Incident at Wandsworth Common 07/08/16

[yorksrob]

That will make it somewhat interesting when working a HST...

Quite !

 

[jopsuk]

That will make it somewhat interesting when working a HST...

Rather depends how quickly such an order is brought in. a simple "in the next four years" perhaps?

 

[SpacePhoenix]

I doubt that the drop light would be sealed shut as afaik that's where the guard dispatches the train from and the guard has to be able to see down the train as it's pulling out of a station

 

[jopsuk]

I doubt that the drop light would be sealed shut as afaik that's where the guard dispatches the train from and the guard has to be able to see down the train as it's pulling out of a station

Depends what stock you're talking about. With 442s they're operating without guards.

Plus, you could have a droplight that required a guard's key to open.

 

[Hadders]

If you seal the droplights on HSTs how are passengers going to open the doors?

You could fit bars across the windows but given the HSTs are being phased out in the next few years it hardly seems worthwhile.

 

[theironroad]

Depends what stock you're talking about. With 442s they're operating without guards.

Plus, you could have a droplight that required a guard's key to open.

Except Monday's ROG 442 tour where the guard was wearing a 3 piece suit complete with a watch chain and a black bowler hat.......lol

 

[causton]

The guard has to be able to see down the train as it's pulling out of a station

These days that is impossible on pretty much any modern stock. Sure, it is nice to see out but by no means necessary.

 

[superkev]

from an earlier reply on this thread. Note the bit about clearances for opening Windows..

As a track engineer, I can assist:

Normally, we try to provide 'full normal' clearances, where everything would be at least 1.625mm away from the nearest running edge.
This can be reduced to 1.624m (yep, 1mm different, great standards!) for things less than 2m long and signals, to 1.470m for OLE masts, and 1.364m for signals located between two tracks with insufficient space for normal clearances.

If clearances are reduced below those levels, we then go into calculations of each train vs each piece of infrastructure so we go into the lists of stock that are 'cleared' for a particular route.

We would try to provide a normal minimum clearance, from a train to a fixed anything, of 100mm.
This sounds low, but has some caveats:
1) This is from the kinematic envelope of the train, not from the static profile. In other words, we use software to determine the train's maximum ever size, taking into account all possibilities such as suspension failed, tilt (where fitted) failed, crush laden, empty, in high winds, full range of speeds etc etc - all things which will affect, ever so slightly, where the train will be in any particular passage. There are all added up and a shape produced for each train that is then used for clearance calculations on each pice of track. It's done for each structure, so takes into account the radius, cant and speed.
2) It's less than the normal clearances as given above, and thus requires specific approval at design stage before it's permitted. However, gaining approval is relatively easy and there are a huge number of instances of these clearances in existence all over the network.
3) The software has a lot of tolerances built in, and it's extremely unlikely that they will all ever happen together, so the actual clearance provided is often 30-50mm+ greater than the number that pops out of the software.

There is also a further requirement that, where trains with openable windows run, this be increased to at least 450mm at the height of the window to allow for persons leaning out. This is not mandated though, only a 'where possible'.

If we can't provide that, we can reduce this to a 'reduced clearance' of 50mm. This requires approval that is harder to get, and must be accompanied by a risk assessment and possibly mitigation measures, such as provision of enhanced track and structure monitoring to ensure movements over time do not compromise this further.

If we are still struggling, we can reduce this to 'special reduced clearance' of 25mm. This is hard to get approved and is nearly always only approved on a temporary basis with good risk mitigation measures in place, such as it being a concrete structure next to slab track.

How this relates to when window bars are provided I'm not sure, but I think it's up to each TOC. It's easy for the TOCs to see what clearances exist on any particular route they are planning to run trains along so I think it's up to them to do their own risk assessments of how many tight clearances they will pass, how fast they will be going, how visible they are etc. I'll leave someone from a rolling stock background to clarify that though - I may be wrong.

 

[talltim]

That will make it somewhat interesting when working a HST...

You could provide a chauffeur service, opening all the doors for the passengers. Keep you out of trouble...

 

[Flamingo]

You could provide a chauffeur service, opening all the doors for the passengers. Keep you out of trouble...

It would take more than that to keep me out of trouble
(Anyway, trouble comes looking for me! )

 

[QueensCurve]

has any train ever come into contact with the gantry?

I don;t in all honesty think that is the question.

Rather, at the time the gantry was installed, there was stock with droplights operating. Probably much more so than now.

It would be a reasonably foreseeable event for someone to lean out of the droplight window. Even if only easing the head out to have a sly peak at the signal aspect, as I have done thousands of times, it becomes a high probability of a fatality at that location.

