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

[millemille]

The 159 fleet was the first to be fitted with emergency sanders in about 1995, and I think the then Wales and West 158 fleet was fitted soon after. I imagine these have since been replaced by something more modern, but I'd be surprised if they were removed without replacement. So I'm also surprised the report doesn't mention them.

The interim report mentions that the sanders will be subject to the investigation and reporting, but it is far too early for RAIB to have investigated the brake/WSP/Sander system fully. They have subcontracted various consultancies to carry out testing of some parts/systems and these are underway.

The sanders fitted to the units involved are a simple automatic fixed rate (nominally 2kg/min) sander, triggered by WSP activity occurring while in a brake step and/or an emergency brake demand and/or the driver's manual sand button being operated. They represent the thinking in place 20+ years ago and do not reflect current best practise.

 

[Grumpy Git]

The sanders fitted to the units involved are a simple automatic fixed rate (nominally 2kg/min) sander, triggered by WSP activity occurring while in a brake step and/or an emergency brake demand and/or the driver's manual sand button being operated. They represent the thinking in place 20+ years ago and do not reflect current best practise.

Have leaves got less slippy in the last 20+ years?

 

[4COR]

Yes, but according to some on here it's not all due to the trees beside the line!

Some folk need to wake up. The biggest force on leaves is gravity (not wind)!

In many situations, yes, leaves do fall somewhere near the tree. However, it's worth reminding ourselves that on the morning of the 31st Oct 2021, there were some severe and damaging winds roughly along a line from Weymouth to Hull (with Salisbury right in the track). We were in the New Forest and had part of a large Oak fall 50 foot through our caravan, and can confirm it was quite hairy... How far do leaves travel in 70+ mph wind?

See: https://rmets.onlinelibrary.wiley.com/doi/epdf/10.1002/wea.4158

 

[rebmcr]

Have leaves got less slippy in the last 20+ years?

No but the engineering designs for sanding have improved.

 

[millemille]

Have leaves got less slippy in the last 20+ years?

No. Sander application and technology has moved on.

For example; the sand that is blown into the wheel/rail interface from the sander is crushed by the wheel rolling over it and it is the crushing/crushed sand that conditions the rail head and improves the co-efficient of friction. BUT as each wheel rolls over the crushed sand the positive conditioning of the sand is attenuated. Data from RSSB research project T1107 shows that sand laid at 2kg/min (which is the typical sand delivery rate in the UK) is, on average, fully attenuated once 2 axles has rolled over it and that wheel slide starts to reoccur on following axles. Increasing that sand delivery rate to 4kg/min - but with a variable rate to ensure that sand density per meter of rail does not exceed a level at which track circuits stop working - and you've got an average of 4 axles running on conditioned rail. Fit a second sander further back down the train and you've now got 8 axles, on average, running on conditioned rail.

The graph below is taken from T1107 and shows the actual difference in stopping performance (55mph to 20mph) on rail with typical leaf fall contamination levels of adhesion.

SFRS = Single Fixed Rate Sanders. Your typical BR era single fixed 2kg/min sanders at axle 3 in the direction of travel, as fitted to 158/159
SVRS = Single Variable Rate Sanders. A single sander at axle 3 in the direction of travel, nominally 4kg/min but varied with speed so as speed drops the sand per meter rate does not exceed 7.5g/meter
DVRS = Dual Variable Rate Sanders. A sander at axle 3 AND at axle 7 in the direction of travel, each nominally 4kg/min but varied with speed so as speed drops the sand per meter rate does not exceed 7.5g/meter



As you can see there is a significant improvement in both the stopping distance and the range of stopping distances when comparing DVRS to SFRS.

 

[172007]

What has always struck me is when as is often with the first train of the day you slide into just about every station and you have say 20 stations, how much sand at 2kg per min or even 4kg is actually being chucked out on that one journey. If the train leaves the depot at 05:10 and arrives back at say 23:30 and the day has been a constant day of wet dry wet dry....... railhead as can happen with a strong drying wind and showers How much sand is used / left.

What kind of capacity does a sand hopper have for an Electrostar or Turbostar 10kg, 30kg or even 50kg?

 

[millemille]

What has always struck me is when as is often with the first train of the day you slide into just about every station and you have say 20 stations, how much sand at 2kg per min or even 4kg is actually being chucked out on that one journey. If the train leaves the depot at 05:10 and arrives back at say 23:30 and the day has been a constant day of wet dry wet dry....... railhead as can happen with a strong drying wind and showers How much sand is used / left.

What kind of capacity does a sand hopper have for an Electrostar or Turbostar 10kg, 30kg or even 50kg?

AFAICR Electrostars are about 30kg.

Hopper capacity is dictated by the space envelope available on the underframe after all the other equipment has been installed. My company specialises in retro-fit sander systems and we aim to get a minimum of 50kg capacity for a metro/suburban route running train.

