Croydon Tram Accident

['CHARD]

Network Rail (Euston):  'Virgin Trains services are removed from the board 2-3 minutes prior to departure; London Midland services are removed 1 minute prior to departure and London Overground services 30 seconds prior to departure for safety reasons.'

 

Virgin locks its train doors up to two minutes prior to departure, so the above NR statement reflects the fact that when it's disappeared from the board you've missed it.  Departure time is the moment the driver applies power.

 

 

 

EDIT: Fat Cont. just basically posted this!

Edited November 20, 2016 by 'CHARD

[andyman7]

These lights have always been provided at Croydon and on other tram systems.  They light up if the system detects a signal passed at danger, and any driver seeing the lights should stop immediately (so as to stop both the SPAD tram and any other that might be in conflict with it).  If they are now operating for all trams then I would think that creates a serious risk of confusion. 

These are beacons, not signals and are not located in relation to the signals as they are on a drive-on-sight section

[Rugd1022]

 

RiverCider wrote, "One of the SPADs that occurred at the depot I worked at involved a driver who regularly swapped for evening turns,

he ended up regularly working a particular service that followed a stopping train so was frequently driving against yellow signals,

until one day one of them was a red",

 

Interesting - What the Marine and Aviation accident investigation psychologists refer to as a "Motor Program Error" - A type of mistake that [almost] only ever occurs in trained and experienced practitioners. Two elements to it - the first is [unreasonable] expectation/anticipation and the second [less obvious] is a built in lack of flexibility that doesn't welcome change in circumstances so a slower response time.

 

 

A few weeks ago on an infrastructure job I was following another train on the Down Slow from West Hampstead all the way to Oakley on double yellows, a distance of around fifty miles. Last week (whilst being assessed by my manager!) I worked a similar train on the WCML from Rugby to Willesden and had double yellows for most of the journey between Hanslope Junction and Wembley. Happens all the time on the freight side of the job.

[Dagworth]

A few weeks ago on an infrastructure job I was following another train on the Down Slow from West Hampstead all the way to Oakley on double yellows, a distance of around fifty miles. Last week (whilst being assessed by my manager!) I worked a similar train on the WCML from Rugby to Willesden and had double yellows for most of the journey between Hanslope Junction and Wembley. Happens all the time on the freight side of the job.

Happens all the time on suburban commuter work too, the run into Liverpool Street used to be very bad, because the signals were so close together there were parts where you would get multiple double yellows rather than G/YY/Y/R. The track is fairly straight and you could see the next half a dozen signals in front of you, ALL showing double yellow, then finally a single yellow and the red.

 

Running into Charing Cross in the morning it was very rare to get anything other than a stream of yellow, or worse - red changing to yellow as you approached.

 

Andi

[jjb1970]

There are two things the investigators need to establish - what happened and why did it happen. The first has been answered quickly, the second may be more difficult. Thankfully they won't stop at driver error, nowadays many investigating and risk management organisations have psychologists on staff to dig deeper into the behavioural and ergonomic aspects. If necessary they'll comb through the drivers life including health, family & financial issues. I'm guessing management are having to answer a lot of awkward questions and that there may be a few nervous people who signed off on the risk assessments. Having gone through this myself on both sides (years ago I was on a ship that suffered a fatality) it is a very stressful experience. On the other side of the table when supporting investigations I've seen the stress lead to breakdowns as people exist under a sword of Damocles, facing possible legal charges. One of the things I tell people to motivate them to take safety seriously is to consider just how seriously messed up your life will be following an incident.

Edited November 20, 2016 by jjb1970

[The Stationmaster]

There are two things the investigators need to establish - what happened and why did it happen. The first has been answered quickly, thw second may be more difficult. Thankfully they won't stop at driver error, nowadays many investigating and risk management organisations have psychologists on staff to dig deeper into the behavioural and ergonomic aspects. If necessary they'll comb through the drivers life including health, family & financial issues. I'm guessing management are having to answer a lot of awkward questions and that their may be a few nervous people who signed off on the risk assessments. Having gone through this myself on both sides (yaers ago I was on a ship that suffered a fatality) it is a very stressful experience. On the other side of the table when supporting investigations I've seen the stress lead to breakdowns as people exist under a sword of Damocles, facing possible legal charges. One of the things I tell people to motivate them to take safety seriously is to consider just how seriously messed up your life will be following an incident.

 

Very true JJB - I know of one person on BR who as the inquiry developed into a major incident decided to resign and then actually left Britain because he was being placed under so much strees with media, and other, 'sabre rattlers' seemingly looking in all directions for scapegoats to prosecute.

