The sign convention of axial load cases is shown in the diagram below. Displacements and forces are positive downwards (for compression loading) and negative upwards (for tension loading). For torsional loadings (where the pile has a torsion applied about its vertical axis) the sign of the loads has no effect. To avoid confusion between axial and lateral load cases where a pile is subjected to torsional loading it twists about the piles vertical axis, as opposed to a rotation under lateral shear and moment loading. Users should adopt either a negative or positive signs for torsions and twists.
It is important to note that displacements and twists take precedence over forces and torsions. For example if a displacement and a force were entered as a load condition for the same grid point then the displacement would override the force and the force input would be ignored. In solving these type of problems it is most efficient, computationally, to use displacement and twist controlled analysis, rather than force-torsion controlled analysis.
Applied loads (Axial and Torsional) are transferred to the pile element bottom and applied displacements (Axial and Twists) are transferred to the pile element middle. In case a load needs to be applied to the top of a pile, it is recommended to create a small top element or use OPILE’s grid generator.
The axial load cases are entered into a table as shown below:
OPILE will automatically generate load cases that are appropriatate for the pile being analysed. Within the load case generation there is an option to optimize the load cases in order to determine the nature of any peaks within the load displacement response.
For piles which are predominantly driven into clay soils, with a residual effect on the TZ curves, it is possible to determine the peak on the pile load-displacement response curve, using the feature in OPILE. The feature works by incrementing through a series of displacements and determining the pile load associated with them. Note that it carries out a finite difference analysis of the pile at each increment and it is not advised to use this feature for piles which gain most of their capacity in sands, or it may take a long time to run with little value in the results, where no distinct peak will be found, rather an ultimate value. Once it finds a peak (no matter how small) the subroutine stops! Example of the routine results is shown in the picture above (from Worked out Example 5).
A necessary input for this type of analysis is the displacement increment, the smaller this increment is the longer the analysis will take. However if the increment is too large then the peak may be skipped. An increment of about 0.1% of the pile diameter usually produces satisfactory results. The other input required is the initial displacement, which if set to a value of greater than zero means that no tension load cases will be generated. This type of peak detection is approximate and may take some adjustment of the displacement interval in order to maximize the accuracy.
Location of initial displacement and the displacement increment input within Load Cases¶
Schematic outline of pile load displacement response¶
An example of pile load displacement response as generated by OPILE¶
OPILE will generate load cases which are sufficient to define the full axial load-displacement response of the pile. These load cases will include sufficient points to encompass the peaks of any response curve.
Due to the complicated nature of pile torsional analysis the automatic load case generation does not generate load cases for torsion/rotation type cases. Generally only very small rotations are required to fail the pile in a torsional manner. The torsional pile response could be used to analyse the response of conductor piles and assess the suitability of hi-torque connectors. The method by which torsional analysis is carried out means it should never be used in combination with any end bearing, otherwise the results that are obtained will be incorrect. Some information on this type of analyses is given in Finnie & Morgan (2004).