Upcoming Games

No games to display

Full list
Add a game

Upcoming Events

No events to display

RAIB discussion

You are here: Home > Forum > Miscellaneous > The real thing (anything else rail-oriented) > RAIB discussion

Page 2 of 3

RAIB discussion 19/02/2015 at 20:43 #69365
Jersey_Mike
Avatar
250 posts
" said:

It would be wrong to just say "oh, he went past a signal, must be his fault" for incidents like this. Why did he go past that signal? Is there a past history of drivers passing that signal? If so, why do they keep going past? Once you determine that, perhaps you can reduce the risk of future drivers passing that signal. After all, that is the aim of inquiries.
As per my original question I am simply surprised that such questions are the subject of a government inquiry as opposed to an internal safety review.

" said:

One thing that struck me about the Metro North derailment was that it blamed a joint bar that failed - fair enough, direct cause. But it completely neglected the reasons why it failed (fatigue is not a "why" but a "what happened"). The only conclusions it could come up with was to do a visual check a bit closer than up to 39(?) feet away - and you can't check for even visible cracks more than a couple of feet away, so what they're really saying is "look for obviously broken joint bars" rather than "detect breaks before they happen". :unsure:
Metal rail components wear out...that's simply what they do. Best you can do is hope to catch the failures before a derailment. Monetary "broken rail" bounties have been very effective on some railroads to encourage better inspection techniques. BTW the official report doesn't contain a photo I obtained that showed the broken rail right after the accident. One broken splice bar was shiny at the break...on the other one the break was rusty. That's why the cause was tracked to poor inspection and maintenance. (MNRR is having a really rough time because all their top management and veteran employees joined in 1983 when the railroad was created and they all retired in 2013 to get their full pension. The FRA is not auditing their internal safety processes (not fun) because they have clearly broken down).

Log in to reply
RAIB discussion 19/02/2015 at 21:14 #69366
Ron_J
Avatar
329 posts
Mike; The Greenford incident was subject to an internal investigation jointly conducted by Network Rail and Chiltern Railways, with trade union representatives on the investigation panel as observers. The RAIB investigation gives an extra level of oversight and it's not unheard of for them to reach different conclusions from the internal railway company investigation.

On a sort of related note, when I first left the signalbox and went to HQ to work in the Ops Safety team I was told by my boss that UIC normalised all the safety data sets (FWI, AFR, PSI etc..) and found that the UK rail industry is the second safest in the entire world after Luxembourg. So that's something I suppose...

Log in to reply
RAIB discussion 19/02/2015 at 22:37 #69371
Peter Bennet
Avatar
5354 posts
" said:

As per my original question I am simply surprised that such questions are the subject of a government inquiry as opposed to an internal safety review.
On a technical and constitutional level the term "Government enquiry" is probably the wrong one to use.

As well as the RAIB there are separate Marine and Air accident investigation branches which, while operating under the wing of the DfT, investigate within a legal framework and more importantly independent of the entities they are investigating so as to ensure impartiality. The fact that it happens to be a public body is more a fact than critical to the matter.

A Government enquiry is a relatively rare thing and set up to look at a specific matter with it's own terms of reference etc.

Peter

I identify as half man half biscuit - crumbs!
Last edited: 19/02/2015 at 22:42 by Peter Bennet
Log in to reply
RAIB discussion 20/02/2015 at 00:03 #69376
GeoffM
Avatar
6274 posts
" said:
" said:

It would be wrong to just say "oh, he went past a signal, must be his fault" for incidents like this. Why did he go past that signal? Is there a past history of drivers passing that signal? If so, why do they keep going past? Once you determine that, perhaps you can reduce the risk of future drivers passing that signal. After all, that is the aim of inquiries.
As per my original question I am simply surprised that such questions are the subject of a government inquiry as opposed to an internal safety review.
As has been stated more than once, independence.


