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

[ComUtoR]

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

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 if I hit a platform at 45mph on a 'Black' day. I won't go flying through the platform ? Would it be a failure of the safety systems if it did ?

So no more defensive driving; brake at your normal markers and using normal brake step (step 2) and it will stop.

I don't brake using 'step 2' and many of us at my TOC don't either. I was taught using 'earlier and lighter' as my normal braking technique. For me, that isn't 'defensive driving'

 

[Nicholas Lewis]

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.

You haven't identified location but this would be the arrangement if there was a substandard overlap between the platform starter and a conflicting move ahead. Some signalling systems allow the signaller to set up a warner route with a reduced overlap which is effectively an approach controlled arrangement to allow the train to draw forward but I believe this is even frowned upon in the latest signalling principles.

 

[millemille]

Are you talking about Sprinters or Electrostars?

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

I will check with our brake engineer tomorrow, but I'm pretty confident when I say, based on my 30 years of rolling stock engineering experience, there's not a single multiple unit in service in the UK that, if equipped with WSP, that doesn't have WSP working in ALL service brake steps.

There did used to be a few BR era units that didn't have WSP in emergency brake: but that was based on the belief that a sliding/locked wheel stopped a train better than a wheel with WSP activity. However a combination of better data and improved WSP algorithms have proved this to be wrong.

I know, from first hand experience, that many multiple units - including some electrostars - don't have any sander operation brake step 1. This has nothing to do with there being no WSP activity to trigger it and everything to do with trying to preserve the sand stored in the hopper.

 

[357]

I will check with our brake engineer tomorrow, but I'm pretty confident when I say, based on my 30 years of rolling stock engineering experience, there's not a single multiple unit in service in the UK that, if equipped with WSP, that doesn't have WSP working in ALL service brake steps.

There did used to be a few BR era units that didn't have WSP in emergency brake: but that was based on the belief that a sliding/locked wheel stopped a train better than a wheel with WSP activity. However a combination of better data and improved WSP algorithms have proved this to be wrong.

I know, from first hand experience, that many multiple units - including some electrostars - don't have any sander operation brake step 1. This has nothing to do with there being no WSP activity to trigger it and everything to do with trying to preserve the sand stored in the hopper.

Sounds good! Will continue this chat via PM and stop drifting off topic

 

[millemille]

So if I hit a platform at 45mph on a 'Black' day. I won't go flying through the platform ? Would it be a failure of the safety systems if it did ?

If you brake at the markers, and using the brake steps you use in normal operation at any other time of year on a day where nothing that could cause low adhesion exists, then the train should stop as though on clean and dry rail.

if the sander system (which is 2 sanders per unit) fails and causes an over run/SPAD then yes, it will be a failure of the safety system if it did.

There is a lot of work going on "behind the scenes" to change the design, maintenance and application of sanders across the rail industry. There's also significant work going on to establish what indication is needed in the cab to give the driver the information they need so that they can have confidence in the sander system and it always working when they need it and/or be informed that it is not working BEFORE they find out by sliding through a platform/red signal.

 

[43066]

I will check with our brake engineer tomorrow, but I'm pretty confident when I say, based on my 30 years of rolling stock engineering experience, there's not a single multiple unit in service in the UK that, if equipped with WSP, that doesn't have WSP working in ALL service brake steps.

Please do report back!

 

[ComUtoR]

It will be a huge mountain to climb to convince the industry that hitting platforms at speed, in low adhesion, is perfectly safe and acceptable. I look forward to the day where it becomes the norm. Probably post retirement but I'll look back with fondness and rose tinted specs.

 

[O L Leigh]

You can get more than 6%g, but that's the assured minimum.

Thank you, yes. I understood.

 

[millemille]

It will be a huge mountain to climb to convince the industry that hitting platforms at speed, in low adhesion, is perfectly safe and acceptable. I look forward to the day where it becomes the norm. Probably post retirement but I'll look back with fondness and rose tinted specs.

That's what the trial on 323's is for; DVRS has worked in controlled conditions with specially selected/trained/supervised drivers but the industry needs to see what happens when normal drivers are left to their own devices in normal service.

 

[mr_moo]

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

Here you go - Latest sectional appendix page to show exactly what is where!

 

[millemille]

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

Only if, as an industry, we can come up with a way of leaf fall debris build up on wheels not causing loss of track circuit activation.

RHTT's fulfil two purposes; adhesion management and signalling operation assurance.

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

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

Weighing the hopper only works when the train is stationary, as soon as you start introducing dynamic movement you can't weigh anything reliably.

