What should GWR do about severe conditions on Dawlish?

[Annetts key]

Unless there is a detailed technical specification, how do you properly define the requirements or test that the train meets the requirements?

Sending a test train through Dawlish is not going to prove anything if there is no storm. And even if you wait until there is a storm, Will that storm be representative of the conditions you want the train to operate in?

I would suggest that a test rig should have been made, using the same high pressure, high volume water pumps that firefighters use. To spray water with the same salt content as sea water over the test train. And there should be multiple hose outlets, spraying water at multiple different angles.

Then the test train should go through/past the test rig at various speeds, including at walking speed, and multiple times in both directions. And with the train being turned so that both sides are tested, or the test rig set up to spray from either side.

That I think would be a reasonable attempt at simulating the conditions At Dawlish.

 

[BayPaul]

Unless there is a detailed technical specification, how do you properly define the requirements or test that the train meets the requirements?

Sending a test train through Dawlish is not going to prove anything if there is no storm. And even if you wait until there is a storm, Will that storm be representative of the conditions you want the train to operate in?

I would suggest that a test rig should have been made, using the same high pressure, high volume water pumps that firefighters use. To spray water with the same salt content as sea water over the test train. And there should be multiple hose outlets, spraying water at multiple different angles.

Then the test train should go through/past the test rig at various speeds, including at walking speed, and multiple times in both directions. And with the train being turned so that both sides are tested, or the test rig set up to spray from either side.

That I think would be a reasonable attempt at simulating the conditions At Dawlish.

No, that's exactly the opposite to the way I'd suggest speccing it! Don't try to work out how to recreate the conditions as a spec, just pass the buck over to Hitachi, and specify that the train must be able to run a normal service through Dawlish in conditions up to XXX, and must perform at least as well as any other train in current service on the route, with significant penalties for any occasion that this isn't achieved in the first 5 or 10 years of the trains life, then let Hitachi work out how to achieve this.

 

[Meerkat]

No, that's exactly the opposite to the way I'd suggest speccing it! Don't try to work out how to recreate the conditions as a spec, just pass the buck over to Hitachi, and specify that the train must be able to run a normal service through Dawlish in conditions up to XXX, and must perform at least as well as any other train in current service on the route, with significant penalties for any occasion that this isn't achieved in the first 5 or 10 years of the trains life, then let Hitachi work out how to achieve this.

Then you dont actually have a definition, and Hitachi will fight you in court if you try to say they didnt meet the spec.
Not forgetting that the DfT wont actually want to pay for a train that will NEVER break down when hit by a wave - if all bidders had come back with over engineered offers to achieve that the DfT would refuse to pay that much.
Just set up a set test, that can then be used by all suppliers - I'm sure one of the universities would love the funding. The film industry can supply designs for chutes and ramps to provide a standard wave thrown over the test track.

 

[AM9]

No, that's exactly the opposite to the way I'd suggest speccing it! Don't try to work out how to recreate the conditions as a spec, just pass the buck over to Hitachi, and specify that the train must be able to run a normal service through Dawlish in conditions up to XXX, and must perform at least as well as any other train in current service on the route, with significant penalties for any occasion that this isn't achieved in the first 5 or 10 years of the trains life, then let Hitachi work out how to achieve this.

Specifications just like that should have been created here are commonplace the world over in the defence market. Even the global commercial shipping and aircraft industry has this type of specification to ensure that suppliers know what they are quoting for, and purchasers know what will be delivered. It's so naïve to think that a legal fight with a supplier will solve the fact that trains are disrupted when they could be manufactured to be fit for purpose.

 

[Vespa]

If you learn the lesson of the village of Hallsands when their shingles was robbed to expand the naval dockyard of Plymouth, the sea came closer and the village eventually got wiped out in a major storm in 1917.

The answer is fairly straightforward, put more shingles and sand on the beach and this will force the sea further out and weaken the force of storm waves as they have much further to go, it's down to money.

