Whilst possibly better protection may have saved the driver of the HST at Carmont, an inspection of the route checking for possible areas of landslip should have been made by Network Rail, prior to any train travelling the route. If a check was made, then I would like to know how Network Rail determined that a landfall would not occur?

You may benefit from reading the report on the accident.

 

[D365]

If a check was made, then I would like to know how Network Rail determined that a landfall would not occur?

How often do you propose these routes need checking? What equipment would the gangs need to carry in order to complete these inspections? What mileage of track would need to be checked each night? Given that track walks would require each line to be secured, how do you propose deploying staff effectively?

 

[RobShipway]

You may benefit from reading the report on the accident.

If you mean i should read chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://assets.publishing.service.gov.uk/media/6329e57de90e0737279c012a/02_2022_Carmont.pdf, then I have read it several times and yes, it does state the crash worthiness of HST power cars and mk3 trains, it all states the following:

"RDG and Network Rail, in conjunction with RSSB, should consider and incorporate all relevant learning from the Carmont accident into the assessment of rolling stock and infrastructure design features that can provide guidance to trains when derailed. Particular features to be taken into account include: a) the risk of derailment from relatively small landslips and washouts b) position of track relative to adjacent ground on which derailed wheels may run (that is, features that can affect the deviation of a derailed train) c) proximity to features with the potential to increase the consequence of an accident (bridge parapets, tunnel portals etc) d) topography likely to increase the extent of vehicle scatter. The above-mentioned assessment should then be used to develop a systemic, risk-based strategy for the provision of additional measures for the guidance of derailed trains that takes into account the appropriate balance between infrastructure-based mitigation and vehicle-based."

Now, note point c) proximity to features with the potential to increase the consequence of an accident.

How often do you propose these routes need checking? What equipment would the gangs need to carry in order to complete these inspections? What mileage of track would need to be checked each night? Given that track walks would require each line to be secured, how do you propose deploying staff effectively?

If the world did not have the consequences of living with global warming, I would say once a week to once a month. But I know many areas around the country that are changing on a daily basis, so would suggest that the areas are checked on a daily basis. You don't necessarily have to walk the track, but if there is road access to the mound above the tracks then check for cracks and how any ground might be hanging on the mounds of the slopes would be sensible. I know that in similar areas in countries across Europe, daily checks are done both by using helicopter in the air and where there is danger points, they are looked at more closely on the ground.

The report does state the following:

"Network Rail should: a) take steps necessary to ensure that all elements of infrastructure constructed in Scotland since 2012 that require routine inspections and maintenance are included in the appropriate asset management processes b) dependent on findings from the above activity, extend the timeframe, to an extent determined on the basis of safety risk, to include work constructed before 2012 c) determine, based on safety risk, the extent to which similar steps are required on Network Rail infrastructure outside Scotland and, if necessary, implement these steps d) conduct an audit review covering the implementation of existing arrangements to identify, report and correct asset database management and data quality issues. (paragraphs 612b, 614b, 614c)"

The area of track even before the accident at Carmont, I believe required routine inspections. When was the last one done before the accident and what was found?

 

[Falcon1200]

But that is where Network Rail, should be checking the routes where landslip is possible prior to any trains travelling on those routes each morning.

What I think @takno was referring to was the timing of Carmont; The accident occurred at 0938, long after any pre-service inspection of the line would have occurred, which would therefore not have prevented it.

 

[RobShipway]

What I think @takno was referring to was the timing of Carmont; The accident occurred at 0938, long after any pre-service inspection of the line would have occurred, which would therefore not have prevented it.

Agreed, but my point is that the inspection should not just be of the line, it should be of the mound beside the line to detect, if any of the land would fall from the mound, which is what is done in many countries around Europe.

 

[Falcon1200]

Agreed, but my point is that the inspection should not just be of the line, it should be of the mound beside the line to detect, if any of the land would fall from the mound, which is what is done in many countries around Europe.

So, are you suggesting that an inspection, not just of the track but of the surrounding land (and not of that one location only but of any possible landslip sites) should be undertaken every day before trains start running? And that this is common practice in Europe?

 

[RobShipway]

So, are you suggesting that an inspection, not just of the track but of the surrounding land (and not of that one location only but of any possible landslip sites) should be undertaken every day before trains start running? And that this is common practice in Europe?

I know that it is common practice in Europe, as one of my ex - girlfriends worked as an engineer on the Swiss Rail network near Lucerne. None of the trains to or from Lucerne where able to go through the mountain peeks without both the line and the peeks themselves having been checked each day for safety.

