Level crossing stupidity...

[Titan]

Haven't Railtrack heard of software upgrades, or do they just ignore them? Not good.

 

You will have to get a Tardis  to find out....

[jim.snowdon]

Upgrading software on something like this is not a trivial exercise. (Nor quite on-topic in this thread, now I come to think of it).

 

I have read that to move from baseline 2 to baseline 3 each train has to be re-certified at substantial cost. (I don't know if that is a purely software upgrade or if there is any hardware change involved as well though).

I would doubt that, but any changes in the software of computer based signalling and train control systems have to be very thoroughly checked, not just that the change works as designed under all foreseeable scenarios, but also to verify that it hasn't affected any other parts of the system as a whole. The result is a lot of testing and considerable expense.

 

Jim

[Coryton]

I would doubt that, but any changes in the software of computer based signalling and train control systems have to be very thoroughly checked, not just that the change works as designed under all foreseeable scenarios, but also to verify that it hasn't affected any other parts of the system as a whole. The result is a lot of testing and considerable expense.

 

Jim

 

I've found where I read it...Wikipedia, and no citation, so take that as you will. It claims around €100k per vehicle to recertify after an upgrade to baseline 3.

[jim.snowdon]

Very useful if it worked properly in the first place

I’ve mentioned this scenario countless times in various ERTMS threads but I’ll put it again here

As it stands now (ie cambrian ertms) on a 97 you can only input one set length for the different types of trains, passenger or freight so if input I’m a class 97 hauling a passenger train the system ‘assumes’ I have the default 12 coach train behind me, I cannot adjust that even if I have only 4 vehicles (such as on the NR test trains), similarly if I put I’m freight it will assume I have the default 10 vehicles behind me (even if I have 20, which has happened in the not too distant past)

Herein lies the problem approaching a speed restriction the train has to be doing the restricted speed 100m before it, as bought down from line speed by the driver within the ‘speed hook’and once over the restriction it will not let you accellerate until 100m beyond the end of the restriction, so a 4 coach ‘passenger’ train going over say a 500m 10kph restriction has to be doing 10kph for 700m PLUS the system then counts 8 ‘invisible’ coaches over the restriction too making a 500m restriction something more like 800-850m long

likewise a ‘freight’ with 20 auto hoppers will be told it’s clear of a restriction when I reality there could be 5-6 wagons still on it!

It is a simple fix apparently by a quick rewrite of the software but no one seems keen implement it

This, and the subsequent post about how the Cambrian ERTMS behaves with regard to approach restrictions at crossings, illustrate the perils of trying to adapt control systems original designed round the characteristics of metro systems to the rather different environment of the mixed traffic railway.

 

Jim

[Coryton]

This, and the subsequent post about how the Cambrian ERTMS behaves with regard to approach restrictions at crossings, illustrate the perils of trying to adapt control systems original designed round the characteristics of metro systems to the rather different environment of the mixed traffic railway.

 

This should be a fairly trivial feature to have programmed in from the start (which doesn't of course mean that adding it now would be easy, depending on how the software is structured).

 

I wonder if its absence is because none of the people involved in specifying ERTMS thought it would be required.

 

I don't know how widespread such restrictions are in the UK or elsewhere. Are they are relatively recent product of the ever-decreasing tolerance to risk at level crossing? It sounds as if they were implemented on the Cambrian after ERTMS.

[jim.snowdon]

This should be a fairly trivial feature to have programmed in from the start (which doesn't of course mean that adding it now would be easy, depending on how the software is structured).

 

I wonder if its absence is because none of the people involved in specifying ERTMS thought it would be required.

 

I don't know how widespread such restrictions are in the UK or elsewhere. Are they are relatively recent product of the ever-decreasing tolerance to risk at level crossing? It sounds as if they were implemented on the Cambrian after ERTMS.

My understanding is that ATO, in all of its various guises, up to and including ERTMS, grew out of metro/rapid transit operations, where although there may be speed restrictions, there are no level crossings getting in the way and train lengths are either automatically known (n times unit length) or fixed (ie all trains are the same length). It may be no coincidence either that some of those in NR who are behind the idea of ERTMS being the railway's saviour came from the rapid transit world, rather than heavy rail operation.