We use a non-contact sensor of a relatively simple design but the software which the sensor reports in to is pretty complex in terms of how it processes the data.

 

[rebmcr]

Only if, as an industry, we can come up with a way of leaf fall debris build up on wheels not causing loss of track circuit activation.

RHTT's fulfil two purposes; adhesion management and signalling operation assurance.

It's certainly not close to reaching full network coverage, but installation of axle counters continues, and one day perhaps will be universal...

Weighing the hopper only works when the train is stationary, as soon as you start introducing dynamic movement you can't weigh anything reliably.

We use a non-contact sensor of a relatively simple design but the software which the sensor reports in to is pretty complex in terms of how it processes the data.

Yeah, knowing how difficult it can be for a computer to be confidently accurate about the real world (e.g. the location of a train for stopping alongside platform doors), I don't doubt that there is quite some sophistication to the measurement!

 

[O L Leigh]

I don't brake using 'step 2' and many of us at my TOC don't either. I was taught using 'earlier and lighter' as my normal braking technique. For me, that isn't 'defensive driving'

Ditto. And driving using this technique doesn't result in delays; at least not on the services I drive.

On this point, however, there is a nagging thought I have in the back of my mind about low adhesion driving that I thought I'd air for discussion.

I recall having a conflab with a driver manager at my former TOC some years back about low adhesion driving. He was banging the drum, telling me that if I thought I was about to slide through somewhere that I should bang the brake straight into emergency. To me this really didn't make a huge amount of sense except on the basis that it ticked a box.

My thinking was that if you were already in a slide it meant that the amount of brake force you were trying to use had already exceeded the limit of adhesion afforded by the conditions, so how would attempting to use even more brake force help you to stop any faster? Surely all you'd do is exacerbate the slide and, by selecting emergency, lose control over the train. I'd shared this thought process with him, but he wasn't going to be distracted from pushing the party line and so I agreed that, yes, if I found myself in a situation like this I would do as he demanded so that he could put it on his form.

I'd parked these thoughts for a while, but the overrun at Stonegate back in November 2010 caused them to resurface. For those who don't recall, this was when a SET service slid for over 3 miles due to a combination of poor railhead conditions and a lack of sand in the leading end of the train. The driver had been using the updated low adhesion driving technique of applying Step 2 brake in order to get the assistance of the sanders but found himself in difficulty approaching Stonegate. The RAIB report into the overrun notes the following;

45 After a pause of 20 seconds, the driver restarted the train and continued in service towards Hastings.

46 In response to the difficulties experienced at Stonegate, the driver changed his driving technique to make use of brake step one, braking much earlier and lighter than previously. Using this technique he made the normal service stops on the uphill section of the line at Etchingham, Robertsbridge and Battle stations.

47 After Battle, the driver returned to the Southeastern driving policy of using brake step two as the initial application. He then had more adhesion difficulties, and had to use the emergency brake on the downhill approaches to both Crowhurst and West St Leonards.

I found it very interesting that, even without sand, going back to the former instructions of "earlier and lighter" the driver was able to make the next three stations without difficulty (no doubt aided in part by the rising gradient). Here was some vindication of my earlier thoughts regarding low adhesion.

Now obviously it's not appropriate to drive like this everywhere on the network. Clearly you can't be doing this on a busy inner-suburban network, but where I work now it does seem that a driver can still utilise this technique without any apparent performance penalties. No doubt this could be replicated elsewhere also.

Thoughts...?

 

[357]

Thoughts...?

I've often thought the same myself, I do brake earlier in low adhesion but still into step two, then if appropriate back to step 1.

In the event of an overrun - even if I am in Emergency - I feel that the company would try to pin the blame on me for "not following PDP" if I don't go straight into step 2.

 

[75A]

I've often thought the same myself, I do brake earlier in low adhesion but still into step two, then if appropriate back to step 1.

In the event of an overrun - even if I am in Emergency - I feel that the company would try to pin the blame on me for "not following PDP" if I don't go straight into step 2.

Sound words,
In my day we had sand available in the 33/73's but very rarely had to use it.

 

[Tomnick]

There did used to be a few BR era units that didn't have WSP in emergency brake...

Something's been niggling away at me for a little while now, and this is a very interesting statement in that respect. Please bear with me, I might ramble a bit, and please forgive me if I've completely missed the mark!

