Derailment at Kirkby (Merseyside) - 13/03/2021

[Intermodal]

Interestingly it is a like for like replacement when the history implies that some improvement might be needed.

Buffer stops are not designed to stop trains travelling at any sort of speed.

 

[edwin_m]

Buffer stops are not designed to stop trains travelling at any sort of speed.

Please see previous posts discussing types that would be more effective at doing so than this one.

 

[Chris M]

Please see previous posts discussing types that would be more effective at doing so than this one.

Most (if not all) of which require more space than is available at this location. Yes the train could stop further down the platform, but that might require platform extensions and/or signalling changes. either of which would be time consuming and expensive. Given that significant changes are planned here in the short term spending significant money on anything other than a like-for-like replacement would be a waste.
The history shows that on average a train will hit the buffers at most once a decade, the replacements should be in place far less time than that.

 

[Intermodal]

Please see previous posts discussing types that would be more effective at doing so than this one.

Yes, I've seen them. None of them would've made any measurable difference at the speed the train was going. If the others are more expensive, I'm not convinced it's a cost effective improvement. A better management plan would be improved TPWS/reduced speeds/non-technical skills management.

 

[Mcq]

Amateur comment here - but thinking of the Moorgate disaster - the result is a speed timed section that is implimented with tripcocks.
I believe these tripcocks are being engineered out in the current referb.
Could some similar system be deployed when platform ends are so close to solid obstacles?
As I say I'm just enquiring - but when this sort of thing happens, Moorgate always comes to mind.

 

[edwin_m]

Most (if not all) of which require more space than is available at this location. Yes the train could stop further down the platform, but that might require platform extensions and/or signalling changes. either of which would be time consuming and expensive. Given that significant changes are planned here in the short term spending significant money on anything other than a like-for-like replacement would be a waste.
The history shows that on average a train will hit the buffers at most once a decade, the replacements should be in place far less time than that.

All most likely true at this location, but an unqualified statement that (all) buffer stops are ineffective at any sort of speed is somewhat misleading.

Yes, I've seen them. None of them would've made any measurable difference at the speed the train was going. If the others are more expensive, I'm not convinced it's a cost effective improvement. A better management plan would be improved TPWS/reduced speeds/non-technical skills management.

They would have made some difference by allowing uniform deceleration over whatever distance was available, rather than a sudden shock on hitting the stop and another one when hitting the walkway. They could perhaps also have been used in conjunction with a frangible surface in the platform walkway behind to increase that distance. This would have reduced the damage to the train end (where the driver might have been had things been a little different). Had the train been full and standing, it would probably have reduced the injuries from people falling over and hitting each other or interior fitting.

From previous posts I thought TPWS was already provided, and if so why it wasn't effective in stopping the train should be a key question for the enquiry. Assuming it is is in fact there, reducing the approach speed might be counter-productive as it risks the driver having to apply power to bring the train to the correct stopping position.

I also agree that assuming the extension of the electric service is extended, this probably will go away, so it's probably not worth doing anything at this specific site. But there may be lessons for other places.

 

[Bletchleyite]

Amateur comment here - but thinking of the Moorgate disaster - the result is a speed timed section that is implimented with tripcocks.
I believe these tripcocks are being engineered out in the current referb.
Could some similar system be deployed when platform ends are so close to solid obstacles?
As I say I'm just enquiring - but when this sort of thing happens, Moorgate always comes to mind.

Presumably TPWS could replace tripcocks - do the 777s have them?

 

[Domh245]

Presumably TPWS could replace tripcocks - do the 777s have them?

It looks like they do - there's a bit of steel plate on the off-side (tripcock side!) of the leading bogie ahead of the shoegear that isn't on the other side - can't see any pictures showing it in detail but with that plate it'd be a surprise if they didn't

 

[Llama]

TPWS is installed, in fact there are two installations approaching Kirkby - one for a permanent speed restriction of 15mph (reduction from 60mph) before the platform, and one at the standard distance before the buffer stop as there are on all dead end platform lines where TPWS is used and passenger trains run. But this shows the weaknesses of TPWS as a system compared to full ATP.

The train was travelling at a speed such that it was below the set-speed for the first set of TPWS OSS loops (a fairly normal approach speed at that point, and those loops are 295 metres before the 15mph PSR starts) but it was still travelling way too fast for the second set of loops to be effective in stopping the train, and if the brake was already applied to the emergency position before the train reached that second set of loops (set at ~10mph) the TPWS would have had no material effect to what was already happening anyway.

It would be possible to employ further, multiple, sets of TPWS OSS loops on approach set at lower and lower speeds. This would effectively force the train to comply with a certain braking curve which would need to be below the set speed of those loops at every installation (or if the train speed was above then the resultant TPWS activation would cause an emergency brake application) but as with everything there's a cost-benefit ratio to be applied.