" said:

One thing that struck me about the Metro North derailment was that it blamed a joint bar that failed - fair enough, direct cause. But it completely neglected the reasons why it failed (fatigue is not a "why" but a "what happened"). The only conclusions it could come up with was to do a visual check a bit closer than up to 39(?) feet away - and you can't check for even visible cracks more than a couple of feet away, so what they're really saying is "look for obviously broken joint bars" rather than "detect breaks before they happen". :unsure:
Metal rail components wear out...that's simply what they do.[/quote]

It's one of the things they do. You can't just shrug your shoulders, mutter "it's old", and walk away, job done, conclusions drawn! Was it manufactured correctly? Was it inspected before it was installed (as far as I can tell, it was not new when installed in this location - which is fine, but were opportunities missed)? All questions that should be asked. If they lead to a dead end, that's fine too, but at least you asked the question.

SimSig Boss
Log in to reply
RAIB discussion 20/02/2015 at 00:35 #69381
RainbowNines
Avatar
272 posts
Even at the most base level, What confuses me is that Mike is from a country where courts apportion blame and award damages left right and centre...

Does that insatiable need for finding someone to blame to sue the seven bells out of them not apply on the railway?

If the bereaved families sue the railway you can sure as hell bet the railway will be looking for damages from (e.g.) the manufacturer, or the maintenance contractor. etc, etc.

And that's before we think about how to stop it recurring...

Log in to reply
RAIB discussion 20/02/2015 at 14:26 #69386
Jersey_Mike
Avatar
250 posts
" said:

As has been stated more than once, independence.
Well of course the RIAB is less dependent than a TOC or NR (or Railtrack before it), but was there a policy choice to move more investigations out of those organizations? I can see that back in the British Rail days there would be less sense of "escalation" since BR was a wing of the government anyway.


" said:

It's one of the things they do. You can't just shrug your shoulders, mutter "it's old", and walk away, job done, conclusions drawn! Was it manufactured correctly? Was it inspected before it was installed (as far as I can tell, it was not new when installed in this location - which is fine, but were opportunities missed)? All questions that should be asked. If they lead to a dead end, that's fine too, but at least you asked the question.
The entire safety regime for track components is one of periodic inspection. Since trains are naturally derailment resistant, defects don't pose the same risk of catastrophic accidents as they do on aircraft. It doesn't matter if there are manufacturing defects because if inspected properly the defects are detected and repaired before they cause a problem. This is why we have a strong commitment to track circuits and also why there is a new (2012) national 60mph speed limit on any track without active broken rail detection.

In the case of Metro North the splice bar had been broken for weeks or months prior to the accident and should have been discovered in an inspection. From a liability standpoint the railroad is the deep pocket for anything beyond a nuisance payment. Also in most cases the components are decades old (example this section of 155 pound Pennsylvania Special rail from 1948 on a 90mph line in Pennsylvania) so good luck trying to sue the manufacturer.

Equivalent levels of safety can be achieved through different paths. For example I think the UK is far more tolerant of SPAD's than the United States is. It was stated in the linked investigation that the signaler was not alarmed by the SPAD because the TPWS system would stop the train. In the US any stop signal violation takes the train out of service, triggers an internal investigation and results in the at fault parties being suspended or fired. It's a major headache every time it happens, which is fortunately not often. Where the UK uses overlaps to mitigate against SPAD's, the US uses approach speed control, two person train operation, rigorous training and increased crash protection.

Getting back to reports, if there are minor problems that are superseded by some major problem, it is probably counter productive to waste decision maker's limited cognitive and economic resources on the minor problems. In this case the major problem was that the driver left his initial terminal with TPWS disabled. Fix that and all the little stuff is covered. Of secondary importance is the GSM-R issues which are a third line of defense past the signls and the TPWS. Beyond that you're into noise. You can't fix the noise because humans aren't robots and will always deviate slightly from the established procedures.

Log in to reply
RAIB discussion 20/02/2015 at 15:19 #69388
Muzer
Avatar
718 posts
I assume active broken rail detection is simply using a track circuit to detect a broken rail? Because that won't detect every type of break. And trains while they may be hard to derail on a straight, are a lot easier to derail on corners... but our inspections, both visual and more technical (with the New Measurement Train and co) are presumably much more thorough than yours (we have a smaller country so it sort of makes sense!), hence the reduced requirement for track circuits/active detection.



