RAIB investigation - West Wickham Incident

[millemille]

This is a well known fault in networkers. You can get interlock with a door obstructed. Happened at Huntingdon and kings cross as well. If you try to open the doors after the driver presses 'close doors' nothing happens. If you try a second time within a couple of seconds the door will open but no hustle alarm sounds and it immediately closes again which it appears is what happened here. Then when you become trapped the rubber seal strips on the doors closes around various obstructions (such as hands and coats and now apparently bags) and seals them firmly into the doors. The driver gets interlock and the train is able to take power.

I take issue with your use of the term "fault".

There is no fault with Networker doors in relation to the Huntingdon and other similar accidents, I say that having spent 20+ years building, commisioning, modifying, supporting and maintaining all EMU Networkers.

The doors of the 365/465/466 units are fully compliant with the regulations that were in place at the time of the relevant design certification. The regulations that are in place now are more rigourous with regards to obstacle detection and withdrawing forces but are not retrospectively applied.

The issue is not, as you suggest, that you obtain interlock with the door obstructed. The door is not obstructed - 1st and 2nd stage interlock is achieved and the active locks are engaged because the door leaves have gone fully home to the closed position with items less than the mininum obs detection triggering thickness - coats, fingers, bag straps etc - trapped in the door leading edge rubbers.

This would be the same for most current powered door designs, even with door rubbers having leading edge obs detection.

The issue is around the withdrawing forces required to extract the trapped items.

The Networker leading edge rubbers dove-tail into each other and once the locks on the leaves are enagaged it is very difficult to extract any trapped fabrice/material unless considerable force, well in excess of current RGS requirements, is exerted.

In the aftermath of Hunntingdon I personally carried out extraction force testing with engineers from FCC and HSBC using a variety of materials and a variety of door air operating pressures to investigate possible simple solutions.

The conclusion is without a complete redesign of the of the door rubbers and leaves it isn't possible to reduce the withdrawing forces to meet current RGS requirments and when you consider the costs of such a modification (and it wouldn't be this modification in isolation because as soon as you embark on such a modification any associated change with the modification all have to
be compliant with current RGS requirements) vs. the consequence it just doesn't make sense to carry out any modification.

Thousands of DOO despatches are made safely every day on the Networker fleets and the MTBF does not warrant modification.

I await the RAIB's report with interets to find out whether, as in every previous entrapment accident involving Networker units, the trapped MOP/passenger either ignored the hustle alarms of actively interfered with the operation of the doors.

 

[Jonfun]

To imply that it was in any way the fault of the passenger that the train drove off whilst she was inconsiderately trapped in its doors is poor. Either there was a fault with the system of dispatch rendering the driver unable to see her on the mirrors or screens on which he does his checks, or the driver saw her and didn't act on it.

Ignoring the hustle alarms, whilst likely to result in a considerable bash from the closing doors, shouldn't result in serious injury or death.

 

[najaB]

Ignoring the hustle alarms, whilst likely to result in a considerable bash from the closing doors, shouldn't result in serious injury or death.

From the description of the incident above, there may not have been any hustle alarm - which makes it even less the passenger's fault.

 

[Bletchleyite]

To imply that it was in any way the fault of the passenger that the train drove off whilst she was inconsiderately trapped in its doors is poor. Either there was a fault with the system of dispatch rendering the driver unable to see her on the mirrors or screens on which he does his checks, or the driver saw her and didn't act on it.

Ignoring the hustle alarms, whilst likely to result in a considerable bash from the closing doors, shouldn't result in serious injury or death.

Indeed, though as najaB says there may not even have been a hustle alarm. One of two things clearly happened - the driver didn't check properly before departing, or the DOO equipment was in some way inadequate or faulty and meant that it appeared to the driver that it was clear when it was not. I will be interested to hear which it was.

 

[sbt]

Well thats all right then. It doesn't matter if someone or something is trapped in the doors, just as long as they aren't (technically) 'obstructed'. Any subsequent injury is therefore nothing to do with any person or organisation concerned with the railway, or at least, if it is, then its the Guard or Driver 'not doing their job' (regardless of how difficult by the systems they are provided to work with make it).

Its also not a 'fault' if, in certain situations, pressing a door open button causes them to open, as expected, before rapidly closing, which is not what people expect, potentially leading to injury. Its meets the regulations and is therefore not a 'fault' and therefore any subsequent injury is nothing to do with any person or organisation related to the railway. Nothing to see here, nothing to fix, move right along, must be the evil passengers conspiring to injure themselves (how much better everything would run if the railway didn't have them).