 

[6Gman]

But that is where Network Rail, should be checking the routes where landslip is possible prior to any trains travelling on those routes each morning. I know when there was landslip on the Uckfield line, between Uckfield and Crowborough, that I believe go reported to Network Rail or the line was checked by network Rail, before any trains left either London or Uckfield, so that the line could be closed. This is something that should have happened at Carmont. Yes, Aslef have the right to make sure that their members are safe. Whilst possibly better protection may have saved the driver of the HST at Carmont, an inspection of the route checking for possible areas of landslip should have been made by Network Rail, prior to any train travelling the route. If a check was made, then I would like to know how Network Rail determined that a landfall would not occur?

I know with leaving near cliffs that get bashed with water by the sea, that there is quite often more recently due to global warming boulders falling from those cliffs, especially over the seven sisters cliff area. But when inspecting the cliffs, you can see where cracks are showing, which is where a boulder is likely going to be breaking off the cliff to the crowd. I would imagine the same would have been the case on the hills above the line at Carmont, if it has been checked before any travel had travelled. This to my mind is what Aslef should be pushing for, rather than saying the issue is totally down to the ex - BR stock and pushing for it's replacement. Yes, the BR stock needs replacing, but it is better to make the lines safe first or whilst new stock is being sorted out, rather than just expecting new stock to just resolve the issue.

It is, of course, not impossible to do two things at the same time.

Improve track safety and enhance the crashworthiness of stock (by withdrawing some stock if necessary).

 

[Bletchleyite]

I know that it is common practice in Europe, as one of my ex - girlfriends worked as an engineer on the Swiss Rail network near Lucerne. None of the trains to or from Lucerne where able to go through the mountain peeks without both the line and the peeks themselves having been checked each day for safety.

There's a world of difference between checking one particularly problematic mountain pass (I guess we're talking about the Gotthard?) and checking every single section of line where there could be a problem (basically every embankment).

 

[edwin_m]

Areas of earthwork or structures considered high risk are frequently now fitted with automated monitoring equipment that will alert relevant people within a few seconds if there is any movement beyond a defined limit. If it's done properly this will be much more effective than a daily inspection, because it's 24/7 but also because the small movements that precede something more serious may not be obvious on a quick visual inspection. It was mentioned on the Nuneham viaduct thread that monitoring equipment was fitted here prior to the closure, but I don't think we know if that was what triggered the closure itself.

With many old earthworks not designed to modern standards, and climate change plus the risk of very localised storms such as Carmont, the bigger problem may now be how to identify which locations are high risk, rather than how to monitor them when they are.

 

[HSTEd]

How many thousands of additional staff will be required for this?

It will cost vast sums of money, and at most could save a small number of lives given the very low number of fatalities and injuries recorded on the railway system from this sort of problem.

If such checks are to be required I would suggest the more likely outcome is mass route closures.

 

[Tetchytyke]

But that is where Network Rail, should be checking the routes where landslip is possible prior to any trains travelling on those routes each morning

No checking is fail safe, it is always just a moment in time. At Carmont the land slip happened between the train heading south and heading back north again a few hours later, so no daily site inspection would have picked it up anyway.

Automated land slip detection is in place in the most dangerous locations already. Identifying those locations is important, but with localised storms anywhere can be dangerous. Embankments, cuttings, bridges. Look at the collapse of the Glanrhyd Bridge in 1987.

Beyond that, the only real solution is to slow everything down after prolonged wet weather. There’s certainly an argument for 20mph or 30mph blanket ESRs after heavy rain. But you can imagine the chaos that would cause.

 

[RobShipway]

How many thousands of additional staff will be required for this?

It will cost vast sums of money, and at most could save a small number of lives given the very low number of fatalities and injuries recorded on the railway system from this sort of problem.

The safety of passengers should not be down to the amount of people use a line in a day, week or month etc.... It should be taken seriously, however many passengers or freight uses a line, as you never know that there could be an event in an area that could take the numbers of passengers from being 200 a day, to being 12,000 a day. A good example of this would be Henley-On-Thames station, which sees very few passengers on a daily basis, accept when the Henley Regatta is on.

Areas of earthwork or structures considered high risk are frequently now fitted with automated monitoring equipment that will alert relevant people within a few seconds if there is any movement beyond a defined limit. If it's done properly this will be much more effective than a daily inspection, because it's 24/7 but also because the small movements that precede something more serious may not be obvious on a quick visual inspection. It was mentioned on the Nuneham viaduct thread that monitoring equipment was fitted here prior to the closure, but I don't think we know if that was what triggered the closure itself.

With many old earthworks not designed to modern standards, and climate change plus the risk of very localised storms such as Carmont, the bigger problem may now be how to identify which locations are high risk, rather than how to monitor them when they are.

