Broadband cavity filter with low insertion loss and design method thereof

Abstract

The invention relates to a low-insertion-loss broadband cavity filter and a design method thereof. The low-insertion-loss broadband cavity filter comprises a cavity, a cover plate covering the cavity, an input interface, an output interface and a plurality of resonators, wherein the input interface and the output interface are formed in the cavity; the resonators adopt a topological structure of a 10-order resonant cavity, and energy exchange is carried out between the adjacent resonators through a coupling structure, so that the insertion loss of the filter within the center frequency range of 1050 MHz to 1450 MHz is not greater than 0.3 dB, the 0.3 dB bandwidth is not less than 400 MHz, the ripple in a passband is not greater than 0.2 dB, and the port voltage standing wave ratio is not greater than 1.3. According to the invention, high-fidelity transmission of signals and high sensitivity of a system are ensured through extremely low insertion loss and in-band ripples; the ultra-wide flat passband is combined with the deep inhibition capability of 40dBc in a plurality of frequency bands, so that excellent frequency selectivity and anti-interference characteristics are provided; and the standing-wave ratio of the full-band inner port is 1.3, so that good impedance matching is ensured. The technical problem that broadband and low loss in a high-performance communication system are difficult to consider at the same time is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of microwave communication technology, and particularly to a low-insertion-loss wideband cavity filter and a design method thereof. BACKGROUND

[0002] Filters are widely used in radio frequency systems, and are classified into LC filters, dielectric filters, cavity filters, microstrip filters, waveguide filters, etc. according to different structural forms. Cavity filters have the advantages of high Q value, low insertion loss, large power capacity, moderate size, etc., and are widely used in communication systems.

[0003] In modern wireless systems such as navigation and radar, filters as key passive devices directly affect the signal quality, anti-interference ability and sensitivity of the entire system. In particular, for application scenarios such as base station receiving channels and repeaters, there is an urgent need for a filter with low insertion loss in a specific frequency band, flat signals in the passband, and effective suppression of out-of-band interference.

[0004] However, although the traditional cavity filter has the advantages of high Q value and large power capacity, when pursuing a wider operating bandwidth, it often faces technical difficulties such as increased insertion loss, deteriorated in-band ripple, and insufficient out-of-band suppression. The filter in the prior art cannot stably control the insertion loss to be below 0.3 dB in a passband as wide as 400 MHz, while ensuring excellent out-of-band suppression characteristics. Therefore, it is a technical problem to be solved in the field to provide a cavity filter that can balance low insertion loss, wide bandwidth, high selectivity and good port matching. SUMMARY

[0005] In view of the problems existing in the prior art, the present application aims to provide a low-insertion-loss wideband cavity filter operating in the 1050-1450 MHz frequency band with low insertion loss and wide passband characteristics. Another object of the present application is to provide a design method for the above-mentioned low-insertion-loss wideband cavity filter.

[0006] To achieve the above-mentioned objects, the low-insertion-loss wideband cavity filter of the present application comprises a cavity (1), a cover plate (2) covering the cavity (1), an input interface (3) and an output interface (4) provided on the cavity (1), and a plurality of resonators (5) arranged in the cavity (1). The resonators (5) adopt a topology structure of 10-order resonant cavities, and the adjacent resonators (5) exchange energy through a coupling structure (6), so that the insertion loss of the filter in the passband frequency range of 1050-1450 MHz (with a center frequency of 1250 MHz) is not greater than 0.3 dB, the 0.3 dB bandwidth is not less than 400 MHz, the in-band ripple is not greater than 0.2 dB, and the port voltage standing wave ratio is not greater than 1.3.

[0007] Further, the coupling structure (6) is a coupling window formed between adjacent resonators (5).

[0008] Further, the input interface (3) and the output interface (4) are connected with the resonators (5) in a direct tap coupling mode, and the insertion depth of the tap probe is configured to achieve an external quality factor In the range of 1500 to 2500.

[0009] Further, the side wall thickness of the cavity (1) is 7.45mm±0.1mm, the end wall thickness is 4mm to 6mm, and the bottom of the cavity (1) is provided with a mounting through hole with a diameter of 2.5mm to 3.2mm.

[0010] Further, the bonding surface of the cover plate (2) and the cavity (1) is provided with an electromagnetic shielding structure.