Sorry, but you attitude gets my goat just as much as the attitude that any accident must be some traincrew, or some other member of railway staff, members responsibility, regardless of how difficult their job has been made. I can understand the irritation at obvious foolishness (although, to those that are not rail savy it may not seem that foolish) but, as a reasonably responsible rail traveller I have been in enough close calls to know that its not always down to the person who gets injured.

 

[FordFocus]

There might be ways to make it work.

To deal with the overlap issue the door numbers could include the 'last two' of the unit number. The probability of a number clash would be quite low and eliminated by a rule forbidding units with the same 'last two' in the the same formation on DOO lines.

To deal with the visability from an angle issue I can see two approaches. The first is to repeat the numbers and use isometric drawing to allow them to be read at an angle. This has been done with road markings. You would require at least two versions, one to be read from each end of the unit, and possibly three, with one to be read directly side on. Testing would have to establish which angle of projection to use in the alphanumerics.

The second, and probably better, approach is not to use alphanumerics at all, or to supplement them. There is no particular reason to use them as all that is needed is something that allows each door to be clearly distinguished and has an identifiable sequence. My first thought would be a system of horizontal lines, one above the other, some thin, some thick, some dashed, some dotted. Possibly other simple geometric shapes could be worked in to the system, provided they could be distinguished from an angle.

Example system, which is not a definitive proposal, but has had a bit of thought put in to it:

Door -
A B C D E F G H I J K L M N O P Q R S T U V X Y Z
- - - : : : : = = = = + + + + / / / / O O O O V V
. - - . - - - . - - - . - - - . - - - . - - - . -
# . - # . - - # . - - # . - - # . - - # . - - # .
# # . # # . - # # . - # # . - # # . - # # . -
# # # # # # . # # # . # # # . # # # . # # # .

Unit Nos -
0 1 2 3 4 5 6 7 8 9
/ X X X X X X X X X
. . / / X X X X X X
# . / # . / / X X X
# # . # # . / . / /
# # # # # # . # . /
# # # # # # # # # .

Unit 37, Door G

: : :
-----
-----
-----
. . .

X X
/ X
/ X
. .

Or


: : : # X X
----- # / X
----- # / X
----- # . .
. . .


(Grey # would not appear in symbol - its there just to get the formatting right)

You will note I have included a dot at the bottom of a symbol to aid the driver in determining if he has read the whole symbol.

Thought provoking way of numbering but I feel I need to explain a bit better on what you see on DOO monitors.

Most DOO cameras see the side of the train but the angle is too shallow and even the image isn't clear enough to see the web address of the TOC or any other decals. A driver is sat in a cab and can be several feet away from the monitors that maybe at an angle slightly. It is most certainly not like been sat in front of a computer. It could also be very, very costly to start moving cameras about on platforms.

Next it can be affected by the open elements. Certain days drivers have to get out of the cab because the DOO monitors have nothing on them because they are wiped out by direct sunlight. Heavy rain, mist or has water droplets on them.

Then you get to the equipment itself, some aren't in HD, monitors can be very dim or too white to see, cameras can be moved by strong wind or vandals or the equipment doesn't work full stop.

Ultimately nothing beats a human eye stood on the platform dispatching a train. Let's call them 'Guards'.

 

[RJ]

This door issue is not the only potentially unsafe thing about Southeastern's Networkers. I've seen some of them rolling back after the doors have been released at Denmark Hill station.

That sounds like a major design/programming fault. I hope the passenger is recovering well.

It would have been a horrific incident - if dragged under a train, the chances of remaining clear of the rail before the train has finished passing over it are probably quite remote.

and DOO is safe because ............. luck ??

I'd await the conclusion of the investigation before criticising the staff or procedure in place. Was it down to the procedure in place being unsafe, or for whatever reason, the procedure in place not being correctly followed?

Well you clearly haven't read the raib prelim report which states the driver didn't see anyone trapped so no, certainly doesn't appear as someone not doing their job properly but feel free to do your normal thing of blindly blaming the staff without having a clue what actually happend!

To be fair, it says the driver didn't see anyone. It didn't state whether or not he/she visually observed the length of the train between closing the doors and moving the train. If they didn't look, that might have been a reason the lady wasn't seen. If the monitors/mirrors are insufficient to see the whole train and there are no platform staff, what is the procedure for dispatch?

 

[A-driver]

I take issue with your use of the term "fault".



