• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Incident at Talerddig, Wales - 21/10/2024

Harpo

Member
Joined
21 Aug 2024
Messages
564
Location
Newport
An inspection of the automatic sanding system fitted to train 1J25 after the accident showed that the sanding hoses on the leading vehicle of this train (which would have been active at the time of the accident) were blocked and apparently unable to discharge sand. (RAIB)
Oh dear.
 
Last edited:
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

12LDA28C

Established Member
Joined
14 Oct 2022
Messages
4,530
Location
The back of beyond
Thought they used a 'Doppler'-type system because of the ERTMS, so axle/wheel rotation isn't necessary to measure speed?

Ah, possibly although if that's the case it should be simple to gain a measurement of the speed at impact which appears to have not been possible.
 

Belperpete

Established Member
Joined
17 Aug 2018
Messages
2,256
Thought they used a 'Doppler'-type system because of the ERTMS, so axle/wheel rotation isn't necessary to measure speed?
I understand that ERTMS has a number of inputs to determine speed and position, so that a failure or mis-reading by one doesn't unsafely affect operation. Presumably it will take the worst case. However, each will have its limitations, particularly at slow speeds. Which of those inputs are used to drive the Speedo, and exactly how much is recorded by the black box, are obvious questions.

I expect that the RAIB will be looking at a number of sources to determine and corroborate train speed, including things like GPS records, and axle counter records.

Hopefully they will be looking into when the sanders got blocked, and whether there was anyway of knowing before they were needed in anger.

The frightening thing for me is just how far the train slipped - not just the 900m beyond the loop, but apparently right through the loop as well. That must be the best part of a mile that the train was out of control. Truly frightening for the driver.
 

InkyScrolls

Established Member
Joined
20 Jul 2022
Messages
1,275
Location
North of England
The frightening thing for me is just how far the train slipped - not just the 900m beyond the loop, but apparently right through the loop as well. That must be the best part of a mile that the train was out of control. Truly frightening for the driver.
It certainly is! And I say that having slid over two miles before...
 

hwl

Established Member
Joined
5 Feb 2012
Messages
7,549
Thought they used a 'Doppler'-type system because of the ERTMS, so axle/wheel rotation isn't necessary to measure speed?
The problem is that you have multiple speed inputs to reconcile:
  1. Doppler radar
  2. Wheel rotation in lead vehicle (sander apparently not working according to RAIB)
  3. Wheel rotation in trailing vehicle (RAIB haven't mentioned that this vehicle sanders weren't working)
Depending on which vehicle the recorder is in will likely mean a difference between 1 & 2 or 1 & 3 being recorded as there may well have been a difference between wheel rotation in the two vehicle on 1J25.
 

Nottingham59

Established Member
Joined
10 Dec 2019
Messages
2,302
Location
Nottingham
I assume there is a sander on each rail? Does this mean that both hoses were blocked and unable to dispense sand?
 

MotCO

Established Member
Joined
25 Aug 2014
Messages
4,744
I could imagine if rainwater got into the hoses, the sand would be solid. How well protected are the hoses or tanks?
 

12LDA28C

Established Member
Joined
14 Oct 2022
Messages
4,530
Location
The back of beyond
Hopefully they will be looking into when the sanders got blocked, and whether there was anyway of knowing before they were needed in anger.

Not sure how they would be able to determine that, as far as the driver is concerned the automatic sanders should be working and doing their job and it would be hard to tell that they're not, whilst the train is in motion.

The sanders on each vehicle should be tested on a train prep which would normally happen before the train enters service each day.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
25,963
Location
Nottingham
I could imagine if rainwater got into the hoses, the sand would be solid. How well protected are the hoses or tanks?
These systems have been around for several years so the design and maintenance ought to have been worked out to keep them functional. Why they apparently weren't in this case is obviously going to be a key question.

Worth noting that the gradient going west from Talerddig is as steep as 1 in 52. That means that if the coefficient of friction is less than 2% it's not possible for a train to decelerate. Less than 1% is by no means unheard of for a wet rail exposed to leaf fall and with no sand applied. So if the rail was in that sort of condition with the sanders inoperative, then once it slid past the end of the loop (which appears to be level) it was inevitable that it would continue for several miles or until it hit something.
 

Towers

Established Member
Joined
30 Aug 2021
Messages
2,251
Location
UK
The sanders on each vehicle should be tested on a train prep which would normally happen before the train enters service each day.
I rather suspect that this might be where a good deal of scrutiny might be directed.
 

fishwomp

Member
Joined
5 Jan 2020
Messages
740
Location
milton keynes
The sanders on each vehicle should be tested on a train prep which would normally happen before the train enters service each day.

