-
Our booking engine at tickets.railforums.co.uk (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!
Collision and derailment near Salisbury (Fisherton Tunnel) 31/10/21
- Thread starter Monty
- Start date
- Status
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R
RailUK Forums
RHTT and SITT are contracted out to FOCs. How do you know the cancelation of the services wasn't due to them? Seems like you like to blame NR every chance you get.
Mcr Warrior
Veteran Member
- Joined
- 8 Jan 2009
- Messages
- 12,180
If it's a predominantly downhill gradient from SY29 for the just under 1km to the tunnel junction, should that particular signal have been displaying red, rather than the much nearer SY31? That would allow more of an over-run, or has the shorter "overrun" distance from SY31 previously always been considered sufficient?
That's absolutely the wrong attitude, which has unfortunately got the railway to where they are nowadays. If I contract something out I am still responsible overall for the matter; doesn't mean I can wash my hands of something and walk away. Just because I contract out to the bottom bidder doesn't mean I absolve myself of any control.
bramling
Veteran Member
I don’t think anyone can really say the layout was deficient in terms of offering braking distance.
With the benefit of hindsight, a delayed clearance arrangement on SY29 might have helped, in that it would have encouraged an earlier brake application, and avoided the situation where drivers delayed their reaction to a double yellow at SY29R - which is itself a potential SPAD trap. For balance, I’d add that a delayed clearance on SY29 is an example of tinkering and adding complexity to what is otherwise a simple layout which appears to have worked satisfactorily for many years. On balance, I’d probably go for it through.
A lot depends on SWR’s PDP. Southeastern’s PDP in force at the time of Stonegate had the driver going straight to step 2, which turned out to be a problem as unknown to the driver the train had run out of sand. I think we can fairly confidently say at Salisbury that an earlier, and probably lighter, brake application would certainly have been desirable. RAIB will no doubt tell us in due course why they believe it didn’t happen.
Last edited:
Huff and puff all you want. Nobody has suggested washing of hands or walking away, but if an operator can't resource a service what exactly can NR do? In the real world (you are welcome to join us) magic wands don't exist!
The trouble is, as people have been trying to politely point out, this is your own opinion, not what it says in the report.
Nowhere do Network Rail provide a guarantee that they will treat all of the track at least once every 24 hours. Nor are they regulated to do so.
The railhead treatment train was only rescheduled because of the other routes in the area being closed, rather than to save money. The gap between treatment trains wasn't much longer than the time 24 hours to which Network Rail 'aspire', only 50%.
Fewer resources are expended on railhead treatment at weekends because railhead treatment is primarily about performance, not safety critical work.
It's unclear if they were actually cancelled anyway, or if there was a shortage of resources or not. There was engineering work nearby which resulted in the need to change the schedule.
Last edited:
Yes you are right. Just saying its not always NR's fault despite what certain posters on here say.
SWR at the time came out publicly and said the drivers actions were more defensive than what they would expect. Other drivers who have spoken to 1L53 driver also confirm 1L53 had the brakes in earlier than demanded/required. I think it’s fair to say the driver done as they were taught to.
The only observation to be made, and something that has been done several times before, is perhaps an urgent call to Salisbury signaller may have been better in preventing the collision but that is purely speculative, and the adrenaline etc and the time to process what was going on may have resulted in exactly the same outcome.
The only observation to be made, and something that has been done several times before, is perhaps an urgent call to Salisbury signaller may have been better in preventing the collision but that is purely speculative, and the adrenaline etc and the time to process what was going on may have resulted in exactly the same outcome.
Annetts key
Established Member
The normal standard minimum distance between a protecting signal and points or a junction is 183 meters (200 yards). This is the standard overlap distance.
Note that if the above mentioned distance is not practical, alternative measures can be used. Hence there are situations where there is less than 183 metres between a protecting signal and points or a junction.
With the above information, you can now calculate the distances for the other signals.
Absolute nonsense.
Annetts key
Established Member
The line speed and and the gradients are all taken into account. As is the braking performance of the worse braked trains that are permitted to use the line (“braking curve”). These feed into the signalling design process. This is then used to determine the minimum required distances between a signal displaying a caution aspect (double yellow or single yellow) and the stop signal (in this case SY31).
Keep in mind that in practice, the type of trains with the worst braking performance would be freight trains, or loco hauled conventional passenger coaches (not disc braked coaches).
