I have to analyze the behaviour of piles under tension loads (settlements...). For the skin friction I have to consider, may I use the same parameters as for the compression loads, or does the fenomenon change so that I have to use coefficients or a completely different model? Thank you in advance,
Simon Naveau brings up an issue that ever so often pops up, but rarely gets to be studied. In my opinion, and more important, according to full scale tests, there is no appreciable difference in shaft shear between piles pulled or pushed. That is, Nature does not distinguish between the direction of movement. The very few, half dozen or so, articles reporting test results from testing piles in pull and push and finding a difference share one aspect: they did not consider that the piles were subjected to residual load. Residual load does not affect the total resistance, but, for a pile tested in push, it overestimates the shaft resistance and underestimates the toe resistance. (The piles were instrumented to separate the shaft resistance from the toe resistance. At the start of the test, the piles were assumed to be free of load --- which assumption or approach is wrong and the cause of the erroneous conclusion of the tests).
Mr. Englund's caution about loading history is important. The capacity of a pile is far from always a well-found value, but subject to some judgment reinforced with one or other definition, usually based on a movement value. This makes the sequence of testing important. For example, the load-movement curve from the pull test will be different when the pile was tested in push first and in pull second as opposed to the other way around and although the shaft resistance is really the same, the slope of the curve and its peak value, if any, differ between the tests.
I wonder about your parenthesis: "(settlements...)". Although "settlement" refers to downward, the context leads me to believe that you refer to upward movement. Are you designing tension piles for a limiting upward movement?
I am not familiar with the book by Fleming et al. However, the common radial contraction/expansion (radial contraction in pull as opposed to radial expansion in push) is without any support in the field. While the concept is qualitatively true, the radial movements are so small that they are of no practical significance. This can be shown by numerical modeling (as opposed to using earth pressure theory for passive and active conditions).