and someone worked out that the stresses in a crank due to
straight torque are

I had reached that conclusion in an earlier posting - and maybe others.
These things are buried so deep in the technical organizational
underbrush that they never seem to reach the light of day. I sure wish
someone in the real know would post to this group.

Your two blade/three blade thing may be correct. Obviously your
observations are. I have been looking at things from only a
theoretical basis.