Signals

THE SIGNAL BOX


Railway signalling discussion

High Speed Derailment Eckwersheim

Signalling outside the UK (but including Northern Ireland), past, present and future

High Speed Derailment Eckwersheim

Unread postby Andrew Waugh » Sat Nov 14, 2015 9:00 pm

Reports of a derailment (apparently) due to excessive speed on the unopened Paris-Strasbourg high-speed line on Saturday at Eckwersheim near Strasbourg. At least ten dead and 32 injured. All were employees of SNCF operating a test run.

http://www.theguardian.com/world/2015/n ... sh-reports

http://www.bbc.com/news/world-europe-34822666

I would assume both train and track would be equipped with ATP. Does anyone know the type?

It's been a terrible 24 hours in France.
Andrew Waugh
Branch line box
Branch line box
 
Posts: 456
Joined: Wed Sep 17, 2008 1:34 am
Location: Melbourne Australia

Re: High Speed Derailment Eckwersheim

Unread postby davidwoodcock » Sat Nov 14, 2015 9:19 pm

I would have thought that unplanned excessive speed was unlikely to be the cause of the accident. The French routinely test their new (and unopened) high speed lines at much higher speeds than will be used when the line opens for normal traffic and it is clear from the photographs that the train concerned was one of the special test units.

The extent of the devastation is surprising given that French TGVs are designed to stay within envelope even after a high speed derailment, and experience to date (and there is about four decades worth) suggests that that does usually happen. Certainly there was a Eurostar which derailed at 300 km/h and the only vehicles that weren't actually on the rails when the train came to a stand were the front and rear power cars even though all 18 of the intermediate passenger cars had passed over a significant obstruction.
davidwoodcock
Rest-day relief
Rest-day relief
 
Posts: 947
Joined: Tue Jul 12, 2011 7:58 am
Location: Champlon, Belgique

Re: High Speed Derailment Eckwersheim

Unread postby Richard Lemon » Sun Nov 15, 2015 8:24 am

The LGV Est is equipped with both TVM430 and ERTMS.

I would guess that this was not operating for the test run, so no excess speed intervention. I thus assume that the driver did not apply the brakes soon enough.

This is purely a guess - all will become clear before long.

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: High Speed Derailment Eckwersheim

Unread postby John Hinson » Sun Nov 15, 2015 9:36 am

So Andrew's question has been answered nicely - there will be nothing else here to say on this accident unless there is a signalling implication in which case correspondents are welcome to contact me to have the topic unlocked.

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6838
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: High Speed Derailment Eckwersheim

Unread postby Richard Lemon » Fri Nov 20, 2015 10:50 am

SNCF has now released an initial report into this accident:

http://www.railwaygazette.com/news/news ... -sncf.html

There is derogation to isolate the Cab Signalling during a Test Run, as the test is deliberately at higher than line speed.

Thus, Route Knowledge is the only protection - there are no line side speed boards on LGVs.

It would appear probable that the number of people in the cab was a distraction for the driver.

I suggest that TGVs in their normal operation are still as safe as ever.

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: High Speed Derailment Eckwersheim

Unread postby davidwoodcock » Fri Nov 20, 2015 1:00 pm

The presence of families aboard on a school day points to this having been a very high speed test run. The SNCF tradition of this being permitted goes back at least 25 years to my knowledge and possibly much longer, and Pepy would unquestionably have known this.

I suspect that it isn't that the cab signalling per se which is isolated but the requirement for the driver not to exceed the displayed speed. Thus I would anticipate that the driver would have received an instruction to slow to perhaps 150 km/h or less (since the test train was actually permitted 176 km/h). However, based on the surprisingly realistic experience of having "driven" the Eurostar simulator, I would say that things happen incredibly quickly even at 300 km/h.
davidwoodcock
Rest-day relief
Rest-day relief
 
Posts: 947
Joined: Tue Jul 12, 2011 7:58 am
Location: Champlon, Belgique

Re: High Speed Derailment Eckwersheim

Unread postby Richard Lemon » Fri Nov 20, 2015 2:05 pm

Good point David - but still no intervention, and with the distraction element?
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: High Speed Derailment Eckwersheim

Unread postby Mike Hodgson » Fri Nov 20, 2015 10:07 pm

Speed limits are usually round numbers - so 176 kph sounds distinctly improbable as a permitted speed.
It sounds more like the result of some absolute calculation, such as the speed above which it could theoretically be expected to leave the rails, with presumably less conservative or no safety margin.