On BR era units, with the Westcode braking system (as fitted on the 159 involved here), the brake application is controlled by two train wires (23 and 24) interfacing with a brake control unit on each vehicle. "Train wire 4" (other sources quote a different number!) provides the feed to these two wires, via the brake controller which energises and de-energises each wire as appropriate. If both 23 and 24 are de-energised, you get a step three brake application. If an emergency brake application is made, "train wire 4" is also cut , and I understand that there's an additional wire that's de-energised as a result (negative return?). My understanding is that the brake control unit on each vehicle, on some units at least, also interacts either with "train wire 4" or the negative return - on some traction that I sign, you get higher brake cylinder pressures with an emergency brake application than you do with a full service, i.e. step 3, application.

If the main res pressure on any vehicle drops below 4.5bar, the main air pressure switch should operate, interrupting "train wire 4" and thus de-energising both train wires 23 and 24, and the negative return, resulting in each brake control unit demanding a full (emergency) brake application, the air of course being supplied from the brake reservoir which should be protected by a non-return valve (but of course now has a finite supply if it's not being replenished). This is an important safeguard in the event of loss of main res pressure as it removes the risk of running out of air to operate the brakes. The last thing that you want in that situation is WSP activity, as that could quickly deplete the air pressure remaining in the brake reservoir and leave you with nothing.

My question is whether the individual brake control units and/or WSP controllers can differentiate between an emergency brake application initiated as a result of "train wire 4" being interrupted by a loss of main res pressure, and an emergency brake application initiated by the brake controller being placed to 'emergency'?

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

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.

There is a difference between restrictions imposed by the interlocking and/or special instructions, and signallers taking matters into their own hands.

In the example you give, which is most likely a restriction imposed by the interlocking, there certainly isn't "zero risk" associated with letting the train into the platform. The train should stop, but it might not stop. It might not even be booked to stop (the interlocking doesn't know or care whether it's booked to stop or not).

Regulating decisions have to be made on the basis of minimising overall delay in the bigger picture, not on the basis of individual assessments of risk like this that should already be accounted for in the signalling design (or special instructions). After all, using an extreme example at a different location, holding an Up train for the 20mph diverging route at a junction to allow a 90mph Down train to run over the straight route first then exposes the Up train to the risk of the following 90mph train being unable to stop before colliding with it in the rear!

 

[357]

My question is whether the individual brake control units and/or WSP controllers can differentiate between an emergency brake application initiated as a result of "train wire 4" being interrupted by a loss of main res pressure, and an emergency brake application initiated by the brake controller being placed to 'emergency'?

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

On my stock putting my handle in emergency breaks the same wire as an emergency brake demand from anything else - wire 13 on my stock, believe some other trains use wire 4 for the same.

 

[Tomnick]

On my stock putting my handle in emergency breaks the same wire as an emergency brake demand from anything else - wire 13 on my stock, believe some other trains use wire 4 for the same.

That's sort of what I'm getting at. You don't want the WSP working away busily if you have lost the main res, but how does it know the difference between that and an emergency brake demand initiated by the driver, and in turn does that mean that you don't get the benefit of the WSP in an emergency brake application? I genuinely don't know, and neither does anyone else that I've asked!

It is possible, I guess, that the WSP system has its own interface with the main air pressure switch on that vehicle, so that it does operate under emergency braking unless it detects that the main res is low.

 

[Annetts key]

It's certainly not close to reaching full network coverage, but installation of axle counters continues, and one day perhaps will be universal...

May be rather a long time off, given that in some locations new track circuits are still being installed, and in other locations, where the signalling is fairly new or where the track circuits are fairly new (read, still have a considerable remaining life span) there would need to be something else to tip the cost/benefit calculations over.

And, it may surprise people, that even in an area which is mostly axle counters, there are some locations where track circuits have been retained. Although it’s hard to work out why.

 

[The Ham]

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.



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

Yes, however by starting at a lower speed (let's say 87mph rather than 89mph that's a speed of 1.5m/s slower, (circa 38.5m/s rather than circa 40m/s) as such even if breaking is constant (which it's not, as it would slow you down more over every 10m at lower speeds) that 1.5m/s reduced speed means that after 26 seconds you'll have traveled about 38.5m less, meaning that you've gained an extra second of braking. Whilst that may not gain you much of a reduction (given the poor breaking conditions), a few extra seconds breaking and a few mph slower could result in the leading train being further ahead and the closing speed of the two trains being reduced at the point of impact.

The other thing to note, if (say) a 2mph reduction was achieved by a brake test at the first non green signal it means that the potential energy is reduced, that means that during the coasting period that it's likely that little more speed would have been lost than was the case. Meaning rather than breaking happening at 2mph slower (the reduction from the brake test) it is likely to be slower by 3mph or more.

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.

