The Behaviour of Plate Anchor Groups in Sand

Wang, T. C. W (1986) The Behaviour of Plate Anchor Groups in Sand. MSc(R) thesis, University of Glasgow.

Abstract

This thesis presents the results of an investigation into the uplift resistance of single anchors and various line groups of anchors embedded in dry sand. Fifty-one tests were conducted using 25mm diameter model plate anchors. Anchor spacings of S = 2B, 4B, 6B, (B - diameter of an anchor) and depth of anchor embedment (D) ranging from 3B to 15B were studied. The author's experimental results are presented in dimension-less factors and ratios so that a possible dimensional similarity between the experimental results and the behaviour of full-scale prototypes can be established. The following general conclusions can be drawn from the author's experimental results. It can be observed that for a given anchor spacing, the ultimate group efficiency (Ef) decreases to a minimum with increasing D/B ratios, and then increases to as large a value as that found with small values of D/B ratios. There appears to be a distinct critical D/B ratio where (Ef) is a minimum value. For example, from the author's test results for a (2x1) anchor group at S/B - 4, the (Ef) value decreases to a minimum of about 70% with increasing D/B ratios, and then increases to as large a value as about 80%. It can also be observed that for a given anchor group size, at anchor spacings of 2B, 4B and 6B, the minimum ultimate group efficiency corresponded to critical D/B ratios of about 8, 10 and 12 respectively. Hence, at anchor spacing of 2B, "shallow" anchors in groups are anchor groups with D/B 8 and "deep" anchor groups have D/B >8. At anchor spacings of 4B and 6B, "shallow" and "deep" anchor groups can be similarly classified. As anchor spacings increase the (Ef) value also increases. However, as group sizes increase the (Ef) value decreases. For anchor groups at very close spacings (S = 2B) , the influence of group configuration on (Ef) is not obvious. The displacement ratio (Ar) is defined as the displacement of the group at failure to the displacement of a single isolated anchor at failure. It can be observed that as anchor spacings and group sizes increase, Ar value increases. However, as D/B ratios increase the Ar value decreases. The distribution of load amongst a group of anchors is non-uniform at failure. Generally the load carried by an anchor is proportional to its distance from the centre of the group where the load is applied. In line groups, the "centre" anchor is shown to carry the smallest load and the "end" anchor the largest load. The author's experimental results and previous theories are compared and discussed for line groups of anchors. The difference between Meyerhof and Adam's predicted (Ef) results and the author's observed (Ef) results are discussed. In this case, Meyerhof and Adam's predicted (Ef) results ranged from 12.7% greater to 49.8% greater than the author's observed (Ef) results. The range of the differences from Meyerhof and Adam's theory is wide and this suggests that Meyerhof and Adam's theory for predicting the behaviour of anchor groups is not entirely satisfactory. The difference between Yilmaz's (first analysis for "shallow" anchors) predicted (Ef) results and the author's observed (Ef) results for "shallow" anchor groups (at D/B = 6) are discussed. In this case, Yilmaz's predicted (Ef) results ranged from 6.5% greater to 7.3% less than the author's observed (Ef) results. In this present investigation a simple analysis has been derived for predicting the ultimate uplift load of "shallow" and "deep" vertical anchors in line groups installed in sand and subjected to static loadings. The difference between the proposed predicted (Ef) results and the author's observed (Ef) results are discussed. In this case, the proposed predicted (Ef) results ranged from 9. 3% greater to 6. 1% less than author's observed (Ef) results. To further investigate the validity of the proposed analysis, the (Ef) test results from Larnach and McMullan's (1975),and Yilmaz's (1971) investigations are compared separately with the predicted (Ef) results derived from the proposed analysis. The predicted (Ef) results derived from the proposed analysis ranged from 12.1% greater to 3. 9% less than Larnach and McMullan's test (Ef) results. In the second case, the proposed predicted (Ef) results derived from the proposed analysis ranged from 6. 5% greater to less than Yilmaz's test (Ef) results. The range of the differences from the proposed analysis is narrower than that from Meyerhof and Adam's theory. This suggests that the (Ef) predictions derived from the proposed analysis gives a better estimate than that derived from Meyerhof and Adam's theory (1968). The predicted (Ef) results for "shallow" anchor in groups (with D/B = 6) derived from the proposed analysis are the same as that derived from Yilmaz's first analysis (1971) for "shallow" anchor groups. The proposed analysis, which is basically an improved version of Yilmaz first analysis, can be used to predict the behaviour of "deep" and "shallow" anchors in line groups installed in dry sand. In Appendix A is an example of how the proposed analysis can be used to design an anchor group system at each stage of the design procedure.

• Item Type: Thesis (MSc(R))
• Qualification Level: Masters
• Keywords: Civil engineering
• Date of Award: 1986
• Depositing User: Enlighten Team
• Unique ID: glathesis:1986-77374
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 14 Jan 2020 09:10
• Last Modified: 14 Jan 2020 09:10
• URI: https://theses.gla.ac.uk/id/eprint/77374