Pull—out tests were conducted using a specially constructed rig on lOOxlOOxlOOmm prism specimens with short embedment length. All tests show a sharp increase in pull—out
load with negligible slip until a "maximum bond force" is developed and the adhesion fails. After this point slip occurs at load levels which depend upon the test variables.
For the case of zero lateral pressure and long embedment length a maximum bond stress of l.9N/mm² is obtained for normal strand embedded in a grout of 57N/mm² compressive
strength. This value is less than that recommended by DD81 but is based upon the true contact area.
The effect of biaxial lateral pressure up to 0.26 times the compressive strength is to linearly increase the maximum bond stress as follows for iN/mm² increase in lateral
pressure.
Normal strand 0.26N/mm² ± 0.04
Dyform strand 0.24N/mm² ± 0.04
Indented strand 0.47N/mm² ± 0.22
The maximum bond stress for the case of zero lateral pressure and 25.4mm embedment are in the ratio 1.00:0.90:1.20 for normal strand, dyform, and indented strand respectively.
The increase in bond due to added length of embedment is not linear. The maximum bond stress tends towards a constant value beyond an embedment length of 50mm.
The maximum bond force increases at an average rate of 0.O7KN for iN/mm² increase in grout strength for up to lateral pressures of 0.26 compressive strength.
An equation is developed which predicts the maximum bond force well, provided values of shrinkage and material constants are known. Comparison with the results obtained by
other investigators shows good correlation for the case of zero lateral pressure. Differences with others using lateral
pressure are due primarily to differences in the loading system. It is shown both experimentally and theoretically that
the torsional stiffness of the strand has little effect on bond.