Some Thoughts on API RP2A for Vertically Loaded Piles

INRODUCTION API RP2A represents one of the very few design guidelines produced by the geotechnical profession The overwhelming success and the worldwide use of API RP2A is the proof of its vital usefulness. The few engineers who over the years have taken the time to establish and improve this recommended practice mustbe thanked by the rest of the profession and urged to continue their crucial, though voluntary, work. Success does not go without controversy and API RP2A has sometimes been under fire. This has led to improvements and has helped in arriving at a well balanced product. It must be recognized in that respect that no one is compelled to use these guidelines since API RP2A is a recommended practice not a code. However this recommended practice has been prepared by some of the most experienced engineers in the geotechnical profession and thus enjoys a worldwide reputation. It is also important to note that API RP2A is a design method not a prediction method; therefore the goal is not to predict the behavior of a foundation s stem within a certain accuracy but to ensure that the behavior of a foundation system will be acceptable and that it can be built for a reasonable cost. In this article, a number of comments related to vertically loaded piles are made with the continued improvement of API RP2A in mind. The topics discussed in this paper include the effect of layering incapacity, the criterion for piles in sand, the plugging of pipepiles, the rate of loading effect, the cyclic loading effect, the load transfer approach, drilled and grouted piles, load and resistance factors, and hydrates. EFFECT OF LAYERING ON CAPACITY A recent study by Briaud and Tucker, (1988) showed that if a method predicts well the capacity of piles totally embedded in clay and if another method predicts well tie capacity of piles totally embedded in sand, the combination of these 2 methods predicts well the capacity of piles through sand underlain by clay but over predicts significantly the capacity of piles through clay underlain by sand. The data base used to arrive at these conclusions was accumulated b the Mississippi Highway Department. It consisted of 9$ compression load tests on 64 squareconcrete piles, 27 H-piles and 7 drilled shafts. The piles diameter averaged15 in. (0.381m) and the length 40 ft (12.2 m). The soils included clays, sands and layered systems. The clays undrained shear strength averaged l.5 tsf (144 kPa), the sands blow count averaged 43 bpi (43 blos per 0.3 m). The 1984 version of API RP2A was used and the calculated capacities were compared to the measured capacities. The measured capacity was defined as the load corresponding to a sett1ement equal to one-tenth of the pile diameter plus the compression of the pileunder that load as if it was a column. The results are shown on Figures 1 to 5 for piles embedded entirely in clay, in sand, in layered soil with the pile tip inclay, in layered soils with the pile tip in sand, and for all the piles in the data base. Figure 6 shows the variation of the ratio of the calculated capacity over the measured capacity for the various soil categories.