 

[etr221]

Signallers have every right to expect a train to stop at a red signal. And adding additional delay into another train may well have additional risks down the line. You cannot run a railway like that

Trains stopping where they are supposed to - at signals which are red or at danger, or wherever - is THE fundamental of railway safe working, to which every thing works - and pretty well every safety improvement has worked to support it: by improving the reliability of the right signal being at red; and the train stopping at it.

In this case the driver's application of the brakes failed to stop the train: the RAIB will doubtless have more to say; for me to say anything more is to theorise without data.

But those questioning the signaller's regulation of the trains concerned perhaps have to answer the question of whether 1F30 would have managed to stop had SY37 been at red? (As it would have been had 1L53 been given priority)

 

[cjmillsnun]

I just can't quite shake off the feeling that braking was left late.

It was a foul night by all accounts, smack bang in the middle of the leaf-fall season, a line with an infrequent service, falling gradient, and a short train. To me there's quite a lot of potential alarm bells there, and I must say I was *very* surprised to read that the driver didn't at least test the water with a light brake application as soon as the double yellow on SY29R was sighted.

Yes the adhesion was particularly bad and the railhead hadn't been treated, but equally hazard anticipation is a part of driving any vehicle, and I can't help but find it a little surprising that the sequence of actions outlined by the RAIB simply doesn't seem to indicate any anticipation. One can only surmise, especially taking into consideration the driver's clearly very extensive experience, is that this location wasn't a location known for issues, but given the characteristics I can't help but find that surprising in itself.

Say the driver did do a light application at the double yellow at SY29R.

It is possible that a perfectly normal braking response may have been observed. Thus satisfied, the driver released the brakes and coasted to the same braking point then applied the brakes and had the same accident.

Railhead contamination is not the same in one area to the next.

The train slid for over a mile. That is seriously scary. The railhead should never have got that bad. No driver could be able to anticipate that.

 

[theking]

Signallers have every right to expect a train to stop at a red signal. And adding additional delay into another train may well have additional risks down the line. You cannot run a railway like that

Excuse me but they do run the railway like that.

Why do they double block trains at certain locations adding delays to the service waiting at a signal when it's perfectly fine to go into the platform.

I know of a certain location where they hold the train 2 signals back on a line with a 45mph limit and the junction the train crosses infront of the platform is 25mph and the train being held will stop at the platform so there is zero risk of letting the train into the platform.

In this risk averse world they should not be stopping 90mph trains for one to cross at 20mph infront of it.

 

[BayPaul]

Say the driver did do a light application at the double yellow at SY29R.

It is possible that a perfectly normal braking response may have been observed. Thus satisfied, the driver released the brakes and coasted to the same braking point then applied the brakes and had the same accident.

Railhead contamination is not the same in one area to the next.

The train slid for over a mile. That is seriously scary. The railhead should never have got that bad. No driver could be able to anticipate that.

In addition to what you have said, and obviously not something that anyone could anticipate, had the driver done such a test, been happy with it, and then applied the brakes at the same point, the accident could well have been very different. The 159 would have arrived at the junction a couple of seconds later due to time lost in the brake test, and so possibly would have ended up hitting the 158 end on, rather than a side-swipe. This would have led to very different, and potentially more serious, consequences, as the energy would have been lost in completely different ways.

 

[bengley]

In addition to what you have said, and obviously not something that anyone could anticipate, had the driver done such a test, been happy with it, and then applied the brakes at the same point, the accident could well have been very different. The 159 would have arrived at the junction a couple of seconds later due to time lost in the brake test, and so possibly would have ended up hitting the 158 end on, rather than a side-swipe. This would have led to very different, and potentially more serious, consequences, as the energy would have been lost in completely different ways.

Indeed - the collision was actually remarkably minor thanks to the side on collision. Nobody died. I have no doubt that if the 159 had hit the rear of the 158 at 55+mph, there would have been a very different outcome.

 

[O L Leigh]

Rather more "ifs" in your post than I'd like, if I'm honest. The theory is sound, but sadly we're not talking about a theoretical situation.

One thing I would point out is that reducing speed under braking isn't liner, this means that the faster you are going the less speed you lose for the same distance of braking compared to lower speeds.

Indeed. But, all other things being equal, it is linear in terms of time as the rate of deceleration would remain constant.

However, in this particular incident all other things were not equal which had a massive effect on the ability of the train brakes to slow the train. As a consequence, braking earlier may not necessarily have resulted in very much difference to the speed of the collision. The problem is that we simply cannot know this.