As such the design of the infrastruction can be questioned.

 

[tsr]

Even if only easing the head out to have a sly peak at the signal aspect, as I have done thousands of times, it becomes a high probability of a fatality at that location.

Perhaps it does, but it's interesting you bring that up because I tend to find I don't need to stick my head out very far at all to see the signal (where any curve of the track actually permits it, which in fact on the route in question can be quite limited). You'd only just notice my nose poking out of the window if you were actively looking that way, it's hardly the case that my head would ever be fully out of the window.

What really has made me wonder in this case is why you would want to have your head out of the window at that point. It's not even the nicest side of the train to be looking out, the chances are that it was running on green or double yellow aspects under normal running there, and you're not likely to be accelerating particularly rapidly or travelling at the highest speed for the GX route - it's just a routine run at about 70ish mph through suburban South London. One of the only times those factors would change is if the train or another in front of it was using a crossover at Balham; on the day in question, at that time, it would have been unlikely. I suppose it could have been accelerating from a controlled signal at Norbury or Streatham North Jn but I couldn't be sure.

 

[QueensCurve]

If you seal the droplights on HSTs how are passengers going to open the doors?

As noted above, you just take out 4 bolts securing the plate over the interiaor door handle. The bolt to secure the droplight already exists. TOCs planning to make futures use of HSTs have a more expensive (if not more boring ) option of replacing the doors with automatic ones.

You could fit bars across the windows but given the HSTs are being phased out in the next few years it hardly seems worthwhile.

Some HSTs are likely to be in use for a while yet.
--- old post above --- --- new post below ---

Perhaps it does, but it's interesting you bring that up because I tend to find I don't need to stick my head out very far at all to see the signal (where any curve of the track actually permits it, which in fact on the route in question can be quite limited). You'd only just notice my nose poking out of the window if you were actively looking that way, it's hardly the case that my head would ever be fully out of the window.

Nor I. I consider myself very careful, but I remain shocked that a signal gantry could be within 260mm (10.4" for Leave voters ) of the moving train.

I don't think in my extensive travels by rail in Britain I have ever knowingly been aware of any signal gantry or OHLE structure that approaches anywhere near that close. Had I been aware that some such existed I would have been much more reticent in using the droplights. :cry:

What really has made me wonder in this case is why you would want to have your head out of the window at that point. It's not even the nicest side of the train to be looking out, the chances are that it was running on green or double yellow aspects under normal running there, and you're not likely to be accelerating particularly rapidly or travelling at the highest speed for the GX route - it's just a routine run at about 70ish mph through suburban South London. One of the only times those factors would change is if the train or another in front of it was using a crossover at Balham; on the day in question, at that time, it would have been unlikely. I suppose it could have been accelerating from a controlled signal at Norbury or Streatham North Jn but I couldn't be sure.

Personally I probably wouldn't have either - and that might be part of the misadventure in this case.

He was however making the sensible precaution of looking out on the side of the train away from oncoming traffic and almost certainly did not, as I would have not, expect a signal gantry to present such a danger.

 

[Sunset route]

Depends what stock you're talking about. With 442s they're operating without guards.

Plus, you could have a droplight that required a guard's key to open.

Apart from the Dailey M-F Eastbourne to London Bridge and vice versa commute scoot, there may be others.

 

[najaB]

Even if only easing the head out to have a sly peak at the signal aspect, as I have done thousands of times, it becomes a high probability of a fatality at that location.

As such the design of the infrastruction can be questioned.

I suppose it depends on your definition of 'high' but, given than none occurred in the years between it being installed and this unfortunate incident, I can't agree that it was high. Higher than average, maybe. Higher than ideal, definitely.

 

[QueensCurve]

I suppose it depends on your definition of 'high' but, given than none occurred in the years between it being installed and this unfortunate incident, I can't agree that it was high. Higher than average, maybe. Higher than ideal, definitely.

It has happened now, at Wandsworth common rather than anything else. By "high" I meant in the lifetime of the gantry.

We must look forwards, not back but who on earth decided to install it so close? And Why?

 

[SpacePhoenix]

Will there be any 442s left in service by the time the RAIB publishes its report? Once the 442s are gone will there be anything left running on the route where a passenger could lean far enough out a window to injure themselves?

 

[sarahj]

Apart from the Dailey M-F Eastbourne to London Bridge and vice versa commute scoot, there may be others.

Plus the 442's that run between Gatwick and Brighton in peak times.

It is funny when a poster tells me the train I was working not more than an hour ago does not have a guard.

 

[bramling]

As noted above, you just take out 4 bolts securing the plate over the interiaor door handle. The bolt to secure the droplight already exists. TOCs planning to make futures use of HSTs have a more expensive (if not more boring ) option of replacing the doors with automatic ones.