[0011] A design method of a low-insertion-loss wideband cavity filter, the design method is used for designing the above-mentioned low-insertion-loss wideband cavity filter, and the design method comprises the following steps: S1. Filter synthesis design: based on the generalized Chebyshev function approximation method, input the center frequency, bandwidth, passband ripple and stopband suppression index, determine the resonator topology structure through iterative analysis, and generate a coupling matrix, wherein cross coupling is arranged between non-adjacent resonators to introduce transmission zeros; S2. Three-dimensional electromagnetic simulation optimization: a resonator model is established using electromagnetic simulation, the tuning structure parameters are optimized through eigenmode analysis, and the coupling structure size is scanned to adjust the coupling coefficient; S3. Input and output coupling design: using tap coupling mode, the probe insertion depth is optimized to achieve the target external quality factor; S4. Overall model integration and optimization: integrate the resonant cavity, coupling structure and port into a complete model, and use an optimization algorithm to perform multivariate iteration with the S parameter curve as the target until the simulation result meets the requirements of insertion loss not greater than 0.3dB and bandwidth not less than 400MHz.

[0012] Further, in step S1, the order of the resonator topology structure is determined through iterative analysis to achieve stopband suppression not less than 40dBc, and the coupling matrix contains cross coupling elements to generate transmission zeros.

[0013] Further, in step S2, the calculation formula of the coupling coefficient is: ; Wherein, is the even mode frequency, is the odd mode frequency.

[0014] Further, the optimization algorithm is an automated process based on gradient optimization or sequential nonlinear programming, and the objective function is defined as the minimization of the difference between the simulated S-parameters and the ideal S-parameters.

[0015] Further, the cavity of the filter performs an electroplating process to reduce conductor loss, the electroplating process is a silver plating process, the thickness of the silver plating layer is not less than 5 μm, and the temperature of the plating solution is controlled in the range of 20 °C to 25 °C.

[0016] The technical advantages of the present application are: high-fidelity transmission of signals and high sensitivity of the system are ensured by extremely low insertion loss and in-band ripple; the super-wide flat passband combined with multiple frequency bands 40 dBc deep suppression capability, providing excellent frequency selectivity and anti-interference characteristics; the full-band in-band port standing wave ratio 1.3, ensuring good impedance matching. The present application solves the technical problem that wideband and low loss are difficult to be considered in high-performance communication systems. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a frequency response simulation curve diagram of a tenth-order Chebyshev bandpass filter; Figure 2 is an ideal frequency response curve diagram obtained based on filter synthesis theory; Figure 3 is an isometric line frame model diagram of a filter resonant cavity assembly; Figure 4 is a three-dimensional line frame model diagram of a filter resonant cavity assembly with a cross-section; Figure 5 is a schematic diagram of three coupling structures adopted by filter input / output ports; Figure 6 is an S-parameter simulation result diagram of a filter circuit; Figure 7 is a schematic diagram of a cavity filter structure; Figure 8 is a machining diagram of a filter cavity structure. DETAILED DESCRIPTION

[0018] The technical solutions of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0019] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0020] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0021] The specific embodiments of the present application will be described in detail below. Figures 1-8 The specific embodiments described herein are intended to illustrate and explain the present application, and are not intended to limit the present application.

[0022] The purpose of the present application is to overcome the technical problems of sharp increase of insertion loss and insufficient out-of-band suppression when expanding the bandwidth of the existing cavity filter, and to provide a low insertion loss wideband cavity filter and a design method thereof. Through its unique comprehensive design, structural optimization and precision manufacturing process, the filter simultaneously realizes 0.3dB of extremely low insertion loss, 400MHz of 0.3dB wide bandwidth, passband edge 40dBc of high out-of-band suppression, and 1.3 of excellent port VSWR, to meet the harsh requirements of modern wireless communication systems for high performance and high reliability of radio frequency front end.

[0023] As Figures 1-8 shown, the low insertion loss wideband cavity filter of the present application comprises a cavity 1, a cover plate 2 covering the cavity 1, and an input interface 3 and an output interface 4 arranged on the cavity 1, and a plurality of (at least two) resonators 5 are arranged in the cavity 1, wherein the resonators 5 are configured to have high unloaded quality factor, and the adjacent resonators 5 exchange energy through the coupling structure 6, so that the insertion loss of the filter at the center frequency of 1250MHz is not higher than 0.3dB, and the 0.3dB bandwidth is not less than 400MHz.

[0024] The bottom of the cavity 1 is provided with at least two mounting structures, which are through holes, the center of the mounting hole is 4mm away from the end face of the cavity 1, and the center distance D2 between the two mounting holes is 23mm. The diameter of the mounting structure through hole is Φ1, and Φ1 is 2.5mm to 3.2mm.

[0025] The center distance of the mounting hole of the input interface 3 and the output interface 4 from the bottom surface of the cavity 1 is H1, which is 31mm. The input interface 3 and the output interface 4 are SMA-KFD type connectors.