Indeed, however within braking calculations there's only so many variables, however they all have an impact on the end speed. If the deceleration rate is the same (even if it's almost zero, which it wasn't as the train did show a bit) but the starting speed is lower the outcome will still be a lower end speed, not least as there would have been a marginally longer duration of braking due to it taking longer to pass over the same length of track Likewise a longer duration (due to starting to brake sooner), even would have some impact add there would have been a longer time for breaking (again deceleration isn't quite zero, so the extra time would have reduced the speed).

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 suggestion is that with a tablet which advises a driver (say) 10 miles out to reduce their speed from 90 to (say) 75 so that they get to a clear signal when they reach the junction then this isn't the stuff of Sci-Fi. As the system would know the timings that would be required to allow the proceeding trains to clear the junction (assuming that they do as expected). However if the proceeding trains don't do as they are expected, by starting the breaking from (say) 75mph rather than 90mph and being (say) 30 seconds later that changes the outcome.

The system has access to more information than the driver. As it know what's supposed to be happening, the driver doesn't know that. As they don't know where the trains ahead of them are (just that there's something which means they can't enter the junction at that point in time). However that doesn't negate the need for the driver to be skilled as there's always going to be the chance that the system doesn't get it right (you don't need to know the history of the Post Office Horizon system to be aware that computers are only as good as the people that program them).

However if the system gets it wrong, the existing signal system would still give the same level of protection that it currently does.

 

[ComUtoR]

Thoughts...?

The 'Theory' behind step 2 braking came out of, as always, an incident (thats how it was briefed to us). There was an overrun of a mile and a quarter that resulted in a collision. Because of the sand not applying in step 1 it was decided to 'force' the unit into a slide to deliberately activate the WSP and the sand. It is also a multi part policy (which people forget) the next part of the process is to not wait or try and control the slide but to stick it straight into emergency. Thus reducing the length of the overrun.

Brake 2 to force the WSP to work and apply sand
Don't 'feather the brake'
Apply emergency sooner than later.

The 'feathering' is a little old school but still applies because dropping the brake then stops the WSP from functioning. (You effectively tell the train to coast). It's about letting the WSP do its job. I've done a RBT at 100mph and the needle dropped to zero. I casually knocked it off and just continued on my way.

Earlier and lighter works as a technique because it gives the Driver more control. Earlier and lighter reduces the initial risk of sliding. Again, the theory being that there is no risk if your not sliding and creeping along. If you suddenly did slide then the overrun distance is minimal. We have had situations where a Driver has hit a platform at 10 mph and gone out the back at 20 !

Our current PDP just says to 'Brake Earlier' and then the rest is down to Driver skill. There has to be a balance between overly defensive driving policy and improved safety systems. If we improve our safety system but then drive around like F1 Drivers then little is achieved. Yes there may be a financial incentive for pushing performance but another reason why Brake 2/Emergency was dropped at my TOC... Those pesky wheel flats from all the sliding !

Again, a reminder, PDPs are not just about braking. Defensive driving polices can also outline where to reduce speeds, outline platform stopping policy, passenger comfort actions, etc etc. All of that is what kills performance from defensive driving policies. I run round all day using earlier and lighter and don't lose time.

 

[theageofthetra]

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

View attachment 110637

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.

This does of course rely on the unit having any sand in it, or it working. On most older stock the driver has no idea of this. I'm certainly interested to see this fully investigated.

== Doublepost prevention - post automatically merged: 24 Feb 2022 ==

If you brake at the markers, and using the brake steps you use in normal operation at any other time of year on a day where nothing that could cause low adhesion exists, then the train should stop as though on clean and dry rail.

if the sander system (which is 2 sanders per unit) fails and causes an over run/SPAD then yes, it will be a failure of the safety system if it did.

There is a lot of work going on "behind the scenes" to change the design, maintenance and application of sanders across the rail industry. There's also significant work going on to establish what indication is needed in the cab to give the driver the information they need so that they can have confidence in the sander system and it always working when they need it and/or be informed that it is not working BEFORE they find out by sliding through a platform/red signal.

Great to know. Thanks for the information

 

[bramling]

The 'Theory' behind step 2 braking came out of, as always, an incident (thats how it was briefed to us). There was an overrun of a mile and a quarter that resulted in a collision. Because of the sand not applying in step 1 it was decided to 'force' the unit into a slide to deliberately activate the WSP and the sand. It is also a multi part policy (which people forget) the next part of the process is to not wait or try and control the slide but to stick it straight into emergency. Thus reducing the length of the overrun.

Brake 2 to force the WSP to work and apply sand
Don't 'feather the brake'
Apply emergency sooner than later.