Some HSTs are likely to be in use for a while yet.
--- old post above --- --- new post below ---


Nor I. I consider myself very careful, but I remain shocked that a signal gantry could be within 260mm (10.4" for Leave voters ) of the moving train.

I don't think in my extensive travels by rail in Britain I have ever knowingly been aware of any signal gantry or OHLE structure that approaches anywhere near that close. Had I been aware that some such existed I would have been much more reticent in using the droplights. :cry:



Personally I probably wouldn't have either - and that might be part of the misadventure in this case.

He was however making the sensible precaution of looking out on the side of the train away from oncoming traffic and almost certainly did not, as I would have not, expect a signal gantry to present such a danger.

Regardless of the sensitivities, the fact remains that he chose to lean out, disregarding a clear and unambiguous warning sign in the process.

Now I will say at this point that I have at times enjoyed the ability to lean out of a droplight window, and still do occasionally, but in so doing I knowingly accept that there is a non-zero risk that something could go wrong. Personally I'd say it's safer to be leaning out on the non-cess side, as although you have trains passing, there's less chance of lineside structures and, worse, overhanging vegetation. I wouldn't lean out more than a couple of inches beyond the side of the train, and never would look anywhere other than forwards.

It's all very well RAIB investigating, but this investigation will be costing public money. There were already mitigations in place which had they been complied with the sad accident wouldn't have happened.

 

[mr_moo]

We must look forwards, not back but who on earth decided to install it so close? And Why?

It will have been an agreement between the signalling engineer, the track engineer and any other engineers with stuff nearby. I do not not know the area but things potentially to be considered are:
- The clearance to trains on the opposite side of the gantry (if there is another track there).
- Clearance to other structures.
- Signal sighting for the signal on the structure that was involved - the red aspect position is very carefully controlled and structures simply need to be made to support it where sighting is best.
- Sighting of other signals in the area, i.e. if signals are close together or if there are other tracks nearby that are sighting through here, the gantry for this signal may need to be positioned to provide a clear line of sight to a different signal. This applies to everything.

Also note when considering clearances it was mentioned, IIRC, that this site was on a sharp curve. Throw thus becomes an issue, as does dynamic movement of the train, so the distance measured by the RAIB is only going to be approximately the distance between the train in question and the gantry, but it highlights that it's sub-standard but within allowable 'with specific authorisation' parameters. Thus, the RAIB will be looking for the records to see why that clearance was approved to ensure it was justifiable.

Note also that track moves around over time, and as long as maintenance keeps the track above allowable parameters, it's generally acceptable. Thus, potentially this was installed at a greater clearance and the track has migrated towards the gantry over time, with tamping and general track settlement/movement. The RAIB would also thus be looking at the maintenance records to see what the history of track quality and track works in the area was.

 

[61653 HTAFC]

All other stock on the lines round there are 20m rather than the 23m 442s. Also the droplight in question is in the middle of the vehicle rather than at the ends as on most mk3s- which means its position on curves is more subject to change.

 

[QueensCurve]

It will have been an agreement between the signalling engineer, the track engineer and any other engineers with stuff nearby. I do not not know the area but things potentially to be considered are:
- The clearance to trains on the opposite side of the gantry (if there is another track there).
- Clearance to other structures.
- Signal sighting for the signal on the structure that was involved - the red aspect position is very carefully controlled and structures simply need to be made to support it where sighting is best.
- Sighting of other signals in the area, i.e. if signals are close together or if there are other tracks nearby that are sighting through here, the gantry for this signal may need to be positioned to provide a clear line of sight to a different signal. This applies to everything.

Also note when considering clearances it was mentioned, IIRC, that this site was on a sharp curve. Throw thus becomes an issue, as does dynamic movement of the train, so the distance measured by the RAIB is only going to be approximately the distance between the train in question and the gantry, but it highlights that it's sub-standard but within allowable 'with specific authorisation' parameters. Thus, the RAIB will be looking for the records to see why that clearance was approved to ensure it was justifiable.

Note also that track moves around over time, and as long as maintenance keeps the track above allowable parameters, it's generally acceptable. Thus, potentially this was installed at a greater clearance and the track has migrated towards the gantry over time, with tamping and general track settlement/movement. The RAIB would also thus be looking at the maintenance records to see what the history of track quality and track works in the area was.

Thank you for the summary of issues.

Noting that track does move, it still seems very close, and - without wanting to speculate - I am not sure any of the other issues would have precluded it being wider.

We await RAIB's findings with interest.

