Great Western Coffee Shop

All across the Great Western territory => The Wider Picture in the United Kingdom => Topic started by: SandTEngineer on November 07, 2017, 19:28:41



Title: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: SandTEngineer on November 07, 2017, 19:28:41
Looks like this will an interesting report when its published.  The outcome could have been extremely serious..... ::)
https://www.gov.uk/government/news/maintenance-train-runaway-near-markinch


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Red Squirrel on November 07, 2017, 19:45:13
Quote
...having oscillated up and down the adjacent track gradients several times.

For some reason, it made me think of this:

https://www.youtube.com/watch?v=zZUJLO6lMhI


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Chris from Nailsea on November 08, 2017, 00:04:10
Looks like this will an interesting report when its published.

With my thanks for your posting that item here, SandTEngineer, I've simply expanded this topic heading - so as to make it easier for any of us to find it here on the Coffee Shop forum again, when that Rail Accident Investigation Branch (RAIB) report is published.


Admin note:
We on the admin / moderator team are keenly aware that the Coffee Shop forum's search facility is now fairly comprehensively 'broken' - but the fact of the matter is, any attempt to 'repair' it, using our very dated software, would probably cause even more problems than would be solved.



Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: SandTEngineer on November 08, 2017, 08:29:34
Looks like this will an interesting report when its published.

With my thanks for your posting that item here, SandTEngineer, I've simply expanded this topic heading - so as to make it easier for any of us to find it here on the Coffee Shop forum again, when that Rail Accident Investigation Branch (RAIB) report is published.


Admin note:
We on the admin / moderator team are keenly aware that the Coffee Shop forum's search facility is now fairly comprehensively 'broken' - but the fact of the matter is, any attempt to 'repair' it, using our very dated software, would probably cause even more problems than would be solved.


Hey CfN, you stole my 'jokey' topic heading....... ::)

Not a problem though ;)


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Chris from Nailsea on November 08, 2017, 10:57:58
Hey CfN, you stole my 'jokey' topic heading....... ::)

Yes, sorry about that, SandTEngineer.  I did consider that when I was renaming this topic, but the problem is, when the RAIB eventually publish their findings - probably in just under a year's time  ::) - we might be hard pressed then to remember the joke and update the appropriate thread. :-[

The original topic heading was 'The Runaway Train'.



Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Chris from Nailsea on January 13, 2018, 06:20:03
The Rail Accident Investigation Branch (RAIB) has released its report (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/672635/180111_R012018_Markinch.pdf) into the runaway of a maintenance train near Markinch, Fife, 17 October 2017.

Quote
Summary

At about 04:25 hrs on Tuesday 17 October 2017, a maintenance train that was clearing leaf debris from the track, hit a tree just north of Markinch station, Fife. The debris from the tree disabled the train’s braking system. The train came to a stop before running away backwards for a distance of about 4.7 miles (7.5 km). The train crew on board made an emergency call to the signaller before jumping off the train, suffering minor injuries.

The train eventually came to a stop at Thornton North junction after running backwards and forwards between two adjacent gradients a total of nine times.

The brakes were fully released when debris from the tree operated the release mechanisms on the brake system distributors underneath both of the vehicles in the train.

The driver was unable to reapply the brakes because the debris from the tree had also separated all three of the brake pipes between the two vehicles in the train.

Recommendations

The RAIB has made two recommendations to Network Rail. The first addresses the risk of a runaway being initiated by debris on the track causing multiple disruptions to the braking system on this type of train. The second recommendation addresses the possibility that similar risks might be present for other short formation trains that operate on its infrastructure.




Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: grahame on January 13, 2018, 08:07:01
The train eventually came to a stop at Thornton North junction after running backwards and forwards between two adjacent gradients a total of nine times.

Which goes to show what marvellous smooth and friction free bearings train vehicles have these days.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Western Pathfinder on January 13, 2018, 09:32:51
It's just a great shame that they don't have a parking brake !.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Chris from Nailsea on January 13, 2018, 21:12:24
Indeed.  :-X

Having posted that earlier link to the full RAIB report (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/672635/180111_R012018_Markinch.pdf), with its rather sanitised 'Summary', I was amazed at some of the more detailed factual account of the incident, later in the report:

Quote
The sequence of events

10 The MPV had left Slateford depot in Edinburgh just before midnight, with the driver and treatment operator on board, and had treated the lines to Tweedbank, Dunfermline and Kirkcaldy before passing through Markinch en-route to Ladybank. At approximately 04:25 hrs, when approximately 0.8 miles (1.3 km) north of Markinch station, the driver saw the outline of the top of a tree across the track ahead. At this moment the MPV was travelling at about 43 mph (70 km/h)

11 The driver immediately pushed the emergency brake plunger, and both the driver and the rail head treatment operator dived to the floor under the control console. It is unlikely that the brakes had time to significantly reduce the train’s speed before it hit the tree. As soon as the train came to a stop, it started to roll backwards, at which point the driver unsuccessfully tried to use the desk controls to stop the train, despite the emergency brake having already been applied.

