Long-term time comparison between frequency standards at NIST and PTB for a test of the validity of local position invariance

The caesium fountain frequency standard CSF1 of PTB has been compared with a hydrogen maser HMNI operated in NIST, Boulder, for more than two years. Two time comparison methods were used for that purpose, two-way satellite time and frequency transfer (TWSTFT) and GPS carrier-phase based frequency comparisons (GPSCP). Standard GPS common view time comparisons performed in parallel were useful to identify the properties of the two methods. Between TWSTFT and GPSCP variations of as much as 10 ns peak-to-peak during the whole period were found. It was nevertheless possible to determine a new limit of the validity of local position invariance (LPI), tighter by almost a factor of 4 than previously possible. The validity of LPI could be tested by comparing two different kinds of atomic frequency standards in the same time-varying gravitational potential /spl Delta/U(t) caused by the earth's annual elliptic orbit around the sun. LPI predicts a null result in such an experiment A frequency variation between CSF1 and HMNI in phase with /spl Delta/U(t) greater than 6.10/sup -6/ of the amplitude of /spl Delta/U(t)/c/sup 2/ can be excluded.