Great Western Coffee Shop

Passengers warned to avoid London Victoria due to severe disruption caused by failure of power supply to signalling.
Thameslink and southern severely affected and Gatwick express suspended.
https://www.bbc.co.uk/news/uk-england-london-44721415 (https://www.bbc.co.uk/news/uk-england-london-44721415)

Someone needs to tell network rail about backup power supplies.

Although in the article it does state that their was an issue with 3 seperate power supplies, so I assume (never assume :))at least one of those was a backup.

BBC News described it as a "signalling error" - really? - and then said no SN/TL/GX trains were running to Victoria. In fact GTR seem to have and be running a planned backup SN/GX service, having TL via East Croydon-London Bridge running. They also have the line via West Norwood back in use, and are now running some Victoria services that way.

However, with the main fast and slow lines through Streatham Common U/S, all services have been cut to half or less so as to fit in what's left. Thameslink seems to be running as normal (whatever that means), so the pictures purporting to be Brighton with no trains can't have been.

I would be very surprised indeed if three genuinely separate power supplies failed at the same time.
Failure of three supplies suggests to me that they were not truly separate but had some single point of failure.
Examples of single points of failure include supplies derived from the same substation, or different supplies but with the cables taking the same route. A single fire or accident with a digging machine then cuts all three cables.

An example of three GENUINELY separated supplies would include,
1) From the local DNO, preferably by a dedicated cable from the DNO substation.
2) An HV supply via private cable along the railway line from an electrically distant part of the DNO network.
3) A converter from the traction supply.

If the signalling centre is not on or near railway property, that precludes options 2 and 3 above, an alternative arrangement could be
1)From the DNO, by a dedicated cable from the nearest substation.
2)A second DNO supply obtained from another substation that is on a DIFFERENT 11KV circuit.
3) A generator.

In all cases great care should be taken to eliminate common points of failure, for example by NEVER putting cables for main and standby supplies in the same trench or cable trough.

Ok I will have to be a little careful what I say as the incident is subject to a formal inquiry which will report back to the NR Board and the Minister.


There are typically in the third rail area the signalling power supply is derived from the traction substations; the 33kV supply system is run as an open ring, a fault on 33kV cable or a DNO outage the ECR can restore.  Substations have 2 33/0.415kV transformers to supply the substation domestic load and signalling power.


In the London area it is usual to have third 415V supply from the DNO street network.

All 3 supplied are controlled by a change over panel which will automatically change form one traction supply transformer to the other in the even of niether of these being available it will switch to the DNO; the ECR can control the change over panel as well, the traction sypply can be parralled so its posibel to change from traction transformer to the other without loss of supply.

The change over panel supplies normally 2 400/650V signal power supply transformers which again supply a change over panel.


Ther Streatham area has recently been resignaled with an auto reconfiguration system along the tack.


Apparantly the problem on this ocassion was one of the change over panels suffering a catostrphic failure, the maintainence teams work soild for 12 hours to hard wire around the destroyed equipment also checking to ensure there was nothing down stream that initated the fault and nothing else was damaged.


The signalling power supplies in the third rail area are generally robust.


As said this incident is subject to a formal inquiry

http://www.dailymail.co.uk/news/article-5920411/Travel-chaos-biblical-scale-commuters-told-AVOID-London.html

The heading to the page has this:

(https://tpc.googlesyndication.com/daca_images/simgad/17578639655038157798)


However below this is the real report:

SECOND day of travel hell: Commuters and Wimbledon visitors are told to avoid Victoria station tomorrow amid signal problems affecting Thameslink, Southern, and Gatwick Express trains

* Hundreds of thousands of passengers are urged not to travel to Britain's second busiest station today
* Frustrated commuters share images of travel disruption and packed carriages as they are forced onto buses
* Dozens of Gatwick Express, Thameslink and Southern journeys to and from London Victoria are cancelled
* Southern has warned beleaguered passengers that 'disruption will continue until at least 7am tomorrow'

There is a temporary fix in place in the power system, I almost certain NR maintainers will be out on standby at key locations through the morning rush hour(s)

…………..and as is always the case, customers are told "don't travel", as the TOCs trouser millions in compensation from Network Rail (aka the taxpayer) ching ching!  >:(

From GTR journeycheck

Chris I know more about this incident than I can say here, my knowledge is based on actual evidence and not press releases

Will you be able to tell us more, eventually ? Or will it remain confidential effectively forever ?

One of your earlier posts suggests that some form of enquiry is to take place, will the results be published or will it be for internal use only.

And of course updates from anyone else would be most interesting.

There is a posiblity some of it may be commerially sensitive, the area was only resignalled in the last 2 years.

The enquiry can take several weeks to complete, the items that cause the fire my need to go to a lab for analysis.

It is highly likely the Secetary of State will want a briefing before it gets released

Has this fault, or a very similar one recurred ?
No effective service from London Victoria again tonight, due to a power failure.

No it was not related to the incident in the OP.  The original incident was due to a loose connection in a power supply change over switchboard, whilst duplication and even triplication of supplies is done there is always one weak link in the chain even these are made as robust as possible.  The incident in the OP was the first such incident for over 30 years in such equipment in the DC electrified area, of which there are several hundred of these types of switch board.

The final report and actions are still commercially sensitive due to the recent re-signalling project.



Yesterdays incident was due to a fault on National Grid 275/400kV network which reflected back onto the railways high Voltage network which caused some of the signalling interlocking to fail safe, due to the time of the event the technicians struggled with heavy road traffic to get to all the locations, there is only a finite number of tech available.   

Thanks for the detailed reply.
I do however find it very surprising that a high voltage grid outage was behind this failure.

High voltage National Grid circuits are usually duplicated and loss of supply therefore very rare from this cause. I  recall a failure some years ago when two almost simultaneous faults blacked out a large part of greater London and North Kent.

However in this case I don't recall any large scale outages to ordinary consumers. It is of course possible that the railway 33Kv system is reliant on a single high voltage grid connection which would be more vulnerable.
In such circumstances I would expect that the low voltage DNO supplies would still be available.

What you may (or may not) notice occasionally is the lights in your house or work place dip, just for a fraction of a second; this can happens when there is a fault on the transmission network or 132kV  33 kV distribution network.


For most people it does not effect them, it may mean your PC reboots, clock on a cook reset etc for some of the traction power systems with signalling supplies derived from the these are designed to fail safe even for a 200mS dip; most times systems self reset within 30 seconds or so, although a train driver may see a black signal or one which has changed aspect to danger, they will stop their train.

All of this adds up from what was 200mS to many minutes of delay.  What happened last week was signal power reset within 2 seconds and most signalling systems reset however there were a few places where the signalling failed to reset and techs had to visit site.

Thanks for the update.
It does however seem to be a poor design that is so vulnerable to events that went largely unnoticed outside the railway.

The fact is the signal power was designed in the area effected last week in the 1980's for that era signalling and interlocking; over the years Signalling have enhanced / upgraded equipment but the power supply at times is seen as just that its a power supply, the view being "its been reliable over the years its never caused problems, the funds are better spent elsewhere where there is a problem"


UPS are being installed, they are expensive to install and to maintain and the former Southern Region has a lot of signal power supply points so it takes time, often waiting for the next major resignalling project or perhaps re-lock if its a big enough project