Design and Practical Application of Soilex Anchors

Design and Practical Application of Soilex Anchors K. R. MASSARSCH '>, K. OIKAWA 2), Y. ICHIHASHI 3>, M. SAT0 4), s. ARONS soW) . .-\~D S. WETTERLING

5 )

11 Geo Engineering AB, Stockholm, Sweden, 2) Kobe Port Construction Office, Kobe, Japan, 3) Kajima Corporation, Tokyo, Japan, 4) Chemical Grouting Company, Tokyo, Japan, 5) Soilex AB, Stockholm, Sweden

~TRODUCTION

Soil anchors are a widely used method for the stabilisation of slopes and retaining structures wd to resist up-lift forces. They offer in many cases technically better, and more economical solution for temporary and permanent structures. An advantage of soil anchors, compared ~ith other foundation methods is, that the capacity of most anchors is checked after installation, Massarsch (1994). Thereby, the design of soil anchors with respect to capacity and deformations can be verified. The interaction between soil anchors, the retaining structure md the soil can be complex and is influenced by many factors. The load distribution behind a ~taining structure depends on the method of excavation and the resulting deflections of the wall. Secondary factors such as construction activities in the vicinity, static and dynamic loads. variations of ground water level or changes of temperature can affect the performance of soil anchors. An important aspect, which is often overlooked by design engineers is, that the interaction between the anchor and the soil is also affected by the type and installation method of the anchor. Similar to piles, it is important to consider in which way the load is Ir.lI1sferred from the anchor to the soil (along the skin of the anchor and/or at the base/head of the anchor), Figure I.

Figure I. Load transfer of conventional friction anchor and Soilex Anchor In the present paper, the design and practical application of the Soilex Anchor will be presented. The design can be carried out similar to that of piles subjected to tension forces. Extensive experience has been obtained from practical applications of the Expander Body for temporary and permanent soil anchors. which has made it possible to verify the design Ground anchoralles and anchored structures. Thomas Telford. London, 1997.

Massarsch, K. R., Oikawa, K., Ichihashi, Y., Sato, M., Aronsson, Sand Wetterling, S. 1996. Design and Practical Application of Soilex Anchors. International Conference on Ground Anchorages and Anchored Structures, 20 - 21 March 1997, Institution of Civil Engineers, London, United Kingdom, 11 p.

218 CONSTRUCTION

concept. Case histories will be presented which describe the innovative applications of the Expander Body. THE EXPANDER BODY The Expander Body, which was developed more than 15 years ago, offers new possibilities for the application of soil anchors . It consists of a thin, folded steel tube which can be installed by means of drilling, driving, vibrating, jacking, or placement in a pre-drilled hole After installation, the Expander Body can be inflated by injection of concrete or grout. In thi~ way, a solid, concrete-filled steel body with pre-determined shape and volume can be created at any depth in the ground. During the expansion phase, and as a result of the high expansion pressure, a densified soil zone is created, which extends to a distance of about two Expander Body diameters. The expansion process increases the density of loose and medium dense granular soil. In cohesive soils, re-compression results in a pre-consolidation effect and improves soil strength and stiffness. The Expander Body can be used to improve the bearing capacity of driven and bored piles (Massarsch and Wetterling, 1993, Terceros et aI., 1995). In the present paper, the applications of the Expander Body for temporary and permanent soil anchors will be presented. Temporary and permanent Soilex Anchors The anchor consists of the concrete-filled Expander Body into which strands or bars art inserted. The strands can be installed individually or in bundles. The load is transferred safely by adhesion between the strands and the concrete inside the steel-protected anchor unit. Alternatively, a tension bar can be screwed into the bottom of the Expander Body, thereby providing a unique load transfer mechanism, which generates only compression forces in the concrete body of the anchor, Figure 2.

Figure 2. Permanent Soilex Anchor with strands or with tension bar. anchored to the base of the Expander Body (fully corrosion-protected) A new type of temporary soil anchor has been developed which has several import:lfl: advantages compared with traditional soil anchors, Figure 3. The anchor strands are foldc~ around the outside of the Expander Body. The volume of the injected anchor can thus nc reduced and the anchor is thus more economical. Injection is easier and faster and tensioninf

MASSARSCH et al. 219

of the Expander Body can be carried out within 24 hours after grouting. Another, very important advantage is that the strands can be extracted after the anchor is no longer needed.

Figure 3. Temporary Soilex Anchor with extractable strands Qualitv control The unique features of the Expander Body make it possible to monitor accurately the grouting process, which provides valuable information regarding the soil conditions and thus the bearing capacity of the anchor after installation. An electronic monitoring system was developed, the main features of which are shown in Figure 4. The grout flow and pressure from the injection pump are recorded and displayed by an electronic measuring unit on site. AU relevant data are stored on a computer and can later be evaluated and interpreted at the office using specially developed software. In this way, all phases of the Expander Body installation can be monitored and documented accurately.

Figure 4. Electronic monitoring system for anchor installation The size and shape of the Expander Body after grouting are known, which eliminates uncertainties regarding anchor design. The soil properties adjacent to the Expander Body can be detenn.ined in situ, based on measurements of the injected grout volume and grout pressure. Based on the concept of the pressuremeter test, a design method has been developed by the Swedish Pile Commission for Expander Body design, (Expanderkroppar, 1988).

