Parametric experiments on stayed columns with slender bipods

Abstract

Stayed columns are competitively light whenever terrestrial, offshore or space applications involve long, lightly loaded columns. They consist of a slender core supported by N light, planar stay frames that are equally spaced around it and consist of pretensioned stays and short arms or bipods. The relatively few published experimental results, which have complemented more numerous theoretical studies, have nearly all used arms that are off-cuts from the core and have been for N = 4, with two of the stay frames omitted to produce a plane frame which is artificially constrained to deflect in its own plane. The experimental results presented are for a column with N = 3, with each stay frame consisting of a very slender bipod and two stays. All connections were push fits except for grub screws which keep the stays tensioned. Thus it was possible to undertake a parametric experimental study which involved finding 203 collapse loads which are summarized compactly. The parameters used were the ratio of bipod height to column length, stay pretension, the plane in which collapse was constrained to occur and initial imperfection magnitude. The main conclusions are that: the collapse load is increased by using several (e.g. four) times the theoretical optimum prestress predicted by ignoring imperfections; the collapse load is then typically 10% below the theoretical critical buckling load of the perfect column; a modified Southwell plot sometimes agrees well with the theoretical critical buckling load; a crude but very simple non-linear method related to the Southwell plot predicts collapse loads accurately enough for most design purposes, and all practical prestress levels and; using slender bipods and push fits is fully justified.