Memo Review Memo: 427 - Antenna Position Calibration Wright, 2002May16 Reviewer: Robert Lucas Date Received: 2002aug29 Review: The report is good, I will just add a few comments. I like the idea of using the multi-frequency capability to solve 2pi ambiguities. This can be very useful if repeatibility of antenna positioning is not good enough, and certainly when a station is used for the very first time. At Plateau de Bure we do not attempt to remove 2$\pi$ ambiguities munually or by enforcing phase continuity, we just perform a least squares minimization inside a predefined search box in the space of baseline coordinates. The minimized function uses the phase determination closest to that measured. We may also fit a time dependent polynomial simultaneously with antenna position offsets. This is not a real help, as instrumental drifts are wery small and atmospheric phase fluctuations mostly occur faster than we can correct this way (we spend typically a few minutes on each source). We may also fit the axes offsets simultaneously, but we only do this on very short baselines and good weather. ALMA antennas can look at some directions with `over-the-top' elevations which should easily allow to separate this effect from antenna position offsets, and measure antenna axes offsets in an absolute way (though this is not really needed, I believe). One point that would need investigation is whether radiometric correction should be used to improve phase determination during antenna position measurements. It is quite possible that the calibration of that correction scheme would use interferometric observations of quasars, as mentioned in Bima memo 78, so it may be not wise to use this phase correction for determination of antenna positions in that case (except for canceling decorrelation effects during the integration on each source).