Soil Reinforcement in LEM Slope Stability
Introduction
Soil reinforcement — geosynthetics (geotextiles and geogrids), soil nails, grouted tiebacks, and end-anchored bars — stabilizes slopes by mobilizing tensile force across the failure surface. In the limit equilibrium framework, each reinforcement element is a straight line defined by its endpoint coordinates, and wherever a trial failure surface crosses a line, a tensile force is applied to the sliding mass at the crossing point.
Three questions determine how that force enters the analysis, and they are independent of one another:
- How large is the force? — governed by the capacity envelope: the tensile strength of the element, the frictional pullout development from each end, and any end anchorage (plates, connections, anchors).
- In what direction does it act? — governed by the Dir setting: tangent to the slip surface (flexible reinforcement) or along the reinforcement's own axis (rigid supports).
- Is it factored by the safety factor? — governed by the Appl setting: active (a known allowable force, not divided by \(F\)) or passive (an ultimate capacity that mobilizes with the soil, divided by \(F\)).
This decomposition follows the convention used by Slide2 and other commercial programs, which allows xslope results to be compared directly against them. The Type column in the input template is a preset over these settings — selecting a support type fills Dir and Appl with the appropriate defaults — not a separate mechanism.
⚠ TODO (figures): this page needs an annotated figure showing a reinforcement line crossing a slice base, with the force drawn both tangent to the base and along the line axis, and a second figure showing the capacity envelope \(T(x)\) along a line with different end conditions.
Capacity Envelope
Force magnitude at the crossing point
The tensile force available at any point along a reinforcement line is limited by three mechanisms, and the available force is the smallest of them:
\(T(x) = \min\left(T_{max},\;\; T_{end1} + T_{max}\dfrac{d_1}{L_{p1}},\;\; T_{end2} + T_{max}\dfrac{d_2}{L_{p2}}\right)\)
where:
- \(T_{max}\) = tensile capacity of the element (rupture limit)
- \(d_1\), \(d_2\) = distances from the point to end 1 and end 2 of the line
- \(L_{p1}\), \(L_{p2}\) = pullout lengths at each end — the distance over which interface friction develops the full tensile capacity
- \(T_{end1}\), \(T_{end2}\) = anchorage capacity at each end: a bearing plate, a facing connection, or an end anchor (default 0)
Special cases:
- \(T_{end} = 0\), \(L_p > 0\) (the classical friction-only taper): tension is zero at the free end and develops linearly over the pullout length. This is the correct model for geosynthetics and for the embedded end of nails.
- \(L_p = 0\): the end is fully anchored — the full capacity is available immediately at the end.
- \(T_{end} > 0\): the end starts at the anchorage capacity and frictional development adds to it. This models a nail with a bearing plate at the wall face, a geosynthetic connected to facing panels, or a bar anchored at both ends. (If \(T_{end} \geq T_{max}\), the end is effectively fully anchored — the tendon governs.)
- Line shorter than \(L_{p1} + L_{p2}\) with no anchorage: the envelopes from the two ends intersect below \(T_{max}\) and only partial tension is mobilized.
Grouted tiebacks with a bonded length. A tieback develops pullout resistance only over its grouted (bonded) length \(L_{bond}\) at the far end, at a bond strength \(b\) (force per unit length); the free length carries whatever force the bond zone can supply. This is expressed in the envelope by entering an effective \(T_{max}' = \min(T_{tendon},\, b \cdot L_{bond})\) and \(L_p = T_{max}'/b\) at the bonded end, with the connection capacity as \(T_{end}\) at the face end.
Per-unit-width convention and spacing
All LEM forces are per unit width of slope. Geosynthetic properties are already per unit width (kN/m or lb/ft), so for them the Spacing column is left blank (or 1). Discrete supports — nails, tiebacks — have per-element capacities (kN per nail) installed at a horizontal spacing \(S\); enter the per-element values and the spacing, and xslope divides all capacity terms (\(T_{max}\), \(T_{res}\), \(T_{end1}\), \(T_{end2}\), and the FEM stiffness \(EA\)) by \(S\).
Force Direction (Dir)
Where a reinforcement line crosses the base of a slice at point \(r = (x_r, y_r)\), the force \(T(x_r)\) is applied to the sliding mass at angle \(\psi\) from horizontal:
- Tangent to slip surface (\(\psi = \alpha\), the slice base inclination) — the default. Flexible reinforcement cannot resist bending; as the sliding mass moves, the reinforcement deforms with it and the force reorients tangent to the slip surface. This is the appropriate (and conservative) assumption for geotextiles and geogrids, and is discussed by Duncan & Wright (2005).
- Axial (\(\psi\) = the inclination of the reinforcement line itself) — rigid supports such as soil nails, grouted tiebacks, and anchored bars carry their force along their own axis; the soil cannot reorient them. UTEXAS/UTEXASED uses this convention, which is why xslope's tangent results for nail problems differ from UTEXASED's (see the reinforced slope sample, where the UTEXASED axial result is FS = 1.646 versus the tangent 1.587).