12 The treatment operator then left the cab and hurried along the MPV’s walkway to operate the emergency brake plunger in the other cab. However, this had no effect and he returned to what had been the leading cab at the time of the collision. While this was happening, the driver operated the emergency alarm button on the GSM-R radio, which made an emergency call to the signaller in Edinburgh. During this call, the driver stated that the train had hit a tree and that it had lost all braking and was rolling backwards out of control.

13 The signaller checked that there were no other trains in the path of the runaway and immediately took steps to prevent any northbound trains from passing Inverkeithing. The next train in the path of the runaway was northbound empty coaching stock that had previously formed the southbound Aberdeen sleeper, which was still south of the Forth Bridge. The signaller held this train at Dalmeny junction, which is 24 miles (39 km) away from Markinch.

14 Having contacted the signaller, and recognising that he was unable to stop the train as it gathered speed, the driver decided that he and the treatment operator needed to get off as soon as possible. He feared that the train could either collide with another train, or derail. The two crew left the cab and jumped from the train onto the side of the track, which was in complete darkness at the time. Both the driver and the treatment operator sustained minor injuries, including cuts, bruises and cracked ribs.

15 The RAIB has calculated that the train had been rolling backwards for about 100 seconds, over a distance of approximately 0.28 miles (450 metres), and had reached about 20 mph (32 km/h) when the crew jumped off. This is based on information from the On-Train Data Recorder (OTDR), and the crew’s account of where they jumped.

16 The train crew checked that each of them was not seriously injured and walked a short distance to the nearest signal (signal ER607). The driver called the signaller from the signal post telephone and told him that they had jumped from the train and that both lines needed to be blocked. The signaller did this and made arrangements for an ambulance to meet them at Markinch station after checking that they were able to walk the mile or so there. The signaller subsequently called the driver to let him know that the train had changed direction and could come back to towards them, and that they would need to keep clear of the track. The train crew were met at Markinch station by an ambulance crew and taken to hospital.

17 The train ran south for a total of about 4.7 miles (7.5 km), reaching a maximum speed of 43 mph (69 km/h), before coming to a stop about 1.1 miles (1.8 km) along the 2.2 mile (3.5 km) long uphill gradient south of Thornton North junction. The train then ran forwards and backwards for a total of eight further movements before coming to a stop at the lowest point at Thornton North junction. The train ran backwards and forwards for a total of 39 minutes from its first stop after hitting the tree.


Wow!  While it's not recorded in that report, I'll bet there were a fair few expletives included in some of those discussions between the crew involved.  :o

Glad to learn that neither of them suffered any more serious injuries than cracked ribs - although that's bad enough.  Well done, guys.  ::)



Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: broadgage on January 14, 2018, 11:57:22
It seems to me that the design of the vehicles was defective in not being equipped with a manually operated parking brake.
One that is applied by turning a wheel and that does not require compressed air or electricity.

Such brakes often operate on only one axle and therefore have limited stopping power, but would have sufficed in this case.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: dviner on January 14, 2018, 13:45:02
It seems to me that the design of the vehicles was defective in not being equipped with a manually operated parking brake.
One that is applied by turning a wheel and that does not require compressed air or electricity.

Such brakes often operate on only one axle and therefore have limited stopping power, but would have sufficed in this case.

If I remember the report correctly, there is parking brake on the MPV, but it can only be activated when the vehicle is stationary - if the vehicle is moving, then that brake won't operate. Bearing in mind that this brake is a parking brake, then that's not a design flaw.

What is a design flaw is that the mechanism to disable the brakes (a feature that is required) was vulnerable to being operated (and jammed in the operated position) through collision with debris on the track.

One of the positive outcomes of this incident is that it highlights a possible failure mode that wasn't considered before.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: stuving on January 14, 2018, 14:27:00
It seems to me that the design of the vehicles was defective in not being equipped with a manually operated parking brake.
One that is applied by turning a wheel and that does not require compressed air or electricity.

Such brakes often operate on only one axle and therefore have limited stopping power, but would have sufficed in this case.

If I remember the report correctly, there is parking brake on the MPV, but it can only be activated when the vehicle is stationary - if the vehicle is moving, then that brake won't operate. Bearing in mind that this brake is a parking brake, then that's not a design flaw.

What is a design flaw is that the mechanism to disable the brakes (a feature that is required) was vulnerable to being operated (and jammed in the operated position) through collision with debris on the track.

One of the positive outcomes of this incident is that it highlights a possible failure mode that wasn't considered before.

I would agree that something conceived as a parking brake is not likely to be useful as an emergency brake. There was yet another brake, but it was linked to the primary air brakes through the train's controls and the distributor valve. That valve turns out to be rather an Achilles' heel of the supposedly fail-safe air braking principle. And both those secondary brakes could not be used on a moving train.

What I feel broadgage was aiming for, but perhaps just missed, is an independent mechanical brake that can be applied by the crew on a moving train. That would necessarily also be a parking brake. But note that the bogies mean that a simple screw-down brake isn't going to work, it needs a motion-compliant connection - either an awkward linkage, or a flexible cable or hose.

That connection needs to be well shielded from assailants such as trees. If you look at one of these MPVs, there is a rather scary (given their duties) amount of dangly stuff around their undersides. Note that the unforeseen failure in this case was a mechanical linkage, in part because it was in a silly place but also because they are hard to protect.

