FemSoilCalc

1. Define the wall x coordinate, domain widths and mesh. Retained soil is on the left; excavation is on the right. y is elevation. Inputs use kN and m.
2. Enter material properties and layer tops in Materials. The first layer starts at ground level and the mesh fits every layer boundary; near-coincident levels require a finer mesh or corrected input.
3. Enter hydrostatic water levels, wall stiffness, normal/shear penalties, friction and explicit wall adhesion (default zero). Horizontal support stiffness and prestress are per metre of wall.
4. Stage 1 describes unexcavated ground. Define subsequent descending excavation levels. Supports are installed at the previous wall position at the start of their activation stage, before that stage's excavation or water changes. An off stage can remove a support or surcharge.
5. A passive-side surcharge stays at its specified surface elevation. Blank means original ground level, so excavation below that level removes the surcharge. Enter the excavation elevation to load a later platform.
6. All stages are calculated in their original order; Analyse selects stages for reporting and the envelope. Warnings check every stage. Review force and moment balance, wall axial/shear/moment actions, support reactions, principal strain and effective-stress mobilisation. Save the input model and export a separate result report after running.

Elevation increases upward. Retained ground is on the left and the excavation is on the right. Positive horizontal movement is toward the excavation. Stored stress is compression-negative. Mobilisation uses compression-positive principal stresses.

Soil elements are constant-strain triangles. The wall is an Euler–Bernoulli beam. Soil nodes on the two faces are separate above the toe and join only at the toe. Results are accepted only when contact equilibrium and global force and moment balance pass.

The domain screen warns if a side width is under 3H or the base is shallower than the larger of H and twice the deeper embedment. H is ground minus formation. The warning does not stop the solve. One finer rerun of the governing reported stage does not replace the published result.

Drained initial stress is effective: pore pressure is subtracted and both horizontal stresses are K₀ times the vertical effective stress. Undrained initial stress is total and uses c as cu. A blank saturated unit weight keeps one bulk weight. A retained rise changes the overburden only when a rise is entered.

Every stage is solved, including an unticked stage. A prop or anchor is installed at the movement already reached at the start of its stage. Unilateral supports drop to zero when the wall moves away from the excavation. An anchor angle is 0 to 80 degrees below the horizontal toward the retained soil.

Elastic soil calculates no yield correction and no factor of safety. If any stage reaches mobilisation 1 or tension, design use is withheld. Drained Mohr–Coulomb uses ψ from 0 to φ, with blank ψ equal to 0, and a tension cutoff. A saved file with no soil model stays elastic. Clear All starts as Mohr–Coulomb. Built-in examples are elastic unless a material is undrained.

Hydrostatic mode imposes a head on each side. Steady seepage needs a permeability on every material, an impermeable wall above the toe, and an equal-head first stage. The heave screen ignores a single layer whose top is at the original ground. Weight over uplift below 1 withholds hydraulic use. Piping is checked only in steady seepage.

Strength reduction stays off unless its switch is ticked. An old numeric value is cleared to zero and does not turn the switch on. The search caps at 3 and does not reduce wall or support stiffness. Diagrams on screen and on the A4 print come from the same figure. The Rankine δ = 0 line is a reference, not a pass or fail. Export Results writes JSON.

Use Load example model. Units are kN and m. A status line that says FEM v6 complete has accepted the reported stages. A status line that says Run failed has accepted nothing from that attempt.

1. Example 1, Braced excavation - dense sand, job FEM-EX-01. Ground 105, formation 100, toe 95.5, bottom 91, retained width 18 m, excavation width 16 m, dx 1.0 m, dy 0.5 m. Medium dense sand from 105 (γ 18, φ 32°, K₀ 0.47) over dense sand from 99 (γ 20, φ 38°, K₀ 0.38). Soil model Elastic. Both water levels 92. One prop at elevation 103, stiffness 50 000 kN/m per m, installed at stage 3, after a dig to 102.5 and before the final dig to 100. Surcharge 10 kPa from x −15 m to −1 m. Design use is withheld. Read movement and mobilisation, and treat moment and prop load as an elastic extrapolation.
2. Example 2, Cantilever wall - loose over dense sand, job FEM-EX-02. Ground 104, formation 100, toe 95, bottom 91, widths 16 m and 14 m. Loose sand from 104 over dense sand from 99.5. No prop. Stage 1 water is 102 on both sides. Stage 2 excavates to 100 with retained water 102 and excavation water 99.5. The domain screen warns and the solve continues. The dense-sand top is below formation, so the final stage fails the heave screen and both hydraulic use and design use are withheld.
3. Example 3, Clay propped cut - undrained, job FEM-EX-03. Ground 106, formation 101, toe 95, bottom 90. Soft clay cu 25 kPa and firm clay cu 55 kPa, both with φ 0, K₀ 1 and drainage undrained. Props at 104.5 (prestress 20 kN/m, stage 2) and 102.2 (stage 3). Retained water stays at 105 while the excavation water follows the dig. A completed run uses the undrained Tresca statement and an equilibrium residual below 1e-7. If mobilisation reaches 1, design use is withheld. A run that does not converge accepts no result.
4. Save Data writes a version 6 file. An older file with no soil model opens as elastic. Clear All starts a new Mohr–Coulomb model. Verification Checks do not validate plastic soil.