[0026] The side wall thickness of the cavity 1 is T1, T1 is (41mm-26.1mm) / 2=7.45mm; and / or, the end wall thickness of the cavity 1 is T2, T2 is (71.66mm-60.66mm) / 2=5.5mm.

[0027] The cover plate 2 is connected with the cavity 1 through a plurality of fixing members, and the mounting hole positions of the fixing members are distributed on the top edge of the cavity 1, wherein the center distance D3 between part of the mounting hole positions and the side edge of the cavity 1 is 5mm.

[0028] The passband of the filter is 1050MHz to 1450MHz, and the in-band ripple in the passband is not greater than 0.2dB. In the frequency range of 1050MHz to 1450MHz, the voltage standing wave ratio of the input interface 3 and the output interface 4 is not greater than 1.3. The suppression of the filter in the stop band of DC-890MHz and 1710-1880MHz is not less than 40dBc. The filter can maintain the insertion loss and 0.3dB bandwidth under the environmental temperature of-40°C to +70°C.

[0029] The resonator 5 is a metal resonant column, and the surface thereof is silver-plated; and / or, the coupling structure 6 is a coupling window formed between adjacent resonators 5.

[0030] Another object of the present application is to provide a design method of a low-insertion-loss wideband cavity filter, which is used for designing the low-insertion-loss wideband cavity filter, and the design method comprises the following steps: S1. Filter synthesis design: based on the generalized Chebyshev function approximation method, input the center frequency, bandwidth, passband ripple and stop band suppression index, determine the resonator topological structure through iterative analysis, and generate the coupling matrix, wherein the cross coupling between non-adjacent resonators is arranged to introduce transmission zero points; S2. Three-dimensional electromagnetic simulation optimization: a resonator model is established by using electromagnetic simulation, the tuning structure parameters are optimized by eigenmode analysis, and the coupling structure size is scanned to adjust the coupling coefficient; S3. Input and output coupling design: a tap coupling mode is adopted, and the probe insertion depth is optimized to realize the target external quality factor; S4. Overall model integration and optimization: integrate the resonant cavity, coupling structure and port into a complete model, and use an optimization algorithm to perform multivariate iteration with the S parameter curve as the target until the simulation result meets the requirements that the insertion loss is not greater than 0.3 dB and the bandwidth is not less than 400 MHz.

[0031] The application forms a complete technical scheme which can be repeated and popularized from theoretical synthesis, simulation optimization to process implementation through the design method of the low-insertion-loss wideband cavity filter.

[0032] The application performs systematic filter synthesis design based on the indexes of a center frequency (F0) of 1250 MHz, a 0.3 dB bandwidth (BW) of 400 MHz, a passband ripple of ≤0.2 dB and suppression of ≥40 dB at specific stopband frequency points (such as 900 MHz and 1800 MHz). The process is the core link of converting abstract indexes into an accurate blueprint that can be physically implemented, and the specific technical means and implementation steps are as follows: In the application, the generalized Chebyshev function approximation method is used to obtain extremely steep out-of-band suppression by accurately arranging transmission zeros (TZs) on the complex plane. The specific implementation is as follows: in the engineering settings of Ansoft Filter Synthesis (or similar tools such as CoupleFil), a new project is created and the type of "bandpass filter" is selected. In the parameter setting window, the following are input in sequence: Central Frequency = 1250 MHz Bandwidth = 400 MHz Passband Ripple = 0.2 dB In the "Stopband Attenuation" or "Transmission Zeros" submenu, specify that transmission zeros are needed at frequencies Fz1=900MHz and Fz2=1800MHz, and set the minimum suppression at the zero points to 40 dB.

[0033] First, the filter order (Number of Poles) is set to N=8, and the synthesis calculation is run. The software will generate the corresponding coupling matrix and frequency response curve; observing the simulation curve at N=8, it is found that the out-of-band roll-off rate is insufficient, and the suppression at 900 MHz can only reach about 32 dB, which cannot meet the requirement of 40 dB, and this result is recorded; the order is increased to N=9, and the synthesis is re-run. At this time, the suppression is improved to about 36 dB, but it still does not meet the standard; the order is set to N=10, and the synthesis is re-run. The simulation result shows that while maintaining the passband performance, the suppression at 900 MHz and 1800 MHz is stable and exceeds 40 dB. Therefore, 10 is determined as the minimum and optimal order under the condition of meeting all performance indicators. This decision is directly supported by the comparison of the quantifiable S parameter curve generated by the software. In order to achieve the required out-of-band suppression characteristics, the filter needs to adopt the topology structure of 10 resonant cavities.

