According to James C. Keith, see appended PDF files, two components of gravitational drag are to be expected on rotating systems, a 1/c5 order drag to be observed preferably on systems of astronomical size, and a 1/c3 order drag preferably observable on millimeter size systems. Observation is limited in either case by experimental resolution of rotational frequency shift. In the early 1970s, Hulse and Taylor [1] determined relative deceleration of a binary neutron star system at 1.5 x 10-12 per second in agreement with theoretically accepted gravitational quadrupole radiation. A relative deceleration of about 2 x 10-11 per second was observed around the same time [2] on a 2.5 mm diameter steel ball at 75 kHz rotational speed. This observation has not yet been accepted by established theoretical science as due to gravitational interaction. The access of experimental results to serious analysis seems to be largely a matter of decision by representatives of theoretical rather than engineeering science.
[1] Astrophysical Journal, Vol. 195, p. L51-L53 (1975)
[2] Physical Review Letters, Vol. 30 (16), p. 753-757 (1973)
Article Significant Deviation of Rotational Decay from Theory at a R...