TBM Cost per mile?

[Tobbes]

I was wondering what the cost per mile was of a TBM bored single track tunnel? @Bald Rick suggested that the TBM was c £10m in 2019, and I was wondering what the total cost of a tunnel about 1.9km / 1.2 miles long would ballpark be?

Many thanks

Toby

 

[WatcherZero]

About £10m as a starting point for a standard gauge TBM is roughly right, there is economies of scale up to about 10-15km as you get more from your investment in a tbm at which point the machine would be exhausted and you would need to buy a second to go further.

Heres a good breakdown for High Speed Rail tunnelling, for a standard gauge rail tunnel reduce the cost by about 25-50%.

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/434516/HS2_Guide_to_Tunnelling_Costs.pdf

 

[Tobbes]

About £10m as a starting point for a standard gauge TBM is roughly right, there is economies of scale up to about 10-15km as you get more from your investment in a tbm at which point the machine would be exhausted and you would need to buy a second to go further.

Heres a good breakdown for High Speed Rail tunnelling, for a standard gauge rail tunnel reduce the cost by about 25-50%.

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/434516/HS2_Guide_to_Tunnelling_Costs.pdf

Thanks very much - c. £15m/km... which makes my 2km tunnel under Wadebridge to a new station site on top of a CPO'd Lidl to allow trains to run through to Padstow a rather pricey option to avoid CPO'ing houses!

I'm still intrigued about where the proposed reopening to Wadebridge would actually put Wadebridge station unless you were going to CPO a small housing estate to get back to the old station.

 

[Bald Rick]

I was wondering what the cost per mile was of a TBM bored single track tunnel? @Bald Rick suggested that the TBM was c £10m in 2019, and I was wondering what the total cost of a tunnel about 1.9km / 1.2 miles long would ballpark be?

Many thanks

Toby

The tbm is about £10m, the cost per mile is about £100m. Portals aren’t cheap. The HS2 doc is a good guide but a little old and the prices need to be lifted.

 

[superkev]

Slightly related but I seem to remember reading probably in Modern Railways that when you took into account the huge costs of legalities, land purchase etc the cost of tunnelling was not much more than over the top.
Definatly the best way go cross the Pennines as its weatherproof.
K

 

[Bald Rick]

Slightly related but I seem to remember reading probably in Modern Railways that when you took into account the huge costs of legalities, land purchase etc the cost of tunnelling was not much more than over the top.
Definatly the best way go cross the Pennines as its weatherproof.
K

You still need all the legalities to build a tunnel, but it is somewhat easier. Across open country it is definitely much cheaper above ground, otherwise HS2 would be in Tunnel the whole way.

 

[Tobbes]

You still need all the legalities to build a tunnel, but it is somewhat easier. Across open country it is definitely much cheaper above ground, otherwise HS2 would be in Tunnel the whole way.

Either way tunnelling under Wadebridge seems unlikely to be vfm...

 

[LNW-GW Joint]

The new Gotthard Base Tunnel project reportedly cost CHF 9.5 billion, or about £6 billion.
That produced twin 35-mile tunnels under the Alps, so 70 miles of tunnel fit for 250km/h.
I think that works out at £114m per mile of single bore.
There were significant portal and realignment works at each end, plus central access shafts, which are included in the cost.
The numerical delivery cost increased by 50% between go-ahead and completion (17 years).
Construction also cost 9 lives.

 

[Ploughman]

The new Gotthard Base Tunnel project reportedly cost CHF 9.5 billion, or about £6 billion.
That produced twin 35-mile tunnels under the Alps, so 70 miles of tunnel fit for 250km/h.
I think that works out at £114m per mile of single bore.
There were significant portal and realignment works at each end, plus central access shafts, which are included in the cost.
The numerical delivery cost increased by 50% between go-ahead and completion (17 years).
Construction also cost 9 lives.

Did they renew the TBM's after 15km as per post earlier.

 

[marko2]

it surely depends on the geology the tunnel passes though. In London is either clay or chalk, which is fairly soft. The Pennines might be a different matter.

 

[Bald Rick]

it surely depends on the geology the tunnel passes though. In London is either clay or chalk, which is fairly soft. The Pennines might be a different matter.

Not really, believe it or not. The costs of tunnelling (once approval is granted) are mostly around setting up the portals, access to the site, disposing spoil and the lining / grouting.

