Delay attribution when power problems occur

Warrior2852 · 9 Aug 2019

Who will be responsible for bearing the cost of the delay claims for this (EDIT: Power cut 9/8/19 causing multiple issues)? It's not the fault of the TOCs, but it isn't really Network Rail's fault either as it is part of a wider outage. Or will it just fall under a category where compensation is not available?

Context added following move from this thread: https://www.railforums.co.uk/threads/ecml-mml-major-power-problems-09-08.188171/

RailUK Forums

Boysteve · 9 Aug 2019

Warrior2852 said:
Who will be responsible for bearing the cost of the delay claims for this? It's not the fault of the TOCs, but it isn't really Network Rail's fault either as it is part of a wider outage. Or will it just fall under a category where compensation is not available?

Wow talk about compo culture!
I suppose passengers claim off the TOC they purchased their tickets from. There will be a set procedure for TOCs to claim the money from Network Rail as they would for any infrastructure problem. I guess Network Rail will have a claim with the Electricity people or will be insured for such a risk.

yorkie · 9 Aug 2019

Warrior2852 said:
Who will be responsible for bearing the cost of the delay claims for this?

The delay attribution will go down to Network Rail.

I have no idea if Network Rail can recoup any costs from anyone else, but I suspect not.

Warrior2852 said:
It's not the fault of the TOCs, but it isn't really Network Rail's fault either as it is part of a wider outage.

It will go down to NR not the TOCs.

Warrior2852 said:
Or will it just fall under a category where compensation is not available?

The train companies will be compensated by Network Rail. Passengers who experience a delay covered by a Delay Repay scheme will be entitled to claim from the TOCs; all claims from passengers are through the relevant TOC and not any third party, irrespective of where the delay is attributed.

Boysteve said:
Wow talk about compo culture!

There is no harm in asking a question! We have plenty of experts here who can answer (along with some people who guess

)

Boysteve said:
I suppose passengers claim of the TOC they purchased their tickets from.

Delay claims for journeys that were made always go to the TOC whose train was delayed which caused the delay. This is irrespective of the root cause of the delay (the TOC concerned will be compensated where appropriate, to a far greater value than any delay compensation claims would ever be!)

The only reason a customer would contact the retailer is if they chose not to travel and this would be a refund request, not a delay compensation claim.

matt_world2004 · 9 Aug 2019

There will likely be a service agreement between nr and national grid due to the nessescity in keeping the railway powered a breach of this service agreement will likely result in a financial reinbursement to NR for breachrs caused by issues outside the railway network. Incases where the breaches are as a result of crime(Eg terrorism or hacking) National Grid may not have to pay anything

ComUtoR · 9 Aug 2019

As more and more compensation is paid out, how much will this affect ticket prices ?

Brissle Girl · 10 Aug 2019

Given a vast proportion of the delays were down to Class 700 series trains not being able to be restarted, thus blocking lines to all traffic, (whereas older electric stock didn’t seem to have the same issues), it’s a debatable point as to whether all the delay minutes should be attributable to NR.

Warrior2852 · 10 Aug 2019

Brissle Girl said:
Given a vast proportion of the delays were down to Class 700 series trains not being able to be restarted, thus blocking lines to all traffic, (whereas older electric stock didn’t seem to have the same issues), it’s a debatable point as to whether all the delay minutes should be attributable to NR.

If I remember correctly the design of the 700s was largely specified by the Department for Transport (of which NR is an arm's length public body), so that may take blame away from GTR (assuming that is where you are suggesting some cost should go to)

Brissle Girl · 10 Aug 2019

I’m certainly suggesting that the TOC would pick up the costs, and who ultimately paid would depend on any contractual agreement with the train supplier (manufacturer or ROSCO). I’m sure the spec didn’t specifically state that it was ok for a unit to sit down after a power interruption, and so I could imagine that if the various parties wanted it there could be quite a debate as to whether the units were not fit for purpose under such a situation. I’d be inclined to think they’re not given other units seem to have managed satisfactorily.

Bald Rick · 10 Aug 2019

This is an interesting one.

Normally any power failure would be NR attributed. And it’s very unlikely there is any means of financial redress between NR and the National Grid - NRs supply contract is with a power supplier (EDF I think) and not NG.

For example, aside from the issues with the 700 units, there were lots of signalling assets which briefly went off line yesterday across the network. Delays due to this will be attributed to NR

However in the case of the 700s / 717s, it gets interesting. The units suffered loss of power, but the power on the OLE didn’t go off. As has been widely reported, the frequency on the Grid dropped below 49Hz. If the failures were caused by the 700s detecting this frequency drop and thus triggering the shut downs, the attribution for the delay will depend entirely on what is in the electrical supply contract between NR and the TOCs, ie if there are fequency parameters.

