Incident at Talerddig, Wales - 21/10/2024

[zwk500]

Are they also not spring loaded/hydraulic points at Talerdig? Admittedly, I haven't been on the line since RETB was phased out, at that time there were quite a few of those type of points on the Cambrian

They would have been replaced with powered points when ETCS was installed, as the loops are now all bi-di (from the map on OTT).

 

[embers25]

I see Salisbury-Exeter is running at reduced speed today due to poor adhesion. Have Network Rail put a more restrictive passing loop procedure in place as a result of this (i.e. the other train can't enter the signal protecting the passing loop before the driver contacts the signaler to confirm they have stopped correctly in the loop?)

 

[thejuggler]

Observation from watching the news earlier is how we have a very poor idea of how fast we are travelling when on transport and any accounts immediately after an indicent must be treat with caution.

Passenger interviewed (who was visibly still shocked), stated the train went from "40, 50 or even 60 mph" before coming to a sudden stop. This part of the clip has now been removed and reports are it is being considered a low speed collision, which looking at the damage is probably correct.

 

[MP393]

I see Salisbury-Exeter is running at reduced speed today due to poor adhesion. Have Network Rail put a more restrictive passing loop procedure in place as a result of this (i.e. the other train can't enter the signal protecting the passing loop before the driver contacts the signaler to confirm they have stopped correctly in the loop?)

The Heart of Wales has also been closed now due to poor railhead conditions so it’s clear decisions are certainly being made differently today, unsurprisingly of course.

 

[Whistler40145]

Has the RHTT visited the Cambrian Line in this year's leaf fall season?

 

[westcoaster]

Technicality explained

I've seen a SPAD mentioned a few times. Remember you can not have a SPAD in ETCS, as there are no signals.

What you have is exceeding Movement Authority, when you pass the End of Authority (Block Marker Boards), the train will enter Trip Mode (same as a Brake demand under TPWS).

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==

Has the RHTT visited the Cambrian Line in this year's leaf fall season?

Believe so.

We also had Storm Ashley on Sunday, high winds and rain

 

[Deepgreen]

this will be interesting to see the outcome of. And also whether ETCS was taking any measures for low adhesion (I don't know enough about how the system works in practice to know what it should have been doing on it's own or with signaller input for low adhesion).

Loss of adhesion is the single train movement issue that no amount of signalling technology can negate. Trains can be stopped if passing a red light, exceeding the line speed, etc., but the fundamental steel wheel on steel rail interface relies on friction and if that is absent, control is lost.

 

[matt]

Has the RHTT visited the Cambrian Line in this year's leaf fall season?

Yes it ran on Sunday night.

 

[Bertiethebobby]

Technicality explained

I've seen a SPAD mentioned a few times. Remember you can not have a SPAD in ETCS, as there are no signals.

What you have is exceeding Movement Authority, when you pass the End of Authority (Block Marker Boards), the train will enter Trip Mode (same as a Brake demand under TPWS).

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==


Believe so.

We also had Storm Ashley on Sunday, high winds and rain

Thanks for that re exceeding Movement Authority that was going to be my question.

Is this ETCS version clever enough that when it detects one train has exceeded its MA that it then cancels the authority of the other approaching train, thus bringing it to a stand?

Regarding the RHTT, someone said earlier in the thread that it had run the night before last.

 

[westcoaster]

Thanks for that re exceeding Movement Authority that was going to be my question.

Is this ETCS version clever enough that when it detects one train has exceeded its MA that it then cancels the authority of the other approaching train, thus bringing it to a stand?

Regarding the RHTT, someone said earlier in the thread that it had run the night before last.

I'm not sure on the technical aspect, as you'd have to factor in many variables.
Time, distance travelled and speed, any overlap distance available.

Crucial one would be time. How long from passing the loop until the collision.

Does anyone know how far passed the loop the Collision was.

 

[Harpo]

What you have is exceeding Movement Authority, when you pass the End of Authority (Block Marker Boards), the train will enter Trip Mode (same as a Brake demand under TPWS).

ETCS will issue a warning if the braking curve is in danger of not being met at any point in a journey. It will intervene (brake demand) to attempt to restore it.