[Oldddudders]

Very true JJB - I know of one person on BR who as the inquiry developed into a major incident decided to resign and then actually left Britain because he was being placed under so much strees with media, and other, 'sabre rattlers' seemingly looking in all directions for scapegoats to prosecute.

 

I recall a TOC MD, a former colleague, being shown tearful in court on tv, after just such a major incident. A level of responsibility I was glad never to hold.  

[wombatofludham]

Why are people going on about signals and SPADs?  Croydon, like Sheffield Supertram, Midland Metro, Blackpool, Nottingham NET and some parts of Metrolink is driven on line of sight.  Even the former rail alignments are classified legally as tramway and the drivers have to drive in a manner to be able to stop if another tram is ahead of them, or some numpty member of the gene puddle walks out in front of them at a stop.  That's why most modern light rail schemes have level pedestrian crossings at stops - and why, for example, Swan Lane level crossing on Midland Metro doesn't have barriers or flashing wig-wag lights, but normal road type traffic lights - because the Railway Inspector designated the whole Metro as tramway.  The only "signals" on these systems are point indicators and the seperate tramway aspect signals at traffic light controlled road junctions (the five white lights in a line type), with the exception of phase 1 of Metrolink which kept lineside signals on the ex-rail formations.

 

The fact this was the first case of a serious fatal derailment on a sharp curve since it opened in 2000 (and indeed the first such accident on any new second generation system since Metrolink began operation in 1992) suggests that calls for expensive rail type intervention systems are a knee-jerk reaction to an awful event, but in the context of millions of miles of safe operation across the country on similar systems with equally challenging curves on their networks, a one off.

 

Light rail/tramways have a completely different operating ethos and legally are defined as something completely different to heavy rail.  Having worked closely with the then HMRI during the construction of Midland Metro it was clear that HMRI saw such systems differently and (with the exception of the Bury-Altrincham Metrolink phase 1 which kept lineside signalling) saw any line of sight driven systems as tramway - which is why there is a fence between Midland Metro and the Jewellery Line and signs at the Hawthorns and Jewellery Quarter stations on leaving the Metro platforms and entering the heavy rail station reminding passengers not to cross the tracks on the Network Rail station, because HMRI was concerned about potential confusion given both systems are segregated running (which given the general level of stupidity of the public was probably entirely justified).

Edited November 20, 2016 by wombatofludham

[Edwin_m]

These are beacons, not signals and are not located in relation to the signals as they are on a drive-on-sight section

I dont understand this comment.  The whole of Tramlink is a "drive-on-sight section" and I don't think I suggested that they were signals.  As others have pointed out, there are already beacons to warn of violation of a stop aspect and I was querying whether the two types of beacon were distinguisable by colour or some other means. Being near a junction, the accident site will have signals and these pre-existing beacons nearby. 

Edited November 20, 2016 by Edwin_m

[D854_Tiger]

Why are people going on about signals and SPADs?  Croydon, like Sheffield Supertram, Midland Metro, Blackpool, Nottingham NET and some parts of Metrolink is driven on line of sight.  Even the former rail alignments are classified legally as tramway and the drivers have to drive in a manner to be able to stop if another tram is ahead of them, or some numpty member of the gene puddle walks out in front of them at a stop.  That's why most modern light rail schemes have level pedestrian crossings at stops - and why, for example, Swan Lane level crossing on Midland Metro doesn't have barriers or flashing wig-wag lights, but normal road type traffic lights - because the Railway Inspector designated the whole Metro as tramway.  The only "signals" on these systems are point indicators and the seperate tramway aspect signals at traffic light controlled road junctions (the five white lights in a line type), with the exception of phase 1 of Metrolink which kept lineside signals on the ex-rail formations.

 

The fact this was the first case of a serious fatal derailment on a sharp curve since it opened in 2000 (and indeed the first such accident on any new second generation system since Metrolink began operation in 1992) suggests that calls for expensive rail type intervention systems are a knee-jerk reaction to an awful event, but in the context of millions of miles of safe operation across the country on similar systems with equally challenging curves on their networks, a one off.

 

Light rail/tramways have a completely different operating ethos and legally are defined as something completely different to heavy rail.  Having worked closely with the then HMRI during the construction of Midland Metro it was clear that HMRI saw such systems differently and (with the exception of the Bury-Altrincham Metrolink phase 1 which kept lineside signalling) saw any line of sight driven systems as tramway - which is why there is a fence between Midland Metro and the Jewellery Line and signs at the Hawthorns and Jewellery Quarter stations on leaving the Metro platforms and entering the heavy rail station reminding passengers not to cross the tracks on the Network Rail station, because HMRI was concerned about potential confusion given both systems are segregated running (which given the general level of stupidity of the public was probably entirely justified).

