Box Tunnel, track lowering, would 3rd rail not have been better?

[sprinterguy]

Is that actually the case? It's always seemed a bit hard to believe for me at least. I've been through there in HSTs and conducted some external observation of the tunnel and was half expecting water to pouring out of the roof from what's been said!

Most of the water that is pumped from the Severn Tunnel each day actually comes from an underground spring, and not from the Severn Estuary above the tunnel:

http://www.railnews.co.uk/news/general/2008/06/03-severn-valley-networkrail.html
Dave Fuller, part of the tunnel’s maintenance team for the last 24 years, says very little seawater enters the tunnel. Most comes from an underground spring, which flooded the tunnel workings during the original construction.

In 2007, 13.86 million gallons of spring water were pumped out every day on average. A pipeline then takes two million gallons a day to a brewery that opened a few miles away in 1979 to take advantage of the tunnel’s water supply. A nearby paper mill, which also used the spring water, closed two years ago.

A further 6.6 million gallons per day on average enters the Severn Tunnel from the farmland above it. About half of the tunnel is situated underneath Monmouthshire, and some rain- water also drains into it from the land on the English side.

I'm not sure of the claims that the tunnel would flood in minutes, though, either. Reportedly when the Great Spring was first encountered during construction in 1879, it was 24 hours later that the tunnel was flooded to river level (and the spring has been sealed behind a headwall since 1881, but evidently still causes major complications). Network Rail have been quoted as saying that the tunnel would be flooded in three and a half minutes if pumping ceased, but I have no idea as to what depth that refers to.

http://www.caldicotu3a.org.uk/The Severn Tunnel - The Story.html
The project began in 1873 and six years later they hit the Great Spring. Only 152 metres separated each tunnel running from the English and the Welsh shore. Within 24 hours the tunnel was flooded to river level, thankfully without any loss of life.

 

[DarloRich]

Is that actually the case? It's always seemed a bit hard to believe for me at least. I've been through there in HSTs and conducted some external observation of the tunnel and was half expecting water to pouring out of the roof from what's been said!

from what i understand without the pumps the tunnel would flood quite quickly!

Rail News: Fighting 14 million gallons of water a day

http://www.railforums.co.uk/newreply.php?do=newreply&p=2366141

 

[edwin_m]

Presumably not a question of the entire tunnel being full of water - to cause a problem it only needs to flood far enough that at the lowest point the track is submerged too deeply to run trains. All the water leaking anywhere in the tunnel will drain to this lowest point, where there would be a sump but as soon as that is full the track will start disappearing quite quickly.

 

[ainsworth74]

All very interesting! Thank you! I had no idea that they hit a spring during construction and that would explain a great deal about the tunnel and its flooding issue.

 

[HowardGWR]

All very interesting! Thank you! I had no idea that they hit a spring during construction and that would explain a great deal about the tunnel and its flooding issue.

Gosh I am genuinely surprised that anyone on this forum would not know about The Great Spring. Next time you are in the library, have a look at the GWR histories. The story of the heroic Diver Lambert is real Boys Own stuff, it's thrilling and gives one the creeps, just imagining what he did.

 

[TheNewNo2]

I've just seen on TV that they lowered the track level of the Box Tunnel in Wiltshire to accommodate future electrification.

I'd have thought it would have been too costly and very risky and as this was once done to a tunnel in Scotland which lead to its collapse burying 2 workers who 35 years later still remain buried.

https://en.m.wikipedia.org/wiki/Penmanshiel_Tunnel

Would it not have been a lot safer and cheaper to use 'third rail' for the tunnel and adapted trains with brushes to use third rail for tunnel sections?

That would kill 2 birds with one stone as it would also allow those trains to operate on the South Eastern networks as well as overhead electric sections.

It wouldn't kill two birds with one stone at all, what you'd do is require 100 new trains to have extra weight fitted because you are worried about something irrelevant, and because you want the trains to go somewhere they would never have need to go.

I'm no expert, perhaps it would have been cheaper to do third rail electrification, but the costs of equipping all the stock with extra shoegear for one single mile of track would be ridiculous. Plus you reduce the maximum speed from 140 to 100.

