Remarks on the suitability of various transonic small perturbation equations to describe three-dimensional transonic flow; examples of computations using a fully-conservative rotated difference scheme