E.R.T.M.S.

[caradoc]

3 hours ago, DY444 said:

I am not aware of any capability for one ROC to take over another in a crisis.  I believe it was discussed at one time but I am not aware it was proceeded with. 

 

I recall that being promoted as an advantage of ROCs but all of us in the operating side knew it was complete nonsense, and could not and would not ever happen; Where would the staff come from, to enable one ROC, at zero notice, to continue its own work and take over another's ? 

 

An issue for me with ROCs is that should one, for whatever reason, go down,  not only is (of course) signalling for its area lost but also the Control staff to manage the resulting disruption would be gone too ! For that reason, and because IMHO communication between Signallers and Controllers should be by recorded means, ie by phone, they do not need to be and should not be in the same place. 

 

[DY444]

24 minutes ago, caradoc said:

 

I recall that being promoted as an advantage of ROCs but all of us in the operating side knew it was complete nonsense, and could not and would not ever happen; Where would the staff come from, to enable one ROC, at zero notice, to continue its own work and take over another's ? 

 

An issue for me with ROCs is that should one, for whatever reason, go down,  not only is (of course) signalling for its area lost but also the Control staff to manage the resulting disruption would be gone too ! For that reason, and because IMHO communication between Signallers and Controllers should be by recorded means, ie by phone, they do not need to be and should not be in the same place. 

 

 

All of that plus a whole host of technical issues to avoid any dependence on equipment housed in the failed ROC. 

Edited May 21, 2022 by DY444

[St. Simon]

On 21/05/2022 at 15:09, caradoc said:

An issue for me with ROCs is that should one, for whatever reason, go down,  not only is (of course) signalling for its area lost but also the Control staff to manage the resulting disruption would be gone too ! For that reason, and because IMHO communication between Signallers and Controllers should be by recorded means, ie by phone, they do not need to be and should not be in the same place. 

Hi,

 

But all that is taken into account during design, everybody seems to think that a random set of people just throw the buildings together without thinking, but it’s not true.

 

Everything goes through a deep assessment of risk (and just safety, but operational, financial, reputational etc etc) during the design process and then mitigations are applied.

 

I don’t know exact details of the mitigations for the scenerio you are thinking of, but I would assume that there are separate power supplies for each floor, separate telephone lines etc. From what I can tell, even if route control and signallers weren’t in the same building, an issue with one of them means that neither can function, so in reality, is it a problem to have them in the same building?

 

Anyway, this is to with ROCs and control rather than ETCS itself.

 

Simon

[uax6]

One of the problems associated with one ROC providing disaster recovery for another was the problem of staffing. Its all very well providing a spare VDU and desk which you can load any config on to, but its a much more difficult thing to train staff to be able to take over any panel in any ROC. Signalling staff have to be proved competent to operate a signalling location, and to do that you need to be exposed to the location fairly frequently (even though the competency time from has been changed from having to work a location at least once every 6 months to once every 12 months) to be able to do it with any level of proficiency.

 

As for diverse routing, I wonder how many of these diverse routes are a different fibre in the same lineside cable... It is more common than you would think....

 

Andy G  

[caradoc]

19 hours ago, St. Simon said:

From what I can tell, even if route control and signallers weren’t in the same building, an issue with one of them means that neither can function, so in reality, is it a problem to have them in the same building?

 

In the ROC I worked in the Signallers and Network Rail Control are in the same building, on the same floor and in the same room, so any catastrophe requiring Signallers to leave the building will equally affect NR Control. However the Signallers there only cover part of the Route whereas Control cover the whole thing where many, many trains would still be running. And the major TOC's Control is co-located, so they would be gone too. 

 

 

 

[phil-b259]

On 21/05/2022 at 10:15, Michael Hodgson said:

 If my understanding is correct there is enough redundancy of hardware and communications systems capability to enable other locations to take over in the event of a disaster at York, as the big banks do with their data centres and ATM networks.

 

NOPE!

 

You forget that even within a ROC with its computerised workstations it is necessary for every signalman to be passed out on a specific workstation (and keep their competence up thereafter by regularly operating it) - As such you cannot suddenly get the signallers at Derby ROC to start running the contents of York.

 

Furthermore signallers workstations are 'hard coded' running dedicated processors and for the most part CANNOT work any other area than the area with which they have been programmed - for this to change you require the wholesale change processor cards, fiddle with the equipment housing address settings etc to and then there is the little matter of getting telecoms to re-organise the telecoms networks so that the workstation is talking to the interlocking.