The 'feathering' is a little old school but still applies because dropping the brake then stops the WSP from functioning. (You effectively tell the train to coast). It's about letting the WSP do its job. I've done a RBT at 100mph and the needle dropped to zero. I casually knocked it off and just continued on my way.

Earlier and lighter works as a technique because it gives the Driver more control. Earlier and lighter reduces the initial risk of sliding. Again, the theory being that there is no risk if your not sliding and creeping along. If you suddenly did slide then the overrun distance is minimal. We have had situations where a Driver has hit a platform at 10 mph and gone out the back at 20 !

Our current PDP just says to 'Brake Earlier' and then the rest is down to Driver skill. There has to be a balance between overly defensive driving policy and improved safety systems. If we improve our safety system but then drive around like F1 Drivers then little is achieved. Yes there may be a financial incentive for pushing performance but another reason why Brake 2/Emergency was dropped at my TOC... Those pesky wheel flats from all the sliding !

Again, a reminder, PDPs are not just about braking. Defensive driving polices can also outline where to reduce speeds, outline platform stopping policy, passenger comfort actions, etc etc. All of that is what kills performance from defensive driving policies. I run round all day using earlier and lighter and don't lose time.

I’ve always found the regimental nature of (some) PDPs unfortunate, and this manifested itself in the Stonegate driver feeling the need to operate the train in a way which was against their better judgement out of concern for something flagging up on a potential download. That is extremely undesirable.

It seems counter-intuitive to spend a considerable amount of money and time training up a driver to a point where at the end of that process they should be pretty skilled, but then lay down to a pretty fine detail such things as what brake position to select. In the absence of a really sophisticated sanding system, earlier and lighter is always going to be preferable in potential low adhesion - it’s better not to get into a slide in the first place than to have to rely on sand. By the same token there is no need for drivers to be adopting that style all the time. Defensive driving is meant to be more about considering, anticipating and mitigating hazards than anything else.

What is the point in having skilled drivers if we then expect them to drive like ATO?

This seems to be more of an issue for the commuter TOCs, I get the feeling the “Intercity” operators seem to have a little more confidence in their drivers.

One way or other, the industry got dealt a decent pack of cards at Salisbury. A near-60mph collision right in front of a tunnel portal (plus a third train nearby) could have ended a *lot* worse.

 

[Annetts key]

One VERY important thing to keep in mind with all the “what if” speculation on braking, speed, timing etc. is that due to the number of variables in play here, the result if there were slight changes to what happened could have been better, or could have been worse. Or may not have made any significant difference.

So I’m not sure that this speculation is particularly useful at the moment. For one thing, we don’t know the full extent of the length of the rail contamination…

The suggestion is that with a tablet which advises a driver (say) 10 miles out to reduce their speed from 90 to (say) 75 so that they get to a clear signal when they reach the junction then this isn't the stuff of Sci-Fi. As the system would know the timings that would be required to allow the proceeding trains to clear the junction (assuming that they do as expected). However if the proceeding trains don't do as they are expected, by starting the breaking from (say) 75mph rather than 90mph and being (say) 30 seconds later that changes the outcome.

The system has access to more information than the driver. As it know what's supposed to be happening, the driver doesn't know that. As they don't know where the trains ahead of them are (just that there's something which means they can't enter the junction at that point in time). However that doesn't negate the need for the driver to be skilled as there's always going to be the chance that the system doesn't get it right (you don't need to know the history of the Post Office Horizon system to be aware that computers are only as good as the people that program them).

However if the system gets it wrong, the existing signal system would still give the same level of protection that it currently does.

I don’t think you understand the way the signalling system works. The system in this location is very likely not a computer based signalling system. Not that it makes any difference anyway as conventional computer based systems follow the same principles as earlier electrical interlocking systems.

The signalling system at this location is a multiple aspect colour light system working to a track circuit block system.

And because it uses track circuits, each of which is a individual fixed length, not even the signalling system knows the exact location of where each train is. All it knows is that a train is somewhere on a particular track circuit section. Track circuit sections can be anywhere between 50 yards long to seven miles long. And the signalling system has no idea of the speed of any of the trains. Or how much time it takes for a train to move through a section. Indeed, if the line is not designed for bidirectional signalling, the signalling system will not even know in which direction a train is moving.

Plus, the junction signal is a ‘controlled’ signal. It will only clear up if the signaller operates it’s control function AND all the conditions required by the interlocking circuitry are proved to be correct (relevant track circuits clear, points ‘called’ to the correct position and detected in the correct position, next signal ahead lit, plus various other controls).

So unless you have a new technology that can predict future events, it’s not possible to predict when the junction signal will clear to a proceed aspect.

