Development of an ultrawideband low-profile circuit analog absorber using resistively loaded tightly coupled dipole arrays

Abstract

Due to developments of modern military radar systems the interest of radar absorbing materials (RAM) is increased. Since pyramid-shaped absorbers used in anechoic chambers and ferrite absorbers are bulky, heavy and thick, they are not preferred to be used in the systems such as aircraft, ships and missiles. The absorber materials must be as light and thin as possible in order to be useful on these platforms because the aerodynamic structures, shapes and performances of the aforementioned systems are directly related to the materials used. It is also important to have wide operational bandwidth. However, increasing the absorption frequency bandwidth causes the absorber structure to be thicker and the absorption level to decrease. Many studies have been carried out to provide solutions to this problem. Some of these studies include the ability of the material to be used to absorb through its electrical and physical properties. The absorption can be achieved through the features of the materials such as dielectric coefficient, magnetic permeability and thickness. In addition, it is possible to design an absorber by using frequency selective surfaces (FSS). In recent years, research has focused on FSS-based absorbers because the bandwidth-to-thickness ratio can be optimized and the operating frequency can be scaled easily. FSS elements can be consists of resistive materials, or they can be loaded with lumped resistors. In both structures, it is seen that there is usually a ground plane in the bottom layer of the absorber and there is an air or another dielectric layer between ground plane and the FSS pattern. Grounded FSS-based absorbers are also called circuit analog absorbers (CAA). The aim of this thesis is to design an ultrawideband and low profile (<0.1λL; λL:wavelength at the lowest operating frequency) resistor loaded CAA. Many of the studies in literature have focused on CAAs loaded with lumped resistors due to their easy production and low cost. Absorbers obtained using various FSS patterns can be single-polarized or dual-polarized (TE and TM), single or multilayer and so on. These structures can be analyzed using electromagnetic simulation programs such as HFSS. The correct configuration of these programs is very important for the reliability of the simulation results obtained. In this thesis, it is given how to do unit cell analysis of an FSS-based absorber in HFSS for this purpose. A previously published design in the literature was remodeled with HFSS and the obtained simulation results were compared with the published results. After the HFSS settings were found to be suitable for simulation, the design phase was started. One of the ways to design an absorber is to use an antenna. Instead of feeding the antenna with electromagnetic wave, if the feeding place is terminated with a resistance suitable for the antenna impedance, the incoming planar wave can be damped on this resistance. A tightly coupled dipole array (TCDA) antenna with a ground plane at the bottom layer is used. In the first phase of the design, Rogers 5880 substrate with relatively low dielectric coefficient, which is available in the industry, was preferred for the absorber operating between 4 and 18 GHz. Absorption is provided in both TE and TM modes, thanks to the tightly coupled dipole (TCD) element placed above and below the substrate. There is a physical surface mounted device (SMD) resistor element in the middle of each element. The size of this resistor has been simulated and the ohm value of the selected resistor has been optimized. The reason why TCD elements are placed above and below the substrate is the difficulty of soldering the resistors of the two elements on the same surface. The equivalent circuit model is utilized for giving the initial value to the dimension of absorber. The determination of the distance separating the conductor pattern on the substrate and the ground plane, which is one of the most important parameters, was analyzed with the obtained impedance equation. Another important parameter is the value of the resistor used. If the TCDA is not matched with a aproppriate resistor value, the absorption characteristic will deteriorate due to impedance mismatch. Other main parameters are the length and thickness of the dipole and the dimensions of the finger-shaped structure that provides the coupling at the dipole ends. The absorption ability of the absorber structure is analyzed by the reflection coefficient obtained with HFSS. A plane wave is sent on the proposed absorber and how much of this signal is returned to the transmitted port is calculated by HFSS. Since the bottom layer of the proposed absorber is completely reflective metal sheet (ground plane), the transmission coefficient is theoretically zero. The absorption level is calculated from the reflection coefficient magnitude so that it can be analyzed how much the RAM absorb the incident wave. According to the simulation results of the first design stage of proposed absorber, it is seen that it absorbs the normally incident wave 90%, between 4.23 - 18.20 GHz for TE mode and 4.42 - 18.92 GHz for TM mode. When simulating, it is necessary to take into account the problems that may arise during the production phase and the points to be considered in order to match the post-production measurement results with the simulation results. For this reason, in the second step of the design, the design was scaled to the 1 GHz - 4.5 GHz frequency region, since the SMD resistor used was not stable in the high frequency region. Since the dimensions of the proposed absorber have changed, the optimization process has been started again. According to the results obtained, it has been observed that the normally incident wave is absorber by 90% in the frequency range of 1.052 – 4.56 GHz for TE mode and 1.06 – 4.60 GHz for TM mode. Then, similarly, Rogers 5880 was changed to Rogers 4003 against the problems that may arise in production due to the fragile and sensitive material of the substrate. The third phase of the design is to simulate again with this substrate. Since the dielectric coefficient changes, the optimization process is more difficult than the second phase. The change in the dielectric coefficient caused a decrease in the bandwidth, and according to the results obtained, it is seen that the normally incident wave is absorbed by 90% in the frequency band 0.99 – 4.00 GHz and 1.04 – 4.21 GHz for the TE and TM mode respectively. In this case, the recommended absorber bandwidth was 1:3.85 (4.00/1.04) and the thickness was 0.093λL. Moreover, simulations were made when the angle of incident wave was between 0° and 45° for analyzing the angular stability. It has been observed that the proposed absorber exhibits stable absorption characteristics between these angles. Furthermore, since foam material is planned to be used as a separator to support the substrate after manufacturing, it has been simulated with Rohacell having dielectric coefficient value of 1.05. According to these results, it has been seen that the proposed absorber fulfills one of the objectives of this thesis. One of the design objectives is reducing the RCS of systems that absorber installed on. The finite element array of absorber (12x12 unit cell) was designed and simulated for this purpose. RCS level of the absorber and the metal plate having the same size and thickness as the absorber, were compared. It was seen that the proposed absorber is capable of to reduce RCS by around 10 dB in the 1 - 4 GHz range compared to metal. Similarly, the RCS reduction performance of the proposed absorber is approximately 7.5 dB better than the LS-18 and LS-24 absorbers which are frequently utilized in the industry. In order to validate the obtained simulation results, the proposed absorber was manufactured and measured. It has been observed that the measured absorptivity agrees reasonably well with simulation result. It can be suggested that the proposed absorber may be used in different applications 5G, stealth technology and so on.