This paper presents a study of the atmospheric refraction and its effect on the light coupling efficiency in an instrument using single-mode optical fibers. We show the analytical approach which allowed us to assess the need to correct the refraction in J- and H-bands while observing with an 8-m Unit Telescope. We then developed numerical simulations to go further in calculations. The hypotheses on the instrumental characteristics are those of AMBER (Astronomical Multi BEam combineR), the near infrared focal beam combiner of the Very Large Telescope Interferometric mode (VLTI), but most of the conclusions can be generalized to other single-mode instruments. We used the software package caos (Code for Adaptive Optics Systems) to take into account the atmospheric turbulence effect after correction by the ESO system MACAO (Multi-Application Curvature Adaptive Optics). The opto-mechanical study and design of the system correcting the atmospheric refraction on AMBER is then detailed. We showed that the atmospheric refraction becomes predominant over the atmospheric turbulence for some zenith angles z and spectral conditions: for z larger than 30{\circ} in J-band for example. The study of the optical system showed that it allows to achieve the required instrumental performance in terms of throughput in J- and H-bands. First observations in J-band of a bright star, alpha Cir star, at more than 30{\circ} from zenith clearly showed the gain to control the atmospheric refraction in a single mode instrument, and validated the operating law.

In this work, we present the first AMBER observations, of the Wolf-Rayet and O (WR+O) star binary system y2 Velorum. The AMBER instrument was used with the telescopes UT2, UT3, and UT4 on baselines ranging from 46m to 85m. It delivered spectrally dispersed visibilities, as well as differential and closure phases, with a resolution R = 1500 in the spectral band 1.95-2.17 micron. We interpret these data in the context of a binary system with unresolved components, neglecting in a first approximation the wind-wind collision zone flux contribution. We show that the AMBER observables result primarily from the contribution of the individual components of the WR+O binary system. We discuss several interpretations of the residuals, and speculate on the detection of an additional continuum component, originating from the free-free emission associated with the wind-wind collision zone (WWCZ), and contributing at most to the observed K-band flux at the 5% level. The expected absolute separation and position angle at the time of observations were 5.1±0.9mas and 66±15° respectively. However, we infer a separation of 3.62+0.11-0.30 mas and a position angle of 73+9-11°. Our analysis thus implies that the binary system lies at a distance of 368+38-13 pc, in agreement with recent spectrophotometric estimates, but significantly larger than the Hipparcos value of 258+41-31 pc.