*Edit to add more information.

 

[najaB]

The answer is fairly straightford, put more shingles and sand on the beach and this will force the sea further out and weaken the force of storm waves as they much further to go, it's down to money.

You're basically right, though it doesn't have to be shingle and sand. Back home we've had a lot of good results building artificial reefs tens to hundreds of metres off-shore and they do double-duty in providing habitat for marine life = better fisheries.

 

[Irascible]

If you learn the lesson of the village of Hallsands when their shingles was robbed to expand the naval dockyard of Plymouth, the sea came closer and the village eventually got wiped out in a major storm in 1917.

The answer is fairly straightforward, put more shingles and sand on the beach and this will force the sea further out and weaken the force of storm waves as they have much further to go, it's down to money.

*Edit to add more information.

You can do the same with large rocks ( and indeed I think they have ) - that takes all the energy out of waves hammering the wall ( or cliff or whatever ) but just by being by the shore you'll get spray in a storm - the energy has to dissipate somewhere, and if it's a wave breaking on something it's usually fairly upwards. Add in the wind...

There are also environmental considerations - how many times have places built groynes to protect their beach & found someone else's beach disappears instead?

Unless there is a detailed technical specification, how do you properly define the requirements or test that the train meets the requirements?

Sending a test train through Dawlish is not going to prove anything if there is no storm. And even if you wait until there is a storm, Will that storm be representative of the conditions you want the train to operate in?

I would suggest that a test rig should have been made, using the same high pressure, high volume water pumps that firefighters use. To spray water with the same salt content as sea water over the test train. And there should be multiple hose outlets, spraying water at multiple different angles.

Then the test train should go through/past the test rig at various speeds, including at walking speed, and multiple times in both directions. And with the train being turned so that both sides are tested, or the test rig set up to spray from either side.

That I think would be a reasonable attempt at simulating the conditions At Dawlish.

Presumably the spec would be "we would like a train which can survive this" - presumably it's up to Hitachi to build the rig to make sure it works. There's an argument already

Aero engines are built to withstand a certain volume of water per second ( a surprisingly huge amount too ) without flaming out, but that I think is defined as a constant stream. Building a train that can run submerged is obviously a bit over the top, the problem is defining your transient conditions to the satisfaction of both engineers & lawyers. Derby would probably have loved this stuff back in the day, gone out with an old Mk1 full of buckets to measure ingress or something

There's a few people involved in rolling stock design here, maybe one of them can chime in.

 

[Vespa]

You can do the same with large rocks ( and indeed I think they have ) - that takes all the energy out of waves hammering the wall ( or cliff or whatever ) but just by being by the shore you'll get spray in a storm - the energy has to dissipate somewhere, and if it's a wave breaking on something it's usually fairly upwards. Add in the wind...

Large rocks breaks the waves up and turns it into smaller splashes that much is true , the energy is not fully dissipated, shingles and sand breaks that energy into smaller splashes and dissipates it gradually by the the time it hits the shore its not noticeable.

Hallsands survived since the 1500s because of the shingle beach that protected the fishing village even through the worst storms, it was only when it was dredged problems started, now there is no shingle at all and the sea comes right up to the shoreline and there is only scattered ruins to remind you of what was once there.

 

[matacaster]

Unless there is a detailed technical specification, how do you properly define the requirements or test that the train meets the requirements?

Sending a test train through Dawlish is not going to prove anything if there is no storm. And even if you wait until there is a storm, Will that storm be representative of the conditions you want the train to operate in?

I would suggest that a test rig should have been made, using the same high pressure, high volume water pumps that firefighters use. To spray water with the same salt content as sea water over the test train. And there should be multiple hose outlets, spraying water at multiple different angles.

Then the test train should go through/past the test rig at various speeds, including at walking speed, and multiple times in both directions. And with the train being turned so that both sides are tested, or the test rig set up to spray from either side.

