K-FACTOR PARADOX.

It is demonstrated how the K-factor concept can give rise to peculiarities that are in direct conflict with the common-sense approach to design. For example, although a buckling analysis may indicate a K-value greater than or smaller than 1. 0 for a leaning column in an unbraced frame, the designer will always adopt K equals 1. 0 for proportioning such columns. However, the K-factors obtained from the buckling analysis are the correct values to be included in the moment magnifier expression. The contradictions created by the K-factor are also apparent in frames with semirigid as well as fully rigid connections. It is proposed that all these anomalies can be eliminated by employing column interaction equations that do not make use of the K-factor.