There is no fault with Networker doors in relation to the Huntingdon and other similar accidents, I say that having spent 20+ years building, commisioning, modifying, supporting and maintaining all EMU Networkers.



The doors of the 365/465/466 units are fully compliant with the regulations that were in place at the time of the relevant design certification. The regulations that are in place now are more rigourous with regards to obstacle detection and withdrawing forces but are not retrospectively applied.



The issue is not, as you suggest, that you obtain interlock with the door obstructed. The door is not obstructed - 1st and 2nd stage interlock is achieved and the active locks are engaged because the door leaves have gone fully home to the closed position with items less than the mininum obs detection triggering thickness - coats, fingers, bag straps etc - trapped in the door leading edge rubbers.



This would be the same for most current powered door designs, even with door rubbers having leading edge obs detection.



The issue is around the withdrawing forces required to extract the trapped items.



The Networker leading edge rubbers dove-tail into each other and once the locks on the leaves are enagaged it is very difficult to extract any trapped fabrice/material unless considerable force, well in excess of current RGS requirements, is exerted.



In the aftermath of Hunntingdon I personally carried out extraction force testing with engineers from FCC and HSBC using a variety of materials and a variety of door air operating pressures to investigate possible simple solutions.



The conclusion is without a complete redesign of the of the door rubbers and leaves it isn't possible to reduce the withdrawing forces to meet current RGS requirments and when you consider the costs of such a modification (and it wouldn't be this modification in isolation because as soon as you embark on such a modification any associated change with the modification all have to

be compliant with current RGS requirements) vs. the consequence it just doesn't make sense to carry out any modification.



Thousands of DOO despatches are made safely every day on the Networker fleets and the MTBF does not warrant modification.



I await the RAIB's report with interets to find out whether, as in every previous entrapment accident involving Networker units, the trapped MOP/passenger either ignored the hustle alarms of actively interfered with the operation of the doors.

Well as I said earlier, the hustle alarm didn't sound as the system on networkers means that once the driver presses close doors, if a passenger tries to open a door by pressing door open more than once within a couple of seconds the doors open and then immediately shut with no hustle alarm which it appears happened here. And it does happen, a few colleagues at work have tried it in the sidings after this report came out.

As for the rubber seals, yes they are compliant but they still are a strong factor in incidents like this and Huntingdon. Almost without doubt Huntingdon would not have happened with a 317 as the coat could easily be pulled free. If you trap you hand in the doors of a 317/321/313 you can't get interlock and can easily pull the hand out with minimal force. However when the lady tapped her hand in the door of a 365 at kings x the other year the train got interlock and she was unable to free her hand.

So it may not be a fault and it may meet standards but it still creates a much higher risk with networkers than other units.

 

[cyclebytrain]

Well as I said earlier, the hustle alarm didn't sound as the system on networkers means that once the driver presses close doors, if a passenger tries to open a door by pressing door open more than once within a couple of seconds the doors open and then immediately shut with no hustle alarm which it appears happened here. And it does happen, a few colleagues at work have tried it in the sidings after this report came out.

I've seen this (or an extremely similar scenario) happen on a 395 too (door open button pushed, door started to open, front wheel of the bike pushed through the gap, door closed on it with no hustle alarm). So presumably even the current standards allow this?

 

[sbt]

Perhaps what we have here is a (misguided) 'feature' intended to allow people to free something that is trapped?

 

[hairyhandedfool]

If I understand correctly, the open button has to be pressed more than once after the driver presses "doors close". If the passenger pressed before that it would open on the first attempt and the hustle alarm should start when the "door close" button is pressed. So, if the passenger opened the door after the driver pressed "doors close" it stands to reason that they must have heard atleast some of the hustle alarm. An exception might be a fault with the hustle alarm.

 

[najaB]

If I understand correctly, the open button has to be pressed more than once after the driver presses "doors close". If the passenger pressed before that it would open on the first attempt and the hustle alarm should start when the "door close" button is pressed. So, if the passenger opened the door after the driver pressed "doors close" it stands to reason that they must have heard atleast some of the hustle alarm. An exception might be a fault with the hustle alarm.

I don't know about the unit involved, but it was posted above that on some units the hustle alarm doesn't sound if the door hasn't been opened.

 

[hairyhandedfool]

Oh okay. Must have missed that.

 

[hassaanhc]

I don't know about the unit involved, but it was posted above that on some units the hustle alarm doesn't sound if the door hasn't been opened.

