133 Antenna and Transmitter Design for a 449 MHz Wind Profiler Radar

Monday, 16 September 2013
Breckenridge Ballroom (Peak 14-17, 1st Floor) / Event Tent (Outside) (Beaver Run Resort and Conference Center)
B. Lindseth, NCAR, Boulder, CO; and W. O. J. Brown, T. Hock, C. Martin, and S. A. Cohn

This paper describes antenna and transmitter design of the new 126-element 7-hexagonal array spaced antenna 449 MHz wind profiler radar at the National Center for Atmospheric Research. With a larger antenna and a higher 8 kW transmit power, higher altitude wind profiles are possible (up to mid-troposphere levels). After initial tests, this system will be available for deployments to scientific field projects. We recently designed and built an 8-kW pulse transmitter. This transmitter is based on combining eight commercially available low-cost 1 kW LDMOS devices. The amplifiers used in this transmitter have an efficiency of about 60%. For operation as a wind profiler transmitter, there are some additional considerations beyond that of a normal pulse transmitter. For example, it is important that the transmitter output very little noise between pulses. This transmitter is 4 times the power of our previous 2-kW transmitter, so we expect a 6 dB increase in system SNR.

The antenna array is comprised of 126 linearly polarized circular patch antennas. They are divided into 7 hexagonal arrays with 18 elements each. The system uses the Full Correlation Analysis Spaced Antenna method to compute horizontal winds. The cross correlation at zero lag between receivers is affected by the array spacing. It is desired to have a cross correlation at zero lag about 0.5 (c12(0)=0.5) for best wind processing. Another design criteria is to optimize the array spacing for low sidelobe levels. Because of the low level of radar returns from clear air turbulence (-150 dBm), returns from clutter targets caused by high antenna sidelobe levels can be significant. Using full-wave electromagnetic simulation in FEKO, the antenna pattern was simulated with many different array spacings to determine the spacing with lowest sidelobe levels and also to compute the cross-correlation at zero lag which is dependent on transmit and receive beamwidth. The simulations show that the 126 element array has a narrow beamwidth which causes a high cross correlation at zero lag of about 0.6. This result indicates that we can optimize spacing of the 126-element array for low sidelobe levels, which will decrease the level of clutter returns received by the system.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner