1S69 Failure Abington 9th September 2024

[68000]

Theres an X thread from someone that was working the incident here: https://x.com/stephen37292ML/status/1833381809482010997

Also an interesting person to follow IMHO

Definitely

 

[Deepgreen]

The 390 has 3 main air compressors.
There is also an emergency coupler in each end should the rescue come from other than a 57. However again the working instructions mean an on call manager has to attend.

So it appears that the couplers can't readily be used because access cannot be gained owing to a failure of the mechanism. Requiring an on-call manager to attend is all very well unless the failure happens in a remote location and the time taken to get him or her there is excessive. Having a well-maintained, working coupling mechanism which can be used by the driver must surely be sensible? It cannot surely be satisfactory for a train to be stranded because the next train ahead or behind cannot attach to it (and, in the latter's case, is blocking access by a rescue loco anyway). The railway has been gradually painting itself into a corner with this issue - the real-life ability for trains to be rescued by others seems to be diminishing steadily.

 

[Peter Mugridge]

Everyone was questioning at the time the wisdom of removing most of the Thunderbird coverage as being a false economy.

I think their point has been well and truly made now...?

 

[DaveyJones]

So it appears that the couplers can't readily be used because access cannot be gained owing to a failure of the mechanism. Requiring an on-call manager to attend is all very well unless the failure happens in a remote location and the time taken to get him or her there is excessive. Having a well-maintained, working coupling mechanism which can be used by the driver must surely be sensible? It cannot surely be satisfactory for a train to be stranded because the next train ahead or behind cannot attach to it (and, in the latter's case, is blocking access by a rescue loco anyway). The railway has been gradually painting itself into a corner with this issue - the real-life ability for trains to be rescued by others seems to be diminishing steadily.

The Pendolino Coupler extension is air driven, if the other comments in the thread are correct, no maintenance regime or driver competence will get a coupler out with no air .

An example of Virgin's vanity hurting the Pendolino fleet still.

 

[LNW-GW Joint]

Everyone was questioning at the time the wisdom off removing most of the Thunderbird coverage as being a false economy.
I think their point has been well and truly made now...?

And yet last week for the failure outside Euston there was a 57 on standby, but no driver (or would have been out of hours, travelling from Crewe was it, or Carlisle).
Either way, it seems this incident was entirely down to the operator (with Alstom maybe implicated in the train failure).
Network Rail plans to install ETCS on the North WCML, giving bi-di functionality, but that must be a decade away.

 

[Bikeman78]

The Pendolino Coupler extension is air driven, if the other comments in the thread are correct, no maintenance regime or driver competence will get a coupler out with no air .

An example of Virgin's vanity hurting the Pendolino fleet still.

Given that loss of air is a common reason for needing assistance, this sounds like an impressive design fail. I've lost count how many sprinters with an air leak or faulty compressor have been rescued by simply attaching another unit.

 

[Dr Hoo]

The Pendolino Coupler extension is air driven, if the other comments in the thread are correct, no maintenance regime or driver competence will get a coupler out with no air .

An example of Virgin's vanity hurting the Pendolino fleet still.

Was the apparent fact that a fault on one compressor knocks out the other two compressors also part of the original design philosophy? Or was the train running around with the two other compressors already failed so that when the ‘last man standing’ went ping, that was game over?
Should there perhaps be a driver-operated foot pump, purely to extend the coupler in the last resort?

 

[GLC]

Should there perhaps be a driver-operated foot pump, purely to extend the coupler in the last resort?

The fitment of foot operated bellows would require an additional new technology uplift in pay for every driver though

 

[Efini92]

The fitment of foot operated bellows would require an additional new technology uplift in pay for every driver though

I think we’ve found the next aslef GS

 

[tpjm]

No time Limit to rectify a fault? We are given 10 minute timeline. After 10 minutes, all resources are used for evacuation and preparation for dragging the failed unit away. The rescue train is also called out after these 10 minutes.

Resources are mobilised within 10 mins, but given the remote location, take time to get to site. The call to DRS to stand up the Class 57 Thunderbird is one of the first things AWC Control do when a unit is stood.

Everyone was questioning at the time the wisdom off removing most of the Thunderbird coverage as being a false economy.

I think their point has been well and truly made now...?