That is the case with the Networkers (definitely 165, 166, 465 and 466, not sure about the 365).
The 455s also have the open button lit while the hustle alarm is sounding (remember their original door buttons did not have lights incorporated), but on those it sounds at all doors regardless of whether the set of doors have been opened or not.

 

[A-driver]

That is the case with the Networkers (definitely 165, 166, 465 and 466, not sure about the 365).

The 455s also have the open button lit while the hustle alarm is sounding (remember their original door buttons did not have lights incorporated), but on those it sounds at all doors regardless of whether the set of doors have been opened or not.

Same with 365s. Not sure about the modified unit with different door sounds.

 

[Bletchleyite]

That is the case with the Networkers (definitely 165, 166, 465 and 466, not sure about the 365).
The 455s also have the open button lit while the hustle alarm is sounding (remember their original door buttons did not have lights incorporated), but on those it sounds at all doors regardless of whether the set of doors have been opened or not.

Yes, it's common on units where the alarms are a retrofit for them to sound at all doors but the buttons remain illuminated. It's the button remaining usable[1] (whether illuminated or not) but the alarm not sounding, as is being suggested here, that is a potentially dangerous flaw, as the passenger is possibly receiving no indication whatsoever that their actions are potentially dangerous.

[1] Yes, it seems you have to press it twice, but if you watch passengers generally they tend to press the button multiple times anyway, so that is of little actual relevance.

 

[RJ]

On Southern's 455s the door will open if the button is pressed when illuminated. Generally there won't be enough time for them to open fully before shutting again once the hustle alarm has started. Still a potential hazard if for any reason, a person can't hear the alarm.

 

[sbt]

I have to say that some of the reaction here has reminded me of the late 80s - early 90s issues with Slam Door locks. The BR line was that the problem was down to passenger misuse, despite at least one of the fatalities being a Guard, and that line was strongly adhered to.

In the end the HSE did an enquiry. Whilst they concluded that drink was a factor in 'a majority of incidents' (ie more than 50%) their survey of incidents in 1991 concluded that in only 13% of incidents was the passenger at fault. Taking a generous view towards BR thats probably an upper limit of 30% of incidents. That leaves approximately 13 deaths and 11 injuries (and possibly up to 16 deaths) per year that were NOT down to passenger misuse that BR were in denial over (hence the introduction of CDL on HSTs).

The conclusion was that, as well as trains with doors departing 'on the catch', tolerances, wear, and maintenance practises meant that it was possible for stock to depart with doors flush with the train side but nothing but the fit of the door in its doorway holding it closed. In that situation anybody leaning on it, as was, and still is, routine in rammed HST vestibules, or when people want to enter the HST toilet, would lead to the door opening and a strong likelihood of someone falling out.

Report, for those interested.

http://www.railwaysarchive.co.uk/documents/HSE_TrainDoors1993.pdf

My point? Yes it sounds like passengers have to carry out a specific set of actions to get this issue to occur but those actions are predictable and there is no advice against them. Plus this is a less dangerous fault that is not cause anything near the number of deaths and injuries door faults were. But to argue that this is not a 'fault' seems an exercise in semantics and denial to me. However, and this is why I brought the Door Lock issue up, that sort of reaction has a history and is probably as predictable as passengers actions when presented with a lit door button. Given that that sort of reaction has a history, everybody who has an interest in rail safety needs to guard against it, despite the number of times the public/press etc. DO blame 'The Railway' for things outside its control.

 

[ComUtoR]

This door issue is not the only potentially unsafe thing about Southeastern's Networkers. I've seen some of them rolling back after the doors have been released at Denmark Hill station.

Not a Networker fault. That was Driver error and disciplinary action was taken.

Hopefully this is just a one off incident! But it's always a concern when anything like this happens.

Not the first and neither will it be the last

The issue is not, as you suggest, that you obtain interlock with the door obstructed. The door is not obstructed - 1st and 2nd stage interlock is achieved and the active locks are engaged because the door leaves have gone fully home to the closed position with items less than the mininum obs detection triggering thickness - coats, fingers, bag straps etc - trapped in the door leading edge rubbers.

Well said and thanks for your informative post.

Well thats all right then. It doesn't matter if someone or something is trapped in the doors, just as long as they aren't (technically) 'obstructed'.

What would prefer ? If there wasn't an allowance for minimal obstruction then the force required to hold the doors closed would most likely crush a persons hand/fingers. There has to be some level of tolerance.

I would suggest that what needs to happen is that there is some form of obstacle detection.