 

[GRALISTAIR]

Loss of adhesion is the single train movement issue that no amount of signalling technology can negate. Trains can be stopped if passing a red light, exceeding the line speed, etc., but the fundamental steel wheel on steel rail interface relies on friction and if that is absent, control is lost.

Well worth repeating. I will check in on the vegetation management thread.

Just waking up to this news in the USA after traveling yesterday. My thoughts are with all those affected. RIP to the deceased.

 

[Egg Centric]

The system will very likely have done its thing but the driver was likely in emergency and deployed the emergency one shot sander before ERTMS tried to intervene.

For anyone else who'd never heard of this, here is a helpful thread.

 

[Efini92]

That is correct.

From what I've been told the collision was due to slippery rails on the bank. Nothing to do with signalling. Although it does seem both units ended on the single line so not sure how one can exit the loop slipping while the other is approaching, and you'd think the points would be set and would cause the slipping train to derail. Unless it happened right on the points

The points were set, the train that slid through would’ve gone through the points forcing them over and allowing it onto the single line.

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==

The system will very likely have done its thing but the driver was likely in emergency and deployed the emergency one shot sander before ERTMS tried to intervene.

The other train was climbing a steep bank, whilst the other one would have increased its speed the further down the bank it got, so the further up the bank the non-sliding train got, the less the impact would have been.

It makes no adjustment for it, that is all on the driver.

Do the TFW 158’s still have 1 shot sand? I thought all the 158’s had this replaced years ago.

 

[High Dyke]

Have you got evidence to back up your guess?

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==

So, speculation on wheel slide, has anyone got any factual evidence or are you guessing, just wait for the official investigation

Inside information.

 

[43066]

Have you got evidence to back up your guess?

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==

So, speculation on wheel slide, has anyone got any factual evidence or are you guessing, just wait for the official investigation

Inside information.

Given the time of year, and the nature of the incident, it would be bizarre if this wasn’t being considered.

 

[Signal Head]

Are they also not spring loaded/hydraulic points at Talerdig? Admittedly, I haven't been on the line since RETB was phased out, at that time there were quite a few of those type of points on the Cambrian

They would have been, originally, though gas-pressure rather than a mechanical spring. I've no idea what's currently installed, there have been instances of those mechanisms being replaced with motored points, certainly in areas controlled by "No Signalman" key token, where the token withdrawal sets the points (and turns off TPWS). I don't know if that's a possibility with this ETCS arrangement.

 

[Unixman]

They would have been replaced with powered points when ETCS was installed, as the loops are now all bi-di (from the map on OTT).

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

 

[Delta558]

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

Points for sidings off passing loops, certainly on the Central Wales Line, are operated by a ground frame and do not move with the operation of the token. They are effectively 'semi-permanently' set towards the normal exit from the loop.

 

[Sly Old Fox]

Really sad to discover somebody has passed away. Best wishes for a speedy recovery to everybody else.

 

[westcoaster]

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

Happened at the other end iirc

Both trains show cancelled at Talerddig due to leaves on the line.

Realtime Trains | 1J25 1831 Shrewsbury to Talerddig | 21/10/2024

Real-time train running information for 1J25 1831 departure from Shrewsbury to Talerddig on 21/10/2024. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

Realtime Trains | 1S71 1909 Machynlleth to Talerddig | 21/10/2024

Real-time train running information for 1S71 1909 departure from Machynlleth to Talerddig on 21/10/2024. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

 

[Lewisham2221]

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

AIUI, the train heading towards Shrewsbury hadn't yet reached the loop.

 

[The Planner]

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

The siding is 52m long and on the up side. Unless its changed recently, the coast bound train goes into the up loop to wait, and the Shrewsbury bound goes on the down, so its of no use.

 

[IrishDave]

One thing that puzzles me, and I hope that someone will tell me, is how the Shrewsbury bound train managed to end up on the single line.
At Talerddig there is a run-off siding that runs for about 200 or so metres parallel to the main line towards Shrewsbury and I would have thought that the points would have been set for the train to go down there, that is until it is safely in the loop.

The only thing I can think of is that the siding's points aren't set by the ETCS system but maybe manually.

Can someone clarify what the situation is?