I'm pretty sure the train, or at least the driver, would be taken out of service on the occasion of a SPAD here, except probably a technical SPAD/SPAR/operational incident/whatever as long as the driver is not too shaken up.



"In this case the major problem was that the driver left his initial terminal with TPWS disabled. Fix that and all the little stuff is covered."

But if you just say it's the driver's fault for leaving with TPWS disabled, you've removed the whole possibility of redesigning the system so that such a stupid possibility isn't possible in the first place, at least without active acknowledgement by the driver so they know they're doing something they shouldn't take lightly. The point of a RAIB report is to fix every little thing that might make such an incident more likely. You keep doing that for a while, and it takes a string of multiple unlikely failures for a major incident to occur. Which is what happened here, when you think about it — the first line failed because the driver didn't stop at the signal. The second line failed because the TPWS didn't stop the train at the signal. The third line failed because the GSM-R display was confusing and so the signaller thought it would be ineffective at stopping the train [apologies if I'm making an error here, I haven't read that report in quite some time]. Every time something relatively major happens, the RAIB report always details a string of failures, and (with a few exceptions) usually suggests something to fix the vast majority of them, providing even more lines of defence.

I think on this point, our record basically speaks for itself. Again, no passenger deaths on trains (caused by a crash/safety incident) since 2007. That crash is coming up to its 8th anniversary in a few days. Given the massive proliferation of crashes throughout the history of our railways even up to quite recently, I think that really says something about our current standards, and long may the record hold.

Last edited: 20/02/2015 at 15:20 by Muzer
Log in to reply
RAIB discussion 20/02/2015 at 16:53 #69390
Steamer
Avatar
3913 posts
" said:
" said:

As has been stated more than once, independence.
Well of course the RIAB is less dependent than a TOC or NR (or Railtrack before it), but was there a policy choice to move more investigations out of those organizations? I can see that back in the British Rail days there would be less sense of "escalation" since BR was a wing of the government anyway.
Railway investiagations have always been handled by an outside (i.e. independent) body, to avoid bias. It's not very credible for a company to investigate itself, too much potential for covering up or using one party as a scapegoat.


" said:
The entire safety regime for track components is one of periodic inspection. Since trains are naturally derailment resistant, defects don't pose the same risk of catastrophic accidents as they do on aircraft.
What nonsense is that? While today's trains are better than those in days gone by, a collision or derailment can still result in deaths or life-changing injuries, as well as causing serious disruption and an expensive clean-up bill. Better crashworthiness mitigates the effects of an accident, but every effort should still be made to avoid the accident in the first place.
Quote:

In the case of Metro North the splice bar had been broken for weeks or months prior to the accident and should have been discovered in an inspection.
So why wasn't it? Either procedures weren't followed, or the procedures were inadequate. The investigation report should examine which of those is the case and suggest steps to prevent it happening again.

Quote:
Equivalent levels of safety can be achieved through different paths. For example I think the UK is far more tolerant of SPAD's than the United States is. It was stated in the linked investigation that the signaler was not alarmed by the SPAD because the TPWS system would stop the train. In the US any stop signal violation takes the train out of service, triggers an internal investigation and results in the at fault parties being suspended or fired. It's a major headache every time it happens, which is fortunately not often.


I don't know how you've got that idea. Everything you've said happens in the UK- the SPAD is investigated, and drivers disciplined or sacked.

Quote:
Where the UK uses overlaps to mitigate against SPAD's, the US uses approach speed control, two person train operation, rigorous training and increased crash protection.
Overlaps are belt-and-braces, designed to reduce the effects should something go wrong. Drivers are still trained to a high standard, and as stated above a SPAD is taken just as seriously now as it was when TPWS etc. wasn't available, if not more so. 'Defensive Driving' rules require them to reduce speed to a certain amount/percentage of line speed when they see a yellow or double yellow signal. Crashworthiness requirements have also become increasingly strict over the years.