I think pilots do a walkaround every time before every take off - certainly seen that on the transatlantics, and quite a check list. Train drivers do a brake test - but on a stationary train of course - anything else? It must have been horrifying to lose braking and be let down by something that is so critical.

Hoping the recommendations that come from the RAIB report will be meaty with appropriate force in maintenance and design.
 

FrodshamJnct

Established Member
Joined
14 Apr 2019
Messages
3,691
Location
Cheshire
I think pilots do a walkaround every time before every take off - certainly seen that on the transatlantics, and quite a check list. Train drivers do a brake test - but on a stationary train of course - anything else? It must have been horrifying to lose braking and be let down by something that is so critical.

Hoping the recommendations that come from the RAIB report will be meaty with appropriate force in maintenance and design.

Train drivers do a running brake test - in the rule book, not much point doing it when stationary.
 

hexagon789

Veteran Member
Joined
2 Sep 2016
Messages
16,465
Location
Glasgow
Train drivers do a brake test - but on a stationary train of course - anything else?
A Running Brake Test, so while on the move, but that will of course only identify what adhesion is like at the point and time where that RBT is made.
 

12LDA28C

Established Member
Joined
14 Oct 2022
Messages
4,530
Location
The back of beyond
I think pilots do a walkaround every time before every take off - certainly seen that on the transatlantics, and quite a check list. Train drivers do a brake test - but on a stationary train of course - anything else? It must have been horrifying to lose braking and be let down by something that is so critical.

Hoping the recommendations that come from the RAIB report will be meaty with appropriate force in maintenance and design.

At maintenance locations a fitter may be tasked with carrying out a train prep, otherwise those duties should be performed by the driver, where included on the diagram/schedule card.
 

357

Established Member
Joined
12 Nov 2018
Messages
1,595
A Running Brake Test, so while on the move, but that will of course only identify what adhesion is like at the point and time where that RBT is made.
It's to demonstrate the train brakes are working not the rail adhesion.

Nor would it show that the sanders are blocked.
Electrostars and Voyagers give the driver a warning when the sand level is below 50% and again when it becomes critically low.

I would hope they would also detect if the level of sand wasn't falling when applied and alert the driver, but I don't know if they do or not.
 

hexagon789

Veteran Member
Joined
2 Sep 2016
Messages
16,465
Location
Glasgow
It's to demonstrate the train brakes are working not the rail adhesion.


Electrostars and Voyagers give the driver a warning when the sand level is below 50% and again when it becomes critically low.

I would hope they would also detect if the level of sand wasn't falling when applied and alert the driver, but I don't know if they do or not.
I've heard it both ways, certainly as a means to ascertain the effectiveness of the brakes though.
 

357

Established Member
Joined
12 Nov 2018
Messages
1,595
At maintenance locations a fitter may be tasked with carrying out a train prep, otherwise those duties should be performed by the driver, where included on the diagram/schedule card.
Or shunter.

I've heard it both ways, certainly as a means to ascertain the effectiveness of the brakes though.
It's useful to test the adhesion if you feel it's required. I've done this today approaching a couple of stations and a signal.

I don't however feel the need to do the full 10mph reduction etc.
 

Taunton

Established Member
Joined
1 Aug 2013
Messages
10,770
Sand absorbing moisture is a longstanding issue. Steam locos had sandboxes against the warmth of the boiler specifically for this reason, but diesels lost this advantage. Are the sanders here just gravity or using compressed air?
 

jfowkes

Member
Joined
20 Jul 2017
Messages
1,046
Is there any basis for arguing that the loop at Talerdding should have had trap points?
 

Dieseldriver

Member
Joined
9 Apr 2012
Messages
997
Is there any basis for arguing that the loop at Talerdding should have had trap points?
If you’re happy to spend the money to fit them at every junction, passing loop on single lines, entrance to single lines in the UK.
 

AndrewE

Established Member
Joined
9 Nov 2015
Messages
5,606
If you’re happy to spend the money to fit them at every junction, passing loop on single lines, entrance to single lines in the UK.
I'm sure it is upthread somewhere, but this is something a risk assessment would allow you to prioritise. So definitely not everywhere, e.g. the downhill direction at Talerddig would seem to be quite a lot more important than uphill...
 

