With a basic DC motor setup of a single coil spinning in a linear magnetic field, the back emf produced is a sin wave, as the change in the angle between the plane of the coil and the magnetic field produces a change in flux.
From what I know, most real motors use a radial magnetic field. I do not see how a back emf is created in this case, as the angle between the field lines and coil plane should always remain perpendicular (in theory, not accounting for differences in field line direction in different parts of the coil plane, and not accounting for the fact that the field is not perfectly radial). This means there should be no flux change and thus no back emf.
What is incorrect in my interpretation?
Note I am considering only half a rotation, because after I imagine the current switches the flux reverses direction very suddenly - though this may be incorrect. I'd appreciate if any answer could cover both the half rotation and full rotation cases.