The centrifugal force reversal and X-ray bursts

Heyl (2000) made an interesting suggestion that the observed shifts in QPO frequency in type I X-ray bursts could be influenced by the same geometrical effect of strong gravity as the one that causes centrifugal force reversal discovered by Abramowicz & Lasota (1974). However, his main result contains a sign error. Here we derive the correct formula and conclude that constraints on the M(R) relation for neutron stars deduced from the rotational-modulation model of QPO frequency shifts are of no practical interest because the correct formula implies a weak condition R R-* > 1.3R(S), where R-S is the Schwarzschild radius. We also argue against the relevance of the rotational-modulation model to the observed frequency modulations.