Pile Negative Friction Eurocode


*DWL = Design Working Life

*DWL = Design Working Life

Eurocode 7 requires consideration of both short- and long-term design situations, to reflect the sometimes vastly different resistances obtained from drained and undrained soils. At first sight, the requirements of EN 1997-1 appear to cut across those of EN 1990. However, it is not difficult to combine these ideas to cater for common geotechnical problems:

Design Real Term Example situation conditions

Persistent Normal Long Buildings and bridges founded on coarse soils and fully-drained fine soils

Short Partially-drained slope in fine soils (with DWL less than 25 years)

Transient Temporary Long Temporary works in coarse soils

Short Temporary works in fine soils

Accidental Exceptional Long Buildings and bridges founded on coarse soils and quick-draining fine soils

Seismic Short Buildings and bridges founded on slow-draining fine soils

3.4.2 Geotechnical actions

EN 1997-1 lists twenty different types of action that should be included in geotechnical design. These include obvious things, such as the weight of soil, rock, and water; earth pressures and ground-water pressures; and removal of load or excavation of ground — and less obvious things, such as movements caused by caving; swelling and shrinkage caused by climate change; and temperature effects, including frost action. [en 1997-1 §2.4.2(4)]

Retaining structures are often subject to a large set of actions, including the weight of backfill, surcharges, the weight of water, wave and ice forces, seepage forces, collision forces, and temperature effects. [en 1997-1 §9.3.1]

Likewise, many slopes are subject to a large set of actions, including previous or continuing movements from vibrations, climatic variations, removal of vegetation, and wave action. [en 1997-1 §11.3.2(P)]

Embankments suffer the erosion effects of overtopping, ice, waves, and rain on their slopes and crests. [en 1997-1 §12.3(4)P]

In situations where structural stiffness has a significant influence on the distribution of actions, for example in the design of raft foundations or pile groups, that distribution should be determined by soil-structure interaction analysis. [EN 1997-1 §6.3(3) and 7.3(4)]

Consolidation, swelling, creep, landslides, and earthquakes can all impose significant additional actions on piles and other deep foundations. When considering these effects, the worst situation may involve upper values of ground strength and/or stiffness. [EN 1997-1 §]

For example, assessment of the upper values of ground strength may be important when determining the potential loading on a pile due to negative skin friction. In this situation, the stronger the ground, the greater the load that acts on the pile. The design, therefore, should consider upper values of strength (the 'superior' values discussed in Chapter 2) when assessing design actions and lower values for resistance ('inferior' values).

3.4.3 Distinction between favourable and unfavourable actions

The Eurocodes make an important distinction between favourable (or stabilizing) and unfavourable (destabilizing) actions, which is reflected in the values of the partial factors yf applied to each type of action. As discussed in Chapters 6 and 7, unfavourable/destabilizing actions are typically increased by the partial factor (i.e. yf > 1) and stabilizing actions are decreased or left unchanged (i.e. yf # 1).

Consider the design of the T-shaped gravity retaining wall shown in Figure 3.3. To provide sufficient reliability against bearing capacity failure, we should treat the self-weight of the wall and the soil on top of its heel (W) as unfavourable (since it increases the effective stress beneath the wall base) — but as favourable for sliding and overturning (since it reduces the effective stress beneath the wall base and increases the clockwise restoring moment about 'O').

The imposed surcharge q is unfavourable for bearing, sliding, and overturning where it acts to the right of the virtual plane shown in Figure 3.3. But where it extends to the left of the virtual plane, it has the same effect as the wall's self-weight W.

Unfortunately, the distinction between favourable and unfavourable actions is not always as straightforward as this. Consider next the vertical and horizontal thrusts Uv and Uh owing to ground water pressure acting on the wall's boundaries. The horizontal thrust Uh is unfavourable for bearing, sliding, and overturning; whereas the vertical thrust Uv is favourable for bearing (since it helps to counteract W), but unfavourable for sliding (reducing the effective stress beneath the wall base) and overturning (since it increases the anticlockwise overturning moment about 'O'). But it is illogical to treat an action as both favourable and unfavourable in the same calculation — how can the horizontal component of the ground water pressure be treated differently to its vertical component?

Retaining Wall Design Example Eurocode

Favourable or unfavourable?

Figure 3.3. Examples of favourable and unfavourable actions

Eurocode 7 deals with this issue in what has become known as the 'Single-Source Principle' (although in fact it is merely a note to an Application Rule):

Unfavourable (or destabilising) and favourable (or stabilising) permanent actions may in some situations be considered as coming from a single source. If... so, a single partial factor may be applied to the sum of these actions or to the sum of their effects. [en 1997-1 §2.4.2(9)P note]

This note allows the thrusts Uh and Uv to be treated in the same way — either both unfavourable or both favourable, whichever gives the more onerous design condition.

The Single-Source Principle has a profound effect on the outcome of some very common design situations, as explained more fully in Chapters 9 to 14. It also precludes the use of submerged weights in design calculations: by replacing the gross weight W and water thrust Uv in Figure 3.3 by the submerged weight W' = W - Uv, the choice is implicitly made to treat both the self-weight and the water thrust as favourable or unfavourable — which does not tally with our discussion of their 'favourableness' given above.

