I have a different proposal for simulating the gravitational field on a flat earth. It is represented in the following sketch:
It has the advantage that there is no need of a potentially infinite (or at least vary long) straight trajectory. Also, once put into rotation, the system has a constant kinetic energy, so there is no need of additional mechanical work to keep the gravitational field.
It's true that the g on the edge of the FE is not strictly perpendicular to the surface, but this can be made sufficiently small by taking a radius of revolution much larger than the radius of the FE.
I have also included the rotation of the FE about its axis perpendicular to its plane and passing through the North Pole. This axis also passes through the center of revolution.
Please note that the cause of the revolution can not be of gravitational origin, because all objects on the FE would be weightless then (like in a rotating satellite).