The mitigations that are in place approaching Kirkby from the Liverpool side were better than the mitigations are when approaching Kirkby from the Wigan side.

 

[Mcq]

I don't know how many times - if ever - since 1975 a train has been halted at Moorgate for travelling too fast (be a bit embarrasing with most of the train still in the tunnel whilst the reset proceedure is invoked) but given the front carrisge is nearly always the fullest, this safety device is most reassuring.
Had the front coach of the Kirkby train been full, we might be looking at something other than what seems to be a basic reinstatement of previous.

 

[61653 HTAFC]

Without wishing to downplay the risks of overruns, there's a world of difference between running through the buffers in open air (though the over bridge and walkway at Kirkby slightly complicates things) and running into the end wall of a dead-end tunnel with low-height rolling stock which has space above allowing over-riding.

 

[Mcq]

Some difference granted - not a world of though.
As I said earlier if the front carriage had been full and not empty - like a buffer vehicle - then not only would this have been very much more serious but - if reports are coreect - the driver wouldn't have been able to escape backwards.
I agree that some layouts do provide enough space to safely overrun - but some, like Kirkby, less so - and Moorgate with its 717s, next to none.
When Moorgate's trip cocks go - what technology will replace them?

 

[edwin_m]

TPWS is installed, in fact there are two installations approaching Kirkby - one for a permanent speed restriction of 15mph (reduction from 60mph) before the platform, and one at the standard distance before the buffer stop as there are on all dead end platform lines where TPWS is used and passenger trains run. But this shows the weaknesses of TPWS as a system compared to full ATP.

The train was travelling at a speed such that it was below the set-speed for the first set of TPWS OSS loops (a fairly normal approach speed at that point, and those loops are 295 metres before the 15mph PSR starts) but it was still travelling way too fast for the second set of loops to be effective in stopping the train, and if the brake was already applied to the emergency position before the train reached that second set of loops (set at ~10mph) the TPWS would have had no material effect to what was already happening anyway.

It would be possible to employ further, multiple, sets of TPWS OSS loops on approach set at lower and lower speeds. This would effectively force the train to comply with a certain braking curve which would need to be below the set speed of those loops at every installation (or if the train speed was above then the resultant TPWS activation would cause an emergency brake application) but as with everything there's a cost-benefit ratio to be applied.

The mitigations that are in place approaching Kirkby from the Liverpool side were better than the mitigations are when approaching Kirkby from the Wigan side.

TPWS at signals is normallly designed so that if a train passes the OSS at just under the set speed, as long as it doesn't accelerate afterwards and the brakes and adhesion are normal it will be tripped at the TSS and come to a halt before any point of conflict. I'm not familiar with the buffer stop installations but I'd expect them to be similar, so this is still an important area for investigation.

 

[Grumpy Git]

Not trying to second guess any investigation here, but is it possible that there was a complete brake failure on the 507?

 

[Llama]

TPWS at signals is normallly designed so that if a train passes the OSS at just under the set speed, as long as it doesn't accelerate afterwards and the brakes and adhesion are normal it will be tripped at the TSS and come to a halt before any point of conflict. I'm not familiar with the buffer stop installations but I'd expect them to be similar, so this is still an important area for investigation.

No, buffer stop installations are different and are in no way related to OSS for any prior speed restriction (or signal).

The TPWS OSS for the 15mph PSR probably have a set speed somewhere between 35-45mph based on the layout, I don't have the set speed figure.

The TPWS OSS being set at ~10mph 55m from the block end would hardly have had time to initiate a brake application if a train goes over them at 25-30mph, which seems to be the case at Kirkby from the footage and other info leaked.

 

[AndyPJG]

Is it still a rule book requirement to do a running brake test using the straight air brake alone (ie not via 'EPB') prior to a dead end termination stop, or do current braking systems preclude this? eg using the straight air brake to stop at Vauxhall prior to arriving at Waterloo.

 

[Llama]

Only a rulebook requirement for LHCS and HSTs to carry out a running brake test before a a dead end platform line.

For multiple units it's 'work to your company instructions'. Our company don't require mandatory running brake tests before bay platforms, they do suggest it's good practice to do so though if required.

Straight air brake I understand to be just a loco brake though rather than on units, I think I know what you're getting at though. Any unit with a dynamic/rheo/regen brake should have that brake deactivated when carrying out a running brake test. Our CAF units have a running brake test button. On 323s the driver manually switches off the regen brake, does their RBT, then reinstates the regen brake afterwards.

 

[noddingdonkey]

Intriguing on those photos that the unit is shown with one door leaf closed. A fault, or for a reason, I wonder?

Wasn't/isn't that standard for the local door on Pacers?