That I think would be a reasonable attempt at simulating the conditions At Dawlish.

Modify a carriage washer to test?

 

[EastisECML]

Would it not be best to just build a new in land line from Exeter to Plymouth and have a local Plymouth to Exeter service serve the Dawlish route?

 

[JohnElliott]

Would it not be best to just build a new in land line from Exeter to Plymouth and have a local Plymouth to Exeter service serve the Dawlish route?

Might be cheaper to build a batch of 9Fs to haul the IEPs in rough weather...

 

[AM9]

Would it not be best to just build a new in land line from Exeter to Plymouth and have a local Plymouth to Exeter service serve the Dawlish route?

Nice try, but no! You don't fix poorly specified trains with tens (or maybe hundreds) of millions of intrastructure works.

 

[najaB]

Would it not be best to just build a new in land line from Exeter to Plymouth and have a local Plymouth to Exeter service serve the Dawlish route?

Long term that might be the best option, but that's going to cost a heck of a lot of money.

 

[Irascible]

Would it not be best to just build a new in land line from Exeter to Plymouth and have a local Plymouth to Exeter service serve the Dawlish route?

I see what you did there we're not going near that route this time.

You've still got to sea-proof the local trains even if the not quite local ones go around ( the distinction between local and not-local can get a bit murky down here ). It's much the same argument as going round to avoid sea wall problems - it's got to be there anyway, so solve the problem by fixing the sea wall. Why not just seal up all of them.

 

[Railwaysceptic]

I see what you did there we're not going near that route this time.

I think he was referring to a new, high speed Dawlish Avoiding Line which would have the huge additional benefit of making Plymouth/Exeter journey times appropriate for the modern age.

 

[mike57]

Salt water and electrical systems are not a good mix, unless systems are totally sealed. Total sealing isnt possible as otherwise everything would overheat. I suspect modern units with a high degree of electronics are more susceptible to issues because there will be more monitoring of things like earth leakage. As far as I know HST are pretty much 'old school' and will probably continue to opertate even if a lot of the current is leaking away where it shouldn't.

One solution would be to build a tunnel over the railway, but that would look pretty awful.

Once salt water has penetrated where it shouldn't you have a serious problem, whilst washing down may help, if any salt is left behind it is going to be a reliability problem going forwards.

Running through unpowered dragged by something else isnt going to help as the salt water will still get in. Also as older stuff gets withdrawn there will be less things that can survive Dawlish when the weather is bad

Really diversion becomes the only option. The problem is going to get worse, its only a matter of time before there is another 2014. so it comes down to a new inland route or a reopening of closed bits via Okehampton to provide a low capacity link when there is bad weather. If there is another sea wall collapse I could see the reopening option being persued, but I cant see the money being found for a brand new route however 'nice to have' that is.

Steam hauled local shuttle service during bad weather maybe Whilst this statement is made tounge in cheek, it would solve the problem, most of the problems occur during the winter which must be a quiet time for the heritage sector.

 

[AM9]

Salt water and electrical systems are not a good mix, unless systems are totally sealed. Total sealing isnt possible as otherwise everything would overheat. I suspect modern units with a high degree of electronics are more susceptible to issues because there will be more monitoring of things like earth leakage. As far as I know HST are pretty much 'old school' and will probably continue to opertate even if a lot of the current is leaking away where it shouldn't.Steam hauled local shuttle service during bad weather maybe Whilst this statement is made tounge in cheek, it would solve the problem, most of the problems occur during the winter which must be a quiet time for the heritage sector. ...

Total sealing of exposed bits is practical if the buyer specifies it for a competant supplier accepts the job. The design has to be completed, tested and qualified just once. It would obviously cost more but the GWR has 276 cars that definitely need to be to that design which is a fair sized production run for the additional cost of salt water spray hardening to be amortised over. The DfT should as the deprtment potentially responsible for the whole class 8xx fleet, should arguably have thought this through and added the development requirement for running under those conditions to the initial IET order, which would then see it as an option for a total vehicle requirement of (as a quick totting up) 1399 vehicles, - as massive order over a very short period even including BR's major re-stocking efforts.