Quote:
Of secondary importance is the GSM-R issues which are a third line of defense past the signls and the TPWS.
There was still a problem with the GSM-R though, which needed to be resolved. What if, next time, a signaller needed to stop trains because of a landslide, or something else, but was too late because the GSM-R was badly designed?

While each accident has a single primary cause, there's normally a catalogue of failures underpinning that cause, which could equally cause a different type of accident.

"Don't stress/ relax/ let life roll off your backs./ Except for death and paying taxes/ everything in life.../ is only for now." (Avenue Q)
Log in to reply
RAIB discussion 20/02/2015 at 17:00 #69391
postal
Avatar
5189 posts
Online
I know that the articles linked from this site are written by someone with an agenda but if what is written is symptomatic of the culture elsewhere in the US rail systems then I am quite content to let Jersey Mike continue to support the US safety regime and travel as many miles as he wants on US trains but I will feel a lot safer confining myself to travel on the UK system.
“In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe
Log in to reply
RAIB discussion 20/02/2015 at 18:20 #69392
Ron_J
Avatar
329 posts
" said:
Railway investiagations have always been handled by an outside (i.e. independent) body, to avoid bias. It's not very credible for a company to investigate itself, too much potential for covering up or using one party as a scapegoat.
I agree with everything else in your post but I must point out that the railway industry has always conducted internal investigations, both into events which have attracted the interest of extrenal bodies (HMRI/RAIB etc..) and into 'minor' incidents which don't require intervention by external bodies. Indeed, managers in the Operations/Standards/Safety department are given detailed Accident Investigation training to conduct such investigations.

It might blow Mike's mind if I mention that as well as internal railway company and RAIB investigations, seperate investigations into accidents/incidents are often carried out by ORR, the government regulator.

Log in to reply
RAIB discussion 20/02/2015 at 19:32 #69394
AndyG
Avatar
1830 posts
" said:
It might blow Mike's mind if I mention that as well as internal railway company and RAIB investigations, seperate investigations into accidents/incidents are often carried out by ORR, the government regulator.
Indeed, as per the ORR investigation into the Finsbury Circus Park debacle after the Christmas works.

I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either.
Log in to reply
RAIB discussion 21/02/2015 at 00:08 #69401
arabianights
Avatar
138 posts
i am america
Log in to reply
RAIB discussion 01/03/2015 at 01:55 #69702
BarryM
Avatar
2158 posts
Mike, is this one of the 10 page NTSB reports you were referring to?

Here is a 58 page pdf report on a Union Pacific collision 25 May 2013. Link

Barry

Barry, Sydney, New South Wales, Australia
Log in to reply
RAIB discussion 03/03/2015 at 16:12 #69793
Jersey_Mike
Avatar
250 posts
" said:
Mike, is this one of the 10 page NTSB reports you were referring to?

Here is a 58 page pdf report on a Union Pacific collision 25 May 2013. Link

Barry
Some reports are better than others. There is clearly a triage process where more newsworthy accidents get longer reports and run of the mill accidents get short blurbs or nothing at all. I'll bet that because this accident collapsed a road bridge they were able to draw resources from the highway investigation division.

Was looking through my e-mails and one RIAB accident report was about the edge of a platform damaging some rolling stock and it got the same sort of attention as something much more serious. Just feels that the safety oversight infrastructure doesn't know how to adjust it's level of effort.

Log in to reply
RAIB discussion 03/03/2015 at 16:29 #69794
TimTamToe
Avatar
654 posts
" said:

Was looking through my e-mails and one RIAB accident report was about the edge of a platform damaging some rolling stock and it got the same sort of attention as something much more serious. Just feels that the safety oversight infrastructure doesn't know how to adjust it's level of effort.
It may sound trivial to you but it could cause serious injury or a fatality to passengers and staff on the platform from flying debris from the train. Just because in the UK we take safety seriously, I don't see why you continuously knock it. As has been mentioned many times, things learnt from these investigations can change the way things are done in the future so that the same things don't happen again. If left next time people may not be so lucky.