172007

Member
Joined
2 Jan 2021
Messages
863
Location
West Mids
Not sure how they would be able to determine that, as far as the driver is concerned the automatic sanders should be working and doing their job and it would be hard to tell that they're not, whilst the train is in motion.

The sanders on each vehicle should be tested on a train prep which would normally happen before the train enters service each day.
On the traction I sign it is immediately obvious if there is sand being delivered to the wheels as the train has a completely different feeling to just WSP working.

That said, not everyone is as sensitive to feeling how a train behaves just like some people have a bad sense of smell, it's all down to genetics and how sensitive your Gluteus maximus is.

Sanders are checked on prep and a driver worth their salt who takes a train out of a maintenance facility in leaf fall season should take the opportunity to test the sanders using test buttons and watching the sand came out.

On the traction I sign it is immediately obvious if there is sand being delivered to the wheels as the train has a completely different feeling to just WSP working.

That said, not everyone is as sensitive to feeling how a train behaves just like some people have a bad sense of smell, it's all down to genetics and how sensitive your Gluteus maximus is.

Sanders are checked on prep and a driver worth their salt who takes a train out of a maintenance facility in leaf fall season should take the opportunity to test the sanders using test buttons and watching the sand came out if there are walking routes to allow it
 

Belperpete

Established Member
Joined
17 Aug 2018
Messages
2,256
Back in BR days it was decided that for trap points on passenger lines, it was no longer acceptable to just tip the train in the dirt after a couple of feet, particularly as this carries a high risk of the train overturning or worse. A requirement was introduced that such trap points have to be provided with an overrun of sufficient length and with suitable retardation capable of stopping the train safely from a likely overrun speed (i.e. without stopping it too quickly with resultant casualties). As a result, BR removed most trap points when resignalling, as it was simpler to provide an overlap instead, as at Tallerddig.

What kind of arrangement could safely stop a train travelling at the kind of speed that one was? At the gradient the line drops beyond the loop, you would soon be building a viaduct to stop the train on.
 

Lurcheroo

Established Member
Joined
21 Sep 2021
Messages
1,214
Location
Wales
Interesting that the speed of 1J25 is still under investigation. No doubt this is due to the fact that as the train was sliding, accurate train speed information was not recorded on the OTDR - in my experience in a similar situation when the wheels 'lock up' under sliding the speedometer reads Zero as the wheels are not rotating.
These units have both the OTDR and the JRU (Juridicial Record Unit) for ERTMS.
Thought they used a 'Doppler'-type system because of the ERTMS, so axle/wheel rotation isn't necessary to measure speed?
They do have Odometery sensors as well as a Doppler radar
The frightening thing for me is just how far the train slipped - not just the 900m beyond the loop, but apparently right through the loop as well. That must be the best part of a mile that the train was out of control. Truly frightening for the driver.
The report says the driver applied service braking for 40 seconds before an emergency application happened so I would think they’ve slid someway before getting to the loop too (but the full report will detail that I’m sure).

The sanders on each vehicle should be tested on a train prep which would normally happen before the train enters service each day.
Just a thought, it is very possible the train was prepped in a location where only an above solebar prep could be carried out that morning, so no testing of sanders.

If you’re happy to spend the money to fit them at every junction, passing loop on single lines, entrance to single lines in the UK.
How many other single lines having passing loops where the driver is permitted to enter at over 30mph and then enters a known low adhesion area on a very step falling gradient ?

Sanders are checked on prep and a driver worth their salt who takes a train out of a maintenance facility in leaf fall season should take the opportunity to test the sanders using test buttons and watching the sand came out.
See earlier point about the possibility of above solebar prep only.
Sand absorbing moisture is a longstanding issue. Steam locos had sandboxes against the warmth of the boiler specifically for this reason, but diesels lost this advantage. Are the sanders here just gravity or using compressed air?

Modern traction (such as TFW’s new 197’s which will replace these 158’s) have heated sand boxes to help dry out sand.
 

12LDA28C

Established Member
Joined
14 Oct 2022
Messages
4,530
Location
The back of beyond
Just a thought, it is very possible the train was prepped in a location where only an above solebar prep could be carried out that morning, so no testing of sanders.

Indeed, that is possible. However that would be rather less than ideal, particularly at this time of year.
 
Joined
24 Mar 2013
Messages
5
Our off train Autumn brief days have discussed doing an RBT well before the usual shut off point in leaf fall for all stations or known stopping points to test the rail head. This isn’t mentioned in the report. Wonder what TFW policy/training is for leaf fall driving?
 

Top