... One solution would be to build a tunnel over the railway, but that would look pretty awful....

There enough protests from those on Riviera and Sea View Terraces that demanded the lowered sea wall was restored in front of their properties so as to preserve their view. That rather tied NR's hands to perpetuating the vulnerability of the trackbed.

... Once salt water has penetrated where it shouldn't you have a serious problem, whilst washing down may help, if any salt is left behind it is going to be a reliability problem going forwards. ...

Hence the answer that salt water should be effectively excluded from sensitive areas in the first place.

 

[MotCO]

How much of the coast is actually affected? If it is not too far, can you install a sea wall out at sea to break the waves further out, very much like harbour walls protect harbours? Could you grow coral or something so that it is a natural defence?

 

[najaB]

How much of the coast is actually affected? If it is not too far, can you install a sea wall out at sea to break the waves further out, very much like harbour walls protect harbours? Could you grow coral or something so that it is a natural defence?

Mentioned up above. It probably will need to be done in the future, though I have no idea how quickly the seabed drops off in that area.

 

[BayPaul]

Mentioned up above. It probably will need to be done in the future, though I have no idea how quickly the seabed drops off in that area.

The 5m depth contour is around 600m from the tracks at the Teignmouth end, and 800m at the Dawlish Warren end.

Total sealing of exposed bits is practical if the buyer specifies it for a competant supplier accepts the job. The design has to be completed, tested and qualified just once. It would obviously cost more but the GWR has 276 cars that definitely need to be to that design which is a fair sized production run for the additional cost of salt water spray hardening to be amortised over. The DfT should as the deprtment potentially responsible for the whole class 8xx fleet, should arguably have thought this through and added the development requirement for running under those conditions to the initial IET order, which would then see it as an option for a total vehicle requirement of (as a quick totting up) 1399 vehicles, - as massive order over a very short period even including BR's major re-stocking efforts.

I wonder if it would have been sensible to spec the trains so they always run 1st class towards London (for example) in bad weather. This would allow all the air intakes and other more vulnerable parts of the train to be on the left hand side when running East, so sheltered by the body of the train.

Then you dont actually have a definition, and Hitachi will fight you in court if you try to say they didnt meet the spec.
Not forgetting that the DfT wont actually want to pay for a train that will NEVER break down when hit by a wave - if all bidders had come back with over engineered offers to achieve that the DfT would refuse to pay that much.
Just set up a set test, that can then be used by all suppliers - I'm sure one of the universities would love the funding. The film industry can supply designs for chutes and ramps to provide a standard wave thrown over the test track.

On the contrary, I think you have a much clearer definition than something based on tonnes of water per minute. In the shipping industry, the concept of a 'reference vessel' is used frequently when speccing in similar circumstances to this, and works very well - it doesn't require gold plating, just for the manufacturer to make sure they understand the issue and build accordingly. For this example, if you were to say 'Must be able to operate through Dawlish in conditions at least as bad as a class 165', then I can't see how Hitachi would be able to argue that they have met the contract when 3 of their trains sat down, and the old stock was running through without issue.

 

[Meerkat]

On the contrary, I think you have a much clearer definition than something based on tonnes of water per minute. In the shipping industry, the concept of a 'reference vessel' is used frequently when speccing in similar circumstances to this, and works very well - it doesn't require gold plating, just for the manufacturer to make sure they understand the issue and build accordingly. For this example, if you were to say 'Must be able to operate through Dawlish in conditions at least as bad as a class 165', then I can't see how Hitachi would be able to argue that they have met the contract when 3 of their trains sat down, and the old stock was running through without issue.

How would that help at all? It would still be near untestable, and really hard to design to.