I'm not suggesting that the driver would be able to guess to brake sooner, but if there was a system which advised that they slowed a little sooner so that is they maintained a suitable speed that they should arrive at the signal with it being clear

Unless the Sirius Cybernetics Corporation allows the rail industry to use the technology employed in their Happy Vertical People Transporters, no technology is going to be able to instruct a driver to apply the brakes earlier for precisely the same reason that prevents drivers reaching the same conclusion for themselves.

The testing at Old Dalby and on the two WMT 323's fitted with the pilot equipment proved that 6%g, or better, stopping could be achieved no matter what the railhead conditions. The concept and equipment now needs to be subjected to normal service conditions with drivers from the normal driving pool to ensure that the results read over.

This looks interesting. 6%g is about half of the retardation that you would expect to get from an emergency brake application under normal conditions, but it might just have been enough to have prevented the SPAD and collision here. However, given that this is still under trial at the moment we're not likely to see the benefits of it for a few years. That being so, I wonder what chance there might be of Nitwit Rail working out some sort of priority system together with flexible pathing for RHTTs to ensure that emphasis is given to those circuits considered of higher priority.

 

[rebmcr]

My company specialises in retro-fit sander systems and we aim to get a minimum of 50kg capacity for a metro/suburban route running train.

Are there any mechanisms (whether produced by your company or by others) which are able to monitor the remaining sand, and alert when it drops below a certain threshold?

 

[ComUtoR]

I know of a certain location where they hold the train 2 signals back on a line with a 45mph limit and the junction the train crosses infront of the platform is 25mph and the train being held will stop at the platform so there is zero risk of letting the train into the platform.

Hither Green ?

Are there any mechanisms (whether produced by your company or by others) which are able to monitor the remaining sand, and alert when it drops below a certain threshold?

Electrostars and Desiro Cities have this feature.

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==

@millemille Are you not worried that increasing braking by 6% would just lead to Drivers going 6% harder and faster. A phenomenon with the introduction of new stock with better braking capabilities is that Drivers then tend to push the limits further. PDPs do help but the squishy thinking machine up the front will realise that they can now hit a platform a few mph fast or hit a PSR (speed limit) they previously couldn't.

 

[millemille]

This looks interesting. 6%g is about half of the retardation that you would expect to get from an emergency brake application under normal conditions, but it might just have been enough to have prevented the SPAD and collision here. However, given that this is still under trial at the moment we're not likely to see the benefits of it for a few years. That being so, I wonder what chance there might be of Nitwit Rail working out some sort of priority system together with flexible pathing for RHTTs to ensure that emphasis is given to those circuits considered of higher priority.

You can get more than 6%g, but that's the assured minimum. The point is though, if you can achieve brake step 2 stopping irrespective of rail head condition why would you need emergency brake, it should be close as possible to normal driving and how often do you need emergency brake in normal service?

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==

Are there any mechanisms (whether produced by your company or by others) which are able to monitor the remaining sand, and alert when it drops below a certain threshold?

Not wishing to blow our own horn, but we are the only sander OEM who has true sand level sensing; in that we measure the remaining sand level and can tell you to within 0.5% of the total capacity of the hopper how much sand is remaining, our competitors use level switches which can only indicate whether the sand is above or below a threshold, not how much above or below the threshold.

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==

@millemille Are you not worried that increasing braking by 6% would just lead to Drivers going 6% harder and faster. A phenomenon with the introduction of new stock with better braking capabilities is that Drivers then tend to push the limits further. PDPs do help but the squishy thinking machine up the front will realise that they can now hit a platform a few mph fast or hit a PSR (speed limit) they previously couldn't.

You may have misinterpreted the data.

It is not a 6% increase in braking, it is braking to achieve 6% of 1g as a retardation rate. Trains are normally timetabled on the assumption that they will always achieve 6%g braking, equivalent to brake step 2. This is where leaf fall season delays come from; trains slide when brake step 2 and normal braking markers are used so defensive driving - light and early braking - prevents the slide but means the train can't keep to timetable.

DVRS allows for "Seasonally agnostic" driving.

 

[43066]

Hither Green ?

20mph there, isn’t it!?

 

[ComUtoR]

You may have misinterpreted the data.

It is not a 6% increase in braking, it is braking to achieve 6% of 1g as a retardation rate.

No I understand the data but I may have communicated my point badly.

On the basic level. If you tell Drivers that their braking performance has increased. They will push it harder.

Through the introduction of 3 new stock types on my part of the network, each time braking performance has increased. Each time that has been met with an increase in stopping incidents because Drivers will push their trains further. If you tell Drivers that their sanders are excellent and braking distances have improved, I can guarantee someone will 'discover' how fast into a platform they can go.

Playing with Desiros; the sanders are very active during normal braking and pretty hyperactive during low adhesion but you can hit a 12-Car platform at 60mph and still stop....

 

[a340egkk]

Hither Green is 20mph at the junction for the Dartford Loop, 60mph through for the Kent fasts and slows. I think the signal prior to the junction signal on the up Kent slow is approach controlled.