 

 

Indeed, the whole point of light rail is that it isn't heavy rail and is not subject to the same safety regime, making it more affordable to build.

 

We notice last week a bus came off the road near Ladbroke Grove, involving serious injury, that could easily have had similar consequences for loss of life but there was no press speculation whatsoever about it. 

 

It was just accepted as one of those things that happens to happen a darn sight more often than a tram crash.

 

Bus drivers can fall asleep at the wheel, with far worse consequences, and with no safety mechanisms for back up.

 

If we start loading light rail costs with loads of additional safety gear, then we are going to end up with fewer of them and then it's another guided bus scheme (and an accident every three months) or no scheme whatsoever all with much worse safety outcomes.

[Dagworth]

Why are people going on about signals and SPADs?  Croydon, like Sheffield Supertram, Midland Metro, Blackpool, Nottingham NET and some parts of Metrolink is driven on line of sight.  Even the former rail alignments are classified legally as tramway and the drivers have to drive in a manner to be able to stop if another tram is ahead of them, or some numpty member of the gene puddle walks out in front of them at a stop.  That's why most modern light rail schemes have level pedestrian crossings at stops - and why, for example, Swan Lane level crossing on Midland Metro doesn't have barriers or flashing wig-wag lights, but normal road type traffic lights - because the Railway Inspector designated the whole Metro as tramway.  The only "signals" on these systems are point indicators and the seperate tramway aspect signals at traffic light controlled road junctions (the five white lights in a line type), with the exception of phase 1 of Metrolink which kept lineside signals on the ex-rail formations.

 

The fact this was the first case of a serious fatal derailment on a sharp curve since it opened in 2000 (and indeed the first such accident on any new second generation system since Metrolink began operation in 1992) suggests that calls for expensive rail type intervention systems are a knee-jerk reaction to an awful event, but in the context of millions of miles of safe operation across the country on similar systems with equally challenging curves on their networks, a one off.

 

Light rail/tramways have a completely different operating ethos and legally are defined as something completely different to heavy rail.  Having worked closely with the then HMRI during the construction of Midland Metro it was clear that HMRI saw such systems differently and (with the exception of the Bury-Altrincham Metrolink phase 1 which kept lineside signalling) saw any line of sight driven systems as tramway - which is why there is a fence between Midland Metro and the Jewellery Line and signs at the Hawthorns and Jewellery Quarter stations on leaving the Metro platforms and entering the heavy rail station reminding passengers not to cross the tracks on the Network Rail station, because HMRI was concerned about potential confusion given both systems are segregated running (which given the general level of stupidity of the public was probably entirely justified).

Not so at all, Croydon Tramlink does indeed have stop/proceed signals at various locations,

There is a report reference earlier in the thread of a collision at New Addington where a tram passed a stop signal at danger in fog and came into collision with another tram https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/412062/060720_R112006_New_Addington.pdf

 On plain line yes trams are driven by sight but SPADs (technically referred to as SPASs - Signals passed at stop)  are very real on trams too.

 

Map of New Addinton showing STOP signals as red circles

 

Andi

Edited November 20, 2016 by Dagworth

[wombatofludham]

Looks to me like they are controlling access to single line sections or access across roads, which is where you would expect such signals to be provided.  Of course it's possible for a tram to over-run a traffic signal (which is what these signals are, the access to the single track section is no different to the traffic signals you sometimes find at single carriageway bridges on the roads, where the signals are controlling alternate lines of traffic) but the point is that there wouldn't be any signalling on the approach to the curve in question and given the fact that the stop or proceed signal is classed as a highway tram aspect, the driver still has to be driving on line of sight, and if he over-runs it, should take action by dropping the emergency brake.

 

The majority of stop and proceed signals are associated with highway traffic signals, so the potential risk of passing one at stop is not of the same order of magnitude as a heavy rail train travelling at speed passing a signal at danger.  Certainly when approving systems like Tramlink HMRI didn't consider the issue merited ATP or TPWS.

[jjb1970]

I don't think there'll be a kneejerk response which threatens the economic viability of tram systems, the ALARP principle considers cost - benefit and if the cost of any further risk controls is disproportionate to the reduction in risk profile then you can quite legitimately draw a line. That said, something often forgotten in the ALARP regime is that ALARP in itself is not sufficient, the residual risk also has to meet the criteria to be considered tolerable.