Instead of isolating the cab signalling for a high speed test run, they ought to have retained the intervention it but specially configured it for these higher speeds, ie this 176, since overspeed when you don't have the usual safety margin is clearly more dangerous than in normal operation. Can that be done without moving balises etc?

I see the drivers have been suspended. My understanding of French practice is that even in steam days they always took a much stricter view of their speed limits than the UK, and a driver was more likely to be sacked for even a slight overspeed.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2478
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: High Speed Derailment Eckwersheim

Unread postby davidwoodcock » Fri Nov 20, 2015 11:14 pm

It is my understanding that calculations for maximum permitted speeds on very high speed test runs took into account horizontal and vertical curvature and pressure wave effects (important at over bridges and through tunnels) but that a safety margin was still included (probably no more than 5%). The drivers would have been given the resulting maximum permitted speed graph. The speeds shown on it would probably have borne little relationship to the planned service speeds for the line. I obviously don't know about this run, but in the past very high speed test runs have been authorised to run at speeds in excess of 500 km/h where the line was suitable - and such speeds have been attained.

My first thought when seeing a range of photos of the accident was that the train was running too fast for the vertical curvature and literally took off. Pepy's statement doesn't seem to totally rule that out although I have only seen an English translation at the moment, he speaks excellent English but I can't imagine that the actual statement was in anything but French.

French drivers were, and are, trained to drive to the timetable, being never more than 2 minutes early or late, but, yes, they were expected to observe speed limits strictly, and this would be even more true of TGV drivers as driving at a speed above the steady-displayed limit would cause the over-ride to cut in and stop the train.
davidwoodcock
Rest-day relief
Rest-day relief
 
Posts: 947
Joined: Tue Jul 12, 2011 7:58 am
Location: Champlon, Belgique

Re: High Speed Derailment Eckwersheim

Unread postby Richard Lemon » Sat Nov 21, 2015 8:18 am

I can add to this that every night the data recorders from TGVs are downloaded, and in the morning the home depot will have a print out of any overspeeds. So, no hiding place. At least, when I worked for Eurostar this was the case, so I assume it still is

As to 176 km/h - and here I am being theoretical - if the maximum line speed was 160 km/h - and that is one of the standard "steps" in the display - and you then authorise a 10% overspeed, you have 176 km/h!

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: High Speed Derailment Eckwersheim

Unread postby Ulf Pålsson » Sat Nov 21, 2015 7:56 pm

The Direction générale de l’audit et des riques of SNCF has released a preliminary report (http://www.sncf.com/ressources/rapport_ ... e_2015.pdf ). It confirms that the ATP system was isolated to make it possible to run the test train with 10 % overspeed. On sections with commercial speed of 320 km/h the test train was running in 352 km/h. In the curve where the derailment took place, the test train speed should have been 176 km/h (commercial speed 160 km/h), but it was in fact 243 km/h.
Ulf Pålsson
Trainee
Trainee
 
Posts: 9
Joined: Sun Jun 9, 2013 12:52 pm

Re: High Speed Derailment Eckwersheim

Unread postby Richard Lemon » Sun Nov 22, 2015 8:28 am

Thank you Ulf - a prompt report.

From the devastation, that anyone survived is remarkable.

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: High Speed Derailment Eckwersheim

Unread postby S&TEngineer » Sun Nov 22, 2015 9:54 am

Moderator Note: Now that it is clear that the incident was caused by overspeed the topic will be locked unless anything else signalling related gets raised at a later date
Regards,
S&TEngineer
-----------------------------------------------------
Out of this nettle, Danger, we pluck this flower, Safety.
Henry IV, Part 1, Act 2, Scene 3
User avatar
S&TEngineer
Moderator
Moderator
 
Posts: 1601
Joined: Fri Nov 9, 2007 8:17 pm
Location: Somewhere in the far South West


Return to Signalling - overseas

Who is online

Users browsing this forum: No registered users and 2 guests