>
\r?\n> One issue with bicycle cranks is that the loads aren't all that well
\r?\n> understood, and they are quite large.
\r?\n>
\r?\n>
\r?\n>
\r?\n> Jan Heine
\r?\n>
\r?\n> That seems evident looking at most cranks. The forces at the pedal end
\r?\n> are primarily torsional about the crankarm centerline axis as the load is
\r?\n> fed in cantilevered out by the pedal, then in the middle part the load path
\r?\n> is almost purely in bending perpendicular to the crankarm centerline in the
\r?\n> rotational plane, then becoming primarily torsional again about the spindle
\r?\n> centerline perpendicular to the axis of torsional stress at the pedal end.
\r?\n> I can't think of a crank made that looks to take the complicated load paths
\r?\n> properly into account. Cranks look to be roughly modeled after automotive
\r?\n> engine's connecting rods more than anything, which is a bad analog as it is
\r?\n> subject to almost none of the torsional stresses a bicycle crank is. I've
\r?\n> sketched what seems to me to be the best solution, but it looks decidedly
\r?\n> strange :D Probably too strange. The answer probably lies forgotten in a
\r?\n> patent drawing from the 1890s.
\r?\n>
\r?\n> Kurt Sperry
\r?\n> Bellingham WA
\r?\n> USA