 

[edwin_m]

No, buffer stop installations are different and are in no way related to OSS for any prior speed restriction (or signal).

The TPWS OSS for the 15mph PSR probably have a set speed somewhere between 35-45mph based on the layout, I don't have the set speed figure.

The TPWS OSS being set at ~10mph 55m from the block end would hardly have had time to initiate a brake application if a train goes over them at 25-30mph, which seems to be the case at Kirkby from the footage and other info leaked.

Thanks for the information. This difference in provision may be something else for RAIB to look at, although TPWS was never really intended to cover buffer stops so perhaps what is done now is the best compromise possible.

 

[Bletchleyite]

Wasn't/isn't that standard for the local door on Pacers?

Yep but never saw it on anything else.

 

[Llama]

It was the standard on Pacers because if both doors opened under local door operation then it would be dangerous to climb up and down when not at a platform, as one of the handrails was fixed to the outside of the door leaf that stayed closed.

 

[73001]

Looks like they've finished and it's open from tomorrow...

We are pleased to confirm that the repair work carried out by Network Rail has been completed and Kirkby line services will resume on Monday 22 March. Timetable information is available here: https://merseyrail.org/plan-your-journey/covid-19-information/covid-19-timetables.aspx

 

[Winthorpe]

Network Rail have posted on Twitter some pictures of the repairs at Kirkby station.

https://twitter.com/NetworkRailLIV/status/1373721901097451537

 

[ainsworth74]

Update via the RAIB regarding the scope of their investigation and a brief summary of events:

At around 18:53 hrs on 13 March 2021, the 18:35 hrs service from Liverpool Central to Kirkby struck the buffers located at the end of the Merseyrail Northern line platform at Kirkby station. The train approached the station at around 42 mph (68 km/h), before an emergency brake application was made, slowing the train before it hit the buffers. The train then derailed and collided with a platform extension which links the Northern line platform to the adjoining Kirkby branch line platform. The derailed train came a stop under a bridge and clear of the branch line, which was not occupied by a train at the time.

Twelve people reported suffering minor injuries as a result of the accident, and significant damage was caused to the train and to railway infrastructure.

Our investigation will seek to identify the sequence of events which led to the accident. It will also consider:

• the performance of the train and the infrastructure
• any factors that may have influenced the train driver’s actions
• the training, supervision and management of Merseyrail drivers
• the processes used to assess and control the risk of terminal platform overruns
• any relevant underlying factors

Our investigation is independent of any investigation by the railway industry, the British Transport Police or by the industry’s regulator, the Office of Rail and Road.

We will publish our findings, including any recommendations to improve safety, at the conclusion of our investigation. This report will be available on our website.

[Withdrawn] Buffer stop collision at Kirkby station

Investigation into a buffer stop collision at Kirkby station, Merseyside, 13 March 2021.

www.gov.uk

 

[Intermodal]

Approached the station at 42mph - scary stuff.

 

[185]

It doesn't say where it was doing 42mph, RAIB's summary is ambiguous and could be construed as rather misleading. The question I'd want to know is what speed was it doing when the front rolled into the station, and was it under the TPWS trip-speed. They may as well write OMGZ! the train was doing 70mph at some point.. (three days earlier probably near Ainsdale).

 

[BayPaul]

It doesn't say where it was doing 42mph, RAIB's summary is ambiguous and could be construed as rather misleading. The question I'd want to know is what speed was it doing when the front rolled into the station. They may as well write OMGZ! the train was doing 70mph at some point.. (three days earlier probably near Ainsdale).

Given it is a very short summary, and they have not yet had time to do their painstaking verification of the facts, I think it has sufficient clarity. 'Approaching the station' does clearly imply very close to the station and at a moment relevant to the investigation. No doubt the exact detail of the speed will be in the final report, but given this will be some time in the future, as usual RAIB are providing only facts that are verifiable, whilst giving a useful indication that speed is likely to be something that the report will focus on.

 

[43096]

It doesn't say where it was doing 42mph, RAIB's summary is ambiguous and could be construed as rather misleading. The question I'd want to know is what speed was it doing when the front rolled into the station, and was it under the TPWS trip-speed. They may as well write OMGZ! the train was doing 70mph at some point.. (three days earlier probably near Ainsdale).

Obviously RAIB think it is pertinent to the inquiry or they wouldn’t have stated it. Rather than demanding information, let them do their job and wait for the full report which doubtless will tell us the details.

 

[2HAP]

I'm getting the impression that it was doing 42 mph at a point where the speed should have been much lower.

 

[185]

wait for the full report

doing 42 mph at a point where the speed should have been much lower.

Speculatively, it probably was, but there should be no room for ambiguity in anything put out by RAIB, including these summarisations - despite this 'not being the full report'
this ambiguity could directly affect subsequent prosecutions.