Plane-strain constant-strain triangles with layer boundaries fitted to the mesh; compression is negative.

Initial vertical total stress is integrated from soil unit weights. Initial effective stresses subtract hydrostatic pore pressure; both horizontal initial effective stresses use K₀.

Total equilibrium uses initial effective stress plus elastic strain increments and the current pore pressure. Excavation removes soil elements; remaining initial stresses release through the exposed boundaries.

Independent soil nodes on both wall faces, joined beneath the toe. Normal contact is compression-only; tangential contact has a friction/adhesion cap (explicit wall adhesion, default zero) and carries plastic slip history. Water traction acts separately from effective soil contact.

Horizontal supports use displacement since installation. Unilateral supports resist positive wall movement towards the excavation; bilateral springs can also reverse their force. Supports and surcharges can be removed at a specified stage. Stiffness and prestress are per metre of wall.

All construction stages are equilibrated, including stages unticked for reporting. Linear and contact force residuals and global force and moment balance must pass before results are accepted.

Wall moments and shears come from beam element end forces. Ux/Uy envelopes include only active soil and wall nodes.

Elastic soil. No factor of safety. Hydrostatic water only. Wall moment and prop load are withheld where mobilisation reaches 1 or effective stress is tensile.

Drained Mohr–Coulomb with a tension cutoff. No factor of safety. Hydrostatic water only. Undrained total-stress Tresca using cu. No excess pore pressure and no factor of safety.

Steady seepage, or imposed heads with a heave screen. No consolidation and no factor of safety. Initial stress is K₀ from the depth under the surface. No slope-stability factor of safety.

Anchor force is axial over the free length. Bond and grout are not checked. Results are per metre of wall. Two plane-strain walls, per metre. A strut is an axial member between the walls, not a waling or a frame.

c–φ or cu reduction after construction at full strength. Not an EC7 partial factor and not a design factor of safety. One finer rerun of the governing stage. Not a full mesh-convergence study.

Side boundaries have fixed horizontal displacement and the base is fixed in both directions. CST elements can be too stiff near incompressibility; ν ≥ 0.49 is rejected. Direct dense solves are limited to small models. These checks do not validate plastic soil. Treat the results as preliminary until independently checked against a validated geotechnical model.

Widths are measured from the wall x coordinate. Retained soil is to the left; excavation is to the right. y is elevation.

The top layer starts at ground level. Every layer top is included in the mesh. Blank saturated unit weight uses the one bulk weight. Blank drainage is drained. For undrained soil, c is cu and φ and ψ are ignored. A blank dilation angle is zero and cannot exceed φ.

Anchor force is axial over the free length. Bond and grout are not checked. Results are per metre of wall. A strut is an axial member between two walls, not a waling or a frame. Blank second-wall properties copy the first wall.

A passive-side surcharge stays at its entered surface elevation. Blank means original ground level; it is removed once excavation passes below it. To load an excavated platform, enter that stage’s excavation elevation. Off stage removes the load before that stage. Imposed hydrostatic heads on each side. No flow under the wall. Steady seepage needs a permeability on every material and equal heads in stage 1.

Every construction stage is calculated in its original order. “Analyse” selects stages for the results envelope; unticked stages still carry their construction history. Supports are installed at the start of their activation stage, at the previous wall position, before excavation or water changes in that stage.

Use these sections to locate the inputs, checks and results shown in the app:

• Project
• Geometry
• Materials
• Wall / Interfaces
• Loads
• Stages
• Summary
• Wall Results
• FEM Results
• Theory