However, never mind, I'll just sit and watch another RED reconstruction at my next safety brief telling me the same things as when this last happened on another DOO train and nothing will change.

Considering the post quoted below I fully believe that is exactly what will happen.

The conclusion is without a complete redesign of the of the door rubbers and leaves it isn't possible to reduce the withdrawing forces to meet current RGS requirments and when you consider the costs of such a modification (and it wouldn't be this modification in isolation because as soon as you embark on such a modification any associated change with the modification all have to
be compliant with current RGS requirements) vs. the consequence it just doesn't make sense to carry out any modification.

Thousands of DOO despatches are made safely every day on the Networker fleets and the MTBF does not warrant modification.

To imply that it was in any way the fault of the passenger that the train drove off whilst she was inconsiderately trapped in its doors is poor...

Ignoring the hustle alarms, whilst likely to result in a considerable bash from the closing doors, shouldn't result in serious injury or death.

I await the RAIB's report with interets to find out whether, as in every previous entrapment accident involving Networker units, the trapped MOP/passenger either ignored the hustle alarms or actively interfered with the operation of the doors.

I can fully attest that passenger completely ignore the hustle alarm. The absolute worst I have witnessed was a Lady who jammed her pushchair (child included) into a set of closing doors.

Ignoring safety systems can and will result in serious injury or death. In this case the hustle alarm may or may not have sounded. I say may not because investigation has proven the issue with the hustle alarm but only the passenger and those in the specified coach can fully state categorically that it did not sound at the time of the incident.

One of two things clearly happened - the driver didn't check properly before departing, or the DOO equipment was in some way inadequate or faulty and meant that it appeared to the driver that it was clear when it was not. I will be interested to hear which it was.

Have you considered a 3rd or potentially a 4th scenario ?

Everything worked as expected and the Driver performed their duties correctly but by random chance and a horrible twist of fate the ACCIDENT occured.

Or

There may or may not have been a blind spot. Again from personal experience there is a point in which a passenger will disappear from the DOO monitor or is about to board but isn't quite in sight just yet. You hit the button, check the signal, obtain interlock, take power....

Not only that you have to consider that once you have taken power its game over nothing is going to then stop you other than a passcom being pulled (you must stop immediately when in a station)

As others have said, The report will reveal a much more precise level of detail.

 

[martynbristow]

Have you considered a 3rd or potentially a 4th scenario ?

Everything worked as expected and the Driver performed their duties correctly but by random chance and a horrible twist of fate the ACCIDENT occured.

Or

There may or may not have been a blind spot. Again from personal experience there is a point in which a passenger will disappear from the DOO monitor or is about to board but isn't quite in sight just yet. You hit the button, check the signal, obtain interlock, take power....

Not only that you have to consider that once you have taken power its game over nothing is going to then stop you other than a passcom being pulled (you must stop immediately when in a station)

As others have said, The report will reveal a much more precise level of detail.

It states her bag was caught. I don't have an expert knowledge of ladies bags but potentially it may have been too small to see and it appeared she was leaving the train. The report will show in more detail soon.

 

[Bletchleyite]

Everything worked as expected and the Driver performed their duties correctly but by random chance and a horrible twist of fate the ACCIDENT occured.

There is no such thing as an accident. A safety issue is always caused by something (a procedure, a physical item, a person...) being deficient - quite possibly something nobody thought of. It may be decided that the accident rate/risk is acceptable, and that level of safety issue will be tolerated, of course.

There is *always* a cause. And that's what the RAIB will establish.

There may or may not have been a blind spot. Again from personal experience there is a point in which a passenger will disappear from the DOO monitor or is about to board but isn't quite in sight just yet. You hit the button, check the signal, obtain interlock, take power....

That would be an example of there being a deficiency in the DOO equipment or procedures. A risk assessment/cost benefit analysis may determine that that deficiency is acceptable (in such things, a life does have a specified financial value, hard though that may be for most people to accept), but it is still a deficiency, just one that is determined not to be cost-effective to solve.

Not only that you have to consider that once you have taken power its game over nothing is going to then stop you other than a passcom being pulled (you must stop immediately when in a station)

Again a possible deficiency (in guard operation with a droplight, the train can be watched out of the platform).

As others have said, The report will reveal a much more precise level of detail.

Indeed, I look forward to reading it.

 

[Jonfun]

The hustle alarm is a safety system to prevent people being clouted by a closing door, not to prevent them being dragged under the train. The safety system to prevent that is a safe method of train dispatch correctly performed. For whatever reason, this safety system failed - it may be driver error but it might equally be faulty DOO equipment.