There is however another theory. If the Grid failure triggered the failure of the ‘shore’ based computers that control the ‘back office’ functions of the fleet, and this somehow triggered error messages to be sent to the fleet, then that is firmly in the TOCs responsibility. The TOC will have arrangements with Cross London Trains about it, one hopes.

Class 170101 · 10 Aug 2019

I would say there will be many arguments within the Delay attribution 'industry' about this one. I can't see NR wanting to take the blame lying down for the Class 700s/717s sitting down because the on board computer said no many minutes after power was restored.

The TOCs may argue about poor regulation during recovery from the incident as well. And with trains in the wrong place as well the delays carried on this morning I can't see NR being too happy to accept those.

Bald Rick · 10 Aug 2019

Class 170101 said:
I would say there will be many arguments within the Delay attribution 'industry' about this one. I can't see NR wanting to take the blame lying down for the Class 700s/717s sitting down because the on board computer said no many minutes after power was restored.

The TOCs may argue about poor regulation during recovery from the incident as well. And with trains in the wrong place as well the delays carried on this morning I can't see NR being too happy to accept those.

I don’t think there will be too many arguments in Track Access Contract terms, it will be completely clear. There may well be arguments between GTR and Cross London Trains (the owner / provider of the Class 700s) and Siemens (manufacturer), but that is outside the Track access contract.

There won’t be any regulation arguments. Nothing was moving, so no regulation decisions to take!

NR will take a hit of course, because through the delay attribution process NR will pay the ‘victim’ TOCs (principally LNER and EMT) for the delays caused to them. And peak services will be at the maximum rate.

hwl · 10 Aug 2019

Class 170101 said:
I would say there will be many arguments within the Delay attribution 'industry' about this one. I can't see NR wanting to take the blame lying down for the Class 700s/717s sitting down because the on board computer said no many minutes after power was restored.

The TOCs may argue about poor regulation during recovery from the incident as well. And with trains in the wrong place as well the delays carried on this morning I can't see NR being too happy to accept those.

Agreed NR will take the hit for the first X minutes on everything then the rest will be argued about with everyone finger pointing at GTR for that...
The real fun will be GTR having big discussions with Siemens et al.

OxtedSignaller · 10 Aug 2019

Remember that GTR is a management contract any revenue or sales they make the vast majority goes to the DfT. They do not have any revenue risk. Therefore majority of the delay minute attributed if any will probably be more a less covered by the DfT if anyone. Siemens are also responsible for maintaining and ensuring trains are fit for service through their managed depots at Three Bridges and Hornsey.

Edders23 · 10 Aug 2019

I would think somewhere along the line either NR or the national grid will have an insurance policy against such an event occurring I would think that LLoyds of london or one of the big insurance companies is probably going to be footing the bill

Chrisgr31 · 10 Aug 2019

Bald Rick said:
However in the case of the 700s / 717s, it gets interesting. The units suffered loss of power, but the power on the OLE didn’t go off. As has been widely reported, the frequency on the Grid dropped below 49Hz. If the failures were caused by the 700s detecting this frequency drop and thus triggering the shut downs, the attribution for the delay will depend entirely on what is in the electrical supply contract between NR and the TOCs, ie if there are fequency parameters.

There is however another theory. If the Grid failure triggered the failure of the ‘shore’ based computers that control the ‘back office’ functions of the fleet, and this somehow triggered error messages to be sent to the fleet, then that is firmly in the TOCs responsibility. The TOC will have arrangements with Cross London Trains about it, one hopes.

It is indeed interesting because presumably there have been previous occasions where there has been a power cut to a Class 700 train, something like the overheads coming down, or an emergency turning off of the overheads. I assume that when that has happened the train hasnt decided not to restart when the power came back on and therefore this instance was a first?

On the question of a "shore" based failure wouldn't that have affected the whole fleet and therefore all of those that were running on third rail would have stopped as well?

The broad implication has to be to me as a layman that something in the dropping of the frequency of the national grid caused an issue to the OLE but not the third rail and that in turn lead to the trains shutting down, and it appears preventing them powering back up. It may be that any "shore" issue only impacted those that were powered down of course.

Elecman · 10 Aug 2019

Chrisgr31 said:
It is indeed interesting because presumably there have been previous occasions where there has been a power cut to a Class 700 train, something like the overheads coming down, or an emergency turning off of the overheads. I assume that when that has happened the train hasnt decided not to restart when the power came back on and therefore this instance was a first?