 

Finally workstations and the associated equipment (including electronic interlockings) are phenomenally expensive - plus tend to change as time goes on. Thus each ROC is basically an empty office floor with signalling workstations progressively added as each individual scheme comes along and pays for their kit to be put inside. They are not - and never have been designed to have any extra redundancy (aside from the usual protection against equipment / communications failures, including diverse routing etc) so if one needs to be evacuated then trains stop till people are allowed back in*

 

* In the Kings Cross incident it was only the nature of relay interlocking design (which distributes the interlocking function over many individual relay rooms rather than a single electronic interlocking installation) that saw remote operating panels able to be provided. Proper re-signalling which eliminates relay based interlockings will see such remote operating panels removed thus eliminating this backup option where it previously existed.

 

In short therefore, a ROC is no different to a mechanical signal box - it cannot control anywhere other than which it is designed to control and cannot be operated by anyone who is not trained to do so.

 

 

[corneliuslundie]

So about the only line still able to run would be the Marches! if if all the signallers were available,

Back in the day, I remember an incident on the Cambridge line when the overhead telegraph lines went down in a storm. The trains were run on time interval, not many but so many still ran. I photographed a passenger train at Whittlesford.

I am afraid that I do not see the point of the amount of centralisation being aimed for by NR.

Jonathan

[phil-b259]

28 minutes ago, corneliuslundie said:

So about the only line still able to run would be the Marches! if if all the signallers were available,

Back in the day, I remember an incident on the Cambridge line when the overhead telegraph lines went down in a storm. The trains were run on time interval, not many but so many still ran. I photographed a passenger train at Whittlesford.

I am afraid that I do not see the point of the amount of centralisation being aimed for by NR.

Jonathan

 

Efficiencies and better able to cope with normal operational problems.

 

Just as one big power station is far more efficient than lots of smaller ones...

 

Its easier to justify backup systems (duplicated communications, uninterrupted power supplies etc) the larger the installation.

 

You also should be able to make savings on mundane things like heating, electricity usage rubbish collection by having a large modern well insulated building than lots of older ones.

 

In staff terms its also helps to train people up so they can cover a number of different workstations - thus having more flexible staffing options if sickness strikes. Yes smaller boxes have a 'relief shift' to cover them but it can take time for them to get to where they need to be if a situation suddenly develops)

 

Also if a big incident develops then you have shift supervisors on hand that can easily step in and help 'double man' the workstation as it were - a bit difficult with individuals working alone or in pairs in smaller boxes.

Edited May 23, 2022 by phil-b259

[St. Simon]

Hi Gents,

 

I know I started off this conversation, but shall we move it to a separate thread?

 

2 hours ago, phil-b259 said:

Proper re-signalling which eliminates relay based interlockings will see such remote operating panels removed thus eliminating this backup option where it previously existed.

To be fair Phil, it is relatively easy to provide a backup workstation or panel on CBI’s, it’s just that they are so rarely used that Operations don’t want to have them as it’s something else they have to maintain for not a lot of gain.

 

Simon

[corneliuslundie]

You take me too seriously!

I agree that there is a lot of sense in centralisation - up to a point. But beyond that the savings of the kind you mention are minimal, and have to be set against increasing complexity. 

And i agree that this subtopic has run its course. 

One thing I am still not clear about, coming back to the topic of the thread, is whether the ECTS system on the Cambrian will have to be updated for the class 197s or whether they can cope with the Cambrian system. If the former that will mean both disruption and that the current 158s will not be available as back-up if the 197s have any problems. And it is not unknown for new stock to have problems.

Jonathan

[PerthBox]

23 hours ago, St. Simon said:

Everything goes through a deep assessment of risk (and just safety, but operational, financial, reputational etc etc) during the design process and then mitigations are applied.

 

I don’t know exact details of the mitigations for the scenerio you are thinking of, but I would assume that there are separate power supplies for each floor, separate telephone lines etc. From what I can tell, even if route control and signallers weren’t in the same building, an issue with one of them means that neither can function, so in reality, is it a problem to have them in the same building?

One of the interesting things at West of Scotland SC at Cowlairs in Glasgow (which is built to the standard NR ROC design) is that there’s no running water if the mains power supply fails and drops onto the emergency generator. An extended power outage in 2019 saw the toilets become completely unhygienic and unusable within a single shift. I presume all the other ROC buildings would be similary afflicted.