 

[bramling]

There is no such thing as an accident. A safety issue is always caused by something (a procedure, a physical item, a person...) being deficient - quite possibly something nobody thought of. It may be decided that the accident rate/risk is acceptable, and that level of safety issue will be tolerated, of course.

There is *always* a cause. And that's what the RAIB will establish.



That would be an example of there being a deficiency in the DOO equipment or procedures. A risk assessment/cost benefit analysis may determine that that deficiency is acceptable (in such things, a life does have a specified financial value, hard though that may be for most people to accept), but it is still a deficiency, just one that is determined not to be cost-effective to solve.



Again a possible deficiency (in guard operation with a droplight, the train can be watched out of the platform).



Indeed, I look forward to reading it.

I agree up to a point, however *no* method of despatch is guaranteed to be completely infallible, especially with a long train on a crowded platform.

I think it's true to say that the act of getting on or off the train is probably the part of the journey (within railway property) that is statistically the part where the passenger is exposed to the most risk. I believe the only other risk that comes close is if they should use an escalator.

The dispatch process and the rolling stock design is arranged to mitigate against this risk as far as reasonably practicable. Whilst this may not hold true for the incident in question as it seems the rolling stock design probably played a part, possibly combined with some issues with the dispatch method, it surprises me how little attention the railway pays to trying to influence passenger behaviour when using train doors. From my experience investigating such incidents, most door injuries (mainly minor), people striking moving trains, or incidents of dragging (either people or property) are caused either when the passenger has tried to board or leave a train whilst the doors are closing, or due to their actions being influenced by alcohol. The risk of the incident not being picked up by the driver increases if the platform edge area is heavily crowded.

In my view the way to prevent most platform/train interface incidents is to have a culture where passengers do not try and board or leave trains when the doors are closing. Whether this is realistic in the real world is another matter, however it sounds like in the West Wickham incident the passenger would not have been in a position of leaving the train with the doors closing had it not been for the described design 'feature' of the Networker train.

Spend some time on any Tube platform in central London, and you won't have to wait long before you see someone trying to board when the doors are closing - often when the next train is just 1 minute behind. Apart from the consequences to the service, passengers should be made fully aware of the risk they are introducing to themselves. For starters they should be fully 'educated' about this when they make complaints about being caught in doors, not just given a fluffy 'customer service' response.

 

[martynbristow]

The hustle alarm is a safety system to prevent people being clouted by a closing door, not to prevent them being dragged under the train. The safety system to prevent that is a safe method of train dispatch correctly performed. For whatever reason, this safety system failed - it may be driver error but it might equally be faulty DOO equipment.

The hustle alarm is there to prevent door conflicts which will reduce such incidents. They are not independent situations. The RAIB are right to point out theres room for improvement. Improving passenger preparedness takes a lot.

I agree up to a point, however *no* method of despatch is guaranteed to be completely infallible, especially with a long train on a crowded platform.

I think it's true to say that the act of getting on or off the train is probably the part of the journey (within railway property) that is statistically the part where the passenger is exposed to the most risk. I believe the only other risk that comes close is if they should use an escalator.

The dispatch process and the rolling stock design is arranged to mitigate against this risk as far as reasonably practicable. Whilst this may not hold true for the incident in question as it seems the rolling stock design probably played a part, possibly combined with some issues with the dispatch method, it surprises me how little attention the railway pays to trying to influence passenger behaviour when using train doors. From my experience investigating such incidents, most door injuries (mainly minor), people striking moving trains, or incidents of dragging (either people or property) are caused either when the passenger has tried to board or leave a train whilst the doors are closing, or due to their actions being influenced by alcohol. The risk of the incident not being picked up by the driver increases if the platform edge area is heavily crowded.

In my view the way to prevent most platform/train interface incidents is to have a culture where passengers do not try and board or leave trains when the doors are closing. Whether this is realistic in the real world is another matter, however it sounds like in the West Wickham incident the passenger would not have been in a position of leaving the train with the doors closing had it not been for the described design 'feature' of the Networker train.

Spend some time on any Tube platform in central London, and you won't have to wait long before you see someone trying to board when the doors are closing - often when the next train is just 1 minute behind. Apart from the consequences to the service, passengers should be made fully aware of the risk they are introducing to themselves. For starters they should be fully 'educated' about this when they make complaints about being caught in doors, not just given a fluffy 'customer service' response.

