Are London's railways falling apart?

[whole diary]

[grahame]

From my Twitter feed this morning, 4.2.2020 ...

Due to a points failure between Moorgate and Finsbury Park all lines are blocked.  Train services running to and from these stations may be cancelled, delayed or diverted via London Kings Cross.

There are urgent repairs taking place to the railway at #NewCross after a track defect was found overnight.  This is blocking the lines to Charing Cross, meaning that no trains can run to Waterloo East of Charing Cross. until at least 08:00

No Thameslink services between Dartford and London Bridge expected until 10:30

Late finish to engineering work at Gunnersbury:
• No District Line Turnham Green and Richmond
• No Overground Richmond and Willesden Junction

until further notice no service on the Overground  South Tottenham to Barking - repairs to the track

[Electric train]

From my Twitter feed this morning, 4.2.2020 ...

Due to a points failure between Moorgate and Finsbury Park all lines are blocked.  Train services running to and from these stations may be cancelled, delayed or diverted via London Kings Cross.

There are urgent repairs taking place to the railway at #NewCross after a track defect was found overnight.  This is blocking the lines to Charing Cross, meaning that no trains can run to Waterloo East of Charing Cross. until at least 08:00

No Thameslink services between Dartford and London Bridge expected until 10:30

Late finish to engineering work at Gunnersbury:
• No District Line Turnham Green and Richmond
• No Overground Richmond and Willesden Junction

until further notice no service on the Overground  South Tottenham to Barking - repairs to the track

Intensely used railway for an increasing amount of operating time with diminishing engineering access hours

[grahame]

until further notice no service on the Overground  South Tottenham to Barking - repairs to the track

From the Barking and Daghenham Post

The Rail Accident Investigation Branch (RAIB^▸ (Rail Accident Investigation Branch)) revealed the cause in its report into the incident, which caused the Barking to Gospel Oak line to be shut for 27 days while repairs were carried out.

According to the report, published earlier this week, the freight train departed Barking shortly after 5.30am on January 23 this year and was just minutes into its journey to Calvert, Buckinghamshire when two adjacent wagons near the rear of the train became derailed as it crossed a bridge between Woodgrange Park and Wanstead Park stations.

The wheel of the second wagon rerailed itself after around 230 metres but the first wagon remained derailed, with the train travelling for around six minutes until it came to a stop.

Although nobody was injured in the incident, the derailment required a significant amount of repair work to be carried out on the track.

This included replacing 39 pieces of rail, 5,300 concrete sleepers and 900 wooden sleepers, as well as removing and replacing 10,000 tonnes of ballast and replacing timbers supporting the track at 10 bridges. Lineside cables and signalling equipment were also repaired.

Investigators carried out an analysis of the track and found that the ?derailment was a consequence of rail movement causing the track gauge (the horizontal distance between the rails) to increase by at least 85 mm, sufficient to allow the right-hand wheel of a wagon to drop inside the right-hand rail?.

The numbers are astonishing ... can't help wondering if an opportunity was taken to deballast as I can't imagine 10,000 tonnes being damaged.

[stuving]

The numbers are astonishing ... can't help wondering if an opportunity was taken to deballast as I can't imagine 10,000 tonnes being damaged.

I would have thought that renewing ballast was just a part of the standard practice for replacing the track - combined perhaps with a concern over what debris of broken track was in it.

Incidentally, I found that in the Wessex Route Strategic Plan these baulks are called "longitudinal wheel timber bridges" or "long wheel timber assets" - baffling terms for which it took a bit of poking about to find explanations.

[stuving]

I was struck by the following, which RAIB^▸ (Rail Accident Investigation Branch) reported but made little of.

The widening of the track gauge was identified, by track recording vehicles operated by Network Rail, six times between March 2019 and January 2020. On three occasions, the track maintenance gangs responded but undertook work in a nearby location due to erroneous GPS data. On three other occasions, the track recording vehicles? software did not report the location of the fault so maintenance staff remained unaware of it.

One of those locations was reported correctly, but the track workers' phone-based GPS led them to a few metres away. The other two were wrongly reported to them by several metres.

It's a standard feature of any train with ETCS^▸ (European Train Control System) that it knows where it is, by fusing data from several sources. GNSS (preferably differential or carrier-phase), odometry, and balises are always used. Other sources might be added if it looks helpful. So I don''t understand why the measurement trains, for which accurate position reporting is vital to their function, don't have that too. In addition, I would expect the location to be translated into the traditional line reference plus chainage form, so local PW^▸ (Permanent Way) staff can relate it to plans and older records. The phone app should be no more than a helpful guide, not the only way they know where to go.

[MVR S&T]

Usual thing, go find a problem, contracter reports all ok at location, problem still there, fault not found, get paid. I have always wanted to find the fault, get it fixed.

[TonyN]

For many years track recording trains dropped whitewash on the sleepers to identify a track fault. The local ganger then repaired it quickley because the Inspector would be along to check.

[stuving]

There was another reference to GPS in that RAIB^▸ (Rail Accident Investigation Branch) report, which I forgot to include:

76  Geometry faults which exceed the intervention limit are also listed, with their global positioning system (GPS) co-ordinates, on a track geometry action report. This is also known as an immediate action report or fault report. Network Rail states that these co- ordinates are accurate to ?1m, providing the track recording vehicle is able to receive a strong enough GPS signal. The signal can be restricted when a train is within a structure such as a tunnel or station. There are no features of this type at the accident site.

The point about integration with an odometer is that position is continuously available, even in tunnels,  to the accuracy of GPS in the best clear reception conditions - or indeed much better, due the the averaging effect.

[Electric train]

For many years track recording trains dropped whitewash on the sleepers to identify a track fault. The local ganger then repaired it quickley because the Inspector would be along to check.

The track recording train now uses video and still images, the video and images, recorded data and exact ELR, mileage and chains or yards is sent to the TME^▸ (Track Maintenance Engineer) (Track Maintenance Engineer) so they can plan what is required