And the very LAST thing you want to do is to mislead a driver. A driver must always obey the aspect of the last signal that he or she saw.

If you want a different system, then the best that can be done is to install ETCS (in cab signalling). In due course, Network Rail (or whatever it’s called by then) will be installing ETCS everywhere. But it will take time.

 

[The Ham]

I don’t think you understand the way the signalling system works. The system in this location is very likely not a computer based signalling system. Not that it makes any difference anyway as conventional computer based systems follow the same principles as earlier electrical interlocking systems.

The signalling system at this location is a multiple aspect colour light system working to a track circuit block system.

And because it uses track circuits, each of which is a individual fixed length, not even the signalling system knows the exact location of where each train is. All it knows is that a train is somewhere on a particular track circuit section. Track circuit sections can be anywhere between 50 yards long to seven miles long. And the signalling system has no idea of the speed of any of the trains. Or how much time it takes for a train to move through a section. Indeed, if the line is not designed for bidirectional signalling, the signalling system will not even know in which direction a train is moving.

Plus, the junction signal is a ‘controlled’ signal. It will only clear up if the signaller operates it’s control function AND all the conditions required by the interlocking circuitry are proved to be correct (relevant track circuits clear, points ‘called’ to the correct position and detected in the correct position, next signal ahead lit, plus various other controls).

So unless you have a new technology that can predict future events, it’s not possible to predict when the junction signal will clear to a proceed aspect.

And the very LAST thing you want to do is to mislead a driver. A driver must always obey the aspect of the last signal that he or she saw.

If you want a different system, then the best that can be done is to install ETCS (in cab signalling). In due course, Network Rail (or whatever it’s called by then) will be installing ETCS everywhere. But it will take time.

I understand that with current systems you can not know exactly where a train is, especially on very long sections.

Likewise you should never bring in a system which overrides the signals.

However I may not have been overly clear.

First up the system would only ever provide a guide to train speed, which would never set the speed faster than a suitable speed for the section of line based on the current signal conditions. For a specific example it may, at the first non green signal (or possibly even before that), have advised the driver to reduce speeds to a maximum of 75mph. This would be a lower maximum speed than the line speed. It's there anything in that which is misleading the driver? Does that stop the driver from thinking, I know that conditions are bad and so I'll go a bit slower than the maximum speed which I've been advised.

It's like if there's a 50mph limit on a smart motorway, if there's thick freezing fog that doesn't stop a car or lorry driver going slower.

Whilst the system may not know the exact location of the trains within the junction the distances are likely to be fairly short and movement times are likely to be fairly similar. However even if something doesn't work out (say a train stops for whatever reason), then the system would update new maximum speeds to the point where a driver comes to a stop at the signal controlling a junction.

On longer run sections the system would know the likely profile of the leading train and advise the following one accordingly. For instance a class 80x following a class 158 would be advised to run at a maximum of 90mph even if the line speed was 125mph.

Of course the above also could be made more accurate by the advisory speed system being aware of which tablets are using it and their locations.

The system wouldn't be a replacement for the driver's skill, rather a maximum speed advisory system. It's goal would be reduced the need for trains to stop at signals and in doing so reduce the energy requirements of the railways (from reducing the account of acceleration which is undertaken). However drivers, much like lorry drivers on variable speed limit roads, still have to be aware of other factors (signals, weather conditions, etc.).

Yes the system isn't going to make it that no train is ever going to stop away from a platform, however by knowing that a train movement is currently underway which prohibits the signal at red from giving clearance for the train approaching to proceed and that it's likely that it'll not do so for at least another 3 minutes and that under normal conditions the train on approach will take 2 minutes to get to the signal. As such it would advise the train to slow to a maximum speed so it then takes 3 minutes 20 seconds to get there.

Now if the train movement underway takes 4 minutes to undertake the movement, the approaching train will still stop at the signal as the driver will still follow the instructions of the signals (like a car should still stop in a 40mph limit of the traffic lights are red). However if it clears it in 3 minutes then the protecting signal (assuming everything is lined up to allow it to do so) would then allow the approaching train to proceed, potentially without the train stopping (although it would still have slowed significantly).

The training behind such a system would be that it's guidance, however if you fail to stop at a signal that would be the same as if you were following it's guidance or not. Likewise if you exceed the maximum speed guidance and something goes wrong then you'd have to have a fairly good reason for doing so (for instance poor braking conditions). Likewise you should know the maximum linespeed and of it advises you to go faster than that, then it should be reported and the linespeed limit adhered to (as systems are only as good as the people programming them and this is only a secondary system).

