of a pile group is usually less than the capacity of a single pile multipled by the number of piles in the group. However, in granular soils, when displacement piles (piles that displace a significant amount of soil during driving) are used, the capacity of a group of piles is equal to or sometimes greater than the capacity of a single pile multiplied by the number of piles in the group. A helix anchor is not a "displacement" anchor and displaces very little soil volume when installed; as a result the density of the soil in the vicinity of the anchor does not increase appreciably.

Many efficiency ("efficiency" is the ratio between the group capacity and the sum of the capacities of the individual piles) formulas are available for the unwary engineer to use in computing the ultimate capacity of a pile group. None of these formulas take into account the factors that actually govern the ultimate capacity of the pile group itself. Likewise, there are no hard and fast rules to use when predicting the pull-out capacity of multiple anchors. Based upon load tests [18] it can only be concluded that multiple anchors are not 100 percent efficient.

Hanna

et al. [45] performed pull-out tests of models in sands. They concluded that when the anchor spacing was approximately 4 diameters, ultimate capacity of the anchor group was nearly 100 percent of the capacity of a like number of individual anchors. Further, they stated that the group theory proposed by Meyerhof and Adams [74] predicted trends of group behavior, but they felt it was in considerable error.

Adams et al. [6] tested a group of four multiple helix anchors in a cohesive soil. The anchors were installed to a depth of 14.5 feet (4.42 m) at the corners of a 3.5-foot (1.1-m) square. Each anchor had 3 helixes each measuring approximately, 1 ft (0.3 m) in diameter. Load tests performed on both the single anchor and the group anchor showed that the group effect was fairly insignificant and the efficiency was greater than 90 percent.

The effects of load cycling and group action on the uplift capacity have been investigated and it was found that if the load levels acting on the anchors are below 50 percent of load capacity, the cyclic loading does not lead to excessive displacements [5].

Facing page: Mobile home with factory installed anchor straps.

6.1 GENERAL

Several empirical approaches are presently used to predict anchor pullout capacity. These fall generally into three categories:

1. Correlation of pull-out capacity with in-situ tests

3. Field tests on full-scale or prototype anchors in similar

soil conditions

6.2 CORRELATION OF PULL-OUT CAPACITY WITH IN-SITU TESTS

6.2.1 Soil and Rock Class Types and In-situ Tests

At present, various soil types have been grouped based on the soils potential ability to provide anchor restraint. What we are about to describe is not a complete soil classification system as was described in section 5.2, but a means of classifying the given soil's ability to hold a certain size anchor. Table 6.1 shows the National Fire Protection Association (NFPA) Standard 501A-1975 [69] soil type definitions. Also shown in this table is the correlation between blow count as measured by the Standard Penetration Test (SPT) [12] and torque in inch-pounds as measured by the "Soil Test Probe (STP)." The "soil test probe" is a patented commercial device, consisting of a helix of 10.75 in (273 mm) overall length 1.25 in (32 mm) major diameter and 0.56 (9/16) in (14.3 mm) minor diameter which is attached to a 0.56 in diameter shaft which is turned by a torque wrench. The torque repaired to insert this probe has been empirically related to anchor pull-out capacity. The test probe is currently being used by industry to predict anchor pull-out capacity and has been further investigated by Fry and Hollander [40, 41].

The Standard Penetration Test is described in section 5.2. It should be noted that the ASTM standard for this test [12] is not explicit enough to ensure that the testing equipment is operated in such a way that energy delivered to the sampler by the hammer does not vary. Until this deficiency is remedied, it is recommended that the following precautions be taken: the hammer used to advance the sampler be raised by wrapping a rope around a pulley (cathead); operators wrap the rope either twice or three times around this pulley; two wraps of rope around the cathead should be used when performing Standard Penetration Tests as it has been found that the number of turns greatly influences the amount of energy imparted to the sampler [62].

Both the SPT and STP tests indirectly measure the shear strength of the soil, one of the key parameters in determining anchor pull-out capacity.