Collision and derailment near Salisbury (Fisherton Tunnel) 31/10/21

[GC class B1]

To clarify, Wheel Slide Prevention (WSP) is a system generally fitted to disc braked vehicles to prevent wheelsets locking when braking in low adhesion conditions. It will apply and release the brakes to prevent the wheelset locking and is similar to ABS on cars. Wheel Slip Prevention is fitted to wheelsets that are motored to prevent the wheelset spinning when under power in low adhesion conditions. There is also a system called WSSP which works on motored axles to prevent both slip and slide.

 

[thedbdiboy]

If, and I stress IF, it was caused by Leaf Fall, what is in place to try taking it into account? Do drivers drive differently regarding technique in Autumn?

There are a lot of mitigations put into place to cover for poor railhead conditions in autumn - rail head treatment trains, services retimed to allow for extra braking distances etc. In that sense, the nature of the recommendations and 'learning points' (as they are termed in the RAIB reports) likely to result from the investigation will depend on whether there were mitigations for this location that were supposed to be carried out but which were omitted or sub-standard; or whether the mitigations themselves were inadequate for the particular location. The intention is not just to address the circumstances at Salisbury but also to identify where else such risks need to be managed better. In a similar fashion to the air industry, rail's enviable safety record is built on this constant learning from accidents and incidents.

 

[WestRiding]

So am I right in saying this technically was a SPAD? Red light comes on, SWR driver (or TPWS) brakes, braking system fails due to low adhesion, SWR ploughs into GWR. Is that how it went down? Or would 'SPAR' be more appropriate?

I believe the RED aspect had been happily showing. It did not revert from a proceed to danger in front of the train. SPAD.

 

[Surreytraveller]

So are we saying the driver moved away against a red rather than encountered a caution aspect, put the brake in and was unable to stop for the red?

The latter situation would be a SPAR I thought, the former a SPAD.

A caution aspect would precede a red, to give the driver their required notice to enable them to brake for the red.
Doesn't sound like the driver moved away against a red, it sounds like they were unable to stop the train at it

A caution aspect would precede a red, to give the driver their required notice to enable them to brake for the red.
Doesn't sound like the driver moved away against a red, it sounds like they were unable to stop the train at it

Being unable to stop at a red is still a SPAD, so long as the required cautionary aspects were displayed

 

[Domh245]

The latter situation would be a SPAR I thought, the former a SPAD.

SPAR is for a signal reverting to red in front of the driver. Various categories depending on why the signal reverted (equipment malfunction, signalling reverting but driver unable to stop in time, or train runaway)

A train passing a "valid" red signal is a SPAD (cat A), regardless of why the train passed the signal. That doesn't mean driver is at fault

 

[Wychwood93]

A SPAR is where the signal goes back to red, and the driver has no chance in stopping, but this was red, and even though it slid past, it would still be a SPAD
Its scary how far a train can slide when conditions are right, or more correctly 'wrong'

A gradient of 1:733 down from approx. Laverstock North to Tunnel Junction - source: my rather old copy of Ian Allan gradient profiles.

 

[Bow Fell]

So are we saying the driver moved away against a red rather than encountered a caution aspect, put the brake in and was unable to stop for the red?

The latter situation would be a SPAR I thought, the former a SPAD.

No;

The driver would have been braking for the 50 PSR and also having received a single yellow and the signal before SY31. No matter who’s fault it is, in the scenario, a SPAD is a SPAD.

A SPAR (or formerly CAT B SPAD) is the if the signal is thrown back for whatever reason (power failure, trespasser etc) and is passed.

 

[hexagon789]

A caution aspect would precede a red, to give the driver their required notice to enable them to brake for the red.
Doesn't sound like the driver moved away against a red, it sounds like they were unable to stop the train at it

SPAR is for a signal reverting to red in front of the driver. Various categories depending on why the signal reverted (equipment malfunction, signalling reverting but driver unable to stop in time, or train runaway)

A train passing a "valid" red signal is a SPAD (cat A), regardless of why the train passed the signal. That doesn't mean driver is at fault

No;

The driver would have been braking for the 50 PSR and also having received a single yellow and the signal before SY31. No matter who’s fault it is, in the scenario, a SPAD is a SPAD.

