Scotrail 334 becomes uncoupled near Uphall (26/11)

[hexagon789]

So I posit: is it possible to operate the uncouple button if a cab is not keyed in?

I would imagine the cab would be entirely 'dead' if not keyed in. Certainly on other units from what I understand only the emergency brake setting on the brake handle or where present 'plunger' work in dead cabs and I think that's only because they 'cut' the brake circuit rather than electrically sending out a brake demand.

 

[4-SUB 4732]

I imagine the train probably wasn't attached properly in the first place. If it was, the emergency brakes would have been applied.
It would be unusual for several things otherwise to go wrong - a coupler failed; automatic brake failed (although it appears to have worked in the rear unit)

Couple of things.

My traction knowledge doesn’t cover the Juniper family but I can say with almost absolute certainty that the train would have come off the depot in the morning and that if it wasn’t coupled properly, it would have shown up a lot earlier than evening peak.

I can also say that whilst something has clearly gone awry to allow Wire 13 to remain energised in the leading unit, the rear unit should remain on emergency brake pressure due to no master key being inserted. This is why the train didn’t ‘roll away’.

 

[4-SUB 4732]

There wouldn't have been any risk to any on coming traffic so please stop that now.
It's not uncommon for the power to go out simple vcb trip happens all the time, simple reset and if power comes your good to go. Driver has no reason to know train has spilt if brakes haven't applied

100% agreed with this.

 

[4-SUB 4732]

How would he check

If the driver took it into Edinburgh and was about to take it back out, he/she would see the same 6 they brought in. If they are taking the train on, they’ll check when they get down to the platform what’s there for them. If it’s 3 vice 6 they’ll notice and stop at the appropriate place on platforms; same for a 6 vice 3. The driver, however, will realistically believe that no matter what is happening if the unit they are driving from is moving and doesn’t come up in a heap, it should be almost impossible for the rear unit to have divided as this should basically leave the two ends of Wire 13 that are ‘closed’ when uncouple is pressed at that end in an ‘open’ position. By this I mean when the couple button is pressed, they will expect to meet up to the other two ends in the other unit. This whole situation is deeply confusing and I’m sure even the most seasoned of traction inspectors or rolling stock engineers would wonder how a modern unit with electric couplers has managed to come apart without both parts stopping dead.

 

[4-SUB 4732]

I'm not sure where to begin with this comment. It makes little to no sense.



There are numerous occasions where I NEVER check my train length when taking over a train.



I have way more questions than any answers. Nothing about it makes any sense.

I’m sorry to say I’m not sure what driver would take over a train and not check length. If what you’re saying is you don’t physically walk back to see that the single 150 that has arrived is formed of 2 coaches then perhaps you wouldn’t ‘check’ as it’s a 150 and it only has one formation on its own (example). That said, any train I took over was always checked by eye even if a cursory glance to make sure it matched the schedule card. If it didn’t, the schedule card with it’s meagre formation info was rapidly scribbled across and then a big fat mental note made...

 

[4-SUB 4732]

Umm not always, a few years ago a pair of Class 317s split upon departing from Hitchin which is why they as in FCC and GTR took the decision not to leave the interconnecting doors between units open.

Anyway, we shouldn't guess at the reasons why this incident happened as that is the job for professionals such as the RAIB not armchair experts!



Class 313s on the Great Northern have a sliding indicator which has 3, 6, 9 cars which show any many carriages are in the consist.



What a load of rubbish!

A driver who is being asked to examine the line does so at a much lower speed, 10mph though tunnels and 20 mph over track defects so seeing as it well might have been a track circuit failure, the signaller will advise the driver of the examining train to proceed at a speed no faster then 20mph in the affected area NOT at linespeed.

So no danger at all!

Those 313 sliders (same as the 317s, 321s, 365s) are only as good as the people who get in the cab and set them correctly before departure. Many a person has left them in ‘3’ with a ‘6’ (etc) and then wondered how they managed that stop short due to over-reliance.