[wombatofludham]

Apparently a 197 is coming to Mach tonight (behind a "Tractor") for static testing to see if it talks to the computers.  Presumably that might explain why the Sunday service last week was caped for a system upgrade and test running.

It should be arriving just before midnight, assuming the move happens.

(Edit) It is just arriving at Mach, so it did move...

 

Edited May 23, 2022 by wombatofludham
Update

[DY444]

9 hours ago, PerthBox said:

One of the interesting things at West of Scotland SC at Cowlairs in Glasgow (which is built to the standard NR ROC design) is that there’s no running water if the mains power supply fails and drops onto the emergency generator. An extended power outage in 2019 saw the toilets become completely unhygienic and unusable within a single shift. I presume all the other ROC buildings would be similary afflicted.

 

That sounds to me like total and utter incompetence by whoever designed the buildings' systems.  If you provide a means for the signalling equipment to remain operational during an extended power outage then you have to ensure that the requirements of the staff who operate that equipment are also catered for during that extended outage.  If you don't then what exactly is the point of providing the extended power backup capability in the first place?

[wombatofludham]

Wales and Borders NR Tweet about their recent Cambrian upgrade which has zero details but might be of interest.

Or not.

Link to Network Rail Twitter tweet on Cambrian ERTMS upgrade (not very detailed...)

[phil-b259]

On 23/05/2022 at 21:06, St. Simon said:

Hi Gents,

 

I know I started off this conversation, but shall we move it to a separate thread?

 

To be fair Phil, it is relatively easy to provide a backup workstation or panel on CBI’s, it’s just that they are so rarely used that Operations don’t want to have them as it’s something else they have to maintain for not a lot of gain.

 

Simon

 

True, but as many ROCs also contain the interlockings providing an on site backup panel wouldn't be much help if the ROC was unusable operator wise.

 

However the nature of relay based interlockings* means they are located at intervals along the controlled route and as such backup control panels located at each interlocking would be unaffected by a ROC outage and are potentially more easy to justify.

 

 

* For the unaware (not you Simon I hasten to add), in railway signalling the interlocking is the final line of defence against dangerous situations occurring. You have a 'non safety critical' and a 'safety critical side' with the interlocking standing between them. Communications from control panels are non safety critical - the panel itself basically does nothing in actually processing the signallers requests - it merely provides the user interface 

 

People can hack into the communications from the control panel to the interlocking (which can be sited many miles apart) all they like but it doesn't matter how many times a set of points is requested to be moved for example, if the interlocking says the track circuit over them is occupied the interlocking will continually ignore the request. On the other hand if Mr Rat chews a bunch of cables between the point machine and the interlocking and some of the cores touch then then the points could well move on their own even with a train passing over them as the interlocking is powerless to intervene (though it can of course react to the movement and put signals to red due to the loss of point detection). 

 

Hence traditionally, relay interlockings were provided at places with points and controlled signals (as opposed to long plain line sections which were set up to work themselves automatically with no need of an  interlocking as such), thus keeping the cables out to the individual equipment as short as possible hopefully reducing the chance for Mr Rat to cause a dangerous situation to occur.

 

With Modern Computer based signalling the use of trackside modules, secure encryption and dedicated links means that the chances of someone being able to inject a 'move a set of points command to 'the safety critical' / after the interlocking side is nigh on impossible - and as such it is more efficient to put all the interlockings in the ROCs along with the control panels. Its still just as safe, but arguably does come with an element of reduced redundancy compared to the old relay based designs (in which the provision of local control panels makes more sense).

 

Edited May 25, 2022 by phil-b259

[corneliuslundie]

I can report that rats definitely do chew electricity cables, but apparently according to our local rodent exterminator they can detect 240 volts and avoid those (and certainly the one which chewed our cables did). Not sure if they can identify the neutral though. Ours stuck to the low voltage cables, which at least were easier to repair. So what voltage are the feeds to point motors?

Jonathan

[phil-b259]

26 minutes ago, corneliuslundie said:

I can report that rats definitely do chew electricity cables, but apparently according to our local rodent exterminator they can detect 240 volts and avoid those (and certainly the one which chewed our cables did). Not sure if they can identify the neutral though. Ours stuck to the low voltage cables, which at least were easier to repair. So what voltage are the feeds to point motors?