 

[swt_passenger]

For all those mentioning the possibility of works offshore, it was looked at and ruled out, there were details in a report in with one of the planning applications, but I don’t remember the basic reasons.

I also think the 2014 repairs will eventually be altered and a similar “recurve wall“ installed to the same height as the station section being done now.

AIUI the 2014 repairs were emergency works, and I suggest a future full planning application will be made and views from ground floors of the adjacent properties won’t be protected, after all views weren’t protected along Marine Parade...

 

[BayPaul]

How would that help at all? It would still be near untestable, and really hard to design to.

It would presumably need a small, but not excessive overdesign. As Hitachi own the trains as well as build them, it could then mean that GWR could claim back significant compensation if they got the design wrong, which would encourage them to get it right. If a testable line is also needed in the contract, then there is no harm in also including something about tonnes of salt water per minute.

 

[Meerkat]

It would presumably need a small, but not excessive overdesign. As Hitachi own the trains as well as build them, it could then mean that GWR could claim back significant compensation if they got the design wrong, which would encourage them to get it right. If a testable line is also needed in the contract, then there is no harm in also including something about tonnes of salt water per minute.

What would cost more - the occassional delay minutes or the extra cost added by Hitachi to overdesign plus cover the financial risk of getting sued?

 

[AM9]

What would cost more - the occassional delay minutes or the extra cost added by Hitachi to overdesign plus cover the financial risk of getting sued?

I'm not one of those who really thinks 'storm hits Dawlish, - England cut off' but if that is important, then "the occasional delay minutes" would seem to be more than an understatement. I don't thinki the 'do nothing' option is necessarily the bes value in view of the worsening climatic conditions that may be hastening in the future.

== Doublepost prevention - post automatically merged: 14 Dec 2021 ==

It would presumably need a small, but not excessive overdesign. As Hitachi own the trains as well as build them, it could then mean that GWR could claim back significant compensation if they got the design wrong, which would encourage them to get it right. If a testable line is also needed in the contract, then there is no harm in also including something about tonnes of salt water per minute.

I don't think that Hitachi own the 802s like they do the 800/801s where the contract is to provide then to run services.

 

[Ashley Hill]

Perhaps the best way to have tested one would've been to park one on the sea wall overnight during a particularly violent storm and see if it started in the morning.
Short term there's perhaps two service options during high seas,not stopping IETs at Dawlish and Teignmouth or terminating short at Exeter and Newton abbot like Voyagers.

 

[najaB]

For all those mentioning the possibility of works offshore, it was looked at and ruled out, there were details in a report in with one of the planning applications, but I don’t remember the basic reasons.

Do you happen to have a link to that report, please?

 

[swt_passenger]

Do you happen to have a link to that report, please?

I’d have to search the Teignbridge planning website again. It was a few years ago I read it…

 

[Irascible]

Large rocks breaks the waves up and turns it into smaller splashes that much is true , the energy is not fully dissipated, shingles and sand breaks that energy into smaller splashes and dissipates it gradually by the the time it hits the shore its not noticeable.

Missed this one - the beach here is shingle, and every time there's a storm most of the beach ends up over the sea wall & in the town! there is a problem of making it stay where you put it too, if it's not a natural beach.

I think he was referring to a new, high speed Dawlish Avoiding Line which would have the huge additional benefit of making Plymouth/Exeter journey times appropriate for the modern age.

Would speed up services between Exeter & Newton Abbott for sure - I think "making Plymouth/Exeter journey times appropriate for the modern age" is overstating the effect somewhat, still got to crawl your way between NA & Pymouth...

 

[EastisECML]

I think he was referring to a new, high speed Dawlish Avoiding Line which would have the huge additional benefit of making Plymouth/Exeter journey times appropriate for the modern age.

Yeah, no interest in that line that used to exist. Just a new direct route between Exeter and Plymouth. I don't think it's good for the railways when there's a direct road but the railway cannot match it.