And let’s be honest, the 365s had a relatively good rate of self-division as well; but it was always said this was why all the ‘tell tales’ on the couplers got painted yellow so drivers could be certain everything had dropped into position correctly. Very rarely do you now hear of anything like this on the GN (although now there’s a load of plastic junk like 387s and 700s about with Dellner couplers that are extremely easy to determine if they are or are not coupled!).

 

[Signal Head]

Strange how the signalling let the front end of the train continue as far as the next stop since the second half of the train was still occupying the platform section surely there would be something in place to recognise that a train has just left a section but that section is still occupied?

I'm curious as to how a train would leave a block section if, to clear the signal in front of said train requires that the whole train has already left the section!

 

[4-SUB 4732]

I think you ought to stop now.



I suspect the former is the most likely cause.

I severely hope someone didn’t. And if they did, they would need to have been in a cab and would have seen what they’d done.

More importantly, I don’t think pressing uncouple works unless you have keyed on. If you do that, the drivers key on also would have made an emergency brake. That’s why this makes this theory very, very unlikely.

 

[LAX54]

Strange how the signalling let the front end of the train continue as far as the next stop since the second half of the train was still occupying the platform section surely there would be something in place to recognise that a train has just left a section but that section is still occupied? Or am I missing something there? And since the platform was still occupied by the rear end of the train that there would be no accident as the signals leading into the platform would be red protecting the platform? And worse comes to worse the next approaching train would be travelling at a speed in-which the driver could stop before he hits the rear end of the train left behind...
First a derailment not long ago and now an accidental uncouple oh dear ScotRail this isnt looking very good...

No perfectly normal, if the split was undetected by the front unit it would proceed normally, the bit left behind would obviously occupy a track circuit, the Signaller would see this, and treat the circuit as defective, and use the next train to examine the line, who would report back that in fact there is a train there ! The Signaller would also find out by calling the next box, or get the driver at the next stop to confirm his train is in fact complete, which of course 99.9% of the time it would be, but we always ask !
I recall many years ago, we had a track 'fail' behind an ECS move, the Signaller stopped it at the next station by maintaining the signal at red, the SPT rang, and before the Signaller had time to ask, the driver started shouting "My Train My Train.....a bit' missing " ! turned out a faulty coupler between a 170 and 153 failed (ice I think) and left the 153 in the middle of nowhere ! (we were about to caution the next passenger service, but when we found out what happened, held it at the previous stn)

 

[LAX54]

I'm curious as to how a train would leave a block section if, to clear the signal in front of said train requires that the whole train has already left the section!

That does not make any sense !

 

[eman_resu]

That does not make any sense !

Makes complete sense. The signalling system is not designed to ascertain that a train has left the section with the same number of vehicles as it entered with, only that a train is not detected within the section.

The signalling system detects the absence of a train (or individual vehicle), and if it cannot detect the absence of a train (or individual vehicle) then will protect that section by not permitting the clearance of the signal in the rear, or indeed any other signals with an overlap in that particular section.

Yes - Axle counters do detect the number of axles, however the number is only used to make sure the same number of axles leaves the section as did enter.

What Signal Head is pointing out, is that to ascertain the completeness of the train, the axles or vehicles would need counting out of the section, which is impossible unless the signal is already cleared...

 

[ComUtoR]

I’m sorry to say I’m not sure what driver would take over a train and not check length.

Quite simply put. I am not required to. If I am booked REL then it's the responsibility of the Driver I am relieving to tell me the train length. There are platforms where I wouldn't be able to see the length of the train and where we only have 1 minute to relieve, going back to check a 12-Car just isn't going to happen.

...700s about with Dellner couplers that are extremely easy to determine if they are or are not coupled!).

700s are fixed length units so if one was coupled then something has gone very very wrong.

 

[Mingulay]

Quite simply put. I am not required to. If I am booked REL then it's the responsibility of the Driver I am relieving to tell me the train length. There are platforms where I wouldn't be able to see the length of the train and where we only have 1 minute to relieve, going back to check a 12-Car just isn't going to happen.