On the question of a "shore" based failure wouldn't that have affected the whole fleet and therefore all of those that were running on third rail would have stopped as well?

The broad implication has to be to me as a layman that something in the dropping of the frequency of the national grid caused an issue to the OLE but not the third rail and that in turn lead to the trains shutting down, and it appears preventing them powering back up. It may be that any "shore" issue only impacted those that were powered down of course.

As DC doesn’t have a frequency the drop in supply frequency to solid state rectifiers is irrelevent

hwl · 10 Aug 2019

2 million squid attributed to 700/717 sit downs...

Taunton · 10 Aug 2019

Delay Attribution (and then payments) is not an "it's not my fault" getout. It is there to incentivise the railway to ameliorate the causes of delay as much, and in particular as QUICKLY, as they are able, rather than easing back for the rest of the day or whatever with a "well, we had a problem earlier" attitude. The quicker you get everything back up, the less attribution there is.

The particular circumstances are not really part of this. And it is up to Network Rail themselves to deal with their supplier, of electricity, just as any other rail industry supplier. Did they discuss with National Grid not being part of the load shedding when grid problems happen? Did they pay extra for this? All part of managing your suppliers. And did the TOC ensure with their train provider that trains restart easily without fitters having to be sent out?

Some things seem beyond control, trees down or flooding, but the industry knows the numbers and that this maybe impacts 0.1% of operations. What the approach is trying to prevent is this in turn impacting 10 times that number of trains because you have no effective contingency plan or anything else to restore operations pronto. Or don't bother.

Just for those defending that last comment, one reads that although Kings Cross was "closed" yesterday, and many were inconvenienced, a couple of 12-car trains did depart in the evening for Peterborough but only those "in the know", which seemingly was principally train crews, got to know about this and managed to board otherwise near-empty trains, plus a handful of passengers who didn't believe the notices.

broadgage · 10 Aug 2019

Elecman said:
As DC doesn’t have a frequency the drop in supply frequency to solid state rectifiers is irrelevant

In the case of a simple rectifier used for say battery charging or to work a DC motor, this is correct.
However in the case of these trains, the DC from rectifying the AC is then inverted to variable frequency AC for the AC traction motors.
The onboard computers monitor this process and shut down the traction package if the mains frequency is out of tolerance. For fear of generating harmonics from the interaction between the mains input frequency and the generated frequency.
Resetting after a low frequency event is reported to require an engineer.
Seems to me like a poor design if a brief drop in frequency disables the trains for the times reported.

hwl · 10 Aug 2019

broadgage said:
In the case of a simple rectifier used for say battery charging or to work a DC motor, this is correct.
However in the case of these trains, the DC from rectifying the AC is then inverted to variable frequency AC for the AC traction motors.
The onboard computers monitor this process and shut down the traction package if the mains frequency is out of tolerance. For fear of generating harmonics from the interaction between the mains input frequency and the generated frequency.
Resetting after a low frequency event is reported to require an engineer.
Seems to me like a poor design if a brief drop in frequency disables the trains for the times reported.

Elecman is talking about the DC 3rd rail supply not the internal DC link which you are...

AM9 · 11 Aug 2019

I berlieve that this is the first time there has been a major power cut where the frequency dropped well outside the specified range. In addition, it is probably the first time that the frequency was allowed to change at a much higher rate than was previously the maximum, (1Hz/sec rather than 0.125Hz/sec) and it may be this that triggered protection action in the trains that had not been experienced before. Don't forget that the class 700s (and class 717s) have a DC bust that is derived from the OLE ac. That ac would probably be a full-wave rectification on the 50Hz, thereby producing a 100Hz ripple on the DC. When powered from 3rd rail at 750VDC, there would be much less ripple as the Network Rail feed equipment would be fed with 3 phase 50Hz, probably giving a 300Hz ripple of much lower amplitude.
The DC bus feeds the traction inverters that feed the motors with variable frequency 3 phase ac. In order for the large switching transients and their harmonics in that process to not interfere sensitive on board and trackside equipment, it is necessary to modulate the frequency of the transients to spread their power across a small frequency band. With 3rd rail operation where the bus is relatively smooth, this is acheived by deliberately varying the inverter switching speed (giving the so called 'space ship' warble. With ac, it seems that the existing 100Hz ripple is sufficient to disperse the energy in the transients without additional modulation, hence the Electrostar type of continuous tone.
Now if that ripple was say 98Hz, as would be the case of Friday's problems when the frequency dropped to 49Hz, the interaction of the inverter transients and the ripple would be different and may have caused the detection circuits to identify a hazardous condition that might affect other safety systems, (think track circuit components or signal comms.).
The maximum rate of frequency change in the National Grid specification until 4-5 years ago was 0.125Hz/second. It has recently been relaxed to eight times that (1.0Hz/second) and the qualification testing of the trains may have been completed when the earlier level was in force. Only speculation but not impossible.