[Derekl]

I think you will find there was a desire by the DfT back in the late 80s to ensure the signs used for first modern tramway in the UK for decades (Manchester) could not be mistaken for conventional road signs. Going for a metric speed limit was a useful in assisting with this.

 

I am pretty sure Metrolink was signposted in MPH as were the speedometers in the Ansaldo trams, visible over the shoulder of the driver from the front seats. Looking at a current cab view from Metrolink, it looks like it is still in MPH - I cannot imagine the line from Vic to Shudehill requiring a 15 kph limit, although 15 mph is possible. I don't know what they have done about the speedometers in the new trams - I suppose it is possible that you could now make them digital and easily convertible.

 

Overall, I think this is a bit of a red herring (the kph/mph issue). As said above, the vehicle and sign posts are consistent. I have never had any difficulty on speed limit issues driving in Europe (although I was once on the spot fined in Bosnia Herzegovina, but that was because I was gong too fast, not because of any difficulty with understanding the limit).

[Edwin_m]

I am pretty sure Metrolink was signposted in MPH as were the speedometers in the Ansaldo trams, visible over the shoulder of the driver from the front seats. Looking at a current cab view from Metrolink, it looks like it is still in MPH - I cannot imagine the line from Vic to Shudehill requiring a 15 kph limit, although 15 mph is possible. I don't know what they have done about the speedometers in the new trams - I suppose it is possible that you could now make them digital and easily convertible.

 

Overall, I think this is a bit of a red herring (the kph/mph issue). As said above, the vehicle and sign posts are consistent. I have never had any difficulty on speed limit issues driving in Europe (although I was once on the spot fined in Bosnia Herzegovina, but that was because I was gong too fast, not because of any difficulty with understanding the limit).

Yes Metrolink still uses MPH as I believe does Supertram.  Incidentally class 395s have a digital speedometer which changes from showing km/h to showing MPH when it goes off HS1 and onto the classic network.  If drivers can cope with that then I don't see a problem with them using a speed unit which is fully consistent across the tram system but happens to be different from what they would use if driving a car. 

[Gwiwer]

 

 

 

Moorgate

There were suggestions made at the time, I believe by persons on platforms 9 and 10 and though I cannot now recall whether they were ever corroborated, that the train actually accelerated into the dead-end platform.  The extent of the impact was such that the train might not have been compliant with the speed restriction.  The dull thud heard and felt in the street above followed by acrid dust emerging from vent shafts is something I shall never forget.

 

 

Looks to me like they are controlling access to single line sections or access across roads

Correct.  Tramlink operates substantially on line-of-sight but has many single-track sections which range from short distances such as in New Addington to the entire central Croydon loop.  These were necessary to squeeze the tramway past existing infrastructure or to get it through the centre of Croydon in streets which could not accommodate a double-track tramway without being closed to other traffic.  It is not possible to see a tram approaching in the other direction in most cases (until it is very close) because the line-of-sight is obstructed by curvature, buildings gradient so the tramway equivalent of stop signals are employed.  Remember the Wimbledon - West Croydon line was single track as a suburban railway when closed for conversion even though it had originally been built double.

[Colin_McLeod]

On the subject of automatically cancelling the warning on double yellows then inadvertently cancelling on a red I never understood why the red did not have a different warning nor do I understand the logic of having a system that includes the very idea of cancelling a red warning

 

Red means stop so why cancel the red warning?

[Dagworth]

On the subject of automatically cancelling the warning on double yellows then inadvertently cancelling on a red I never understood why the red did not have a different warning nor do I understand the logic of having a system that includes the very idea of cancelling a red warning

 

Red means stop so why cancel the red warning?

The whole AWS system only has two indications, Clear or Warning. Clear can only be given by a Green light/Off distant signal.

Warning is given by any signal showing less than a Clear warning as described above.

 

Red signals are also capable of showing warning aspects and so are fitted with AWS. 

Note that two aspect red/green colour light signals that would be in place of semaphore Stop signals may well not have AWS at all, semaphore Stop signals are not fitted. (They cannot show a Warning aspect)

 

Edit: As to cancelling the warning, if the warning is not cancelled it will instigate an uncontrolled brake application which is never a good thing except in emergency.