The only real benefit OLE has in this case, over third rail, is that the OLE is going to stay energised long enough to have a chance of getting a unit out of a flooding tunnel, if the third rail trips, the unit may well be stuck with water rising about it, whilst an AC unit has at least half a chance of getting itself out before the traction motors become submerged or other component failure occurs.

The irony of that statement is, of course, that 25kV AC units 314208 and 314212 may have been the last electric units to suffer flood damage in this manner when Glasgow Central Low Level flooded in 1994.

Weren't there a few damaged 319s last year when that water main burst in the Thameslink tunnels?

 

[Radedamer]

Box, Severn, Badminton and other tunnels are all having track lowered, as are various bridges. To install a third rail system at one would be nonsense.

 

[jmbill]

I've just seen on TV that they lowered the track level of the Box Tunnel in Wiltshire to accommodate future electrification.

I'd have thought it would have been too costly and very risky and as this was once done to a tunnel in Scotland which lead to its collapse burying 2 workers who 35 years later still remain buried.

https://en.m.wikipedia.org/wiki/Penmanshiel_Tunnel

Would it not have been a lot safer and cheaper to use 'third rail' for the tunnel and adapted trains with brushes to use third rail for tunnel sections?

That would kill 2 birds with one stone as it would also allow those trains to operate on the South Eastern networks as well as overhead electric sections.

Based on the photographs and video I've seen, it looks like NR have simply taken the track bed back down to its original Brunel level (by about 12 inches), rather than excavating out several feet of tunnel footings. Presumably over the last 60 years or so, each time a ballast drop has been done and the track has been tampered, it's caused it to rise up a bit each time. So they seem to be taking the track bed back down to its original level. And of course while the track is up, it makes sense to renew other things like drainage, signalling and do a detailed structural / geological survey so the tunnel will be in good shape for another 100+ years.

I heard that the tunnel clearance was fine at the western end, but tolerances were a bit tighter at the London end and at one point in the middle. Hence the need to drop the track level. It's one of several locations where they are reducing the trackbed as part of electrification.

I understand that Box Tunnel requires pumping to remain dry - adding third rail sounds like a recipe for interesting times.

Some of my family used to live in Box village, so it's a patch I know fairly well. I believe the tunnel drains naturally, rather than relying on any pumping gear. It is above the water table and slopes slightly from the Corsham end towards the Bath end. There's a public footpath that runs parallel between the railway line and Peter Gabriel's recording studio. Partway along is a culvert where the drain water from the tunnel comes out. During the summer months it is dry but during the winter months it flows out at quite a significant rate down some stone steps and into the By Brook.

Historically I don't think there has been any issues with flooding in the tunnel. The only weather-related incident was a lorry smashing into the parapet of the A4 road bridge many years ago, which subsequently caused a lot of masonry to fall onto the track below and damage a passing train a short time later. That's why the two sides of the road bridge are a different style.

Undoubtedly tunnel safety standards have improved. BR did a mock tunnel evaluation in Box during the early 90's, with a Pacer traversing up the 'wrong line', carrying firefighting equipment and stretchers. I have some photographs of this I'll upload to Flickr when time permits.

 

[LexyBoy]

I think it I heard it said that they pump ten million gallons a day!

Some back of the envelope calculations:

Volume of the tunnel, assuming a 13 foot radius cylinder 4.25 miles long is 89 million gallons, so it would take over a week to flood even at this rate.

Secondly, if we assume the energy of pumping this water is simply the change in gravitational potential of lifting 10 million gallons of water by 50 metres (probably overestimating the depth, but ignoring frictional and all other losses), then this would require 450 kW of pumping power, all day every day. Sounds expensive - all those electricity bills would pay for a lot of this .

 

[najaB]

Volume of the tunnel, assuming a 13 foot radius cylinder 4.25 miles long is 89 million gallons, so it would take over a week to flood even at this rate.

Yeah, but it doesn't need to be completely full to ruin someone's day.

Sounds expensive - all those electricity bills would pay for a lot of this .

I like the way you think! Speaking seriously though, you have to wonder if it wouldn't be possible to line the tunnel somehow.

 

[Philip Phlopp]

Yeah, but it doesn't need to be completely full to ruin someone's day.
I like the way you think! Speaking seriously though, you have to wonder if it wouldn't be possible to line the tunnel somehow.