The Up Refuge Siding (to give the run-off siding its proper name) would only be relevant as an overrun if the trains were crossing under left-hand running; in this case, the trains were planned, and signalled, to cross under right-hand running.

The Down train (1J25 to Aberystwyth) was booked to stop at Talerddig from 19:24½ to 19:27½, while the Up train (1S71 to Shrewsbury) was booked to stop at Talerddig from 19:25½ to 19:26½. The Timetable Planning Rules explicitly state that "The first train into Talerddig MUST use the up loop." Therefore the Down train would have been planned to use the Up Loop, and based on the map on OTT (see screenshot below), that's how the trains were signalled.



== Doublepost prevention - post automatically merged: 22 Oct 2024 ==

The siding is 52m long and on the up side. Unless its changed recently, the coast bound train goes into the up loop to wait, and the Shrewsbury bound goes on the down, so its of no use.

Whichever train is booked to arrive first is booked into the loop, but the timetable varies a little from hour to hour - most of the time the Down train is booked to arrive first and therefore goes in the Up Loop, but not always (e.g. 1J23/1I28 cross the other way round).

 

[Unixman]

The Up Refuge Siding (to give the run-off siding its proper name) would only be relevant as an overrun if the trains were crossing under left-hand running; in this case, the trains were planned, and signalled, to cross under right-hand running.

The Down train (1J25 to Aberystwyth) was booked to stop at Talerddig from 19:24½ to 19:27½, while the Up train (1S71 to Shrewsbury) was booked to stop at Talerddig from 19:25½ to 19:26½. The Timetable Planning Rules explicitly state that "The first train into Talerddig MUST use the up loop." Therefore the Down train would have been planned to use the Up Loop, and based on the map on OTT (see screenshot below), that's how the trains were signalled.

View attachment 167714

== Doublepost prevention - post automatically merged: 22 Oct 2024 ==


Whichever train is booked to arrive first is booked into the loop, but the timetable varies a little from hour to hour - most of the time the Down train is booked to arrive first and therefore goes in the Up Loop, but not always (e.g. 1J23/1I28 cross the other way round).

I did wonder which side of the loop the trains were running (I did see that the loop is bi-di) and thank you very much for the clarification. I wonder if that rule might change now with the refuge siding being set for Shrewsbury bound trains.

But thank you again.

 

[The Planner]

I did wonder which side of the loop the trains were running (I did see that the loop is bi-di) and thank you very much for the clarification. I wonder if that rule might change now with the refuge siding being set for Shrewsbury bound trains.

But thank you again.

It would make things a bit tight at the Aberystwyth end, but doable.

 

[zwk500]

I did wonder which side of the loop the trains were running (I did see that the loop is bi-di) and thank you very much for the clarification. I wonder if that rule might change now with the refuge siding being set for Shrewsbury bound trains.

But thank you again.

AIUI from the earlier posts it was the down train that arrived first and slipped, so the side of the loop it used is not a factor in this collision.

 

[Krokodil]

The Heart of Wales has also been closed now due to poor railhead conditions so it’s clear decisions are certainly being made differently today, unsurprisingly of course.

Neither the HoW, Borderlands, nor the Conwy Valley cancellations are related to last night's collision. Trains are simply losing too much time to maintain the service.

 

[Ron Perry]

As I recall, from many years back, Talerddig is at the line's summit where surely the 1 in 52 down to Llanbrynmair hasn't yet begun, and trees etc are relatively sparse. As the down train was booked to cross there it seems odd that it proceeded on to the single line as a result of poor adhesion. Once on the single line and a steep gradient with more vegetation, however, I can see how poor adhesion could have contributed to a collision.
My second pont is that it is very difficult to assess speed of a train on continuously welded rail and with few lineside lights as points of reference. The last time I travelled by rail through Llanbrymair I recall sections of 40 ft jointed rail!

 

[blueberry11]

Looking at this Wikipedia page, this is the first crash in Wales that resulted in a passenger death since 1987 (Glanrhyd Bridge collapse) and the UK since 2020 (Stonehaven, Scotland).

Edit: I thought it was 1950 because I was relying of ctrl+f looking for 'Wales' on that page