3.4.4 Should water pressures be factored?

According to Eurocode 7, for ultimate limit states:

design values [of groundwater pressures] shall represent the most unfavourable values that could occur during the design lifetime of the structure. [EN 1997-1 §]

whereas for serviceability limit states:

design values shall be the most unfavourable values which could occur in normal circumstances. [en 1997-1 §]

To many engineers, the first definition conjures up the idea of 'worst credible' groundwater pressures, i.e. the most adverse water pressures that are physically possible during the structure's working life. The second definition suggests less severe conditions, e.g. the most adverse water pressures that are likely to occur without something exceptional happening. Extreme water pressures 'may be treated as accidental actions'.

In many cases, water pressures are calculated from an assumed water level, which should therefore be the most unfavourable water level that could occur. Eurocode 7 goes on to say:

Design values of ground-water pressures may be derived either by applying partial factors to characteristic water pressures or by applying a safety margin to the characteristic water level [en 1997-1 §]

The interpretation of this Application Rule is an issue that provokes strong opinions from the engineers with whom we have discussed the matter. Figure 3.4 shows some of the possible interpretations of §

Highes! possible Highest normal water level water level

Highes! possible Highest normal water level water level

Figure 3.4. Effects of factoring water pressures

Consider a gravity wall retaining water as shown in Figure 3.4(a). Based on a knowledge of the ground's hydro-geology, we might identify the highest water level expected behind the wall 'in normal circumstances' and the highest possible water level 'during the design lifetime of the structure'. On Figure 3.4, the characteristic water pressures are shown by triangle (b) and possible design pressures — depending on your interpretation of § — by 'triangles' (c)-(f).

In Figure 3.4(c), the water pressure is regarded as being at its design value when its level is at its highest possible and hence no factor is applied to it (i.e. y = 1.0). In Figure 3.4(d), the water pressure is treated as a permanent action and factored by yg = 1.35.1 In Figure 3.4(e), the additional pressure due to water rising from its highest normal to its highest possible level is treated as variable and factored by yq = 1.5, while the remaining pressure (hatched) is treated as permanent and factored by yg = 1.35. In Figure 3.4(f), all the water pressure is treated as variable and factored by yq = 1.5.

The differences between (d), (e), and (f) are relatively minor and of less importance than selecting suitable water levels in the first place. However, the choice between (c) and one of (d) to (f) depends on how you answer the question should water pressures be factored? This question provokes strong debate. For many geotechnical engineers, it is illogical to apply a partial factor to a quantity whose ultimate value is relatively well known (particularly if the highest possible water level is placed at ground surface). For others, it is illogical to treat water pressures any differently to other actions, especially effective earth pressures, which are normally factored by yg. From a practical point of view, applying factors to effective earth pressures but not to water pressures would make numerical analysis extremely difficult (if not impossible).

Two arguments favour the approach of applying factors to water pressures. First, structural engineers have traditionally applied partial factors between 1.2 and 1.4 to retained liquid loads4 and ground water pressures5 (1.2 where the maximum credible water level can be clearly defined; otherwise 1.4). This practice is continued in Eurocode 1 Part 46 for liquid induced loads during tank operation (where yf = 1.2) and in the head Eurocode7, which gives yg = 1.35 for permanent and yq = 1.5 for variable ground- and free-water pressures. If the design omits these factors, then it must make compensating adjustments to the structure's partial material factors in order to attain the same level of reliability.

fChapter 6 discusses the Design Approaches in which these factors apply.

Second, it is common for geotechnical engineers to perform numerical analyses using unfactored parameters and then to apply a factor of safety to the resultant structural effects, such as bending moments and shear forces. In doing so, the analyst has implicitly applied the same partial factor to water pressures as to effective earth pressures. If you perform a calculation with the partial factor yg = 1.35 applied to effective earth pressures only (and not to water pressures), you will get different bending moments and shear forces from those obtained from the numerical analysis.

The main argument against applying factors to water pressures is that it results in physically unreasonable values, a feature that is exacerbated by the Eurocodes' use of factors between 1.35 and 1.5, rather than the traditional 1.2-1.4.

One approach that provides a balance between providing reliability and maintaining realism in the design is shown in Figure 3.5.

Ground level Highest normal Highest possible

\ water level water level

Ground level Highest normal Highest possible

\ water level water level

Characteristic Design Water Design Water

Condition 1 Condition 2

Figure 3.5. Recommended treatment of water pressures for design

Characteristic Design Water Design Water

Condition 1 Condition 2

Figure 3.5. Recommended treatment of water pressures for design

When partial factors yg > 1.0 are applied to effective earth pressures, then pore water pressures should also be multiplied by yg > 1.0 but calculated from highest normal (i.e. serviceability) water levels — i.e. no safety margin is applied. On Figure 3.5, we have called this 'Design Water Condition 1'.

Alternatively, when partial factors yg = 1.0 are applied to effective earth pressures, then pore water pressures should also be multiplied by yg = 1.0 but calculated from highest possible (i.e. ultimate) water levels — after an appropriate safety margin has been applied. Figure 3.5, this is 'Design Water Condition 2'.

The table below summarizes this approach and indicates the relative magnitudes of the water thrusts acting against the wall.

Limit state DWC*

Partial factor Yg

Safety margin Ahw

Water thrust

Characteristic -

+1 -2


  • Marco Beyer
    What is favourable unfavourable eurocode?
    8 years ago
  • semolina
    What is the difference between favourable and unfavourable action in eurocodes?
    6 years ago

Post a comment