700s are fixed length units so if one was coupled then something has gone very very wrong.

It’s fortunate it’s was this class of train and not a 380 or 385 with connecting gangway as you would have 2 open ends to the track . Bit more dramatic for passengers.

 

[whhistle]

There seems to be a lot of silly comments in this thread :/
There's a lot of "if the situation was like this then..." and "if the situation was like that then...". But the situation isn't like that.
It would be better to keep to the facts we have, rather than involve different scenarios. Page 2 and part of 3 are some of the most confusing to read in any thread I've read on here - including my own!

Although I wonder why there was no electricity in the rear part of the train?
I would have thought the pan would be up on the rear unit to provide it with power (thinking 350s here).
If there was no power, I guess no tail light at the rear... unless that works on some sort of battery by itself? Unlikely? But then I've heard some Pendolinos drop their pan (or used to) if someone sneezed, so perhaps a rough pull away might have activated some sensor to lower the pan. Interesting that the destination board id still alight, which suggests some power was on the 2nd unit.

I think they work for the super fast branch of the RAIB because they've already found the root cause and unlike the RAIB have also found who to blame all within a few hours. Must say I'm quite impressed.

Let's not posit wild/silly theories - good to see the ruk investigation branch have swung into action at such short notice!

*like* - to both.

Well, if a driver can proceed at an inappropriate speed when they know a section to be occupied, I'm sure they can when they don't know!

By that logic though, a train driver can drive any speed at any point.
Yes, systems will stop them eventually, but they can still physically do it...
Not really a comparison though as it was a different train/driver/environment/situation.

But you shouldn't be suspecting anything to be the cause until the professionals have done their investigation.
I propose the couple button was never pressed. Or the uncouple button was pressed in error during the attachment.

So you're telling one member not to suspect anything, then suspect something yourself... in the same post?
Sorry but I'm with @DarloRich here - some of your posts in this thread have been... interesting to say the least.



I guess another question is whether the train will be "driven" away or towed (I guess from the other end)?

 

[Quakkerillo]

It’s fortunate it’s was this class of train and not a 380 or 385 with connecting gangway as you would have 2 open ends to the track . Bit more dramatic for passengers.

Although in that case, a passenger from the first half of the train would surely have noticed and pulled the emergency break handle before the train had actually left the station.

 

[Surreytraveller]

Although in that case, a passenger from the first half of the train would surely have noticed and pulled the emergency break handle before the train had actually left the station.

The vast majority of passengers will, assuming they notice, ignore what's going on around them.
Or they'll worry about getting into trouble for activating emergency brakes

 

[Surreytraveller]

Quite simply put. I am not required to. If I am booked REL then it's the responsibility of the Driver I am relieving to tell me the train length. There are platforms where I wouldn't be able to see the length of the train and where we only have 1 minute to relieve, going back to check a 12-Car just isn't going to happen.



700s are fixed length units so if one was coupled then something has gone very very wrong.

If you' re taking another driver's word for something, who gets put on the improvement plan when you stop short and release the doors if there's no recording of the conversion you've had?

 

[ComUtoR]

If you' re taking another driver's word for something, who gets put on the improvement plan when you stop short and release the doors if there's no recording of the conversion you've had?

It's part of the handover process. If the Driver didn't tell me, I'd ask. We have to put a lot of trust and faith in the system and protocols such as these are what makes for a 'safe' railway. Just as you would have to put faith in the 'computer' There are many methods of working and there are overlapping rules and policies in place to prevent stop shorts and plenty of mitigation steps to ensure you are aware of your train length.

If we took the approach of trust no-one and never trust the computer then you would have to physically go back and count each coach when you relieved someone. This is highly impracticable. It would take me about 15 minutes to relieve another Driver.

Both Drivers would be in trouble because there are other procedures in place that need to be adhered to. However, if the relieved Driver gave out an incorrect train length and the relieving Driver stopped short. As long as they followed the correct procedures and the stop short could not have been prevented then only the relieved Driver would be in trouble. Yes this relies on all parties telling the truth.