 

Andi

Edited November 20, 2016 by Dagworth

[david.hill64]

I don't think there'll be a kneejerk response which threatens the economic viability of tram systems, the ALARP principle considers cost - benefit and if the cost of any further risk controls is disproportionate to the reduction in risk profile then you can quite legitimately draw a line. That said, something often forgotten in the ALARP regime is that ALARP in itself is not sufficient, the residual risk also has to meet the criteria to be considered tolerable.

Unfortunately ORR seems to have thrown ALARP out of the window with its stance on electrification clearances so we may find ourselves in a whole new and very costly world.

[PatB]

I was under the impression that the discussion here of SPADs was more with regard to the psychological factors contributing to those accidents which may be generically termed "driver error" than on specific details of this crash in particular. Understanding of the reasons why drivers may fail to act upon warnings, whether by signals, AWS, fixed speed restriction signs or even general appearance of the route ahead is important if such failures are to be minimised in future.

 

Speaking as someone who has had a professional interest in transport safety matters (not rail) and a continuing amateur interest in the subject, I have found all the contributions here to be most instructive.

[jjb1970]

Unfortunately ORR seems to have thrown ALARP out of the window with its stance on electrification clearances so we may find ourselves in a whole new and very costly world.

I find their approach on that to be disappointing and indicative of lazy (or incompetent) decision making which will threaten the economic viability of electrification business cases. There is nothing to stop them using risk assessment and engineering analysis to demonstrate functional equivalence using smaller air gaps. To be honest, whilst I fully support electrification, as a tax payer I also see no reason to inflate project costs to spare NR the effort of doing a bit of analysis.

[Edwin_m]

The whole AWS system only has two indications, Clear or Warning. Clear can only be given by a Green light/Off distant signal.

Warning is given by any signal showing less than a Clear warning as described above.

 

Red signals are also capable of showing warning aspects and so are fitted with AWS. 

Note that two aspect red/green colour light signals that would be in place of semaphore Stop signals may well not have AWS at all, semaphore Stop signals are not fitted. (They cannot show a Warning aspect)

 

Edit: As to cancelling the warning, if the warning is not cancelled it will instigate an uncontrolled brake application which is never a good thing except in emergency.

 

Andi

The AWS system was developed with absolute block signalling in mind and was intended to be used only at the Distant signal.  Even at the time there was quite a bit of multiple-aspect signalling operational or planned, but despite not really allowing for this I imagine AWS has paid for it self several times over in accidents avoided.  The addition of TPWS partly remedies the deficiencies in AWS.  Incidentally red/green signals in colour light areas now tend to be fitted with AWS because of the risk of driver confusion if they aren't. 

Edited November 21, 2016 by Edwin_m

[Bernard Lamb]

I recall a TOC MD, a former colleague, being shown tearful in court on tv, after just such a major incident. A level of responsibility I was glad never to hold.

I never appeared in court and never had that level of job, but I was once involved in a situation where there was a fatality on a job with which I was directly responsible. It was very upsetting and not something I would want to go through again. My colleagues and myself were cleared of any responsibility for the incident. I will not give too much detail, but after the enquiry it did lead to a change in the law.

My deepest sympathy to all those affected by this event.

Bernard

[The Stationmaster]

I find their approach on that to be disappointing and indicative of lazy (or incompetent) decision making which will threaten the economic viability of electrification business cases. There is nothing to stop them using risk assessment and engineering analysis to demonstrate functional equivalence using smaller air gaps. To be honest, whilst I fully support electrification, as a tax payer I also see no reason to inflate project costs to spare NR the effort of doing a bit of analysis.

 

Slighty OT but I think the problem was that ORR failed to get a derogation sorted (probably because they were not aware of what was happening or took no notice of it) thus they are hoist on a petard of their own making from which the whole industry will suffer.  And if they start behaving in that way in respect of this incident yet more over-reaction could possibly result in some extremely expensive requirements.

[The Stationmaster]

I never appeared in court and never had that level of job, but I was once involved in a situation where there was a fatality on a job with which I was directly responsible. It was very upsetting and not something I would want to go through again. My colleagues and myself were cleared of any responsibility for the incident. I will not give too much detail, but after the enquiry it did lead to a change in the law.

My deepest sympathy to all those affected by this event.

Bernard

I would not recommend 'sitting in the hot seat' at an Inquiry conducted by one of the previous style of HMRIs.  Don't get me wrong - they were absolutely excellent and brilliant at their job of trying to seek out the facts and giving every consideration to the witnesses but it is not a 'nice' situation in which to find yourself.  And I can tell you from experience that it is strange how memory does, or doesn't work.

 

Mind you in later years it could be very beneficial if you found yourself at the other side of the inquiry process and were asking the questions instead of answering them.