If anything, it would be easier in the rock beneath the Pennines, as you don’t have to worry about ground settlement as much. The bigger issue there will be water, and for comparison there is more than enough of that under London.

 

[LNW-GW Joint]

Did they renew the TBM's after 15km as per post earlier.

Looks like they used 4 TBMs (2 in each tunnel) and each one bored about 15 miles of tunnel.
No idea if they were renewed en route, but I seem to remember that the Channel Tunnel undersea TBMs each stopped twice for a month or so for a refit.
It involved replacement of the cutting heads and repairs and improvements to the spoil train.
The Channel Tunnel was largely through shallow soft chalk which I imagine might be similar to the Chilterns, rather than hard alpine rock at great depth on the Gotthard.

 

[Grumpy]

The new Gotthard Base Tunnel project reportedly cost CHF 9.5 billion, or about £6 billion.
That produced twin 35-mile tunnels under the Alps, so 70 miles of tunnel fit for 250km/h.
I think that works out at £114m per mile of single bore.

Apologies but isn't £6bn divided by 70 approx £86m? Or am I missing something?

 

[stuu]

The tbm is about £10m, the cost per mile is about £100m. Portals aren’t cheap. The HS2 doc is a good guide but a little old and the prices need to be lifted.

How long does a tunnel need to be for it to need a portal structure - i.e. ventilation and controls and whatever else they include?

 

[LNW-GW Joint]

Apologies but isn't £6bn divided by 70 approx £86m? Or am I missing something?

It certainly is.
I must have mixed up the CHF and km conversions somewhere.
But either way, not massively at variance with £100m/mile estimate.
You'd expect to get more economies of scale with a long tunnel.
And £10m for a replacement TBM is practically noise on the total figure.

 

[Domh245]

How long does a tunnel need to be for it to need a portal structure - i.e. ventilation and controls and whatever else they include?

A portal is just the entry/exit from the tunnel, so even a hypothetical metre long tunnel would need two of them (unlikely as a 1m long tunnel is to exist!)

 

[stuu]

A portal is just the entry/exit from the tunnel, so even a hypothetical metre long tunnel would need two of them (unlikely as a 1m long tunnel is to exist!)

Yes and no, if you look at contracts like HS2 and Crossrail they have separate contracts for portal structures, which are AFAIK much more significant than just the the bit at the mouth of the tunnel. They cost £50m or more so there must be more to it than simply the beginning and end of the tunnel

 

[Domh245]

I can't see any standalone contracts for tunnel portals on HS2. Crossrail have a handful of standalone ones (Royal Oak, Victoria Docks, Pudding Mill Lane), whilst Woolwich is lumped in together with the station box there. The expensive (>£50m) ones at Pudding Mill and Royal Victoria Docks both included moving the DLR around to fit the portals in, which will go some way to explaining the cost. Royal Oak Portal was in the £5-25m category, although I can't see anything about what the actual final cost was

I also stand corrected about a hypothetical metre long tunnel, as HS2's document says that they can be several hundred metres long to transition from ground level to the tunnel level! Quoting:

Portals are approximately 30m wide structures that allow trains to travel between ground level and the bored tunnel headwalls (and vice versa), at gradients no steeper than 3%. The depth of the portal (rail level) at the tunnel’s headwall must be at least twice the diameter of the tunnel, or about 20m, below ground level in order to safely commence bored tunnelling. In flat topography portal structures may be approximately 700m long, but where tunnels pass under a hillside and ground levels are rising, the portal structures will be shorter. Portal structures are typically constructed of concrete and comprise parallel concrete ground retaining walls with a concrete base. In some circumstances the portal is also required to be propped close to ground level using either permanent individual concrete props or a roof slab. A specific design feature associated with high speed trains travelling through tunnels is the need to control the air pressure pulse that is created in front of the train as it enters a tunnel. This is achieved through the introduction of ‘porous’ portal hoods, comprising enclosed portal structures with openings.

 

[Bald Rick]

How long does a tunnel need to be for it to need a portal structure - i.e. ventilation and controls and whatever else they include?

Others have answered, but it depends. Typically a portal is either diving down to the tunnel, or the ground is rising up around it. The former are more expensive, but not that much.

 

[Taunton]

Apart from basic construction costs, tunnels also come with additional operating costs, for drainage (rain and groundwater), flood protection, ventilation, fire protection, emergency lighting, evacuation procedures, etc. Regulations for these have increased notably in recent times.

