IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 42, NO. 6, JUNE 2014
1515
High Pulsed Power Compact Antenna for High-Power Microwaves Applications Sébastien B. Pottier, Franck Hamm, Dominique Jousse, Patrick Sirot, Friedman Tchoffo Talom, and René Vézinet TABLE I
Abstract— Thales Communications and Security has developed, under Commissariat à l’Énergie Atomique et aux Énergies Alternatives contract, a new concept of compact antenna able to radiate very high pulsed power levels in C-band and X-band. These antennas could be powered by relativistic sources, such as magnetrons or backward wave oscillators (BWO). This antenna principle is based on an array of helixes, whose dimensions are linked to the desired gain and central frequency. Pulsed power—typically a few hundreds of megawatts—feeds the antenna via a circular waveguide in TM01 mode. These works were successfully verified with a first example of such an antenna, developed in X-Band and tested with a BWO delivering 10-ns long and 500-MW peak pulses, repeated at 100 Hz.
R EQUIREMENTS L IST
Index Terms— Array antenna, helical antenna, high power microwave (HPM), insulation.
I. I NTRODUCTION HE last 50 years have seen an unprecedented and unparalleled growth in the development and use of electronics, both in military and civilian applications. Associated with this has been an increase in concerns over electromagnetic (EM) compatibility. In this context, the developments in highpower microwave (HPM) source technology have seen major reductions in size as well as increase in efficiency [1]. The vulnerability of electronics to the generated waveforms has also been proven. High pulsed power antennas (HPPAs) are key devices to develop HPM systems. Depending on the feeding HPM source (magnetrons, backward-wave oscillators (BWO), magnetically insulated line oscillators (MILO), Marx generators, etc.), HPPAs can provide very intense electric field (E-field) levels at hundreds of meters covering narrow band to ultrawideband spectrums. Such antennas must have a gain as high as possible, side lobes as low as possible, withstand input power as high as possible, and be very compact. This paper presents the optimization of an emerging technology for X-band applications, and compares the calculations with the experimental results obtained by feeding the antenna with a
T
Manuscript received October 30, 2013; revised March 7, 2014 and April 18, 2014; accepted April 25, 2014. Date of publication May 20, 2014; date of current version June 6, 2014. This work supported in part by the Commissariat à l’Energie Atomique et aux Energies Alternatives and in part by the Direction Générale pour l’Armement. S. B. Pottier, F. Hamm, D. Jousse, P. Sirot, and F. T. Talom are with Thales Communications and Security, Gennevilliers Cedex 92700, France (e-mail:
[email protected]; franck.
[email protected];
[email protected]; patrick.sirot@ thalesgroup.com;
[email protected]). R. Vézinet is with Commissariat à l’Energie Atomique et aux Energies Alternatives, Gramat F-46500, France (e-mail:
[email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPS.2014.2321416
500-MW peak power BWO. The proposed antenna is based on an array of 120 individual low-profile helical antennas fed by the coupling of a TEM-mode wave traveling in a doublelayer line onto the loops. The TM01–TEM mode conversion is carried out at the input of the antenna. This paper is organized as follows. The concept and architecture of the antenna are given in Section II. The design and calculated performances are presented in Section III. The lowlevel and high pulsed power experimental results are shown in Section IV. II. C ONCEPT AND A RCHITECTURE A. Design Constraints The antenna technology choice and detailed architecture were led by the technical requirements at stake. B. Technology Choice To reach the defined goals, [R1]–[R11], different technologies were analyzed and compared. Table I presents a synthesis of this comparison with the following order from columns 2 to 6: 1) parabolic antenna [T1]; 2) slotted waveguide array [T2]; 3) Vlasov antenna (with lenses and electron bandgap) [T3]; 4) reflect array antenna [T4]; 5) radial transmission helix array [T5].