The direction affects each solution method the same way the pile force does: the force is resolved into components normal and tangential to the slice base — \(T\sin(\alpha - \psi)\) normal (zero for tangent) and \(T\cos(\alpha - \psi)\) tangential — and for moment-based methods it contributes a moment about the circle center through its real moment arm at point \(r\). For tangent reinforcement on a circular surface the force is tangent to the circle and its moment arm is exactly \(R\), which is why the classical formulation reduces to a bare \(\sum P\) in the OMS and Bishop denominators. The per-method equations are given on the OMS, Bishop, Janbu, force equilibrium, Spencer, and Morgenstern-Price pages.
Force Application (Appl)
Two conventions exist for how a support force enters the factor of safety, and published solutions use both — so the choice is exposed per line:
- Active (Slide2's "Method A", the default): the force is a known, allowable working load. It is applied to the driving side of the equilibrium equations and is not divided by \(F\) — the factor of safety applies to the soil strength only. Appropriate for pre-tensioned supports (tiebacks) and whenever the entered capacity already carries its own safety factor.
- Passive (Slide2's "Method B"): the force is an ultimate capacity that mobilizes together with the soil strength. It is added to the resisting side and is divided by \(F\). Appropriate when the support only develops force as the soil deforms (nails, geosynthetics in some formulations) and the entered capacity is unfactored.
The distinction matters numerically: on the classic Duncan & Wright tieback example (their Fig. 6.34), the same 9,000 lb/ft support gives FS = 1.51 active and FS = 1.32 passive. It also changes what you should enter in the \(T_{max}\) column — an allowable force for active, an ultimate force for passive.
Support Type Presets
The Type column fills Dir and Appl automatically (either can be overridden by typing over the value):
| Type | Dir | Appl | Typical use |
|---|---|---|---|
| Geosynthetic | Tangent | Active | geotextile / geogrid layers |
| Nail | Axial | Passive | drilled and grouted soil nails |
| Tieback | Axial | Active | pre-tensioned grouted anchors |
| Anchor | Axial | Active | end-anchored bars |
Leave Type blank for a generic tensile line with the defaults (Tangent, Active) — the behavior of earlier versions of xslope.
Which sheet models my support?
| Support | Sheet | Settings | Why |
|---|---|---|---|
| Geotextile / geogrid | reinforce | Tangent, Active | flexible; reorients with the soil |
| Soil nail | reinforce | Axial, Passive, \(T_{end}\) = plate capacity | tension-dominated |
| Grouted tieback | reinforce | Axial, Active, \(T_{end}\) = connection capacity | pre-tensioned tension member |
| Micropile / pile / pier | piles | \(H\) (user or Ito-Matsui), \(V_{cap}\)/\(M_{cap}\) | shear and bending govern, not tension |
| Facing weight (shotcrete) | lloads | \(L\) at the face, \(\delta = -90°\) | a load, not a resistance |
LEM vs. FEM
The same reinforcement lines drive both engines, but the mechanics differ:
- LEM applies the capacity envelope value as a prescribed force at the crossing point, in the Dir direction, factored per Appl. The residual strength \(T_{res}\) is not used — LEM has no strain compatibility, so there is no notion of an element loading past peak.
- FEM models each line as tension-only truss elements whose force emerges from displacement compatibility; the same capacity envelope caps each element's allowable force. An element that reaches it yields and holds that force (elastic-perfectly-plastic) — unless \(T_{res}\) has been filled in, in which case it drops to that residual (bounded by the end anchorage in anchored zones). Dir and Appl have no meaning in the FEM. See Soil Reinforcement in FEM.
For typical stiffness values (\(E\), \(Area\)) and guidance on pullout lengths by reinforcement type, see the FEM reinforcement page — the same table serves both engines' inputs.
Typical Anchorage Capacities
Approximate ranges for the \(T_{end}\) columns, for preliminary estimates only:
| End condition | Typical capacity | Notes |
|---|---|---|
| Soil nail bearing plate | 50-150 kN (10-35 kip) per nail | plate punching or facing flexure governs |
| Geosynthetic facing connection | 30-80% of \(T_{max}\) | per connection test data (wrap-around, bodkin, panel) |
| Tieback anchor head / connection | tendon capacity | usually the tendon governs, not the head |
| Free (no plate) | 0 | the friction-only default |
Capacities are per element; with a Spacing entry they are converted to per-unit-width automatically.
References
Duncan, J.M., & Wright, S.G. (2005). Soil Strength and Slope Stability. John Wiley & Sons.
Rocscience Inc. Slide2 Documentation — Support: Active/Passive Force Application; Define Support Properties.
Wright, S.G. (1999). UTEXAS4 — A Computer Program for Slope Stability Calculations. Shinoak Software, Austin.