Log in to reply
RAIB discussion 03/03/2015 at 17:49 #69797
Jersey_Mike
Avatar
250 posts
" said:

It may sound trivial to you but it could cause serious injury or a fatality to passengers and staff on the platform from flying debris from the train. Just because in the UK we take safety seriously, I don't see why you continuously knock it. As has been mentioned many times, things learnt from these investigations can change the way things are done in the future so that the same things don't happen again. If left next time people may not be so lucky.
Proper risk assessment is how one takes safety seriously. Trust me, what the United States saves on rail safety flies out the window when it comes to anti-terrorism related "security" efforts. I once saw a great chart that details much time off a typical life a whole host of threats and activities take. Terrorism was something trivial like 5 minutes. Riding in road vehicles was something on the order of 2-3 years. Smoking was something like 6 years and the number 1 threat to one's life was simply being poor, at 9 years. I'm sure that last figure is less in countries with larger social safety nets, but all the surface level safety gains could be lost due to decreased availability of rail transport and lower economic growth.

BTW, from the way that speed cameras are constantly reviled yet continually expanded, I figured most of the health and safety culture was being crammed down the public's gullet by a largely unaccountable bureaucracy that was mostly interested in their own officiousness. Has watching Top Gear given me the wrong impression?

Log in to reply
RAIB discussion 03/03/2015 at 21:20 #69801
Muzer
Avatar
718 posts
Yes, if you want a fair and balanced opinion of the general public, Top Gear is pretty much famously NOT the place to go
Log in to reply
RAIB discussion 03/03/2015 at 23:32 #69808
GeoffM
Avatar
6274 posts
" said:
Has watching Top Gear given me the wrong impression?
Never let the truth get in the way of a good - er - TV show.

SimSig Boss
Log in to reply
The following users said thank you: postal, ozrail
RAIB discussion 04/03/2015 at 00:51 #69814
ozrail
Avatar
197 posts
Have a look at the ATSB website atsb.gov.au (I don't know how to do a link) and the report into ASBs that is similar to the UKs T12 Rule, which I think has been changed to Handbook 21. Anyway. On page 11 it purports to show the Homebush control centre (similar to IECC), but the photo is in fact the RMC the Train Control centre that had no control over any trains or Signalling. How could they get it so wrong?
Last edited: 04/03/2015 at 00:52 by ozrail
Log in to reply
RAIB discussion 04/03/2015 at 09:19 #69816
BarryM
Avatar
2158 posts
I suggest the guys in Canberra have never been to the Homebush/Strathfield complex.

Barry

Barry, Sydney, New South Wales, Australia
Log in to reply
RAIB discussion 21/05/2015 at 23:18 #72580
Muzer
Avatar
718 posts
(I'm assuming this is the place to post for discussion of RAIB reports).

Quote:
RAIB has released its report into the runaway and subsequent collision near Loughborough Central station, Great Central Railway, 12 May 2014.

R042015_Loughborough_Central.pdf

My question is: Why does the class 37 not have fail-safe automatic brakes? I would have thought those would be pretty standard; some system whereby zero pressure in the system means the brakes are fully applied (perhaps with some clever spring mechanism or something). Is this just mechanically impossible/unfeasible, or is there another good reason this isn't present? Is this present on many/any locos? I was always under the impression that automatic brakes were generally fail-safe, but clearly (in the long term; maybe they are in the short term) they aren't.

Log in to reply
RAIB discussion 22/05/2015 at 01:51 #72581
BarryM
Avatar
2158 posts
It is not Muzer. The original is down on this front page still named All RAIB Reports by Tallington. All the reports on this page should be moved. AndyG, could you move them please?

Barry

Barry, Sydney, New South Wales, Australia
Log in to reply
RAIB discussion 22/05/2015 at 03:26 #72582
GeoffM
Avatar
6274 posts
" said:
It is not Muzer. The original is down on this front page still named All RAIB Reports by Tallington. All the reports on this page should be moved. AndyG, could you move them please?