 

[ComUtoR]

20mph there, isn’t it!?

I had allowed for an 'enthusiasts' interpretation of the junciton layout and speeds...

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==

Hither Green is 20mph at the junction for the Dartford Loop

Recheck your routes....

 

[rebmcr]

Not wishing to blow our own horn, but we are the only sander OEM who has true sand level sensing; in that we measure the remaining sand level and can tell you to within 0.5% of the total capacity of the hopper how much sand is remaining, our competitors use level switches which can only indicate whether the sand is above or below a threshold, not how much above or below the threshold.

Wonderful! After typing my question I wondered whether it is as simple as weighing the hopper!

 

[cjmillsnun]

In addition to what you have said, and obviously not something that anyone could anticipate, had the driver done such a test, been happy with it, and then applied the brakes at the same point, the accident could well have been very different. The 159 would have arrived at the junction a couple of seconds later due to time lost in the brake test, and so possibly would have ended up hitting the 158 end on, rather than a side-swipe. This would have led to very different, and potentially more serious, consequences, as the energy would have been lost in completely different ways.

That’s a very good point and thankfully not something the RAIB have to look at.

 

[357]

I don't know about sprinters, but on the Electrostars I drive, sanding and WSP is only available in power/brake step 2 and above.

As a result drivers are taught to go straight into step 2 when driving in low adhesion.

So step 2 later, rather than step 1 at the single yellow, could be because in low adhesion the train would stop better with the use of sand and WSP instead of just sliding.

Again, the above is based on my stock and I have no idea if the stock involved here works the same way!

 

[43066]

I had allowed for an 'enthusiasts' interpretation

Always the best interpretation. Line speed for HSTs was 130mph, right?!

 

[a340egkk]

I had allowed for an 'enthusiasts' interpretation of the junciton layout and speeds...

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==



Recheck your routes....

Hither Green Junc. is 20mph over the junction (and on the down, until clear of the country end of HG) if routed to/from the Dartford Loop (i.e. Sidcup), 60mph if not (or 15mph if you're going from the UKS to UKF IIRC)?

 

[43066]

Hither Green Junc. is 20mph over the junction (and on the down, until clear of the country end of HG) if routed to/from the Dartford Loop (i.e. Sidcup), 60mph if not (or 15mph if you're going from the UKS to UKF IIRC)?

Down sidcup is 60mph from the London end of HG platform 6 (20mph only across the junction) IIRC.

 

[millemille]

No I understand the data but I may have communicated my point badly.

On the basic level. If you tell Drivers that their braking performance has increased. They will push it harder.

Which is exactly, within the bounds of safety, that the industry needs.

Low adhesion related delays cost the industry and wider society ~£350M per year. The majority of this sum is penalties for late trains and late trains are generally caused by defensive driving. Defensive driving; Braking early and light to prevent, as far as is possible, slides from occurring (and if they do, leaving room to cope with the reduced stopping performance) costs money.

It has been proven on the test track at Old Dalby and on closed track on the Cross City line in Birmingham, as well as some limited in-service operation, that DVRS gives the potential for the train to be driven as though it is running on clean, dry rail no matter what the rail head conditions. So no more defensive driving; brake at your normal markers and using normal brake step (step 2) and it will stop.

It doesn't allow the train to stop any better than it would do on clean dry rail.

== Doublepost prevention - post automatically merged: 23 Feb 2022 ==

I don't know about sprinters, but on the Electrostars I drive, sanding and WSP is only available in power/brake step 2 and above.

As a result drivers are taught to go straight into step 2 when driving in low adhesion.

So step 2 later, rather than step 1 at the single yellow, could be because in low adhesion the train would stop better with the use of sand and WSP instead of just sliding.

Again, the above is based on my stock and I have no idea if the stock involved here works the same way!

Small point. WSP works in any and all brake steps. It's the sanding that is restricted to certain brake steps.

 

[357]

Small point. WSP works in any and all brake steps. It's the sanding that is restricted to certain brake steps.

Are you talking about Sprinters or Electrostars?

Our training material specifically states no WSP in step 1 on our Electrostars.

 

[ComUtoR]

Hither Green Junc. is 20mph over the junction

I could have sworn they moved that 20 to before the station on the up too. Apologies if that isn't correct. No more route assessments for me !!

Down sidcup is 60mph from the London end of HG platform 6 (20mph only across the junction) IIRC.

On the Down they moved the 20 board to the London end

 

[PG]

It has been proven on the test track at Old Dalby and on closed track on the Cross City line in Birmingham, as well as some limited in-service operation, that DVRS gives the potential for the train to be driven as though it is running on clean, dry rail no matter what the rail head conditions.

Would this, potentially, reduce the requirements for routes to be treated so frequently by the RHTT?