Gravity gradient bias in the EPF experiment

Abstract

AbstractThe Eötvös  Pekár and Fekete (EPF) equivalence test was aimed to check composition dependence of the gravitational force. It was designed to be free from any bias related to the possibly time-varying ambient gravity field. Interestingly in 1986 Fischbach and his colleagues found a systematic composition dependency in the EPF data. This discovery induced an intense debate, and even though the effect was unreproducible, the cause behind is still unknown. Now we found, however, that time-varying gravity gradients may have biased the results in spite of the experimenters’ protocol to avoid it. Moreover, even in a constant, but inhomogeneous gravity field the gravitational force was necessarily different on the EPF samples that had different shapes. These issues are serious: we demonstrate that time-varying ambient gravity alone can fully reproduce the EPF results. Hence, we propose a remeasurement with a careful control of gravity gradient bias. We also discuss other more precise equivalence tests in connection with the above bias.