A 158 or 159 with more a modern braking system in normal circumstances should not need to start braking when the driver is at the viewing point of the signal showing double yellow.
Poor railhead conditions can be a serious cause for concern, and hence safety.
Poor railhead conditions may cause or contribute to:
- Wheel slip, and hence cause “rail burn” where the top of the rail is deformed, and which weakens the rail,
- Wheel slide, increasing the stopping distance during braking,
- Track circuits to malfunction, with the signalling system not detecting the presence of a train (track circuit showing clear when occupied - SCWO), which is a “unprotected wrong side failure” and may result in the signal(s) that should be protecting the train showing a proceed aspect (which could be green) instead of a red aspect, or could allow points to be moved ahead of, or under a passing train.
Last edited:
PDP is what, please? Apologies if it's explained a little up-thread. Thank you! (I know what a PDP-11 is, but that's not what you're referring to for sure .....)
Professional Driving Policy
Yes they did. Read the post I was quoting.
Reminds me of the Southall accident. There, a formal procedure for the signaller to be informed if the AWS was inoperative, presumably to add extra leeway, was not followed, and the train was just signalled from high speed to a stand with a freight crossing immediately ahead. Here, the rails on a downhill on a poor adhesion day had not been treated for well beyond the usual limit, yet the signaller was not informed and so ... the train was signalled from high speed to a stand with two trains crossing immediately ahead.
See the similarities?
norbitonflyer
Established Member
Indeed - the word "aspire" speaks volumes
It appears from OTT Maps to be a ground signal controlling movement from the down line to the up line over the trailing crossover which is shown just to the east of Laverstock Junction
Last edited:
The standard overlap is 200 yards, which can be reduced if the linespeed is lower (and even further for a route that's approach controlled).
Provided the required overlap is provided, there is no reason why approach control or double red protection need be provided.
Risk assessment in the past may have recommended the fitment of TPWS+ (a second set of overspeed loops on approach to a high-risk signal in advance of a conflicting junction), particularly given the circumstances of the significant decrease in linespeed just before the signal.
That said, I don't know whether that was fitted to any of the signals in the area, and even if it had, it probably wouldn't have made any material difference to the accident (just as the ordinary TPWS overspeed loop didn't help, as the emergency brake was already applied).
norbitonflyer
Established Member
Undoubtedly, even though we should note that much of the damage was caused by the emergency services in getting him out.
The train was already late. The driver did not know whether the signal would clear as he approached. If it did, an earlier brake application would have meant that he would slowed more than necessary and have delayed the train further. He slowed as much as should have been necessary to stop at SY31, whilst maintaining as much momentum as possible in case it did clear. Especially over a busy junction like that, taking more time than you should is likely to delay other trains - the driver was not to know whether a train bound for Southampton was waiting to cross behind him (instead of, as it happened, ahead of him). Would it be useful or a distraction to have something like OTT maps visible in the cab so drivers can see the situation ahead in rather more detail than the signals alone can?
One point which perhaps needs to be covered somewhere, but possibly not in the RAIB report, is an enquiry into the misinformation put out immediately after the accident by apparently official sources jumping to conclusions.
Last edited:
I think even the most diligent and well-intentioned driver could end up easily drawn in to an avoidable SPAD if tasked with thinking ahead and predicting other trains' movements.
edwin_m
Veteran Member
I was rather surprised not to have seen any mention of sanders being either fitted or working in the Salisbury collision.
On the other hand, braking earlier would mean reaching the following signals later, by which time the signal protecting the junction might have cleared and the train would have been able to continue at a higher speed. Short of a connected driver advisory system, there's no way of knowing the best course of action.
Absolutely not useful. You should always drive to what the signals are showing, without second guessing what they're going to do. That's how you end up with SPADs - 'this is always a single yellow then when I go round the corner the next signal has always cleared to green so I won't have to stop' - except the time it doesn't and you pass that signal at red. Having a system (which is an unnecessary distraction) encouraging drivers to make an assumption about the route ahead is a bad idea.
PG
Established Member
What other trains are doing is the business of the signaller and they have the oversight of the wider picture.
Drivers are only driving their own train, the responsibility for other trains should remain solely with the signaller.
Grumpy Git
On Moderation
At least one of the photographs in the actual report show what a farce lineside vegetation management is, when there are trees overhanging the rails. I wonder how many miles of lineside could be cleared for what this accident has cost (never mind the injuries to the 158 driver)?