If you can setup equipment in the right areas, it would be cheaper than having to hire people in to be checking the areas concerned. But as you state, the bigger problem is identifying which locations are high risk. Once, you have identified an area is high risk, then a decision needs to be made how to monitor that area. If automated camera and land movement monitors can be setup great, but the issue is where do you setup equipment on a line that say goes round mountains or hills.

No checking is fail safe, it is always just a moment in time. At Carmont the land slip happened between the train heading south and heading back north again a few hours later, so no daily site inspection would have picked it up anyway.

Automated land slip detection is in place in the most dangerous locations already.

Beyond that, the only real solution is to slow everything down after prolonged wet weather. There’s certainly an argument for 20mph or 30mph ESRs after heavy rain.

Fair point, but surely if on the line there is sensors added to detect landslip/fall, then surely that could be setup to change the proceeding lights to red to hold the train and let the signal centre know that there has been an incident that needs investigating?

 

[HSTEd]

The safety of passengers should not be down to the amount of people use a line in a day, week or month etc.... It should be taken seriously, however many passengers or freight uses a line, as you never know that there could be an event in an area that could take the numbers of passengers from being 200 a day, to being 12,000 a day. A good example of this would be Henley-On-Thames station, which sees very few passengers on a daily basis, accept when the Henley Regatta is on.

But the reality is that the railway already takes safety seriously.

How many people have been killed or injured in the last twenty years as a result of landslips or similar problems?

That is the absolute maximum number of people that would have been saved by a massive new inspection regime.
And, ultimately, lives are not considered to be infinitely valuable.

A mass inspection regime would cost hundreds of millions to billions a year to operate given the number of staff that would be required. This is especially true as failure to complete an inspection would close the line, so large amounts of cover would be required to ensure it all gets done every day.

 

[RobShipway]

But the reality is that the railway already takes safety seriously.

How many people have been killed or injured in the last twenty years as a result of landslips or similar problems?

That is the absolute maximum number of people that would have been saved by a massive new inspection regime.
And, ultimately, lives are not considered to be infinitely valuable.

A mass inspection regime would cost hundreds of millions to billions a year to operate given the number of staff that would be required. This is especially true as failure to complete an inspection would close the line, so large amounts of cover would be required to ensure it all gets done every day.

Fair comment, but if we go to something more local to where I live which is the landslip between Uckfield and Crowborough. Now the The Argus reporting the https://www.theargus.co.uk/news/19662768.landslips-fallen-trees-cause-train-disruptions-flooding/ on the 21st October 2021.

For those that may not be able to read the link, it states the following:

"
FALLEN trees, landslips and subsequent train cancellations are causing major railway disruptions.

Network Rail have announced widespread disruptions across Sussex and the surrounding counties due to ongoing weather chaos.
The disruption will affect trains from Southern, Southeastern and Thameslink.

A landslip overnight means no trains are running between Crowborough and Uckfield.
“Our engineers are inspecting the damage and as you can see there is some debris on the tracks that needs to be removed,” a Network Rail spokesman said."

Now, I am not sure what time the landslip occurred between Uckfield and Crowborough and whether it affected any trains on the route. During the same night a fallen train blocked the Brighton Mainline at Balcombe. Now, I am presuming that the trains due to the weather at the time where going at slower speeds or someone in the areas reported the issues or was this detected by devices on the line?

My point being even if it is not a person, then cameras or devices checking the line would be sufficient. However, nothing can beat the human eye for picking up on issues.

 

[D365]

My point being even if it is not a person, then cameras or devices checking the line would be sufficient. However, nothing can beat the human eye for picking up on issues.

In most scenarios, a human is very unlikely to detect a landslip that hasn’t already happened. So I’m not sure what your argument is based upon.

 

[edwin_m]

If you can setup equipment in the right areas, it would be cheaper than having to hire people in to be checking the areas concerned. But as you state, the bigger problem is identifying which locations are high risk. Once, you have identified an area is high risk, then a decision needs to be made how to monitor that area. If automated camera and land movement monitors can be setup great, but the issue is where do you setup equipment on a line that say goes round mountains or hills.

Indeed. Basically I'm saying that NR is already doing what you suggest, but generally using automated monitoring rather than sending someone out to take a look each night (and bearing in mind it's usually dark before the first train of the day, so either humans and visible light cameras would be of limited use). The problem of identifying which ones are high risk is the same, whichever method is then proposed to monitor them.

Mountains and hills aren't really a problem for automatic monitoring - they are probably worse for a human who might be at considerable risk when climbing up the side of a cutting with a live railway at the bottom. In fact crude systems of that type have been in use for many decades in the Highlands, provided with special signals to alert drivers to rockfalls.