You'd think so, but I'm told DfT are pushing for less Thunderbird coverage! DRS have been unable to resource the Rugby TB for some time now, but the 57s are on borrowed time!

Was the apparent fact that a fault on one compressor knocks out the other two compressors also part of the original design philosophy? Or was the train running around with the two other compressors already failed so that when the ‘last man standing’ went ping, that was game over?
Should there perhaps be a driver-operated foot pump, purely to extend the coupler in the last resort?

You can get a coupler out without the Pendo having a working compressor.

 

[Efini92]

You can get a coupler out without the Pendo having a working compressor.

Providing the pendo still has sufficient air.

 

[bahnause]

Providing the pendo still has sufficient air.

Every single traction I sign has some kind of mechanical backup to open the front and extend the couplers. It is usually a very awkward and slow process which is probably fun to watch, but not to execute.

 

[Efini92]

Every single traction I sign has some kind of mechanical backup to open the front and extend the couplers. It is usually a very awkward and slow process which is probably fun to watch, but not to execute.

390’s have a staubli connection so the assisting loco can supply the air to the coupler.

 

[deltic]

Are additional staff needed for a train to train transfer? And if so why given the number of staff already on board?

 

[tpjm]

Providing the pendo still has sufficient air.

I was more thinking about the assisting loco

Are additional staff needed for a train to train transfer? And if so why given the number of staff already on board?

Yes. Not all staff on board will be PTS trained, and train to train is rarely as straight forward as throwing a ramp out between two units. Consider customers climbing down a ladder, walking across ballast and back up into another unit.

 

[QueensCurve]

The "decision" probably didn't take that long but getting all the resources to manage a controlled evacuation in time (even if train-to-train) undoubtedly took longer.

This is why bidirectional signaling, while seen as an unaffordable lucure, is actually essential. As soon as the decition was taken things could have been moving straight away.

Avanti 1S69 failure ongoing at Abington this evening. Queue of trains trapped in the rear. These 390s seem to be dropping like flies these days

Which is disappointing. Considering the sets have distributed power, one might expect there to be enough redundancy and diversity for them to keep moving. Is there any review into the failures with a view to preventing them?

Indeed, and if I had a pound for every incident in which full WCML bi-di signalling would have made a massive difference to the delays incurred I would be richer than Elon Musk.....

Me too.

I remember a breakdown at exactly this point in the 1970s, on the north side of the climb to Beattock.

The long, steep, climb makes this a black spot for failures. Even if only the bank was bidirectionally signalled it would improve things significanly.

 

[muz379]

Is there any review into the failures with a view to preventing them?

No doubt any incident which causes significant delay to be attributed to set failure will be investigated and any measures that can be put in place to prevent reoccurance will be . But that kind of stuff generally wont make it to the public domain .

 

[Efini92]

No doubt any incident which causes significant delay to be attributed to set failure will be investigated and any measures that can be put in place to prevent reoccurance will be . But that kind of stuff generally wont make it to the public domain .

I believe it was a power supply issue to the compressors that caused the failure, not the compressors themselves.

 

[AndrewE]

I believe it was a power supply issue to the compressors that caused the failure, not the compressors themselves.

but isn't that something else which would be "distributed?" i.e. each compressor would have a software-controlled switch off a main power bus. So more likely a software faut in the equipment sending a "compressors on" message than a loss of power to an actual power line running to the compressors the length of the train.

 

[Efini92]

but isn't that something else which would be "distributed?" i.e. each compressor would have a software-controlled switch off a main power bus. So more likely a software faut in the equipment sending a "compressors on" message than a loss of power to an actual power line running to the compressors the length of the train.

I’m not sure, we were talking about it this morning and it was mentioned there was an issue with batteries not charging.

 

[Falcon1200]

No doubt any incident which causes significant delay to be attributed to set failure will be investigated and any measures that can be put in place to prevent reoccurance will be .

I attended many such reviews, into all sorts of massively disruptive events on the WCML in Scotland, and every time made the point regarding bi-directional signalling. Difference made it none.

 

[AndrewE]

I’m not sure, we were talking about it this morning and it was mentioned there was an issue with batteries not charging.

so how does that fit with post #68?