Narrow gauge stock needed...

 

[jopsuk]

Going back to the original post, I am always amused somewhat by enthusiasts (and I am one- not a professional) making posts like this. I mean, stop and think for a moment- they might, just, have considered that already? Especially given the extensive enquiries and learning from that disaster.

 

[Trog]

Speaking seriously though, you have to wonder if it wouldn't be possible to line the tunnel somehow.

Apparently once the tunnel was finished the drain was sealed off, leaving the water of the Great Spring sealed behind the tunnel walls. The pressure built up to about 200PSI and bricks started to pop out of the walls. At which point it was decided that pumping the water out might not be such a bad idea after all.

 

[Bald Rick]

Some back of the envelope calculations:

Volume of the tunnel, assuming a 13 foot radius cylinder 4.25 miles long is 89 million gallons, so it would take over a week to flood even at this rate.

But assuming the tunnel follows a classic under river profile, ie down gradient, short level section, up gradient, and if that short level section is 100 metres long, then flooding that section to the crown would be not much more than a million or two gallons.

So still a few hours, but that's only a couple of hundred minutes.

 

[HowardGWR]

Based on the photographs and video I've seen, it looks like NR have simply taken the track bed back down to its original Brunel level (by about 12 inches), rather than excavating out several feet of tunnel footings. Presumably over the last 60 years or so, each time a ballast drop has been done and the track has been tampered, it's caused it to rise up a bit each time.

Yes, (interesting post by the way) I see that the same applies at the track next to Sydney Gardens entering Bath, further down. Two photos, a half century apart, which I think I saw in the Broad Gauge Society Newsletter*, shewed the build up of the ballast graphically.

*You never know; the world may see sense eventually.

 

[Llanigraham]

But assuming the tunnel follows a classic under river profile, ie down gradient, short level section, up gradient, and if that short level section is 100 metres long, then flooding that section to the crown would be not much more than a million or two gallons.

So still a few hours, but that's only a couple of hundred minutes.

I don't think the bottom level is that long!

 

[swt_passenger]

I don't think the bottom level is that long!

About 200 yards, no more than 300?

Network rail archive drawing here: http://www.networkrailmediacentre.co.uk/resources/diagram-of-severn-tunnel

 

[Trog]

The severn tunnel has gradients of 1 in 90 and 1 in 100, ignoring that there must be a vertical curve and or short level section for the change in the gradient. A flood 20'-6" deep would be needed to fill the tunnel at its deepest point this flood is going to then be 20.5*90 + 20.5*100 = 3895 feet long with an average depth of 10'-3". As the tunnel arch springs 7 feet above the rails and the tunnel is 26 feet wide the cross sectional area is going to be about 460 square feet. So ~ 260 square feet half full as the bottom is wider than the top.
260*3895 gives 1012700 cubic feet at 6.23 gallons to the cubic foot that is 6.31 million gallons. The Great spring can run at 30 million gallons a day so that is water to the roof at the bottom of the tunnel in just over five hours.

200yards flat 600*460 = 276000 cubic feet = 1719480 gallons = about 1.4 hours to fill.

So about 6.5 hours to roof.

 

[najaB]

About 200 yards, no more than 300?

From that diagram it appears to be a little over 12 chains. Never mind, reading it wrong.

 

[Murph]

And I'm sure most here will be aware of the Severn Tunnel, which if the pumps and the backups fail would flood in a matter of minutes.

Is that actually the case? It's always seemed a bit hard to believe for me at least. I've been through there in HSTs and conducted some external observation of the tunnel and was half expecting water to pouring out of the roof from what's been said!

Perhaps it's a question of what's meant by "would flood in a matter of minutes". There's a lot of discussion about how long it would take for water to reach to top of the tunnel at the lowest section. In railway terms, however, what's important is when the water reaches the top the rails at the lowest point, which might well be much closer to "a matter of minutes", and then the line would be "flooded" for operational purposes.

 

[furnessvale]

In railway terms, however, what's important is when the water reaches the top the rails at the lowest point, which might well be much closer to "a matter of minutes", and then the line would be "flooded" for operational purposes.