My suggestion for something that is simple enough that it might happen is this:
  • Add a hydraulic pump operated by a handwheel or lever, either in the cab or just behind it (and not requiring anything to be found and attached to make it work)
  • Connect this to the hydraulic brake system, either in place of or as well as the existing pump,
  • Check the routing of the pipes and hoses for vulnerability, and move or protect as need be.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: broadgage on January 14, 2018, 16:22:51
Indeed, I was stating that in my view that the vehicle should have had a basic manually applied brake, operated by turning a wheel or in some similarly direct way, not reliant on anything else nor interlocked with anything that could prevent its use.

Whilst the stopping power would be limited, it would have prevented this accident. Note that after striking the tree, that the train continued forwards by momentum, stopped and then ran backwards under gravity.
Just before stopping, or just after starting to run back the speed would have been very low  and will within the stopping power of a brake primarily intended for parking.

To avoid damage in normal use from someone omitting to release the parking brake before setting of, an interlock could be fitted to prevent power being taken until the hand brake was released.
But no interlock or control of any kind should ever stop the handbrake from being applied.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: dviner on January 14, 2018, 17:34:43

My suggestion for something that is simple enough that it might happen is this:
  • Add a hydraulic pump operated by a handwheel or lever, either in the cab or just behind it (and not requiring anything to be found and attached to make it work)
  • Connect this to the hydraulic brake system, either in place of or as well as the existing pump,
  • Check the routing of the pipes and hoses for vulnerability, and move or protect as need be.


Quote
41 The train is also fitted with a hand brake2
, which is hydraulically operated via a
button on the control desk. However, this is fitted with an interlock to prevent its
use while the train is moving; the hand brake is designed to prevent a stationary
train from moving, not to stop a moving train. Therefore, the driver was unable
to use this desk control after the train had started rolling backwards. There is
a manual pump on the exterior of the cab that the driver can operate to apply
the hand brake, using a handle that is kept in the cab. However, this is slow to
operate and may not be fully able to stop a moving train, and the driver had to
consider the possibility that this was also inoperative. Use of the hand brake
while moving is not a documented or approved procedure, but this was the only
brake that would still have been operable, albeit it was not designed to stop a
moving train.




Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: stuving on January 14, 2018, 17:59:43

My suggestion for something that is simple enough that it might happen is this:
  • Add a hydraulic pump operated by a handwheel or lever, either in the cab or just behind it (and not requiring anything to be found and attached to make it work)
  • Connect this to the hydraulic brake system, either in place of or as well as the existing pump,
  • Check the routing of the pipes and hoses for vulnerability, and move or protect as need be.


Quote
41 The train is also fitted with a hand brake2
, which is hydraulically operated via a
button on the control desk. However, this is fitted with an interlock to prevent its
use while the train is moving; the hand brake is designed to prevent a stationary
train from moving, not to stop a moving train. Therefore, the driver was unable
to use this desk control after the train had started rolling backwards. There is
a manual pump on the exterior of the cab that the driver can operate to apply
the hand brake, using a handle that is kept in the cab. However, this is slow to
operate and may not be fully able to stop a moving train, and the driver had to
consider the possibility that this was also inoperative. Use of the hand brake
while moving is not a documented or approved procedure, but this was the only
brake that would still have been operable, albeit it was not designed to stop a
moving train.

Yes, I think I misread that bit about the "hand brake" (operated by as button?), and there was a way of applying it without getting down onto the ground (though that's not 100% clear). It also reminds me of one step I left off my list:
  • Check the brake itself and ensure it can stop a train without being so damaged as to lose its effectiveness (though being damaged in itself is not a problem).   

Of course if the current "hand" brake passes all those checks (and the RAIB report hints it might) it becomes just a matter of training.


Title: Re: Runaway maintenance train - Markinch, Fife - 17 Oct 2017
Post by: Electric train on January 14, 2018, 20:11:35
I think the key is the second of the recommendations - "The RAIB has made two recommendations to Network Rail. The first addresses the risk of a runaway being initiated by debris on the track causing multiple disruptions to the braking system on this type of train. The second recommendation addresses the possibility that similar risks might be present for other short formation trains that operate on its infrastructure".

The braking system on these trains will be made up of standard freight waggon braking components, I doubt if the distributors and reservoir's are located anywhere unusual.

.............. (Its been best part of 40 years since I did a train braking course  :o  )Train air brakes rely on a res on the individual waggon / coach to apply the brake via the brake cylinder (unlike lorry brakes that have springs and air is used to hold them off.  The distributor valve is a proportional valve that converts the drop in train pipe pressure to an applied pressure to the cylinders.

   



This page is printed from the "Coffee Shop" forum at http://gwr.passenger.chat which is provided by a customer of Great Western Railway. Views expressed are those of the individual posters concerned. Visit www.gwr.com for the official Great Western Railway website. Please contact the administrators of this site if you feel that content provided contravenes our posting rules ( see http://railcustomer.info/1761 ). The forum is hosted by Well House Consultants - http://www.wellho.net