Not forgetting that this is specific to my TOC and happens on a daily basis. We drive any length between 2 and 12 and flip unit lengths multiple times on a diagram. Yes it can go wrong but that is the whole point of investigation.

 

[TheSel]

I don't think the average person could move a train somehow.

On the contrary. I would suggest that the vast majority of trains are driven by average persons. Possibly experienced, and hopefully well trained ones, admittedly, but average in all meaningful respects.

 

[DanDaDriver]

If you' re taking another driver's word for something, who gets put on the improvement plan when you stop short and release the doors if there's no recording of the conversion you've had?

When you relieve another driver then you’ll be told about any defects, and how long your train is.

And you trust your colleague. That’s just how it works.

Sometimes it’s as much as “8 on, it’s a good’un.”

If you’ve got 2mins to set up your cab, put your number in the OTDR as well as all the other things, you don’t have time to walk down the train to count what’s on. (And you don’t have to.)

 

[westcoaster]

Just a quick looking the web at 334 pictures.
When did they change them from auto couplers to dellner couplers.

Also if you look and compare pics with dellner couplers something bright yellow is missing (guide horn).

Edit it might still be there very hard to tell.

 

[Cliveblackpool]

If the stranded unit did not show on the track circuiting - leaves etc, or no track circuiting on that part of track, could the next train have had a green light, unless they have axle counters?

 

[Aictos]

Let's not posit wild/silly theories - good to see the ruk investigation branch have swung into action at such short notice!

Indeed hence my remark about armchair experts, people posting such wild and silly theories!

No, because he wouldn't be talking to the driver of the "split" train. The only driver he is going to speak to is the driver of the train that is going to examine the line.
And I can assure that the words about "proceed with caution, being prepared to stop at any obstruction" will most certainly be said, and when the tapes are listened to after the event if they aren't there then it is a severe rollicking.

Indeed, couldn't have said it better myself!

 

[JN114]

If the stranded unit did not show on the track circuiting - leaves etc, or no track circuiting on that part of track, could the next train have had a green light, unless they have axle counters?

Yes, in as much as any train can disappear from track circuits anywhere. But this point is now being very strongly overplayed. It’s so incredibly rare that such wrong-side failures happen as it is, never mind coinciding with this very rare coupling issue.

 

[theironroad]

On the contrary. I would suggest that the vast majority of trains are driven by average persons. Possibly experienced, and hopefully well trained ones, admittedly, but average in all meaningful respects.

...and some of us are below average lol

 

[theironroad]

When you relieve another driver then you’ll be told about any defects, and how long your train is.

And you trust your colleague. That’s just how it works.

Sometimes it’s as much as “8 on, it’s a good’un.”

If you’ve got 2mins to set up your cab, put your number in the OTDR as well as all the other things, you don’t have time to walk down the train to count what’s on. (And you don’t have to.)

+1
Probably the one thing mentioned in every handover at my place if all else is good.

 

[DarloRich]

I severely hope someone didn’t. And if they did, they would need to have been in a cab and would have seen what they’d done.

More importantly, I don’t think pressing uncouple works unless you have keyed on. If you do that, the drivers key on also would have made an emergency brake. That’s why this makes this theory very, very unlikely.

apologies - i meant to write the former is more likely than the later! tired brain.

 

[bahnause]

I once lost a unit on the line between two stations. They have reacted as planned (50% braking force for the leading unit, emergency brake from the trailing unit and the according fault message). Had to go on a walk to the second unit to attach it again.

 

[LAX54]

Exactly this. A collision is very rarely the result of one error. Error 1 - train isn't attached properly, train divides. Error 2 - signaller doesn't realise train has divided. Error 3 - next train examining the line travels too fast.
It happens.

That would be 95% Signaller error then ! but ALL highly unlikely

 

[LAX54]

or.....up to now..... an unknown design fault ?