Abstract: In this note a comment on the paper "Modularizing, Minimizing, and Interpreting the K & H Fault Tree" by M. O. Locks is presented, and a revised and simplified analysis method for non-coherent fault trees is introduced. It is shown that all of the PIs for primary fault tree can be obtained if the logical polynomial, which consists of all MCSs for the inverse, is inverted directly. The results of inverting (and minimizing) from all MCSs, or all PIs of minimal form, or all PIs for the inverse, are the same. Of course, each set must be a complete set, and any importance or size restricted set is not adequate. It has also been pointed out that the insertion of Quine's consensus operation in the process of Nelson's "double complementing" is not neccessary, and is unsuitable.