 

Jim

[Coryton]

My understanding is that ATO, in all of its various guises, up to and including ERTMS, grew out of metro/rapid transit operations, where although there may be speed restrictions, there are no level crossings getting in the way and train lengths are either automatically known (n times unit length) or fixed (ie all trains are the same length). It may be no coincidence either that some of those in NR who are behind the idea of ERTMS being the railway's saviour came from the rapid transit world, rather than heavy rail operation.

 

Jim

 

Well, perhaps that is the explanation and as usual when it comes to prototype railways I only know what I've read and experienced as an interested outsider.

 

However, whatever it grew out of so far as I know ERTMS is the result of representatives from EU railway systems getting together to decide what they all wanted/needed out of a common cab-signalling system.

 

I'd like to think that the end result of a long process would be based on what people thought they wanted at the time.

[Fat Controller]

This should be a fairly trivial feature to have programmed in from the start (which doesn't of course mean that adding it now would be easy, depending on how the software is structured).

 

I wonder if its absence is because none of the people involved in specifying ERTMS thought it would be required.

 

I don't know how widespread such restrictions are in the UK or elsewhere. Are they are relatively recent product of the ever-decreasing tolerance to risk at level crossing? It sounds as if they were implemented on the Cambrian after ERTMS.

The restrictions on approaching certain types of level crossings have applied for a long time, certainly before ERTMS/ETCS was even dreamt about. As I understand it, the idea of ERTMS was a political one, put forward by certain MEPs, keen to see 'Open Access' to European railways, with their many different signalling systems, and not something originating from the railways themselves.

[Coryton]

The restrictions on approaching certain types of level crossings have applied for a long time, certainly before ERTMS/ETCS was even dreamt about. As I understand it, the idea of ERTMS was a political one, put forward by certain MEPs, keen to see 'Open Access' to European railways, with their many different signalling systems, and not something originating from the railways themselves.

 

The idea of ERTMS may indeed have been political, but I would like to think that the committees who decided on the implementation included people who actually knew what was required.

[Fat Controller]

The idea of ERTMS may indeed have been political, but I would like to think that the committees who decided on the implementation included people who actually knew what was required.

I would hope that were the case, even if progress thus far suggests this may not have been the case.

[david.hill64]

The idea of ERTMS may indeed have been political, but I would like to think that the committees who decided on the implementation included people who actually knew what was required.

.

 

It certainly is political as it is borne out of the interoperability directives. The trouble is that when you place engineers from 28 countries in the same room, most of whom want not to have to change their operating regimes, and a few of whom desperate to ensure their domestic supply industries are not disadvantaged, it takes forever to agree the specifications. Then the systems have to be designed and manufactured, tested and approved. Getting approval of safety critical hardware platforms and associated firmware and software is vastly expensive. And if one supplier is ahead of the game then you need to change the specifications to give chance for the competition to catch up.

Or am I being a little cynical?

[Wickham Green]

.

 

It certainly is political as it is borne out of the interoperability directives. The trouble is that when you place engineers from 28 countries in the same room, most of whom want not to have to change their operating regimes, and a few of whom desperate to ensure their domestic supply industries are not disadvantaged, it takes forever to agree the specifications. Then the systems have to be designed and manufactured, tested and approved. Getting approval of safety critical hardware platforms and associated firmware and software is vastly expensive. And if one supplier is ahead of the game then you need to change the specifications to give chance for the competition to catch up.

Or am I being a little cynical?

Probably not cynical enough - you forgot to mention that technology moves on and by the time anything DOES actually get approved it's horribly out of date ........

[david.hill64]

Probably not cynical enough - you forgot to mention that technology moves on and by the time anything DOES actually get approved it's horribly out of date ........

True: I think I have posted before that metro signalling systems are more advanced than mainline ones. They have a shorter life cycle.

[jim.snowdon]

True: I think I have posted before that metro signalling systems are more advanced than mainline ones. They have a shorter life cycle.

But just like main line systems, their useful working life seems to be getting shorter with every generation.

 

Jim

[david.hill64]

But just like main line systems, their useful working life seems to be getting shorter with every generation.

 

Jim

True. This is a real issue for the signalling suppliers. Hardware becomes obsolete very quickly. Changing hardware is a vastly expensive commitment, hence the move to common platforms for main line and metro systems.

 

At least with metros there are no interoperability requirements so the suppliers can use their standard in-house products tailored to individual system requirements. When you have to get international agreement to sanction specification changes it is bound to be a long process.