 

[edwin_m]

It is possible that successive tampings and minor realignments have moved the track slightly closer to the gantry over the years, although there should be checks in place to ensure that this does not reduce any clearances to unacceptable values. I wonder also if the longer wheelbase of the 442 stock made the clearance slightly less on the inside of the curve than it would have been with traditional slam door stock (but the 442 is probably narrower too, which might cancel out the extra throw).

Also the 260mm quoted by RAIB must have been measured with the train standing still. As mentioned above, with the train moving there will be some sway and lateral movement in the suspension which could have reduced the clearance by, I guess, several tens of millimetres. There is probably no way of proving what the dynamic clearance actually was at the time of the accident, although it should be possible to establish a minimum that it might have been.

 

[BRX]

Clearance to lineside structures is constantly monitored, isn't it (for example I notice that steam specials etc are often rerouted due to "out-of-gauge" concerns that arise since they were originally planned)? So if the track had "migrated" towards this gantry over the years there must have been record of this.

It wasn't really that long ago that lots of slam-door commuter stock (with droplights at all points along the length of carriages) was operating across south London so unless this gantry was installed within the last 10 years or so it does seem a little alarming that it's that close.

 

[mr_moo]

Clearance is monitored regularly, yes. Track doesn't (usually!) migrate fast so inspections are done every few months for the key lines and less frequently for less key lines, depending on various criteria such as tonnage, frequency, consequence and speed.

Whilst it may have not been that long since slam doors stock with droplights was phased out, don't forget that it's possible this clearance did exist even before then. As I said before, the standards do state that 450mm *should* be provided *where possible*, but it's possible that this was gantry was one of the 'not possible'. Without knowing the history of the gantry and track and the risk assessments I wouldn't know.

As the RAIB has now given enough information to identify the gantry in question, I've had a look at it using aerial mapping. One thing I note is that the track there is up on an embankment. The flat area at the top of the embankment for track workers to walk along looks quite narrow.

It's pure speculation, but it's possible that a decision was made when designing it that moving the support leg further away from the track would mean that track workers walking along the track here would need to pass on the track side of the gantry, placing them very close to the third rail whilst also probably holding onto a big metal gantry, thereby increasing risks for track staff. Thus, moving it closer may compromise the clearance for anyone leaning out of the window, which they shouldn't be doing and they have reasonable sight lines on the approach as it's on the inside of a curve, but has an overall safety benefit for track staff and possibly drivers if they needed to get out and walk up towards the next gantry if their train had failed. I would assume that this area is red zone banned today but it probably would not have been when the gantry was installed - it looks like an old gantry.

260mm is plenty of clearance for normal structures and the 450mm requirement only applies where trains with openable windows normally run. I don't know what other stock uses the line but if most stock doesn't have this and even this one only has the one door that is for guard's usage only officially, the risk would probably be deemed low enough so as not to require the 450mm to be adhered to.

It's also noteworthy that this is on the inside of a curve at 60mph so the dimension to the train in question is likely to be greater than the static 260mm value, unless of course it had a failed suspension bag or something.

It would not surprise me if the RAIB published a safety digest instead of conducting a full investigation as there is, in my opinion, little in the way of new safety lessons to be learned here. This is, again, IMHO, a simple tragic accident where a low probability event (in terms of risk assessment and engineering decisions) still occurred.

 

[Muzer]

Didn't slamdoor stock on the Sussex route have window bars?

 

[yorksrob]

Didn't slamdoor stock on the Sussex route have window bars?

Only on the East Grinstead line. Everything else didn't.

 

[MichaelAMW]

Only on the East Grinstead line. Everything else didn't.

I would say that you're somewhere between right and wrong there. Yes, the need for window bars related only to the East Grinstead line but it didn't have a dedicated set of units, so those with the bars could and did turn up all over the place. In practice, I think - just from memory - that the VEPs were fitted but the "mainline" stock, the CIGs, weren't and so couldn't go down there.

 

[yorksrob]

I would say that you're somewhere between right and wrong there. Yes, the need for window bars related only to the East Grinstead line but it didn't have a dedicated set of units, so those with the bars could and did turn up all over the place. In practice, I think - just from memory - that the VEPs were fitted but the "mainline" stock, the CIGs, weren't and so couldn't go down there.

That's probably true.

I certainly don't recall seeing anything on the coastway with window bars (except occasional thumpers).

 

[QueensCurve]

Personally I'd say it's safer to be leaning out on the non-cess side, as although you have trains passing, there's less chance of lineside structures and, worse, overhanging vegetation.

Given only a 100mm (4") clearance to passing trains, high closing speeds and the pressure pulse that is hardly a safe practice.