Jonathan

 

Down my way its 120VDC or 110VAC. Older installations can use 50VDC

 

However it should be noted that as all railway signalling circuits are deigned to be 'earth free' / floating with respect to earth, part of the regular maintenance tasks is to test all power supplies for how much is leaking through to earth - thus hopefully identifying damaged cables before you get to a stage where two bare cores can touch...

Edited May 25, 2022 by phil-b259

[iands]

2 hours ago, phil-b259 said:

 

Down my way its 120VDC or 110VAC. Older installations can use 50VDC

 

However it should be noted that as all railway signalling circuits are deigned to be 'earth free' / floating with respect to earth, part of the regular maintenance tasks is to test all power supplies for how much is leaking through to earth - thus hopefully identifying damaged cables before you get to a stage where two bare cores can touch...

Hi Phil,

 

I also remember years ago that Mr Rat seemed to favour signalling cables over telecom cables because the insulation was different; the signalling cables being more 'rubber' based insulation (a bit on the softer side), whereas the telecoms cables were more polyurethane based (a bit on the harder/tougher side).  

[SamThomas]

On 15/07/2011 at 10:42, Oldddudders said:

And the roads. I do not see either being changed in my lifetime - 30 limits becoming 50 kph, 70 limits being re-signed as 120? You don't need to be much good at risk assessment to see where that might lead. It's stupid enough that UK official fuel economy figures include 56 mph - 90 kph, which is the usual out-of-town single carriageway road limit in Europe, but is not a speed anyone is required to drive at in the UK!

56mph/90k - it is for HGV's.

[jim.snowdon]

6 hours ago, corneliuslundie said:

I can report that rats definitely do chew electricity cables, but apparently according to our local rodent exterminator they can detect 240 volts and avoid those (and certainly the one which chewed our cables did). Not sure if they can identify the neutral though. Ours stuck to the low voltage cables, which at least were easier to repair. So what voltage are the feeds to point motors?

Jonathan

Fortunately, they seem to steer well clear of our 750V DC traction cables, although there can be all sorts of problems caused by mechanical damage where these cables have to pass under tracks in cable management sleepers.

6 hours ago, phil-b259 said:

However it should be noted that as all railway signalling circuits are deigned to be 'earth free' / floating with respect to earth, part of the regular maintenance tasks is to test all power supplies for how much is leaking through to earth - thus hopefully identifying damaged cables before you get to a stage where two bare cores can touch...

This is inclined to cause chaos when the earth-free 750V traction system gets into contact with the signal rail in single rail track circuit areas (and results in many very rude words being directed by S&T towards those in charge of the CRE. :( )

 

[Dava]

5 minutes ago, jim.snowdon said:

Fortunately, they seem to steer well clear of our 750V DC traction cables, although there can be all sorts of problems caused by mechanical damage where these cables have to pass under tracks in cable management sleepers.

This is inclined to cause chaos when the earth-free 750V traction system gets into contact with the signal rail in single rail track circuit areas (and results in many very rude words being directed by S&T towards those in charge of the CRE. :( )

 

I recall Frank Zappa recorded an album called Hot Rats. Probably no connection as he was in California.

 

Dava

[wombatofludham]

2 hours ago, Dava said:

I recall Frank Zappa recorded an album called Hot Rats. Probably no connection as he was in California.

 

Dava

Unless he had bashed a couple of 25s slogging their way over Talerddig with a fully loaded summer Saturday London to Pwllheli service.  I expect they would have got a bit toasty.

[corneliuslundie]

Back more or less on topic. does anyone know how long the class 197 is staying at Machynlleth for the static testing? I am hoping someone will manage a photo.

Jonathan

[APOLLO]

On 23/05/2022 at 21:53, PerthBox said:

One of the interesting things at West of Scotland SC at Cowlairs in Glasgow (which is built to the standard NR ROC design) is that there’s no running water if the mains power supply fails and drops onto the emergency generator. An extended power outage in 2019 saw the toilets become completely unhygienic and unusable within a single shift. I presume all the other ROC buildings would be similary afflicted.

 

Easily solved !!

 

 

Carlisle Bog Junction !!!!!!

 

Brit15

[ess1uk]

Over the weekend under test conditions a train arrived at Moorgate platform 9 in ETCS Level 2 Full Supervision mode.