Is this PDA result geotechnically credible?
Consultants and owners who engage and rely on third party providers of dynamic pile testing services should ideally assess the validity of the reported test results. There are many aspects to evaluation of the test results, and in due course, I intend to share my thoughts and recommendations on various aspects of this review process. One of the fundamental aspects is to evaluate the geotechnical credibility of the provided solution. An analysis which is not geotechnically credible calls into question the reliability of the assessment, and also limits the degree to which the results might be useful for interpolation/extrapolation to other pile penetrations or locations. The distribution of resistance along the pile shaft and the base is particularly important for example for piles which might be subject to future scour, or for piles which must resist critical tension loads.
Typically, when engineers do assess geotechnical credibility they compare unit shaft resistance and unit end bearing with the design values inferred previously from insitu test results or laboratory test results from the nearest borehole or sounding (e.g. CPT). Of course, there are inevitable uncertainties due to stratigraphic variability. But more importantly, the strength of the ground insitu can only be inferred from the test results, and the relationship of those strengths to the achieved shaft resistance or end bearing values is yet another level of uncertainty, as clearly demonstrated by Bengt Fellenius in his report on the pile testing prediction event conducted at the Bolivian Experimental Site for Testing Piles 3rd Bolivian International Conference on Deep Foundations. The predicted load movement curves for 94 international participants are compared with the actual response of the bored pile, and the distribution of capacity predictions is shown to be normally distributed with a coefficient of variation of 24%.
My point here is that assessment of geotechnical credibility may be subjective, and there is no fixed and absolute reference. I should also emphasize that geotechnically credible does not automatically imply that the solution is correct, or the correct static load-movement behaviour is predicted. Again, much fertile ground for future discussions!
These complications aside, my own review practice is to focus particularly on the predictions at and just above the pile toe.
Assuming the pile toe is embedded several diameters/widths within a layer, and not located above a significant transition (e.g. founded at the top of a dense sand layer, or on rock) then there should be a sensible relationship between the unit shaft resistance and end bearing. For example, and using simple empirical relationships for sand such as shaft (kPa) = 2N and end bearing (kPa) = 200N, then one would expect the end bearing/shaft resistance ratio to be 100 - give or take. This sits comfortably with a typical friction ratio for sand (fs/qc) of 1% - from which one would again expect an end bearing/shaft resistance ratio of 100. Similarly for clays, assuming for sake of argument fs = 0.75 su and qb = 9su, then the expected end bearing/shaft resistance ratio would be 12, which would equate to a CPT friction ratio of 8.33%, which is reasonable for certain clays.
Unfortunately, there can be a complication with dynamic pile testing that the amount of displacement generated at the toe is insufficient to fully mobilize the end bearing. The mobilized end bearing may therefore be an underestimate, and in that instance, the computed end bearing/shaft resistance ratio may be misleadingly high.
An approach I have often used in the past is to evaluate the inferred base modulus which avoids the problems with incomplete mobilization of base resistance. Base modulus can be computed using the elastic solution for a deep circular loaded area in an elastic halfspace by Nishida (1966). Those with a good knowledge of Wave Equation analyses such as CAPWAP (Rausche F et al 1985) and TNOWAVE (Middendorp P 2004), will appreciate that the pile toe quake parameter is generally reliably modelled, from which I conclude that the base modulus assessment is primarily dependent on the interpreted end bearing. The inferred modulus values can be compared with a range of empirical recommendations in the literature (e.g. Kulhawy and Mayne 1990).
I have used base modulus assessment with good success over many years to verify the geotechnical credibility of base response, whether or not full mobilization has been achieved. I will write further about Nishida’s approach in the context of a large diameter open ended steel pipe pile for which comparison was possible against an instrumented static load test pile.
My base modulus spreadsheet (which implements the Nishida solution) is available for download here. This spreadsheet is provided with my compliments but with no guarantee of being free of errors, or of being relevant to your particular project or application! That is for you to verify.
References
Fellenius, B.H., 2017. Report on the B.E.S.T. prediction survey of the 3rd CBFP event. Proceedings of the 3rd Bolivian International Conference on Deep Foundations, Santa Cruz de la Sierra, Bolivia, April 27-29, Vol. 3, pp. 7-25.
Kulhawy, F.H. and Mayne, P.W., 1990. Manual on estimating soil properties for foundation design, Final Report. Electric Power Research Institute-EL-6800 Palo Alto, CA, pp.1493-6.
Middendorp, P., 2004. Thirty years of experience with stress wave simulation applications based on method of characteristics (TNOWAVE). In International conference on the application of stress–wave theory to piles, 7, Lumpur, Malaysia (pp. 165-175).
Nishida, Y., 1966, January. Vertical stress and vertical deformation of ground under a deep circular uniform pressure in the semi-infinite. In 1st ISRM Congress. International Society for Rock Mechanics and Rock Engineering.
Rausche, F., Goble, G.G. and Likins Jr, G.E., 1985. Dynamic determination of pile capacity. Journal of Geotechnical Engineering, 111(3), pp.367-383.