Over the years, radar system developers and integrators have tried many different approaches to compensate for some of the shortcomings of early high-frequency technologies.
 |
Radar has as rich a history as any application area in this industry, dating to the early days of the last century. As with many early electronic systems, it was exclusively based on vacuum tubes in its early days, eventually making use of solid-state amplification as transistors developed as a viable alternative. In tracing the development of early radar systems into many different applications, the engineering ideas that attempted to overcome such device limitations as frequency stability were both innovative and impressive.
For example, in September 1965, concern over achieving better antijamming properties in their radar systems had the US Department of Defense (DoD) exploring the development of frequency-agile radar systems. The frequency-agile radars were projected to have greater range than fixed-frequency radar systems operating at the same power level. They were also thought to provide heighted aiming accuracy in fire-control radar systems and reduced bearing error in search radar systems. Along with the radar systems, DoD funding was targeting the development of more agile frequency sources, including magnetrons from Raytheon and Amperex (Fig. 1).
One of the concerns for creating these more agile vacuum tubes was the need for an automatic-frequency-control (AFC) circuit that could stabilize frequency with each rapid change. Amperex, for example, had developed their Fast-Tracking Local Oscillator (FTLO) system, which would use a backward wave oscillator to send a signal into the magnetron cavity about 50 s ahead of a frequency change, so as to analyze the cavity's cold resonant frequency for the sake of making adjustments when the frequency was changed. A General Electric approach, designated the frequency agility radar modification by engineered retrofitting of systems (FARMERS) method, used either a monitoring servo or a positional transducer to get a frequencyproportional signal from the tube. A servo with 30-ns or better response time would monitor the tube directly from its output. The positioning transducer would yield an output proportional to the tuner location for making adjustments to the output frequency.
 |
The January 1966 issue reported on a millimeter-wave radar system developed by North American Aviation (Columbus, OH). The 10-kW, 3.2-mm pulsed radar system was transportable (Fig. 2) and featured an intermediate frequency of 150 MHz with 100-MHz bandwidth. The IF amplifier gain was variable to as high as 100 dB. The radar featured nitrogen-pressurized waveguide and variable polarization. It could control pulse width from 4 to 250 ns at a pulse repetition frequency (PRF) of 1 kHz.
The following year, the January 1967 issue featured a panel discussion on which electron tubes were best suited for use in phased-array radar systems. Crossed-field amplifiers (CFAs) were considered better for systems that must move around or be portable, while traveling wave tubes (TWTs) were thought to provide more gain per device and were better suited for fixed applications. The CFAs were smaller and lighter than TWTs, with considerably higher device efficiency, and so were the preferred amplification device for portable radar systems.
In March 1967, the Cornell Aeronautical Laboratory (Buffalo, NY) announced the development of a coherent-detection radar system with separate transmit and receive antennas, funded by a contract from the Air Force Rome Air Development Center. The system incorporated a 9375-MHz klystron with 40 mW CW output power for both the transmit and the local oscillator (LO) power. Transmitted signals were reflected by both moving and nonmoving surfaces, with the moving surfaces shifting the phase of reflected returning signal to an equivalent phase shift in the 60-MHz IF. This low-power radar system was thought to have applications in hospitalsspecifically, for noncontact sensing to remotely tell if a patient's heart had stopped.
A news story in the February 1973 issue reported on a highly sensitive radar system claimed at the time to be able to detect a house fly at altitudes to 10,000 ft. Developed by the Naval Electronics Laboratory (NELC) in San Diego, the ground-based, frequency-modulated-continuous-wave (FMCW) radar operated with 200 W power at 2.8 to 3.0 GHz and a beam angle of 4.5. The radar was capable of detecting airborne crickets during a demonstration in the California desert. The radar system relied on a YIG oscillator to achieve highly linear tuning from 2.8 to 3.0 GHz. The system's sounder antennas consisted of two 10-ft.-diameter dishes built into pits in the ground and surrounded by microwave absorber material to achieve good isolation and low ground clutter (Fig. 3).
By the following year, the February 1974 issue would report on early work in automotive collision-avoidance radar systems. Based on a baseband radar detection sensor (BARDS) system from Sperry, work performed at the Sperry Research Center (Sudbury, MA) led to a system with transmitter and separate receiver mounted behind an automobile's front grill. A step-recovery-diode circuit in the transmitter was used to generate unipolar 200-ps pulses. Radiated energy in the form of triangular pulses extended from 20 MHz to 20 GHz, with an effective beamwidth of about 70. One of the challenges for the design team was to minimize the false-alarm rate for the collision-avoidance system. Sperry had estimated a price of $50 to $60 per system.
 |
Reasonable price was also a goal that year for a team working on fire control radars, which had set a price goal of $200,000 per system in production quantities. The team included Hughes Aircraft Co. (Culver City, CA), Rockwell International (Anaheim, CA), and Westinghouse (Baltimore, MD). The team wanted to create a fire-control radar along the lines of the modular WX-200 from Westinghouse, which could be readily adapted for use in different aircraft. The WX-200 was a range-gated pulsed Doppler system with average power level of 770 W and peak power of 100 kW. The system weighed 350 lbs and consisted of seven modules or line-replaceable units (LRUs). Output power came from a Litton TWT with 100-kW peak output power at 1% duty cycle, 0.85-s duty cycle, and 9030-Hz PRF.
Later that year, Raytheon would deliver an AN/TPN-19 precision approach radar (PAR) to Elgin Air Force Base, FL. The system showed 100% success in tracking aircraft approaching a runway in heavy rain (5 inches/hour or more). The older AN/ FPN-16 radar could only track 6% of the same aircraft under those conditions. The PAR antenna was a hybrid limited-scan design that used a small phased array to illuminate a portion of a much larger hyperbolic reflector. The system employed a total of 824 3-b ferrite phase shifters.
In November 1974, a Data General NOVA 800 minicomputer was used to control the 16 basic functions of an AN/ TPQ-39(V) digital instrumentation radar. Built by RCA's Missile and Surface Radar Division (Moorestown, NJ), this was thought to be the first radar designed and built around a solid-state minicomputer. RCA was scheduled to deliver three units to the Army, Navy, and Air Force as part of a tri-service, $3.7 million contract. The mainframe of the computer had slots for 17 modular boards, including a slot for the main part of the computer and its 16k core memory (16,000-word computer).
The radar system operated at C-band with 250 kW output power at a PRF of 640 Hz. The noise figure specified at the receiver's mixer input was 6.5 dB. A five horn monopulse-type feed and six-foot-diameter parabolic disk were used for form a 2 beam. A single operator was able to run the system with three joystick controllers, or could select an automatic mode for "hands-off" control. The radar system could track a 6-in.-diameter metal sphere to a range of 45,000 yards.