Defiantly the platform interface is a serious risk but nothing has been done to deal with it.
Having been elsewhere the train platform interface here needs a lot of work! The gaps at some stations are huge.
This morning I had a problem getting on the train because the guy infront of me decided to hang around the door way and I nearly slipped on the step :/
The layout of the train and overcapacity is a problem. Getting on and off trains with short stop times and heavy loadings makes these situations worse.

 

[bramling]

The hustle alarm is there to prevent door conflicts which will reduce such incidents. They are not independent situations. The RAIB are right to point out theres room for improvement. Improving passenger preparedness takes a lot.

Defiantly the platform interface is a serious risk but nothing has been done to deal with it.
Having been elsewhere the train platform interface here needs a lot of work! The gaps at some stations are huge.
This morning I had a problem getting on the train because the guy infront of me decided to hang around the door way and I nearly slipped on the step :/
The layout of the train and overcapacity is a problem. Getting on and off trains with short stop times and heavy loadings makes these situations worse.

There are a few measures that passengers can take that will greatly reduce their chances of being involved in a "platform/train interface" incident.

1) Don't attempt to board or alight from a train whilst the doors are closing, or the chimes/alarm is sounding. This is the main precursor to most incidents.

2) If alighting from a train, be mindful that the driver can't see inside the train. In most situations this should not present a problem, but might if you are late getting up, or if you are encumbered in some way, or if someone else has blocked your path to the door.

3) Once off the train, move as far away from the train as possible.

4) Don't try and board a train where there is clearly not room for you to fit on. If unable to board, step well back from the train.

5) If waiting on a platform for a subsequent service, stand as far back from the train as possible. Don't let the desire to 'hold' a spot put you in a hazardous situation. Standing too close also blocks the driver or guards view down the side of the train.

6) Make sure your belongings are secure, and not likely to be knocked or blown out of your hands.

7) Finally, if you have special requirements then speak to staff first, and where possible try and position yourself at the front of the train so the driver can see you. Even if you make eye contact with the driver, don't assume he understands what your special needs may be. I've investigated a few staff or wheelchairs get caught in doors because people have assumed the driver knows what they're planning to do, without having had a conversation and come to a full understanding.

 

[talltim]

I take issue with your use of the term "fault".

There is no fault with Networker doors in relation to the Huntingdon and other similar accidents, I say that having spent 20+ years building, commisioning, modifying, supporting and maintaining all EMU Networkers.

The doors of the 365/465/466 units are fully compliant with the regulations that were in place at the time of the relevant design certification. The regulations that are in place now are more rigourous with regards to obstacle detection and withdrawing forces but are not retrospectively applied.

The issue is not, as you suggest, that you obtain interlock with the door obstructed. The door is not obstructed - 1st and 2nd stage interlock is achieved and the active locks are engaged because the door leaves have gone fully home to the closed position with items less than the mininum obs detection triggering thickness - coats, fingers, bag straps etc - trapped in the door leading edge rubbers.

This would be the same for most current powered door designs, even with door rubbers having leading edge obs detection.

The issue is around the withdrawing forces required to extract the trapped items.

The Networker leading edge rubbers dove-tail into each other and once the locks on the leaves are enagaged it is very difficult to extract any trapped fabrice/material unless considerable force, well in excess of current RGS requirements, is exerted.

In the aftermath of Hunntingdon I personally carried out extraction force testing with engineers from FCC and HSBC using a variety of materials and a variety of door air operating pressures to investigate possible simple solutions.

The conclusion is without a complete redesign of the of the door rubbers and leaves it isn't possible to reduce the withdrawing forces to meet current RGS requirments and when you consider the costs of such a modification (and it wouldn't be this modification in isolation because as soon as you embark on such a modification any associated change with the modification all have to
be compliant with current RGS requirements) vs. the consequence it just doesn't make sense to carry out any modification.

Thousands of DOO despatches are made safely every day on the Networker fleets and the MTBF does not warrant modification.

I await the RAIB's report with interets to find out whether, as in every previous entrapment accident involving Networker units, the trapped MOP/passenger either ignored the hustle alarms of actively interfered with the operation of the doors.

I'd say there was a fault, not with the functioning of the doors and associated equipment against the spec, but with the spec. Otherwise, why would you have even had to look into it?
I'd also say that you may be putting HSBC/Train operators in a dangerous reputational (and maybe legal) position, by saying that the issue (I won't call it a fault) with doors on Networkers was known about, and that a solution was possible, but was discounted as being too expensive, especially considering how unpopular banks and train operators are with seemingly all the press, right or left wing.
--- old post above --- --- new post below ---

There are a few measures that passengers can take that will greatly reduce their chances of being involved in a "platform/train interface" incident.