220 CONSTRUCTION

ANCHOR DESIGN The bearing capacity of the anchor can be designed using conventional geostatic methods. However, it has been found that the ultimate (failure) load can be estimated reliably based on the cone penetration test (CPT). The bearing capacity is composed of the base resistance. p. and the skin resistance, Ps' (I)

Because of the shape and installation process of the Expander Body, the bearing capacity is governed in most cases by the base resistance. The base resistance is mobilised gradually and reaches the ultimate load at deformations on the order of 10% of the Expander Body diameter. The skin friction is usually mobilised at smaller deformations and is almost independent of the diameter of the Expander Body. In most soils, the bearing capacity of Soilex Anchors is ductile, compared to the brittle behaviour of friction anchors. The electric CPT can measure the tip resistance, the sleeve friction and the local pore water pressure simultaneously. The total resistance P u of the Expander Body can be calculated from:

(2) where: qc: cone resistance of the CPT, ~ : end bearing area of the Expander Body, K, : reduction factor, which depends on soil type, cf. Table 1, f,: sleeve friction from CPT, A,: skin area of the Expander Body and k2: reduction factor for tension load. The reduction factor k, takes into account the geometric differences between the cone penetrometer and the Expander Body and is dependent on soil type, cf. Table 1. If the skin friction is not measured. the value of the local sleeve friction f, can be estimated from: (3)

where FR is the friction ratio (%), which depends on soil type (cf. Table 1) and on the streSS conditions of the soil deposit. It should be noted that the estimated value gives a first approximation only and needs to be verified by further geotechnical investigations. The reduction factor k2 takes into account the reduced bearing capacity of piles and anchors subjected to tension loads and is typically k2 = 0,8. Table 1. Typical values of bearing capacity correction factor and friction ratio FR Soil type

Particle Size, 0 50

k,

~m)

. Clay Clayey silt Silt Silty sand Fine sand Sand Gravel

0.001 0.005 0.01 0.05 0.1 0.5 1

Friction ratio, FR ~

0.8 0.7 0.6 0.5 0.5 0.5 0.5

5 4 3 1 0.8 0.5 0.3

In many countries, other types of penetration tests are used. such as the Standard Penetration Test (SPT), heavy ram sounding (SRS) or Weight sounding. Therefore, for prelimin~

MASSARSCH et al. 221

estimates, the bearing capacity of Expander Bodies can be determined using conversions between the respective penetration test . and the cone penetration test. Several correlations have been published between SPT N-value (blows/ft) into CPT g,value (MPa). The following correlation proposed by Robertson et al. (1983) is used, Table 2: a.c =a N

(4)

where a. is an empirical factor, based on the correlation proposed by Robertson et al. (1983). Table 2. Correlation between CPT (MPa) and SPT (N-value), Robertson et al. (1983) Soil type Clay Clayey silt Silt Silty sand Fine sand Sand Gravel

0'0 (mm)')

a

0.001 0.005 0.01 0.05 0.1 0.5

0.1 0.15 0.2 0.3

1

O.B

0.4

0.55

I) particle size of 50% passing in the grain size curve The corresponded q,-value can be calculated from SPT N-values and is inserted in equation (2). For a reliable correlation between CPT and the N-value, it is important to know the soil type and in-situ stress conditions on the site. PERMANENT SOILEX ANCHOR, SODERTALJE, SWEDEN At the city of S6dertiilje, south of Stockholm a channel links the Baltic Sea with lake ·'Miilaren". Permanent sheet pile walls of type Larssen IV were installed in order to increase the stability of the slopes along the channel. These were retained with 125 fully corrosionprotected, permanent Soilex Anchors with tension bar, cf. Figure 2. The channel follows an esker which passes through the central part of S6dertiilje and has a very complex geotechnical structure, consisting of material with alternating layers of gravel, sand and fine sediments of silt and clay. The geotechnical conditions vary both in the vertical and horizontal direction. Occasionally, boulders were encountered. This caused significant difficulties when assessing the bearing capacity of the Soilex Anchors. Because of the difficult and variable soil conditions, extensive field investigations, including ram sounding and Swedish Weight Sounding were performed. Because of the occasional, high penetration resistance (hard layers and boulders), electric cone penetration tests could not be used. However, based on a typical Weight Sounding record, the equivalent cone penetration record could be established, Figure 5. Generally, the relative density of the soil deposit increases with depth, even though local variations exist. Elements of fine sand, silt and clay occur both close to the .surface as well as at greater depths. The 10-12 m long sheet piles were driven into the dense sand and gravel, using a vibration hammer which was operating at 38 Hz. A drill rig type Atlas Copco ROC 60 I was mounted on a barge in the channel from where the Expander Bodies were installed at an angle of30°. The required length of the anchors was on average 16 m, at a depth of 12 m below the ground surface. At some locations the soil was too dense and had to be pre-drilled. Soilex anchors, consisting of EB 512 (expanded diameter500 mm and initial length 1.2 m) with double-

222 CONSTRUCTION corrosion protected bottom coupling and pipes 4>88.9x5 nun were used. A GWS-bar 4>32 mm was screwed to the bottom coupling of the Expander Body. The GWS-bar was sand-blasted. coated with epoxy tar and grease and encapsulated into a plastic (PEM) hose. The bar ':Vas inserted into the pipe and Expander Body and finally covered with concrete. This installation procedure was conducted in order to obtain a full-proof corrosion protection. Cone resistance, MPa 10 0.0 -2.5

13

15

18

20

I

c

I I I

-5.0

-;-=;:

i

I

I

I

I

~

I

I

I

I

I

I

E

,

I

i

, :

~ i ~

· 12.5

j

i

.:;;;;;;

i

-15.0

I

i

:

I

i

P

,

i

!

Figure 5. Equivalent cone penetration record, based on Swedish Weight Sounding test The injection pressures varied typically between 1.5-3.0 MPa. The data acquisition system illustrated in Figure 4, was used to measure grout flow and injection pressure. A typical recording ofvolurne and pressure for the Expander Body, EB 512 is shown in Figure 6. ~----------~-----------------------r--~-,175 150

2,5 (;'

c-

2

...

1,5

~ Q)

100

=.. Q)

:::J

75

!:.! c-

50