Croydon Tram Crash

[Taunton]

The glass may be unbreakable but if the lightweight body flexes when thrown on its side the glass may well fall out in one piece.

The doors are fully glazed and are effectively windows as well, and are not even held in place by any rigidity of the body, but just by the plug mechanism top and bottom, which is certainly not designed to hold the door in position when the vehicle is sliding on its side.

 

[littlerock]

As regards the Vigilence system, I was simply repeating a comment made to the media, after the accident by Andrew Braddock of Tram UK who said "On the tram control lever there is a button which the driver presses at least every 30 seconds. If that button is not pressed the brakes are applied automatically. It makes no difference if the driver is powering up or braking, he has to keep pressing that button. It's a routine that you get into. If the driver lets go, the lever springs back to the neutral position, automatically applying the brakes, also a Klaxon sounds. "

Survivors report the train did not brake and no one heard a klaxon, so the Vigilence system, if it was on the train, did not have time to activate.

 

[455refurb]

The BBC is now running the story about trams allegedly speeding around the Sandilands Junction curve prior to the accident, based on a complaint made to customer services about the same journey on 31st October that the Facebook user had posted about.
http://www.bbc.co.uk/news/uk-england-london-37979121

Shantall Singh's email to Tramlink (31 October):

"As the tram went around the bend at the junction between Lloyds Park and Sandilands the driver from my point of view missed the bend or he was going too fast. I thought I was going to die and leave my two-year-old son. I do hope this matter will be looked into because I am left with a fear for travelling in the tram."

Tramlink customer services reply (2 November):

"We are sorry to hear of this unfortunate incident and we would like to reassure you that we will be fully investigating this. There are a couple of trams it could be and to identify the driver and to view the CCTV from the tram we need to have these details as soon as possible."

The article doesn't say whether the complainant provided the further information requested.

Based on the various anecdotal reports in the press, and the two seperate reports of the incident on the 31st October, it does paint a worrying picture of speeding on the network which, arguably, should have been nipped in the bud by the management at First. It will be interesting to read the RAIB interim report as they will no doubt have downloaded data recorders from other trams.

 

[jon0844]

When there are multiple, public, posts querying the speed you do begin to wonder what was going on along there.

Of course it might well be the same driver, so not any sign of competitiveness between staff. Or it might be nothing but people just feeling the tram was going too fast around a curve.

I do wonder if it's possible that drivers got to thinking that 12mph was unnecessarily slow and mess room talk had made some conclude that it was possible to go faster. Still doesn't explain why the tram was going THAT much faster though.

I hope this interim report comes out soon as there's so much to speculate about.

 

[507 001]

As regards the Vigilence system, I was simply repeating a comment made to the media, after the accident by Andrew Braddock of Tram UK who said "On the tram control lever there is a button which the driver presses at least every 30 seconds. If that button is not pressed the brakes are applied automatically. It makes no difference if the driver is powering up or braking, he has to keep pressing that button. It's a routine that you get into. If the driver lets go, the lever springs back to the neutral position, automatically applying the brakes, also a Klaxon sounds. "

Survivors report the train did not brake and no one heard a klaxon, so the Vigilence system, if it was on the train, did not have time to activate.

Yes I got the impression that it wasn't just something you made up

I'm afraid Mr Braddock is wrong. As I said, the driver has to keep his or her hand on the TBC. There is nothing that requires pressing every 30 seconds, nor does it spring back to the coast position.

If he works for tram uk, he's probably been playing around in the cabs of T68 and T69 vehicles, they did have spring loaded TBCs.

The glass may be unbreakable but if the lightweight body flexes when thrown on its side the glass may well fall out in one piece.

The doors are fully glazed and are effectively windows as well, and are not even held in place by any rigidity of the body, but just by the plug mechanism top and bottom, which is certainly not designed to hold the door in position when the vehicle is sliding on its side.

Believe me it doesn't need the body to flex for a broken flexity window to fall out.

 

[Bletchleyite]

Believe me it doesn't need the body to flex for a broken flexity window to fall out.