 

[ComUtoR]

It's there anything in that which is misleading the driver? Does that stop the driver from thinking, I know that conditions are bad and so I'll go a bit slower than the maximum speed which I've been advised.

The training behind such a system would be that it's guidance,

This is the biggest problem. 'Guidance and advice' is open and even if the DAS advised to go 75mph you will still have trains going linespeeds and ignoring the DAS altogether. You mention cars and motorways so consider the orange speed limit signs that give 'advice' They are mostly ignored and you burn past them. Even on a smart motorway when the limit changes you get Drivers blasting past the reduction and the red circle ones are mandatory. How many times do you look at your SatNav/Waze etc. and think, na, I'll take the next left instead because you don't believe the computer or from experience you know that a quick left takes a little used shortcut.

Humans are not very good at following 'guidance' However, they can be very good at following an instruction. I know that when I leave a certain station that I will get checked down to a Red at the junction because there will be another service going across me. I don't need a speed advisory here because its part of my route knowledge and something experienced every trip. Giving a Driver more information isn't always a good option either. There needs to be a balance between overload and underload. Humans are also very good at filtering out information. Something that will only offer 'advice' can easily become background noise and get filtered out anyway.

 

[43066]

View attachment 110659

Here you go - Latest sectional appendix page to show exactly what is where!

I’m sure that has changed, and that the 60 board used to be before platform 6 heading down (which is what I think @ComUtoR was referring to).

Apologies as off topic but interesting nonetheless.

On longer run sections the system would know the likely profile of the leading train and advise the following one accordingly. For instance a class 80x following a class 158 would be advised to run at a maximum of 90mph even if the line speed was 125mph.

This is already largely done through route knowledge, especially in four aspect areas. Eg if you know you’re following a 100mph EMU, running at 95 - 100mph results in the signals stepping up to green in front of you. It’s possible to fine tune this very accurately, e.g. slowing down to circa. 40mph following an all stops service works well to avoid encountering a red at all, and also avoids constant acceleration and deceleration.

The most useful thing on my patch would be knowing what the next signal aspect is on the long three aspect sections, because if you pass one at single yellow you must assume the next is at danger, and creep for almost a mile towards it.

 

[Poppysdad]

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 ==


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 ==


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.

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 ==


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 ==


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.

Ref the sand level sensing you are not strictly correct regarding measurement of sand. You are correct in so much as older trains had a sensor which triggered an indication when the sand level reached a pre determined level but this is not the case with more modern trains with sanding level monitoring from many different suppliers which measures the amount of sand in the sand hopper.

== Doublepost prevention - post automatically merged: 25 Feb 2022 ==

Something's been niggling away at me for a little while now, and this is a very interesting statement in that respect. Please bear with me, I might ramble a bit, and please forgive me if I've completely missed the mark!

On BR era units, with the Westcode braking system (as fitted on the 159 involved here), the brake application is controlled by two train wires (23 and 24) interfacing with a brake control unit on each vehicle. "Train wire 4" (other sources quote a different number!) provides the feed to these two wires, via the brake controller which energises and de-energises each wire as appropriate. If both 23 and 24 are de-energised, you get a step three brake application. If an emergency brake application is made, "train wire 4" is also cut , and I understand that there's an additional wire that's de-energised as a result (negative return?). My understanding is that the brake control unit on each vehicle, on some units at least, also interacts either with "train wire 4" or the negative return - on some traction that I sign, you get higher brake cylinder pressures with an emergency brake application than you do with a full service, i.e. step 3, application.

If the main res pressure on any vehicle drops below 4.5bar, the main air pressure switch should operate, interrupting "train wire 4" and thus de-energising both train wires 23 and 24, and the negative return, resulting in each brake control unit demanding a full (emergency) brake application, the air of course being supplied from the brake reservoir which should be protected by a non-return valve (but of course now has a finite supply if it's not being replenished). This is an important safeguard in the event of loss of main res pressure as it removes the risk of running out of air to operate the brakes. The last thing that you want in that situation is WSP activity, as that could quickly deplete the air pressure remaining in the brake reservoir and leave you with nothing.

My question is whether the individual brake control units and/or WSP controllers can differentiate between an emergency brake application initiated as a result of "train wire 4" being interrupted by a loss of main res pressure, and an emergency brake application initiated by the brake controller being placed to 'emergency'?

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


There is a difference between restrictions imposed by the interlocking and/or special instructions, and signallers taking matters into their own hands.

In the example you give, which is most likely a restriction imposed by the interlocking, there certainly isn't "zero risk" associated with letting the train into the platform. The train should stop, but it might not stop. It might not even be booked to stop (the interlocking doesn't know or care whether it's booked to stop or not).