Sadly the report is silent about that.
There was also concern about sliding while braking on contaminated rail when disk brakes first came in. The goo gets rolled onto the rail head and also onto the wheels causing a loss of adhesion. This doesnt happen with tread brakes as the brake cleans the wheels. While I would not want to go back to tread brakes, maybe something is needed to clean the wheels.
Another issue is the improved bogie/wheel design. We dont see hunting* now. But hunting cleaned the wheels/railheads.
*This is where the bogie 'hunts' from side to side as the wheel/rail interface is unstable. The side to side hunting was thought to clean railheads.
No. Because the similarities are in your mind, not in the facts written in the report.
There's no evidence here to suggest that the correct procedure wasn't followed. Not yet anyway.
I would imagine it would depend on what the system was telling the driver.
For example it should be possible to advise a driver to slow in advance of a signal so that as long as that they hit their new target speed when they need to that they then are able to pass through a non red signal without actually needing to stop. Even though it's currently red.
However the system would need to be smart enough of what's going on in the wider network so that the driver's don't have to second guess that.
If you want a human check on the order of trains that could still be undertaken by the signaler, the driver could be told the wider picture. However I suspect that potentially more information than they need.
Surreytraveller
On Moderation
- Joined
- 21 Oct 2009
- Messages
- 2,810
Even the signaller doesn't have the wider picture. Control would have that. Control would know what the drivers are doing, and which trains they will be relieving, which could cause them to get the signaller to alter what they've already decided to do
bramling
Veteran Member
You wouldn’t have catch points - they are provided to prevent vehicles running away on gradients.
Trap points could be provided, but you run the risk of creating a serious derailment. The railway tends to provide them in places where a collision would be exceptionally serious - such as before viaducts. They aren’t generally provided elsewhere for passenger moves.
Annetts key
Established Member
These ideas of providing a driver with additional information so that they can decide how to drive their train are all well and good, but no one has a working crystal ball. No one can predict the future with 100% certainty. So no one can tell what will actually happen. Any number of things may happen.
So it’s best not to try to circumvent the existing system. If a signal is showing a caution aspect, a driver must alway assume that the red signal ahead will still be red when the train gets to it and so control their train appropriately.
This is important, because occasionally points may fail, track circuits or axle counters may fail. The train ahead may stop due to wheel slip, or mechanical or electrical reasons. Or any number of other possible reasons…
So drivers should never try to second guess what or when a signal may, or may not clear up to a proceed aspect.
In this particular case, railhead contamination looks like it caused wheel slide, which reduced the effectiveness of the brakes.
Look again at the pictures in the RIAB report.
Now if you don’t know, imagine that contamination is oil…
As every motor racing enthusiast knows, oil on a race track is highly dangerous, as it can cause a loss of grip and cause a driver to loose control. No one jokes about wrong type of oil…
It’s about time that railhead contamination was treated with the same respect. Despite what some on here and elsewhere say about cutting all the trees down, this will not solve all the problems. Leaves will find there way on to the railway from elsewhere.
This is not the place to speculate on possible solutions. You can always do that elsewhere. So in this topic, let’s stick to talking about the known facts. And wait for the full report to be published.
So it’s best not to try to circumvent the existing system. If a signal is showing a caution aspect, a driver must alway assume that the red signal ahead will still be red when the train gets to it and so control their train appropriately.
This is important, because occasionally points may fail, track circuits or axle counters may fail. The train ahead may stop due to wheel slip, or mechanical or electrical reasons. Or any number of other possible reasons…
So drivers should never try to second guess what or when a signal may, or may not clear up to a proceed aspect.
In this particular case, railhead contamination looks like it caused wheel slide, which reduced the effectiveness of the brakes.
Look again at the pictures in the RIAB report.
Now if you don’t know, imagine that contamination is oil…
As every motor racing enthusiast knows, oil on a race track is highly dangerous, as it can cause a loss of grip and cause a driver to loose control. No one jokes about wrong type of oil…
It’s about time that railhead contamination was treated with the same respect. Despite what some on here and elsewhere say about cutting all the trees down, this will not solve all the problems. Leaves will find there way on to the railway from elsewhere.
This is not the place to speculate on possible solutions. You can always do that elsewhere. So in this topic, let’s stick to talking about the known facts. And wait for the full report to be published.
You mean trap points. And no, not at the speed that the train was going. What may have then happened can be speculated about in a different topic.
Last edited:
- Status