Barry
Moved. I'll rename the thread as well, to avoid confusion. This was split off from some heated discussion that started in that thread.

SimSig Boss
Log in to reply
The following user said thank you: BarryM
RAIB discussion 22/05/2015 at 07:05 #72583
sorabain
Avatar
72 posts
" said:
(I'm assuming this is the place to post for discussion of RAIB reports).

Quote:
RAIB has released its report into the runaway and subsequent collision near Loughborough Central station, Great Central Railway, 12 May 2014.

R042015_Loughborough_Central.pdf

My question is: Why does the class 37 not have fail-safe automatic brakes? I would have thought those would be pretty standard; some system whereby zero pressure in the system means the brakes are fully applied (perhaps with some clever spring mechanism or something). Is this just mechanically impossible/unfeasible, or is there another good reason this isn't present? Is this present on many/any locos? I was always under the impression that automatic brakes were generally fail-safe, but clearly (in the long term; maybe they are in the short term) they aren't.
I was a bit surprised as well. I'd heard in the past that air brakes were "fail-safe" such that if a train ended up split then the brakes would be automatically applied. In my mind I assumed it took air pressure to take the brakes off... after reading a bit about this, and doing a little bit of googling it looks like the "fail-safe" part relies on a local reservoir of air to apply the brakes, i.e. still takes air pressure to apply the brakes, and once (or if) that local reservoir ends up drained the brakes will release again. Yikes.

I'm not sure why they're designed this way. At first glance it seems liked a relatively complicated design to include these local air reservoirs (e.g. http://www.treehugger.com/energy-disasters/what-caused-train-disaster-not-brake-failure.html). I assume there's a great reason for it since I trust engineers to seek simple solutions where possible, but the reason is beyond me at the moment.

Log in to reply
RAIB discussion 22/05/2015 at 20:57 #72598
kbarber
Avatar
1708 posts
" said:
" said:
(I'm assuming this is the place to post for discussion of RAIB reports).

Quote:
RAIB has released its report into the runaway and subsequent collision near Loughborough Central station, Great Central Railway, 12 May 2014.

R042015_Loughborough_Central.pdf

My question is: Why does the class 37 not have fail-safe automatic brakes? I would have thought those would be pretty standard; some system whereby zero pressure in the system means the brakes are fully applied (perhaps with some clever spring mechanism or something). Is this just mechanically impossible/unfeasible, or is there another good reason this isn't present? Is this present on many/any locos? I was always under the impression that automatic brakes were generally fail-safe, but clearly (in the long term; maybe they are in the short term) they aren't.
I was a bit surprised as well. I'd heard in the past that air brakes were "fail-safe" such that if a train ended up split then the brakes would be automatically applied. In my mind I assumed it took air pressure to take the brakes off... after reading a bit about this, and doing a little bit of googling it looks like the "fail-safe" part relies on a local reservoir of air to apply the brakes, i.e. still takes air pressure to apply the brakes, and once (or if) that local reservoir ends up drained the brakes will release again. Yikes.

I'm not sure why they're designed this way. At first glance it seems liked a relatively complicated design to include these local air reservoirs (e.g. http://www.treehugger.com/energy-disasters/what-caused-train-disaster-not-brake-failure.html). I assume there's a great reason for it since I trust engineers to seek simple solutions where possible, but the reason is beyond me at the moment.
Simply, in the event of a failure, you want to be able to move the blessed thing (under suitable precautions of course) to get it clear of running lines and get traffic moving. Even more so rolling stock, that needs the brakes releasing for shunting. I can assure you that shunting 'with the air in' (as we put it in the Brent) is a lot slower than pulling the strings and loose shunting; we would only do it if we had a good long raft (sometimes we'd get a good load of Dover ferryvans on the front of the up Bescot). If we'd had to shunt everything with the air in, the Brent would've been completely blocked by the end of day one.

Log in to reply