Regulating decisions have to be made on the basis of minimising overall delay in the bigger picture, not on the basis of individual assessments of risk like this that should already be accounted for in the signalling design (or special instructions). After all, using an extreme example at a different location, holding an Up train for the 20mph diverging route at a junction to allow a 90mph Down train to run over the straight route first then exposes the Up train to the risk of the following 90mph train being unable to stop before colliding with it in the rear!

 

[dciuk]

But no interrogation or investigation of the signallers actions.

If they didn't route a train that was 18 late infront of the SWR then this all could have been avoided.

While on this occassion it was only due to late running of the GWR service that both trains reached tunnel junction around the same time, it is quite common (at least was when I used to often travel on this route as a passenger a few years ago that the Waterloo to Exeter service would be held before tunnel juncton for a Portsmouth to Cardiff service as it was a scheduled connection from Portsmouth to Cardiff service to the Exeter Service at Salisbury. If the Portsmouth to Cardiff was running a few minutes late the Exeter Service would sometimes enter Salisbury ahead of it (presumably where doing so would not delay the Cardiff service further) but would then be held as Salisbury while the Cardiff service arrived and departed. On this occassion by delaying the Exeter at Tunnel Junction makes sense as by not doing so would have further delayed both services leaving Salisbury.

 

[Poppysdad]

Something's been niggling away at me for a little while now, and this is a very interesting statement in that respect. Please bear with me, I might ramble a bit, and please forgive me if I've completely missed the mark!

On BR era units, with the Westcode braking system (as fitted on the 159 involved here), the brake application is controlled by two train wires (23 and 24) interfacing with a brake control unit on each vehicle. "Train wire 4" (other sources quote a different number!) provides the feed to these two wires, via the brake controller which energises and de-energises each wire as appropriate. If both 23 and 24 are de-energised, you get a step three brake application. If an emergency brake application is made, "train wire 4" is also cut , and I understand that there's an additional wire that's de-energised as a result (negative return?). My understanding is that the brake control unit on each vehicle, on some units at least, also interacts either with "train wire 4" or the negative return - on some traction that I sign, you get higher brake cylinder pressures with an emergency brake application than you do with a full service, i.e. step 3, application.

If the main res pressure on any vehicle drops below 4.5bar, the main air pressure switch should operate, interrupting "train wire 4" and thus de-energising both train wires 23 and 24, and the negative return, resulting in each brake control unit demanding a full (emergency) brake application, the air of course being supplied from the brake reservoir which should be protected by a non-return valve (but of course now has a finite supply if it's not being replenished). This is an important safeguard in the event of loss of main res pressure as it removes the risk of running out of air to operate the brakes. The last thing that you want in that situation is WSP activity, as that could quickly deplete the air pressure remaining in the brake reservoir and leave you with nothing.

My question is whether the individual brake control units and/or WSP controllers can differentiate between an emergency brake application initiated as a result of "train wire 4" being interrupted by a loss of main res pressure, and an emergency brake application initiated by the brake controller being placed to 'emergency'?

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


There is a difference between restrictions imposed by the interlocking and/or special instructions, and signallers taking matters into their own hands.

In the example you give, which is most likely a restriction imposed by the interlocking, there certainly isn't "zero risk" associated with letting the train into the platform. The train should stop, but it might not stop. It might not even be booked to stop (the interlocking doesn't know or care whether it's booked to stop or not).

Regulating decisions have to be made on the basis of minimising overall delay in the bigger picture, not on the basis of individual assessments of risk like this that should already be accounted for in the signalling design (or special instructions). After all, using an extreme example at a different location, holding an Up train for the 20mph diverging route at a junction to allow a 90mph Down train to run over the straight route first then exposes the Up train to the risk of the following 90mph train being unable to stop before colliding with it in the rear!

Minor point but the brake system fitted to the Class 158's/159's was not the Westcode system supplied by Westinghouse Brakes but was a different system albeit it similar supplied by Davis and Metcalf. The energisation and de-energisation of the trains wires is in a grey code sequence.

Regarding the air supply then you are correct regarding the MR pressure switches will initiate an emergency brake application but as the MR pressure drops the compressors will attempt to replenish the MR system. Whilst I couldn't say for certain on DMU's of this generation the compressors are mounted on the engines and driven by the engine and hence compressor output can vary depending on engine speed. The compressors were also lubricated using the engine oil system which caused lots of problems with oil carryover into the air system.

Modern DMU's now tend to have a separate electric motor driven compressor which gives constant air delivery as it is independent of Diesel engine speed speed with the big advantage of not relying on the engine oil to lubricate the compressor.