Crash on The Cambrian

[johnofwessex]

https://www.bbc.co.uk/news/articles/c5y0yg7m8meo

I hope there are no casualties

Is there a loop there?

[johnofwessex]

https://www.walesonline.co.uk/news/wales-news/caersws-train-incident-sees-major-30197641#3620025

[melmerby]

Crash occured just outside the loop at Tallerdig

1J75 formed of 158841 on the 1831 Shrewsbury to Aberystwyth collided with 1S71 formed of 158824 on the 1909 Machynlleth to Shrewsbury at approx 1925

No fatalities

 

EDIT

Currently (2330) both trains are sitting in the loop,

 

EDIT 2

Sadly later it was announced a passenger had died, with 15 people injured.

Edited Tuesday at 10:31 by melmerby

[MartinRS]

Information on the  ETCS Level 2 signalling system, used on the Cambrian line, can be found here (scroll down to United Kingdom).

[wombatofludham]

One person dead from a heart attack. 15 others with minor injuries. According to the Cambrian News reports the driver of one train reportedly entered the passenger saloon and shouted to people to brace for impact. Reports on line of poor railhead conditions after yesterday's storm, a RHTT was in the area last night and was due again tonight. One train apparently slid past the block marker and into the path of the other train, photos on Twitter show a head on collision apparently at about 20mph.

 

Extremely sad to learn of the death on local news media confirmed by the police.

[rogerzilla]

Sounds a bit like the Salisbury Tunnel Junction crash, where a train also kept sliding uncontrollably due to autumn leaf sludge.

[Covkid]

Very very sad news indeed, and the first passenger death for a while - whether a result of the collision or not is not known

 

For those not aware Talerddig is a loop at the top of hill climbing from Caersws.  Once through the loop the single line falls away quite steeply through Llanbrynmair and round a steep 30mph curve through Commins Coch. 

 

The route was signalled conventionally until around 1987 or1988 when RETB (radio electronic token block) was being introduced. RETB gave way to ERTMS which is the for runner of ETCS.  ETCS is basically "space invaders" for the cab with drivers observing conventional signals if there are any, or block markers which are linked to the radio system. 

 

From what has been in the public domain it appears that the Down train (to Aberystwyth) could not stop in the loop and carried on back onto the single line whilst the up train was on the last few yards of the steep climb into the passing loop.  

[big jim]

Thoughts are with the crew and passengers on board, sad to hear of the death on waking up this morning

 

looking at the pics that are starting to circulate the driver was lucky to get away with just injuries in my opinion, I wonder if it will bring into question the crash worthiness of sprinters now as happened with the HSTs following the crash in Scotland, certainly might speed up the introduction of 197s on the Cambrian which is taking an age 

from personal experience the tallerddig area can get particularly hairy railhead wise but it was normally coming up the bank from Machynlleth, iirc we used to water spray at 30mph going to macc then sandite at the same speed on the way back between commins coch and tallerddig, most of rest of the line was done at the usual 60mph apart from a few other short 30mph locations 

 

 

Edited Tuesday at 08:02 by big jim

[locoholic]

After the Salisbury Tunnel crash, did Network Rail identify places where poor autumn railhead conditions were safety-critical, and was Talerddig on that list?

[Boris]

Still very sad news for one family and unpleasant for those involved.

[KeithHC]

Also after Salisbury is it the 158/159 that’s the problem. Do they have a form of ABS or could you use cadence braking.

 

Keith

Edited Tuesday at 08:19 by KeithHC
Spelling

[Wickham Green too]

Can't help wondering why the second train was still on the move if the first wasn't proven to have stopped in the loop.

[Jonb82]

27 minutes ago, KeithHC said:

Also after Salisbury is it the 158/159 that’s the problem. Do they have a form of ABS or could you use cadence braking.

 

Keith

they have wheel slide protection, our driving instructions for these in leaf fall  are initial step two brake, let the WSP work, use more brake force if necessary then place in emergency if you’re not going to stop, obviously approaching known low adhesion areas at lower speed also.

 

Trying any other braking technique will result in you potentially not being exonerated from the incident after investigation.

Edited Tuesday at 08:45 by Jonb82

[big jim]

What’s bothering me is the reports say the crash was near llanbrynmair not tallerddig and I’ve heard unconfined reports that the trains are 3/4 mile away from the loop on tallerddig bank

 

That’s a hell of a slide if true 

 

 

[dibateg]

It's just on the west side of Tallerddig village where the main road comes close to the line on a bend.

