I want to describe one system (two massive bodies synchronously rotating around each other in circular orbits). Bodies move counterclockwise. This system is well known - Kepler's problem - but with the only difference that the speed of propagation of the gravitational force is finite and, as predicted by general relativity, equal to the speed of light. This additional factor significantly changes the behavior of the rotating two bodies.
Let's look at this two-body system (attached file). Let us assume that at time T the bodies are located at points A (heavy body) and B (light body). Knowing the velocities of bodies in their orbits and the distances AO and BO, it is possible to determine the exact position of bodies in orbits at delayed time Tr = R/c = AO/c:
Arc length AA' = AO*v_h/c, arc length BB' = BO*v_l/c
where v_h and v_l are the velocities of heavy and light bodies. Why it is possible to determine the exact location of bodies at a retarded time is explained in: https://www.qeios.com/read/R1KXRJ
So, when the heavy body is at point A, the gravitational force, which ensures the rotation of the bodies, was emitted by the light body from a point with arc length BB*.
(arc BB*)= (arc BB")*OB/(OB + AO)
Let us assume for simplicity that this force comes from point B'.
If in the classical Kepler problem the gravitational forces are of a central type - they are always directed along a line connecting the points of instantaneous location of bodies - line AB, then at a finite speed of gravity the forces are of a non-central type. The force attracting a heavy body acts along the line AB', and the corresponding other force acts along the line A'B.
It can be seen that an additional tangential component of the gravitational force appears.
F_tan(B->A) = AB' * sin alpha, where sin alpha ~ alpha ~ v/c
The important thing is that this force accelerates bodies. Consequently, the angular momentum and energy of each body must increase. All calculations are given in the specified file.
This is a strange result, since the gravitational force is of a conservative type - no energy is spent to create this force. Thus, there is no energy loss to increase the energy of the system. We have a classic example of a perpetuum mobile.
My question is: if it is not possible, then why is it not possible?