0093-3813 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
1516
IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 42, NO. 6, JUNE 2014
TABLE II T ECHNOLOGY C HOICE M ATRIX
The technologies in Table II were indicated to be the following. 1) OK: when suitable to comply with a requirement. 2) OK*: when to be optimized to be compliant with a requirement. 3) Risk: when at risk to be compliant with a requirement. 4) NOK: when unreachable. The circular polarization could only be obtained with the last technology using an array of helixes, as well as the compliance with requirement [R10]. The other technologies were indicated as NOK for requirement [R5]. The technology that appeared to be the most suitable for the whole set of technical performances and constraints is using an array of helixes powered by a radial transmission line [3]–[5]. The antenna that is presented in this paper was developed with this technology and architecture. C. Detailed Architecture The antenna is composed of a double-layer TEM transmission line, as shown in Fig. 1. The EM wave traveling in this line feeds magnetic field (H-field) probes. These probes individually supply the circularly organized helixes, which form the antenna array. The supplied TM01 mode was converted into TEM mode at the input of the double-layer transmission line, as described in the mode conversion theory [2]. The mode converter was designed to lower the E-field. The transmission line was chosen to be double layered to divide the power on a larger surface before interacting with the H-field probes and also to lower the E-field. No element was placed in the lower part of the line to enable the best transmission of the EM wave by avoiding reflection phenomena. In the upper part of the transmission line, the probes supplying the helixes were chosen to be loops to avoid the use of monopoles creating E-field enhancement. The magnetic flux couples with the loops to drive the radiating array. No material was added at the center of the antenna to absorb the remaining traveling wave. The antenna was designed so that the energy
Fig. 1.
Antenna architecture.
is coupled to the H-field probes before reaching the center of the upper part of the transmission line. The radiating elements of the array were chosen to be short helixes because of their gain level and frequency bandwidth. The entire antenna must be evacuated to increase the electrical field breakdowns level. A high-frequency window was not required at the input of the antenna, as it was directly connected to the output of a source also set under vacuum. This avoided any risk of electrical breakdown at a vacuum interface. The low vacuum level could be achieved through the use of a secondary vacuum pump. This pump could either be connected to the input of a test primary vacuum pump or to the input of the HPM source vacuum pump. The antenna was closed with a radome. The use of such rectangular and circular helical arrays pulsed by different kinds of double-layer transmission lines was thoroughly described in [6] and [7]. III. D ESIGN AND P ERFORMANCES A. Subassemblies Design and Materials Choices The helical elements were designed for the central frequency. To reach the targeted gain the elements had to be synthesized into a circular array. The array was organized to minimize the mutual interactions between helixes and to optimize the gain and lower the side lobes level, as explained in [3]–[5]. With the allowed surface, the array was composed
POTTIER et al.: HIGH PULSED POWER COMPACT ANTENNA
Fig. 2.
1517
3-D antenna full model.
of five circles and 120 helixes. The main lobe was oriented in the direction perpendicular to the array. The optimization of the antenna directivity according to each helix position was performed through the use of the following: D(θ0 , φ0 ) = 2π 0
4π|F(θ0 , φ0 )|2 π 2 dφ 0 |F(θ, φ)| sin θ · dθ
Fig. 3. Antenna return loss without and with its radome and support elements.
(1)
where D(θ , φ) θ φ
directivity of the array antenna; elevation angle; azimuth angle
F(θ, φ) = E 0 (θ, φ)
N
In e j [kρn sin θ cos(φ−φn )+δn ]
(2)
n=1
where F(θ , φ) E 0 (θ , φ) N ρn k In δn
field pattern of the array antenna; pattern of the single helical antenna; number of elements of the array antenna; distance from the origin to the element; wave number; excitation amplitude of the nth element; excitation phase of the nth element
δn = −kρn sin θ0 cos(φ0 − φn ).
Fig. 4. Simulated antenna gain without and with its radome and support elements.
(3)
The H-field probes and their orientations were designed to power the array with the distribution law, which had been determined previously. Except the radome, the antenna was totally metallic. The inner surface treatments were chosen to lower the desorption level while pumping. The material of the radome and its support elements were chosen as a compromise between their EM characteristics, dielectric permittivity (εr ) and loss tangent (tanδ), and their desorption levels. B. Simulated RF Performance The RF performance was calculated with a 3-D model, as shown in Fig. 2, using CST Microwave Studio. The dielectric radome and its support elements were considered in this model.
The return loss was calculated between 8.5 and 10 GHz without and with the antenna radome and its support elements. The defined port was located at circular waveguide level. In both cases, the antenna had an acceptable return loss (