[brushman47544]

As only an occasional UK rail user these days, it never ceases to amaze me how overgrown the railway is now compared to the past. It really is time to strip ALL growth away and return to vegetation free railway corridors as they were up to around the 80s. Any loss of life is tragic but more particularly when they are avoidable through better vegetation clearance.

[Reorte]

10 minutes ago, brushman47544 said:

As only an occasional UK rail user these days, it never ceases to amaze me how overgrown the railway is now compared to the past. It really is time to strip ALL growth away and return to vegetation free railway corridors as they were up to around the 80s. Any loss of life is tragic but more particularly when they are avoidable through better vegetation clearance.

Could be but IMO that's very much a case of wait to see what the reports say, including how likely something similar is to happen again.

[martin_wynne]

41 minutes ago, dibateg said:

It's just on the west side of Tallerddig village where the main road comes close to the line on a bend.

 

???

 

That's a long way from the loop:

 

 

 

[spamcan61]

11 minutes ago, martin_wynne said:

 

???

 

That's a long way from the loop:

 

 

 

That map aligns with the photos online showing the train next to a road on a bend.

[Stuart]

I grew up travelling on the Esk Valley line where both of the stations with passing loops (Castleton Moor and Glaisdale) had overruns in each direction which would prevent a train entering the single line unless properly signalled.

 

Why do current passing loops have nothing to prevent inadvertent overruns?

 

Are such overruns so rare as to be not worth preventative infrastructure?

 

Are they incompatible with electronic signalling?
 

Very curious.

 

Stu

 

 

[Hroth]

1 hour ago, brushman47544 said:

As only an occasional UK rail user these days, it never ceases to amaze me how overgrown the railway is now compared to the past. It really is time to strip ALL growth away and return to vegetation free railway corridors as they were up to around the 80s. Any loss of life is tragic but more particularly when they are avoidable through better vegetation clearance.

 

It would be expensive to clear 40+ years of lineside growth back to clear cuttings, there's a heck of a lot of fairly mature trees that will take time to fell and would possibly require planning permission to fell. Then again, any felling would be restricted to times of year outside the nesting season.

 

Also, pressure groups like The Woodland Trust would raise a stink about precious trees being cut down, despite the context.  They made a nuisance of themselves when a limited de-boughing was undertaken a few years ago to try and reduce leaf-fall onto the tracks.

 

An accountancy led cost cutting measure by BR back in the 80s has now come back to bite Network Rail in the bum.

 

Edited Tuesday at 11:09 by Hroth
spelin

[KeithHC]

2 hours ago, Jonb82 said:

they have wheel slide protection, our driving instructions for these in leaf fall  are initial step two brake, let the WSP work, use more brake force if necessary then place in emergency if you’re not going to stop, obviously approaching known low adhesion areas at lower speed also.

 

Trying any other braking technique will result in you potentially not being exonerated from the incident after investigation.

Thank you Jon for your answer. I am not a railway person so do appreciate the input made by the professional railway staff on here both those working and those retired. 
 

Keith

[Michael Hodgson]

15 minutes ago, Stuart said:

Are such overruns so rare as to be not worth preventative infrastructure?

 

Are they incompatible with electronic signalling?
 

 

 

It really doesn't matter what sort of signalling is in place as regards this type of accident.

Signals don't stop trains.  Brakes stop trains (or sometimes fail to - the laws of physics apply concerning friction etc).

 

If you have a trap at the end of a loop, a train unable to stop will derail or hit a stop block.  On a passenger lines there's a trade off between the probability of a head-on collision (nasty) and the much greater probability of a (albeit less nasty) derailment when there is no other train about.  The traditional resolution of that dilemma is that we trap freight lines but not passenger loops and we don't allow two trains to enter passing loops at the same time.

Edited Tuesday at 11:14 by Michael Hodgson

[martin_wynne]

19 minutes ago, Michael Hodgson said:

If you have a trap at the end of a loop, a train unable to stop will derail or hit a stop block.

 

Not necessarily -- a sand drag is quite effective.

 

[Jeremy Cumberland]

There are other ways of stopping trains, such as sand drags and uphill grades, but it seems the train in this accident was travelling at 20 mph, which is a huge amount of energy to dissipate.

 

Slower speed overruns, where the driver has made a misjudgement rather than finding a complete loss of adhesion, are generally accommodated by the signalling system imposing a gap between conflicting train movements rather than physically trying to stop an overrunning train. Where distance separation is impracticable, then there might be trap points to prevent a collision, but trap points almost guarantee a derailment for all but the lowest speed/shortest overruns, and I doubt that derailing the overrunning train in yesterday's accident would have resulted in a better outcome.