IanXC · 11 Aug 2019

Bald Rick said:
There won’t be any regulation arguments. Nothing was moving, so no regulation decisions to take!

Assuming everything was in the correct place when it came to a stand! If a 700/717 had been incorrectly regulated already and hence failed in a position it shouldn't have been in....

Ken H · 11 Aug 2019

AM9 said:
I berlieve that this is the first time there has been a major power cut where the frequency dropped well outside the specified range. In addition, it is probably the first time that the frequency was allowed to change at a much higher rate than was previously the maximum, (1Hz/sec rather than 0.125Hz/sec) and it may be this that triggered protection action in the trains that had not been experienced before. Don't forget that the class 700s (and class 717s) have a DC bust that is derived from the OLE ac. That ac would probably be a full-wave rectification on the 50Hz, thereby producing a 100Hz ripple on the DC. When powered from 3rd rail at 750VDC, there would be much less ripple as the Network Rail feed equipment would be fed with 3 phase 50Hz, probably giving a 300Hz ripple of much lower amplitude.
The DC bus feeds the traction inverters that feed the motors with variable frequency 3 phase ac. In order for the large switching transients and their harmonics in that process to not interfere sensitive on board and trackside equipment, it is necessary to modulate the frequency of the transients to spread their power across a small frequency band. With 3rd rail operation where the bus is relatively smooth, this is acheived by deliberately varying the inverter switching speed (giving the so called 'space ship' warble. With ac, it seems that the existing 100Hz ripple is sufficient to disperse the energy in the transients without additional modulation, hence the Electrostar type of continuous tone.
Now if that ripple was say 98Hz, as would be the case of Friday's problems when the frequency dropped to 49Hz, the interaction of the inverter transients and the ripple would be different and may have caused the detection circuits to identify a hazardous condition that might affect other safety systems, (think track circuit components or signal comms.).
The maximum rate of frequency change in the National Grid specification until 4-5 years ago was 0.125Hz/second. It has recently been relaxed to eight times that (1.0Hz/second) and the qualification testing of the trains may have been completed when the earlier level was in force. Only speculation but not impossible.

Thats all very nice. But the trains should be capable of a hard reboot -i.e everything (yes, everything) shuts down, waits 20 sec, then reboots. It should start up from that. What we need to know is why they didnt.

Ken H · 11 Aug 2019

Chrisgr31 said:
It is indeed interesting because presumably there have been previous occasions where there has been a power cut to a Class 700 train, something like the overheads coming down, or an emergency turning off of the overheads. I assume that when that has happened the train hasnt decided not to restart when the power came back on and therefore this instance was a first?

On the question of a "shore" based failure wouldn't that have affected the whole fleet and therefore all of those that were running on third rail would have stopped as well?

The broad implication has to be to me as a layman that something in the dropping of the frequency of the national grid caused an issue to the OLE but not the third rail and that in turn lead to the trains shutting down, and it appears preventing them powering back up. It may be that any "shore" issue only impacted those that were powered down of course.

was it a power cut that disabled these trains, or was it the unstable frequency?

Elecman · 11 Aug 2019

Ken H said:
was it a power cut that disabled these trains, or was it the unstable frequency?

Given the train suffers a power cut every time it goes through a Neutral section that is pretty unlikely to be the cause, rapid dropping of the system frequency is a far more likely cause

matacaster · 11 Aug 2019

A little off topic (sorry mods), but just a quick question.

Some have mentioned a frequency related 'warble' noise, is this the same as the whooping noise AC only 323's make when accelerating?

petergb · 13 Aug 2019

Post #123 of the ECML/MML power problems thread suggests Class 700 has had previous issues with Neutral sections, though:

ComUtoR said:
700s have been having changeover issues at City/Farringdon and through Neutral sections for a long time and have already caused incidents and evacuations. Nothing has so far triggered an RAIB investigation.

https://www.railforums.co.uk/threads/ecml-mml-major-power-problems-09-08.188171/page-5#post-4146248

(apologies for formatting, I'm still trying to find my way round the editor!)

Delay attribution when power problems occur

Member

RailUK Forums

Member

Forum Staff

Established Member

Established Member

Established Member

Member

Established Member

Veteran Member

Established Member

Veteran Member

Established Member

Member

Member

Established Member

Established Member

Established Member

Established Member

Member

Established Member

Veteran Member

Emeritus Moderator

On Moderation

On Moderation

Established Member

On Moderation

Member