P. J. Napier [1], J. Cheng [2], D. T. Emerson [2], M. A. Gordon [2], J. B. Lugten [3], J. M. Payne [2], W. J. Welch [3], D. P. Woody [4]
October 20, 1995
Keywords: antennas, convetional antenna, slant axis antenna, off axis antenna, electromagnetic performance, polarization, sensitivity, antenna phase stability, fast switching, close packing,
This report compares two existing 8m designs for the Millimeter Wave Array (MMA) antennas. One is a conventional elevation-over-azimuth design derived from the 6m antennas used in the BIMA Array at Hat Creek, California. This design uses steel for both the mount and the back structure. The other is a slant-axis design using an offset parabaloid as the primary mirror, a combination seldom if ever used in radio astronomy. This design uses steel for the antenna mount and carbon fiber reinforced plastic (CFRP) for the back structure. Both designs use cast-aluminum surface plates, milled by a numerically controlled machine and stress-relieved by thermal cycling, to form the primary mirror.
This report lists the antenna specifications required by the MMA, describes the structural elements of both designs, evaluates the performance of the designs, and estimates their costs.
This report finds that although either design would be appropriate for the MMA, each has deficiencies. The conventional design has a strong dependence of phase errors with temperature and marginal pointing performance due principally to using a yoke to support the primary surface. The offset/slant-axis design has poor polarization and field-of-view performance unless equipped with supplemental mirrors to alter the optical path from the surface to the receiver location, a modification that reduces sensitivity. Both designs fall slightly short of the performance requirements needed to calibrate phase errors rapidly.
As evaluated, we estimate each antenna to cost to be $1.3M and $1.5M, respectively, for the conventional and offset/slant-axis designs.
At this writing, this report recommends the conventional design for the MMA antennas. Our evaluation presumes that the phase errors will be measured every few minutes and the pointing errors, every 30 minutes. In this case the conventional design *just* meets specifications. If this frequency of calibration cannot be met or if there is a desire for operation at higher frequencies, then a better design may result from combining the best features of the two designs considered here, that is, a conventional reflector on a CFRP backstructure on a slant-axis mount.
Whatever the choice, the committee strongly recommends that a prototype antenna be tested before committing to purchasing the 40 antennas needed for the MMA, especially if the choice is a offset/slant-axis design that would be new to radio astronomy.
[1] NRAO/NM
[2] NRAO/TUC
[3] BIMA
[4] OVRO
View an HTML version of MMA Memo #145.
Last modified: 09 December, 1999
kweather@nrao.edu