1) Don't attempt to board or alight from a train whilst the doors are closing, or the chimes/alarm is sounding. This is the main precursor to most incidents.

2) If alighting from a train, be mindful that the driver can't see inside the train. In most situations this should not present a problem, but might if you are late getting up, or if you are encumbered in some way, or if someone else has blocked your path to the door.

3) Once off the train, move as far away from the train as possible.

4) Don't try and board a train where there is clearly not room for you to fit on. If unable to board, step well back from the train.

5) If waiting on a platform for a subsequent service, stand as far back from the train as possible. Don't let the desire to 'hold' a spot put you in a hazardous situation. Standing too close also blocks the driver or guards view down the side of the train.

6) Make sure your belongings are secure, and not likely to be knocked or blown out of your hands.

7) Finally, if you have special requirements then speak to staff first, and where possible try and position yourself at the front of the train so the driver can see you. Even if you make eye contact with the driver, don't assume he understands what your special needs may be. I've investigated a few staff or wheelchairs get caught in doors because people have assumed the driver knows what they're planning to do, without having had a conversation and come to a full understanding.

Don't press the Open button and expect the doors to remain open long enough for you to get out...

 

[martynbristow]

There are a few measures that passengers can take that will greatly reduce their chances of being involved in a "platform/train interface" incident.

1) Don't attempt to board or alight from a train whilst the doors are closing, or the chimes/alarm is sounding. This is the main precursor to most incidents.

2) If alighting from a train, be mindful that the driver can't see inside the train. In most situations this should not present a problem, but might if you are late getting up, or if you are encumbered in some way, or if someone else has blocked your path to the door.

3) Once off the train, move as far away from the train as possible.

4) Don't try and board a train where there is clearly not room for you to fit on. If unable to board, step well back from the train.

5) If waiting on a platform for a subsequent service, stand as far back from the train as possible. Don't let the desire to 'hold' a spot put you in a hazardous situation. Standing too close also blocks the driver or guards view down the side of the train.

6) Make sure your belongings are secure, and not likely to be knocked or blown out of your hands.

7) Finally, if you have special requirements then speak to staff first, and where possible try and position yourself at the front of the train so the driver can see you. Even if you make eye contact with the driver, don't assume he understands what your special needs may be. I've investigated a few staff or wheelchairs get caught in doors because people have assumed the driver knows what they're planning to do, without having had a conversation and come to a full understanding.

As much as I agree with you, I cannot say that most customers understand this :/
Cluttering in door ways is becoming a big problem

 

[Peter Mugridge]

The hustle alarm is there to prevent door conflicts which will reduce such incidents.

Travelling on 455 706 this afternoon, it now makes a separate alarm signal when the doors are cut in, so we now seem to have alarms for both the doors opening and closing... While the sounds are different for each purpose, this does seem to me to be introducing an avoidable potential for confusion which may in time result in people ignoring the door closure alarms?

 

[cjmillsnun]

There are a few measures that passengers can take that will greatly reduce their chances of being involved in a "platform/train interface" incident.

1) Don't attempt to board or alight from a train whilst the doors are closing, or the chimes/alarm is sounding. This is the main precursor to most incidents.

2) If alighting from a train, be mindful that the driver can't see inside the train. In most situations this should not present a problem, but might if you are late getting up, or if you are encumbered in some way, or if someone else has blocked your path to the door.

3) Once off the train, move as far away from the train as possible.

4) Don't try and board a train where there is clearly not room for you to fit on. If unable to board, step well back from the train.

5) If waiting on a platform for a subsequent service, stand as far back from the train as possible. Don't let the desire to 'hold' a spot put you in a hazardous situation. Standing too close also blocks the driver or guards view down the side of the train.

6) Make sure your belongings are secure, and not likely to be knocked or blown out of your hands.

7) Finally, if you have special requirements then speak to staff first, and where possible try and position yourself at the front of the train so the driver can see you. Even if you make eye contact with the driver, don't assume he understands what your special needs may be. I've investigated a few staff or wheelchairs get caught in doors because people have assumed the driver knows what they're planning to do, without having had a conversation and come to a full understanding.

Hmmm. In regards to number 4. I have started to walk away from a full train at New Street (I was going to wait for the next one), only to be pushed on by platform staff.

 

[Bletchleyite]

What? Did you report the assault to Police?