Are they single or double glazed? I don't think anyone has been able to answer that yet. Double-glazed windows would hold their integrity better.

 

[SpacePhoenix]

How long would the data from the data recorders normally be kept for before being deleted?

 

[matt_world2004]

Even if the data recorders have been wiped couldn't tfl work out the trams speed based on the time it stopped at the two adjacent stops . They are recording that information to monitor the contract.

 

[TC60054]

Are they single or double glazed? I don't think anyone has been able to answer that yet. Double-glazed windows would hold their integrity better.

Single glazing on CR4000s iirc.

 

[talldave]

Single glazing on CR4000s iirc.

If the glass is laminated then it's made of more than one layer of glass though.

 

[2HAP]

BBC reporting that the driver in the October 31 incident was not the driver in the 9 November accident.

 

[507 001]

Are they single or double glazed? I don't think anyone has been able to answer that yet. Double-glazed windows would hold their integrity better.

Single glazed.

 

[Bletchleyite]

Single glazed.

Cheers (and to the other person who replied).

That will mean far poorer structural integrity, then. Bonded double glazing is the strongest (though is out of favour at present because a broken window requires the unit out of service for a day for the glue to dry).

It's educational how much bus bodies distort with single glazing compared with double.
--- old post above --- --- new post below ---

If the glass is laminated then it's made of more than one layer of glass though.

Indeed, though that doesn't provided the same level of structural integrity that true double glazing provides (the sealed air gap has an effect). And laminated double glazing would be three or four layers.

 

[theageofthetra]

On modern cars (not sure about PSV or Commercials) the front and rear glass forms an important part of the vehicles structural integrity -is this the case on this type and the newer vehicles operating in Croydon?

 

[rebmcr]

Cheers (and to the other person who replied).

That will mean far poorer structural integrity, then. Bonded double glazing is the strongest (though is out of favour at present because a broken window requires the unit out of service for a day for the glue to dry).

It's educational how much bus bodies distort with single glazing compared with double.


Indeed, though that doesn't provided the same level of structural integrity that true double glazing provides (the sealed air gap has an effect). And laminated double glazing would be three or four layers.

If Croydon wanted particularly crashworthy vehicles, the route would be heavy rail. I think the focus should be on signalling failsafes that prevent this kind of incident in the first place.

 

[neonison]

If Croydon wanted particularly crashworthy vehicles, the route would be heavy rail. I think the focus should be on signalling failsafes that prevent this kind of incident in the first place.

.
Absolutely. Crash-worthiness is all well and good but there are a limited number of situations when it is relevant to light rail. At the risk of irritating the "line-of-sight" purists, I hope that one recommendation would be to have automated systems in place to moderate speed to the extent that a roll-over derailment cannot occur.

 

[littlerock]

Yes but the RSSB analysis concluded the majority of fatalities occur when passengers are thrown out of the window when a train comes off the tracks and overturns. So they recommended laminated glass. It is not clear if this was what was on the Croydon train or how it was installed. But in an incident in Ireland on the same model of tram, it was reported that it was the shattering kind of glass that you used to get (?) in car windscreens.

Irish news report: while the UK Office of Rail and Road has confirmed the Croydon tram, a Bombardier CR4000 Number 2551 was not fitted with a safety protection systems that would apply the brakes automatically if the tram was going too fast, such a system is fitted to the Alstom-built, Citadis trams used by Luas in Ireland.

Mr Gerry Madden, MD of Transdev, said the system used by Luas was known internationally as an automatic train protection or automatic train control.

“Each tram is fitted with a system called vigilance device, the system requires a regular activation by the tram driver otherwise an audible alarm is generated in the cab and the brake will be applied automatically” said Mr Madden.

In addition Mr Madden said the tram maximum speed is limited by the system to 80km/h, in excess of 80km/h a brake will be applied.

“Maximum authorised line speeds are defined considering the maximum design speed, the maximum legal speed and a safe operation speed, according [to] line of sight principle. Curves have a maximum speed defined taking into account the radius of the curve,” he said.

He said “any recommendations that may arise out of the ongoing investigation into the Croydon accident” would be considered.“There have been no Luas tram crashes whereby the tram has tilted over,” he added.