[Colin_McLeod]

Red and green flags have not been updated in years!

[big jim]

I like the way this thread has drifted for a couple of pages, it goes to show we haven’t had any news worthy level crossing incidents for a while!

[jamie92208]

Red and green flags have not been updated in years!

 

Neither has the pilotman's armband I saw one in use near Bournemouth a year ago and it was identical in design to the previous one I saw at Settle in 1966.  The technology that had changed was the clothing of the wearer, from big black mac to all over orange.

 

Jamie

[big jim]

I do believe they are no longer permitted to wear a red flag around the left arm any more though in the absence of an armband

[meil]

Red and green flags have not been updated in years!

They have, we use a yellow flag now!

[phil-b259]

Red and green flags have not been updated in years!

 

 

 

They have, we use a yellow flag now!

 

The correct position on NR is that persons appointed to act as a handsignaller (i.e. a person taking the place of a signal) must NOT under any circumstances show a green flag or light to a driver when giving permission to pass a defective signal. Regardless of whether it is a 4 aspect colour light or a traditional semaphore stop signal (which obviously can only show a red or green indication when it is working), permission to proceed is given by showing a yellow flag / lamp to the driver. This is on the basis you only need handsignalers if the fixed signalling is inoperative in some way and as such you don't have the normal safeguards in place - therefore drivers should be proceeding cautiously anyway.

 

By contrast, most Heritage railways still use the steam era approach where red or green flags are used at semaphore stop signals or a yellow flag at a distant signal.

 

The only people authorised to have green flags / lamps on NR are level crossing attendants who take the level crossing barriers under local control due engineering works or a failure preventing the normal operation of the crossing. In these situations, once the barriers are down and it is safe for the train to proceed, then the level crossing attendant will show a green flag / lamp to the driver.

[Fat Controller]

I like the way this thread has drifted for a couple of pages, it goes to show we haven’t had any news worthy level crossing incidents for a while!

I have seen reference to a few in the past week, the most notable being one in Cumbria, where a drunk driver turned on to the tracks at a crossing, and then drove for about 150 metres along the line. This was at Waltham Nurseries UWG. Not sure where this is exactly, as I couldn't find it in Quail.

[beast66606]

I have seen reference to a few in the past week, the most notable being one in Cumbria, where a drunk driver turned on to the tracks at a crossing, and then drove for about 150 metres along the line. This was at Waltham Nurseries UWG. Not sure where this is exactly, as I couldn't find it in Quail.

 

Between Foxfield and Millom at 42m 62ch where Carnforth is 0m

[The Stationmaster]

They have, we use a yellow flag now!

 

And you have for the past 30years if it involves passing a signal at danger!

 

What happened was rather straightforward but one of those odd occasions when logic reared its head.  When we (I as a member of the working group) were working on the complete revision of the Rule Book in 1985 somebody had put in a question via one of the group members asking why we used a green flag to handsignal a movement past a semaphore stop signal at danger but a yellow flag to do the same thing at a multiple aspect colour light signal and wouldn't it be better to apply the same handsignal to both.  Clearly - revisiting our discussion - we could not authorise a green handsignal at a multiple aspect colour light as it could be very misleading to a Driver so we decided to standardise on the use of a yellow handsignal to signal to a Driver that he was allowed to pass any sort of stop signal at danger and that went into the reissue of the Rule Book (which I think was actually published in 1988 as we were also revising the Block Regulations and the General Appendix).

 

The fact that yellow means the Driver should proceed cautiously is completely irrelevant and didn't even enter our discussion as there was no need for it to do so,.  The handsignal indicates to a Driver that he can pass that stop signal at danger and proceed towards the next stop signal which, as the Rule Book tells him, he could expect not to be displaying a proceed indication or aspect.  if there is any reason to caution a Driver regarding the state of the line ahead that would be done before he is given the handsignal authorising him/her to pass the signal at danger - different matter entirely and nothing to do with the colour of the handsignal.

 

So it is now a bit over 30 years since yellow handsignals have been used to authorise a Driver to pass a stop signal at danger and I doubt that it would even occur to anybody to use a green handsignal as there must now be not a large number of folk about who remember the old Rule in resect of semaphore stop signals (and there aren't anything like the number of semaphore stop signals there were 30 years ago. 

[jim.snowdon]

Which makes it all the stranger that anyone should use a green flag/lamp handsignal under any circumstances.

 

Jim