 

[squizzler]

There is a lot of comparative data for costs of doing things in different countries on Alon Levy's excellent blog. Here is a post comparing the costs of tunnelling (and other structures) for metro construction in New York and Paris.

Tunneling: about $150 million per km vs. $90 million, a factor of 1.7
Stations: about $750 million per station vs. $110 million, a factor of 6.5
Systems: about $110 million per km vs. $35 million, a factor of 3.2
Overheads and design: 27% of total cost vs. 15%, which works out to a factor of about 11 per km or a factor of 7 per station

 

[InOban]

Very different geology.

 

[squizzler]

To which I would ask: How is the geology different and are the easier conditions be to be found in Paris or NYC?

The linked blog - Pedestrian Observations - is strongly of the opinion that the cost differential is for too great to be explained by differing local conditions. The author's work is thorough and covers most aspects of HSR and metro construction costs. Contributors to the various electrification threads lamenting UK costs will find mileage (pun not intended) in his national comparison of wiring costs per km.

 

[DarloRich]

Shall we just agree that the sum is more that a shed load but less than a truck load?

it is surely so varied based on geology, depth, width, water table, ventilation, fire suppression etc etc etc

 

[OxtedL]

Shall we just agree that the sum is more that a shed load but less than a truck load?

Going by the estimates on this thread, even a truckload of credit cards with £10,000 spending limits would not buy you very much tunnel.

 

[Bald Rick]

There is a lot of comparative data for costs of doing things in different countries on Alon Levy's excellent blog. Here is a post comparing the costs of tunnelling (and other structures) for metro construction in New York and Paris.

Tunneling: about $150 million per km vs. $90 million, a factor of 1.7
Stations: about $750 million per station vs. $110 million, a factor of 6.5
Systems: about $110 million per km vs. $35 million, a factor of 3.2
Overheads and design: 27% of total cost vs. 15%, which works out to a factor of about 11 per km or a factor of 7 per station

Very different geology.

Different geology, different station sizes, different systems, different politics.

 

[squizzler]

Different geology, different station sizes, different systems, different politics.

It is not helpful to say that without making some attempt at quantifying what these differences mean. The beauty of the 'Pedestrian Observations' is that its author is a skilled mathematician who can interrogate the data from projects over the world to find out what it is that is accounting for the difference in costs. He is also articulate enough to make it understandable to a lay audience. For amateur scholars of transport provision, of which were are many on this forum, it is a great resource and I make no apologies for keeping bringing it up.

Some variables like the geology are what nature deals you, but the good news is that ground conditions seem not to account for as big a difference as people might think. I got the impression that building around a legacy system (like what UK must do in London) adds more cost.

Stuff we do have control over like the institutional culture and building practices seem to account for the lions share of cost differences. If we have the data to show what savings might be on the table, we can choose whether or not we want to change to implement them. This blogger has lived in big cities both siders of the pond and has had the chance to compare numerous national cultures. Of course there are those who would happily keep paying over the odds rather than admit 'Johnny Foreigner' might be doing things better than us, this blog is not for them

 

[stuu]

Others have answered, but it depends. Typically a portal is either diving down to the tunnel, or the ground is rising up around it. The former are more expensive, but not that much.

I've obviously got the wrong end of the stick somewhere, as I believed 'portal structures' to be more than just the bit where the track enters the tunnel, such as at Pudding Mill lane for crossrail where there is a big cut and cover section and the ventilation plant etc.

 

[MarkyT]

I've obviously got the wrong end of the stick somewhere, as I believed 'portal structures' to be more than just the bit where the track enters the tunnel, such as at Pudding Mill lane for crossrail where there is a big cut and cover section and the ventilation plant etc.

Portal in tunnels is a special usage for the entrance opening. Portal frame is a rigid integrated structure of column and beam, usually in steel, used for spanning a wide area such as in warehouse construction. Historically, portal usually refered to an important doorway in a major building. Must be related to French 'porte' for door.

 

[Bald Rick]

I've obviously got the wrong end of the stick somewhere, as I believed 'portal structures' to be more than just the bit where the track enters the tunnel, such as at Pudding Mill lane for crossrail where there is a big cut and cover section and the ventilation plant etc.

You’re quite right, but for tunnels through a hill or mountain, the track doesn’t need to dive down. The tunnel entrance is still a portal though