 

[Taunton]

I'm still waiting to hear why check rails were not installed on the curve because I think they would make it significantly less likely for the vehicle to derail.

The wheel profile of trams is significantly less than on railway vehicles, because they have to be able to run on street sections with grooved rails. This therefore seems to make them more susceptible to derailment than railway vehicles with full depth flanges.

 

[Bletchleyite]

Presumably the Luas uses old-Metrolink-style railway block signalling off the street sections, then? That said, the description sounds like a deadman-type device and a fixed maximum speed, not what I'd call ATP/ATC which would require full speed control and some kind of block signalling.

FWIW, some kind of TPWS-style speed and "red light" enforcement would not be precluded by the use of line-of-sight operation. It just wouldn't be able to prevent rear-endings because traditional blocks aren't used.

 

[edwin_m]

Yes but the RSSB analysis concluded the majority of fatalities occur when passengers are thrown out of the window when a train comes off the tracks and overturns. So they recommended laminated glass. It is not clear if this was what was on the Croydon train or how it was installed. But in an incident in Ireland on the same model of tram, it was reported that it was the shattering kind of glass that you used to get (?) in car windscreens.

Irish news report: while the UK Office of Rail and Road has confirmed the Croydon tram, a Bombardier CR4000 Number 2551 was not fitted with a safety protection systems that would apply the brakes automatically if the tram was going too fast, such a system is fitted to the Alstom-built, Citadis trams used by Luas in Ireland.

Mr Gerry Madden, MD of Transdev, said the system used by Luas was known internationally as an automatic train protection or automatic train control.

“Each tram is fitted with a system called vigilance device, the system requires a regular activation by the tram driver otherwise an audible alarm is generated in the cab and the brake will be applied automatically” said Mr Madden.

In addition Mr Madden said the tram maximum speed is limited by the system to 80km/h, in excess of 80km/h a brake will be applied.

“Maximum authorised line speeds are defined considering the maximum design speed, the maximum legal speed and a safe operation speed, according [to] line of sight principle. Curves have a maximum speed defined taking into account the radius of the curve,” he said.

He said “any recommendations that may arise out of the ongoing investigation into the Croydon accident” would be considered.“There have been no Luas tram crashes whereby the tram has tilted over,” he added.

The trams on Luas are Alstom Citadis, similar to the Nottingham ones, but Croydon has Bombardier CR4000s which are a different design from a different manufacturer.

The method of calculating permitted speeds mentioned is in principle exactly the same as in Croydon, and I can't see it having any bearing on the accident.

Although Automatic Train Protection is mentioned here, the description goes on to describe a vigilance device and a speed limiter that simply enforces the top speed of the tram. Croydon trams are also limited to 80km/h although I think they just refuse to go any faster rather than applying the brakes, and they have a spring-loaded switch on the traction/brake controller which may be less sophisticated than the vigilance device mentioned in the quote.

Neither of those devices is remotely describable as ATP, which suggests the quote may have been garbled - do you have a link to the original?
--- old post above --- --- new post below ---

I'm still waiting to hear why check rails were not installed on the curve because I think they would make it significantly less likely for the vehicle to derail.

The wheel profile of trams is significantly less than on railway vehicles, because they have to be able to run on street sections with grooved rails. This therefore seems to make them more susceptible to derailment than railway vehicles with full depth flanges.

Both of the above would be true in the case of a flange climb derailment, where the lateral force is enough to push the flange up and over the outer rail. In a flange climb the vehicle usually leaves the track at a tangent to the curve but upright, though of course something may cause it to overturn after that.

We await the report but the fact the tram was on its side suggests an overturning derailment where the whole vehicle tipped bodily sideways while still on the track. Neither check rails nor deeper flanges make any difference to this type of derailment, because the inner wheel is simply lifted off its rail as the vehicle tips over.

 

[Taunton]

Neither check rails nor deeper flanges make any difference to this type of derailment, because the inner wheel is simply lifted off its rail as the vehicle tips over.

