Which Tunnel is Safer: Less Vertical Cover vs More Vertical Cover?

[ank_gl]

Background I joined a power company in the hydro division and I am currently working in the planning department. I have a bachelor's degree in mechanical engineering & unfortunately most of the things I am facing everyday are related to civil engineering, lots & lots of construction terms & all.(I spend most of the time reading IS codes for concrete & other civil stuff)

Important topic I had a discussion with a colleague(a civil engineer) about tunnel stability with respect to its depth. If the rock mass properties are exactly same & the tunnel dimensions are also same, he says that a tunnel which has lesser vertical cover will be much more safe than one with more vertical cover, because it has to support lesser mass(rock mass between the roof of tunnel & the natural surface land). I have a different take on this, I think the lower tunnel will be much more safer because a lot less tensile stress occurs at the roof of tunnel(I don't know much of soil(or rock) mechanics though).

Safer tunnel => at lesser risk of caving in with same magnitude of lining.

Question to civil engineers: Which explanation is correct?

To Mods: shift it to general engineering if you think so, I thought otherwise.

 

[Nick Bruno]

This is interesting. I am not a civil engineer, but I too graduated with my bachelors in mechanical. In my opinion, you can have a tunnel with no lining. Let's say the top of the tunnel is arched and it holds itself together. If a box culvert is placed in the tunnel obviously the only stress on the top of the box culvert would be from its own weight. I think the strength of the soil around the tunnel and the compactness of the soil greatly determines the stress on the lining. if you dig a big hole, put in a culvert and throw loose soil on top the weight may collapse the lining or culvert or whatever you may want to call it.

In other terms, I think soil can be viewed as a composite, which has different strengths in different directions. I think average strength values could be used with some FEA software and you can predict the stress of the tunnel with minimal accuracy.

Other than that, I am not sure how those civil engineers do what they do.

 

[ank_gl]

This is interesting. I am not a civil engineer, but I too graduated with my bachelors in mechanical. In my opinion, you can have a tunnel with no lining. Let's say the top of the tunnel is arched and it holds itself together. If a box culvert is placed in the tunnel obviously the only stress on the top of the box culvert would be from its own weight. I think the strength of the soil around the tunnel and the compactness of the soil greatly determines the stress on the lining. if you dig a big hole, put in a culvert and throw loose soil on top the weight may collapse the lining or culvert or whatever you may want to call it.

In other terms, I think soil can be viewed as a composite, which has different strengths in different directions. I think average strength values could be used with some FEA software and you can predict the stress of the tunnel with minimal accuracy.

Other than that, I am not sure how those civil engineers do what they do.

Hi Nick

unfortunately I couldn't visualize much of what you said. Anyways, I base my point on two facts:

1. I assumed (being a mechanical engineer) the cover as a uniform load & calculated the stresses at the tunnel roof, turns out that tensile stress is inversely proportional to the cover. But I am not sure if the caving in of the roof can be modeled as a tensile failure, it seems more of a shear failure, but that too I calculated & it turns out to be constant with cover. I think it is getting hard to visualize, let me know if I should put up a drawing & some rough calculations.

2. When I was a kid, I used to play in sand & make sand hills. I remember that deeper I tried to dig in a tunnel, stronger it was against caving in & more I went towards the top, easily it would fail.

Anyways, I am going to the COO after I write this whole thing.

In other terms, I think soil can be viewed as a composite, which has different strengths in different directions. I think average strength values could be used with some FEA software and you can predict the stress of the tunnel with minimal accuracy.

I think soil is pretty much isotropic. Do you mean rock or soil? I am not sure about the FEA thing, internal pressure problems are pretty much easy to handle, external loading problems are more like experiment & write stuff.

 

[Nick Bruno]

1. drawings and rough calculations would be nice to look at. I am confused on the configuration of the tunnel. If the roof caves in and fails due to bending, of course this is a tensile failure because sigma = my/I and there is tensile stress on either side of the neutral axis.

2. I remember going over what to do if you are considering the ground in my FEA class. When modeling the ground in FEA it is important to make the soil domain large enough so that the effect of the load is negligable at the edges. For example, if you are standing on soil, the first 10 elements (near the top) may deflect and then the deflections become excessivley small. In other words, I don't think the depth of the tunnel effects its strength at all, just the load it sees from the top. Applied loads on the top are seen less the deeper the tunnel goes, but then again, more Earth is on the top initially which presses down.

As for the soil, I live in Arizona and the tunnels I have seen are blown out of the side of mountains, not sand. There is a lot of granite and dirt. Assuming this is the case for your Earth matrix, the "soil" may be viewed as orthotropic, rather than isotropic like sand.

Like I said before, this is an interesting topic and I want to know your results when you find some!

 

[FredGarvin]

I would think the soil/rock loading on the tunnel would induce bending stresses on the roof of the tunnel and thus the tensile failure mode.

 

[ank_gl]

I talked to some civil guys, turns out that tunnels don't fail like that at all, provided they are at a sufficient depth. Usually its a rock mass failure, lose rock starts falling from the tunnel roof, if it isn't nicely "gripped" by the rock.
Tunnel stability depends upon the rock mass property. Class I rock(very few cuts, negligible shear zone, high compressive strength etc) can do without a lining(from stability point of view), whereas class V rock(bad rock type) needs a lot of shotcreting & bracing arches to prevent collapsing of roof.
As I suspected, if assumed that rock mass is monolithic & bla bla, the tunnel at more depth is safer in terms of tensile failure & equally good in shear failure. But that is seldom the case(as in case of insufficient rock cover).

I expected a few civil engineers here

NOTE: i don't know much of civil technical terms, so I might be wrong in some definitions.