Pile retaining wall design methods - US/Canada

[DougHole]

Hi guys,

just out of curiosity, just wondering what everyone in the US and Canada are using for piled wall basement design?

By this i mean bored reinforced concrete piles spaced at generally less than 3D centres e.g. 600 mm diameter piles at 1800 mm centres? and temporary props (raking props/flying shores) and permanent props (basement and ground floor slabs).

Are you using WALLAP, LPILE, hand calcs??

 

[Guest262653]

I'm based in Europe, but I design deep excavations (6 or 7 underground storeys, diaphragm walls or concrete piled walls, steel struts and/or ground anchors) always with a combination of three design levels:
- hand calcs to feel the problem and establish the geometry;
- beam model with uncoupled elasto-plastic springs, taking into account the construction sequence (using a VBA program in an Excel spreadsheet) to get a better force envelope in the wall and to check the evolution of anchor forces;
- 2D plane strain FEM analysis of the problem (Plaxis, Phase2 or equivalent software) to properly model seepage, get the final design forces, assess wall and surface displacements, etc. If no sensitive structures exist behind the wall, usually a simple Mohr-Coulomb material model is sufficient.

I've never used 3D analysis. Yet.

For one or two basement storeys, I usually stop at level 2.

 

[DougHole]

thanks Avscorreia - can you clarify what you mean by "beam model with uncoupled elasto-plastic springs, taking into account the construction sequence (using a VBA program in an Excel spreadsheet) to get a better force envelope in the wall and to check the evolution of anchor forces;"?

 

[Guest262653]

It's basically an extension of a beam on elastic foundation FEM model. This is pretty much a standard for dedicated embedded wall programs (Deepex, Geo5, Deltares, Frew, etc).

Breaking it down:
- Finite element beam model representing the wall;
- Soil is modelled as elasto-plastic springs. Their behaviour is independent of each other. Maximum force is dependendent on effective stress and direction of movement (active or passive);
- Anchors or struts modelled as elasto-plastic springs;
- Construction sequence is modelled by different load steps/increments and spring activation/deactivation.

For a detailed explanation, take a look at the manuals of the softwares I mentioned or at "Computational Geomechanics", Verrujt, 1995 or "Deep Excavation, Theory and Practice", Chang-Yu Ou, 2006.

 

[DougHole]

ok thanks. It sounds very much like WALLAP which is what i am used to using.In WALLAP, the basement floor slab props and temp props (raking/anchors, etc.) are modelled as springs. you put in your free length, youngs modulus, pre-stress, angle, cross sectional area, etc. and it calculates an equivalent spring