 

[SansHache]

The Class 390 trains are very effective at self-recovery, hence the reason why Virgin got rid of most of their Thunderbird locos more than 10 years ago. However, there are vulnerabilities, mostly related to the air system and the 25kV distribution circuits.
On most occasions when there is a major failure such as this, it is a combination of events that coincide to prevent the train being able to continue to limp along.
All systems have redundancy built-in, but if a train is already in a degraded condition it may not be effective.
If a train has a major air leak then there is not a lot you can do to keep it on the move. The trains were originally designed to be able to couple to rescue another failed unit, but you are then left with the problem of what do you do with a 500m length train with more than 1000 passengers on board.
The important lesson is to learn from such incidents and to ensure that the possibility of them repeating is minimised.

 

[Efini92]

so how does that fit with post #68?

I don’t think foot operated bellows would’ve helped

I think that’s what the power supply issue was, there was no power from the batteries. Will wait and see what comes out.

 

[Toby268]

The Class 390 trains are very effective at self-recovery, hence the reason why Virgin got rid of most of their Thunderbird locos more than 10 years ago. However, there are vulnerabilities, mostly related to the air system and the 25kV distribution circuits.
On most occasions when there is a major failure such as this, it is a combination of events that coincide to prevent the train being able to continue to limp along.
All systems have redundancy built-in, but if a train is already in a degraded condition it may not be effective.
If a train has a major air leak then there is not a lot you can do to keep it on the move. The trains were originally designed to be able to couple to rescue another failed unit, but you are then left with the problem of what do you do with a 500m length train with more than 1000 passengers on board.
The important lesson is to learn from such incidents and to ensure that the possibility of them repeating is minimised.

Speaking of things repeating themselves. Another 390 failed outside of Euston this evening with a loss of line light again. Thankfully it managed to set back. Service cancelled and passengers on buses from Preston.

 

[AndrewE]

I don’t think foot operated bellows would’ve helped

I think that’s what the power supply issue was, there was no power from the batteries. Will wait and see what comes out.

sorry, typo for #78. I'm surprised that an EMU with live wires overhead is stopped in its tracks by a battery issue which stops the compressors working. At least a foot pump would have got the pan up and you would think that then the mains would be able to power the compressors.

 

[Efini92]

sorry, typo for #78. I'm surprised that an EMU with live wires overhead is stopped in its tracks by a battery issue which stops the compressors working. At least a foot pump would have got the pan up and you would think that then the mains would be able to power the compressors.

I did think you were joking at first
The compressors aren’t fed directly from the 25kv. I’m not sure if they are fed direct from the auxiliary converters or from batteries. I will dig out the manual and have a look.

 

[SansHache]

Speaking of things repeating themselves. Another 390 failed outside of Euston this evening with a loss of line light again. Thankfully it managed to set back. Service cancelled and passengers on buses from Preston.

Hopefully not a repeat of the incident that affected 390117 at Camden on Sunday 31st August. The train had just departed Euston when the OLE tripped and it was stranded in the dead section. The driver of another train reported a flash from the OLE near Primrose Hill tunnel and a "passer by" reported the wires were flapping in the wind. Four and a half hours later the OLE was re-energised as no evidence of any damage had been found. The train set off under its own power five hours after it came to a stand and the line reopened at line speed. Possibly a very expensive incinerated pigeon?

 

[Horizon22]

Hopefully not a repeat of the incident that affected 390117 at Camden on Sunday 31st August. The train had just departed Euston when the OLE tripped and it was stranded in the dead section. The driver of another train reported a flash from the OLE near Primrose Hill tunnel and a "passer by" reported the wires were flapping in the wind. Four and a half hours later the OLE was re-energised as no evidence of any damage had been found. The train set off under its own power five hours after it came to a stand and the line reopened at line speed. Possibly a very expensive incinerated pigeon?

I doubt nothing specific happened considering it was 4.5 hours later.

 

[SansHache]

I did think you were joking at first
The compressors aren’t fed directly from the 25kv. I’m not sure if they are fed direct from the auxiliary converters or from batteries. I will dig out the manual and have a look.

The three compressors are powered by 400V 3-phase supplies from three different Auxiliary Converters.
There is also an Auxiliary Compressor operated from the 110V battery supply to raise the local pantograph and close the VCB when initially powering up the train. However, at least one of the main compressors has to operate to charge the train air supply to release the brakes