A SPAR (or formerly CAT B SPAD) is the if the signal is thrown back for whatever reason (power failure, trespasser etc) and is passed.

I've evidently misunderstood what constituted a SPAR then, I thought everything that wasn't previously Cat A was now a SPAR, ie anything that wasn't 100% the driver's fault including low adhesion preventing stopping in time.

 

[Efini92]

SY29 in rear also has TPWS (Trainstop and Overspeed) because of the conflict at Laverstock North Junction. The approach speed to that signal is 90, according to the Appendix. The 50 PSR towards Tunnel Junction begins at Laverstock North.

Cheers, I mistakenly thought it was 50 on approach.

I’d be interested to know if the emergency brake application caused by the TPWS intervention actually made the slide worse?



Once TPWS intervenes with a emergency brake application the driver loses all control and is basically a passenger.

Unlikely, If the train was in a full slide the only difference would be that the driver couldn’t release the brakes until the TPWS has timed out.
The braking instructions for trains with WSP is to leave the brake in rather than take it off and try and gain rotation.
It will be interesting to know if the sander hoppers had any sand in them.

 

[BrianW]

Already done!

Thank you SLF2021 for putting us in the picture and passing on thanks. It's very easy to get caught up in our own concerns- delay, refunds etc rather than concern for others and being grateful we are ok.

 

[Efini92]

I've evidently misunderstood what constituted a SPAR then, I thought everything that wasn't previously Cat A was now a SPAR, ie anything that wasn't 100% the driver's fault including low adhesion preventing stopping in time.

Im pretty sure you’re correct.
Under the old way it would be a cat D

 

[hexagon789]

Im pretty sure you’re correct.
Under the old way it would be a cat D

So in simple terms while previously a Cat D, a Cat D is not treated as a SPAR under the present system then?

 

[Efini92]

So in simple terms while previously a Cat D, a Cat D is not treated as a SPAR under the present system then?

A spad is just a cat A now. A spar is B,C or D

 

[hexagon789]

A spad is just a cat A now. A spar is B,C or D

So why isn't a low adhesion SPAD (previously B for unable to stop before signal) now a SPAR if Cats B-D were converted to SPARs?

I'm getting really confused!

 

[WestRiding]

Im pretty sure you’re correct.
Under the old way it would be a cat D

A catD SPAD was for such as a runaway wagon passing a signal at danger, not attached to any locomotive etc. Like what happened at Wrenthorpe sidings the other week,

 

[GB]

Im pretty sure you’re correct.
Under the old way it would be a cat D

It wouldn't have been a Cat D as that was reserved for spads that didn't have any traction unit attached or not in use. Eg a stabled train/wagon/set that had runaway.

 

[45669]

A SPAR is where the signal goes back to red, and the driver has no chance in stopping, but this was red, and even though it slid past, it would still be a SPAD
Its scary how far a train can slide when conditions are right, or more correctly 'wrong'

Several pages back there was a link to a video on YouTube about a case involving a SPAD caused by a train sliding on slippery rails; it took over a mile to stop. For anyone that has joined this thread recently and who may not have seen it, here it is again:

 

[43096]

It will be interesting to know if the sander hoppers had any sand in them.

Good question. From what has been said elsewhere, my understanding is that 158/159 sets with sanding fitted automatically apply sand if the brake is in step 2 or above and wheelslip is detected by the WSP system.

 

[GB]

So why isn't a low adhesion SPAD (previously B for unable to stop before signal) now a SPAR if Cats B-D were converted to SPARs?

I'm getting really confused!

It is still a SPAD (cat A) because the signal was displayed correctly and in the correct sequence for the train to stop at it. As someone said up thread, SPAD (cat A) does not mean its the drivers fault.

 

[the sniper]

The poor journalists heads will be spinning reading this...

 

[WestRiding]

Its hard trying to explain Railway Signalling to non Signallers

 

[Efini92]

It wouldn't have been a Cat D as that was reserved for spads that didn't have any traction unit attached or not in use. Eg a stabled train/wagon/set that had runaway.