With the Mersey tunnel I believe it is around 45 minutes if both end pump sets failed. As far as I know this has never happened although we did have to rely on Birkenhead only when renewing the Liverpool side pumps in 1971-72.

The drainage adits are designed so that all the water can flow to one end for disposal before tunnel invert level is reached.

 

[DarloRich]

Perhaps it's a question of what's meant by "would flood in a matter of minutes". There's a lot of discussion about how long it would take for water to reach to top of the tunnel at the lowest section. In railway terms, however, what's important is when the water reaches the top the rails at the lowest point, which might well be much closer to "a matter of minutes", and then the line would be "flooded" for operational purposes.

Exactly - the whole tunnel doesn't have to flooded

 

[rebmcr]

this would require 450 kW of pumping power, all day every day. Sounds expensive - all those electricity bills would pay for a lot of this .

Considering that a single Class 390 is rated at around 6MW and there are dozens of those thundering up and down all day, not to mention every single other EMU in service, 450kW is negligible.

 

[Murph]

Considering that a single Class 390 is rated at around 6MW and there are dozens of those thundering up and down all day, not to mention every single other EMU in service, 450kW is negligible.

Your point is quite valid, but not entirely a full comparison as written. 6MW is the peak energy usage for a 390, so average usage will be considerably less.

Another way of putting 450kW in perspective is that it's around 1/3 of a Valenta 12RP200's power output.

 

[Bald Rick]

Some back of the envelope calculations:

Secondly, if we assume the energy of pumping this water is simply the change in gravitational potential of lifting 10 million gallons of water by 50 metres (probably overestimating the depth, but ignoring frictional and all other losses), then this would require 450 kW of pumping power, all day every day. Sounds expensive - all those electricity bills would pay for a lot of this .

What's the going rate for 1kWh in industrial use? 10p? So £45 / hr. Just over £1000 a day.

And what's the going rate for selling decent spring water? 20p/tonne? (Compare to the price of Evian, or even tap water). 10m gallons is 45,000 tonnes. So £9000 a day.

Net £56k/week. Nearly £3m a year.

Nice little earner.

 

[HowardGWR]

What's the going rate for 1kWh in industrial use? 10p? So £45 / hr. Just over £1000 a day.

And what's the going rate for selling decent spring water? 20p/tonne? (Compare to the price of Evian, or even tap water). 10m gallons is 45,000 tonnes. So £9000 a day.

Net £56k/week. Nearly £3m a year.

Nice little earner.

Indeed, one hopes they collect it from the spring and not the rest of the 'effluvia'. Shades of the Del Boy episode.

As a footnote, the day the Great Spring, er, sprung, the rivers of that part of Monmouthshire emptied, simply dried up.

 

[DynamicSpirit]

I think it I heard it said that they pump ten million gallons a day!

At the risk of sounding very ignorant... Are there no completely waterproof and stable materials suitable for lining tunnels, which could therefore be used to remove the need to keep pumping most of the time [*]? (I can believe that 100 years ago, nothing suitable would have been available, but surely today there must be something?)


[*] Though I'm guessing that you'd still have the problem of rainwater entering the tunnel ends.

 

[D365]

As I understand it, it's not just the problem of water ingress; without piping the fluid elsewhere, pressure would simply build up and risk the tunnel structure caving in.

 

[Llanigraham]

As a footnote, the day the Great Spring, er, sprung, the rivers of that part of Monmouthshire emptied, simply dried up.

I think that is a bit of a "story" that isn't actually true.
Very little of the tunnel is in Monmouthshire and the Great Spring is believed to run in line with the River Severn but in the opposite direction.
--- old post above --- --- new post below ---

As I understand it, it's not just the problem of water ingress; without piping the fluid elsewhere, pressure would simply build up and risk the tunnel structure caving in.

Correct!

 

[RichmondCommu]

As I understand it, it's not just the problem of water ingress; without piping the fluid elsewhere, pressure would simply build up and risk the tunnel structure caving in.

Well yes but in all fairness that should be pretty obvious. However when I read on here that river water appears to leak into the ELL tunnel then that does raise concerns. If the entire tunnel was to fail that would surely have the potential for large loss of life and suggests that the tunnel was not dug deep enough. I'm not aware of water being pumped out of the Dartford Tunnel as an example.