I believe that a check rail on the outer rail would inhibit this, because if the vehicle does tilt significantly and pivot about the outer wheeltreads, the outer wheel flanges would be inhibited from tilting further by starting to rub, and then be restrained, by the outer check rail. Even better if the check rail can be somewhat proud of the running rails, as much as the suspension etc allows.

 

[Signal Head]

I believe that a check rail on the outer rail would inhibit this, because if the vehicle does tilt significantly and pivot about the outer wheeltreads, the outer wheel flanges would be inhibited from tilting further by starting to rub, and then be restrained, by the outer check rail. Even better if the check rail can be somewhat proud of the running rails, as much as the suspension etc allows.

I can't see that working at all, the flanges are too small to provide anything like enough purchase/leverage to resist the overturning force. It would be like trying to prise up an elephant using a 5p piece.

 

[Crossover]

I believe that a check rail on the outer rail would inhibit this, because if the vehicle does tilt significantly and pivot about the outer wheeltreads, the outer wheel flanges would be inhibited from tilting further by starting to rub, and then be restrained, by the outer check rail. Even better if the check rail can be somewhat proud of the running rails, as much as the suspension etc allows.

I can't see that working at all, the flanges are too small to provide anything like enough purchase/leverage to resist the overturning force. It would be like trying to prise up an elephant using a 5p piece.

I can't help but think that given the forces involved, the checkrail would be more likely to give way first, if the flange doesn't just get free of the checkrail anyway

 

[Signal Head]

Both of the above would be true in the case of a flange climb derailment, where the lateral force is enough to push the flange up and over the outer rail. In a flange climb the vehicle usually leaves the track at a tangent to the curve but upright, though of course something may cause it to overturn after that.

I have wondered if the overturn occurred, wholly or partially, as a result of the derailed tram hitting the 6' rail of the opposite track.

 

[littlerock]

here is the link

http://www.irishtimes.com/news/irel...as-says-trams-have-fail-safe-brakes-1.2864624

 

[Shaw S Hunter]

I have wondered if the overturn occurred, wholly or partially, as a result of the derailed tram hitting the 6' rail of the opposite track.

Indeed, there are various possibilities which I don't think have been raised in this discussion. That is probably because the precise location of the accident meant that the actual point of derailment was obscured from the view of any members of the public so we have no pictures of it.

Was the tipping solely due to the excessive speed or was there a track component failure under the forces caused by the excessive speed? Doubtless the investigators already have a good idea about this so maybe there will be some clues in the initial interim report due out this week sometime.

 

[craigybagel]

here is the link

http://www.irishtimes.com/news/irel...as-says-trams-have-fail-safe-brakes-1.2864624

Seemingly the MD has never looked into the cab of one of the LUAS Citadis', or looked out the front window. 70kmh top speed (as is the case with all Citadis' as far as I'm aware) both for the trams and for the lines they travel on.

 

[BestWestern]

Croydon trams are also limited to 80km/h although I think they just refuse to go any faster rather than applying the brakes, and they have a spring-loaded switch on the traction/brake controller which may be less sophisticated than the vigilance device mentioned in the quote.

Such a system sounds like a invitation for drivers to simply throw the lever into maximum and tune out when on the fast sections, minimal attention required as there is no need to actively regulate the speed. I wonder if that may have played a part.

 

[bradders1983]

Sorry if this has been discussed, but not going back through 36 pages to check: cant agree with them naming the driver. He could be innocent as far as we all know.

 

[edwin_m]

Indeed, there are various possibilities which I don't think have been raised in this discussion. That is probably because the precise location of the accident meant that the actual point of derailment was obscured from the view of any members of the public so we have no pictures of it.

Was the tipping solely due to the excessive speed or was there a track component failure under the forces caused by the excessive speed? Doubtless the investigators already have a good idea about this so maybe there will be some clues in the initial interim report due out this week sometime.

The relevant experts usually don't have major difficulties determining the sequence of events during a derailment by looking at things like witness marks on the wheel and rail. If there was a track failure that would probably be obvious too. So I would expect the next RAIB report to shed some light on this question, or if not that certainly the final report.