I seem to remember it just saying a cat D was a runaway. However after reading your second reply I’d have to agree.

It is still a SPAD (cat A) because the signal was displayed correctly and in the correct sequence for the train to stop at it. As someone said up thread, SPAD (cat A) does not mean its the drivers fault.

 

[brad465]

So am I right in saying this technically was a SPAD? Red light comes on, SWR driver (or TPWS) brakes, braking system fails due to low adhesion, SWR ploughs into GWR. Is that how it went down? Or would 'SPAR' be more appropriate?

I believe the RED aspect had been happily showing. It did not revert from a proceed to danger in front of the train. SPAD.

Nice to see that the statement clearing up all the confusion and speculation about what caused the collision has only led to more confusion in the form of SPAD/SPAR or whatever terms exist for going past a red/danger signal.

 

[swt_passenger]

Will there be any internal questions as to why a Network Rail spokesman apparently announced that the signalling system had been “knocked out”?

 

[KendalR]

The poor journalists heads will be spinning reading this...

For any journalist snooping on these forums I suggest they read the Rail Delivery Group's 'Adhesion Working Group' report: Managing Low Adhesion Sixth Edition, January 2018

All 351 pages.

 

[Sunset route]

I hope this helps, the SPAR term came after the alterations.

 

[Adversane]

BBC South Today news
View attachment IMG_0052.mov

 

[Signal Head]

The three junctions (Tunnel Junction, Laverstock North, and Laverstock South are all under the protection of a T3 occupation. Before any train service will be allowed to operate, the investigation evidence gathering has to conclude. The BTP have to hand the scene over to the RIAB. RIAB may well do further investigation. They then have to hand the scene back to Network Rail. Then the trains need to be removed. Then extensive testing of the signalling system will be carried out. But if the points and track are damaged too much, they may have to have some work done on them first.
Once all the investigation testing has been done, then the damaged track and point work needs to be repaired or replaced. The tunnel walls and roof need to be inspected and if needed, repaired.

I would be very surprised if all this has been completed before Saturday morning.


Technically speaking, all objects are in motion. The question is really the relative motion between two objects.

Looking on Google Earth/Streetview there appears to be a signalling equipment case ('location case' ) adjacent to the junction points. That space is now very likely occupied by the rear coach of the GWR 158, with whatever remains of the loc case and its contents underneath it.

Given that the whole of the east end of Salisbury panel probably works though that case (line cable terminations and power at the very least), I doubt there will be any operational signalling that side of the tunnel until it has been replaced with a new one.

This also probably explains the early reports of signalling being 'knocked out' - the main 650V signalling supply will pass through it, depending on the damage that could have been shorted out and blown the supply fuses further back.

 

[swt_passenger]

Looking on Google Earth/Streetview there appears to be a signalling equipment case ('location case' ) adjacent to the junction points. That space is now very likely occupied by the rear coach of the GWR 158, with whatever remains of the loc case and its contents underneath it.

Given that the whole of the east end of Salisbury panel probably works though that case (line cable terminations and power at the very least), I doubt there will be any operational signalling that side of the tunnel until it has been replaced with a new one.

This also probably explains the early reports of signalling being 'knocked out' - the main 650V signalling supply will pass through it, depending on the damage that could have been shorted out and blown the supply fuses further back.

Thanks for the explanation. I guess NR’s initial announcements were probably before the whole timeline was confirmed, ie when they were still thinking there were two phases to the incident?

 

[Signal Head]

Thanks for the explanation. I guess NR’s initial announcements were probably before the whole timeline was confirmed, ie when they were still thinking there were two phases to the incident?

I think that's very likely what's happened. The signalling failure as a consequence of a single collision, rather than the consequence of one and the cause of another.

I'm also not very impressed by whoever was sounding off about there being a "major signalling flaw", as reported in the press and apparently attributed to "a senior NR engineer" (or 'engineer').

Is this the press making stuff up, or has someone really been irresponsible enough to come out with that stuff, I wonder?