Pass of Brander stone signals - Wikipedia

en.wikipedia.org

The Pass of Brander stone signals, also known as Anderson's Piano, are a series of railway signals situated in the Pass of Brander, between Loch Awe and Taynuilt stations on the Oban branch of the West Highland Line in Scotland. A screen of wires, linked to semaphore signals, are erected on the mountainside alongside the railway. In the event of one or more wires being broken, signals in each direction are automatically placed at 'danger'. They are part of a warning system that advises train drivers to exercise caution in the event of a rock-fall, and cover a section of track that is just over 4 miles (6.4 km) in length.

 

[yoyothehobo]

It can be nigh on impossible to spot a landslip location before it happens, you can use a lot of data modelling using LiDAR, rainfall analysis, slope angle, geology, slope height etc.. to give you a risk score relative to everything else, but geotechnics can be such an inexact science, especially as the cost to get detailed data is so high and even then its not always the most accurate information (look at design calculations in Eurocode 7 for examples).

Embankment issues can be detected by the measurement train by the effect they have on the track, but modelling for cuttings is incredibly difficult.

The high risk sites already have inspection regimes in place for adverse weather.

What more do you want?

We generally live by risk being As Low As Reasonalby Practicable

 

[greyman42]

How often are point checked at high speed junctions such as Colton Jc. near York? There was a fault on this junction recently which was fortunately spotted before any incidents occurred.

 

[JKF]

At some point I suspect they will start doing track inspections with drones, it would be quite easy to have something similar to drive assist in cars that can follow the rails maybe at a height of around 1m, and report obstructions and faults in real-time. This could run ahead of trains if necessary. I guess the issue at the moment is the flight time possible with battery power.

 

[edwin_m]

At some point I suspect they will start doing track inspections with drones, it would be quite easy to have something similar to drive assist in cars that can follow the rails maybe at a height of around 1m, and report obstructions and faults in real-time. This could run ahead of trains if necessary. I guess the issue at the moment is the flight time possible with battery power.

The measurement trains are fitted with cameras and pattern recognition to identify faults automatically, and can run at normal speeds without a possession. If an inspection is needed that can't be done that way or by reviewing the video, a drone at that altitude the line would need a line closure anyway, and an operator nearby, so there's not much benefit over sending a person.

Some inspections are done by helicopter, such as looking for hotspots with a thermal camera. A drone could do this, but for a long section of line a single helicopter run is probably more efficient than repeated launchings of drones along the route. That might change if autonomous long-range drones become available and legal, that can fly a pre-planned route along the railway and report back any problems.

 

[Bald Rick]

How often are point checked at high speed junctions such as Colton Jc. near York? There was a fault on this junction recently which was fortunately spotted before any incidents occurred.

Depends what you mean ‘checked’. There’s various levels of inspection depending on the component parts of the points, the type of equipment, and the linespeed / tonnage over them.

 

[greyman42]

Depends what you mean ‘checked’. There’s various levels of inspection depending on the component parts of the points, the type of equipment, and the linespeed / tonnage over them.

Does anyone know how the recent fault at Colton Junction was spotted?

 

[Bald Rick]

Do you know how the recent fault at Colton Junction was spotted?

yes!

 

[greyman42]

yes!

Could you enlighten us?

 

[WAO]

I would be happier if the Civil Engineer had the last word not a budget controller wanting a paper exercise business plan.

At least then we could have done no more if unforeseeable disaster strikes.

WAO

 

[HSTEd]

I would be happier if the Civil Engineer had the last word not a budget controller wanting a paper exercise business plan.

At least then we could have done no more if unforeseeable disaster strikes.

The problem with that system is the Civil Engineer has little reason not to balloon the budget with huge numbers of risk control measures to protect themselves and their reputation.

Trying to cover every possible eventuality is a natural impulse for an engineer but it is not a great way to run a railway.
It will lead to ballooning costs, schlerosis and eventually to collapse.

 

[Bald Rick]

Could you enlighten us?

unfortunately not.

I would be happier if the Civil Engineer had the last word

He or she does.

 

[WAO]

The problem with that system is the Civil Engineer has little reason not to balloon the budget with huge numbers of risk control measures to protect themselves and their reputation.

Trying to cover every possible eventuality is a natural impulse for an engineer but it is not a great way to run a railway.
It will lead to ballooning costs, schlerosis and eventually to collapse.

(Chartered) Engineers are educated and trained to produce cost effective solutions that involve judgement as well as calculation.

I am not convinced that the Carmont works that preceded the disaster were to an engineer's entire satisfaction, either in design or construction.

In the end you have to trust your Doctor. The Engineer has no less status.

WAO