[0034] The coupling matrix is a square matrix (for a 10-order filter, it is a 12x12 matrix), and the rows and columns of the matrix correspond to the input port (P1), the output port (P2) and the 10 resonant cavities (C1 to C10) of the filter; the diagonal elements (M ii ) represent the self-resonant frequency offset (normalized value) of each resonant cavity. When initially synthesized, they are usually set to zero or a value close to zero, indicating that all cavities are tuned to the same center frequency. The non-diagonal elements (M ij , i≠j) accurately quantify the strength of the energy coupling path, and each non-diagonal element M ij in the matrix is a coupling coefficient. The values of adjacent couplings (such as M 12 , M 23 ,..., M 9,10 ) (for example, M 12 =0.048) determine the passband width. Physically, it is directly mapped to the size of the adjacent resonant cavity coupling window (such as width W and height H). Subsequent electromagnetic simulation will optimize the physical size of the window with this value as the target. The value of the cross-coupling (such as M 1,9 ) (for example, M 1,9 =-0.015) is the key to generating transmission zeros and achieving high suppression. The negative sign indicates that the coupling is inductive or capacitive, and there is a phase reaction with the main coupling path. The physical realization of this cross-coupling is achieved by opening a specific cross-coupling window or using a probe / loop coupling structure between non-adjacent resonant cavities (such as the first cavity and the ninth cavity).

[0035] The input / output coupling Q value calculation formula of the application is: ; wherein, and are the first two element values of a normalized low-pass prototype filter, is the relative bandwidth. The value of is calculated by Ansoft Filter Synthesis software, which provides a preferred embodiment for achieving the filter performance. The Q value directly determines the size of the feed structure. For the direct tap-coupling method, the value is strongly related to the depth (d) of the probe inserted into the cavity. By the formula ∝1 / (d²) or more accurate electromagnetic simulation relationship, it can be determined that the initial probe depth d required to achieve the target (about 1980) is about 5.5 mm. In subsequent three-dimensional electromagnetic simulation (such as HFSS), this depth will be used as the starting point for parameter scanning fine-tuning.

[0036] The fine simulation of high-Q resonant cavity of the present application is a systematic modeling and parameterized analysis using three-dimensional electromagnetic simulation software ANSYS HFSS. The core of this process is the application of the eigenmode solver, and its technical connotation lies in calculating the inherent electromagnetic resonance characteristics of the resonant cavity under no external excitation, including the resonance frequency and the intrinsic quality factor (Q value). The specific implementation steps are as follows: First, the physical model of the single cavity is accurately established in the HFSS software. This is not a simple geometric body, but a parameterized model strictly following the processing design. Cavity modeling, create a cuboid model to represent an independent chamber of the filter, and its initial size is estimated and set according to the 1 / 8 wavelength of the center frequency; resonant rod and tuning screw modeling, create a cylinder on the center axis of the cavity as the main resonant rod. The key is to create an elongated cylinder with a diameter consistent with the actual screw (such as M3 thread) directly above it to accurately represent the metal tuning screw. The distance between the end of this screw and the top of the resonant rod (i.e. the insertion depth) is set as the key variable Tune_Screw_Length. When defining the material properties, set the material properties of all metal parts (cavity, resonant rod, tuning screw) to ideal conductor (PEC) or specify silver to simulate high electrical conductivity surface. This is the premise for accurately estimating the no-load Q value and thus relating to the low insertion loss performance.

[0037] Parameter scanning simulation of the length of the tuning screw is the core operation of fine design, and its specific technical means are as follows: Select Eigenvalue Solver in HFSS. Set the solution mode as "Fast Sweep" or "Interpolation Sweep" and add a parameter sweep analysis item; specify the sweep variable as Tune_Screw_Length defined previously. Set a reasonable sweep range, for example, from 20 mm to 28 mm with a step of 0.5 mm or less for high precision curve; then, invoke the Eigenvalue Solver to perform simulation calculation. This solver calculates all eigenmodes of the resonant cavity at each specific screw length and their corresponding resonant frequencies and intrinsic quality factors (Q values) by solving the Helmholtz equation. After the simulation is completed, the software outputs a detailed data list and a relationship curve of the resonant frequency of the fundamental mode (i.e., the lowest order mode) versus the screw length.

[0038] By analyzing the simulation results of the above parameter sweep, the corresponding relationship between the tuning screw length and the resonant frequency is determined. The analysis shows that the resonant frequency monotonically decreases with the increase of the screw length. To accurately adjust the fundamental mode resonant frequency to 1250 MHz, the length of the tuning screw needs to be optimized to 24 mm. At this length, the screw is screwed in about half of the total height of the cavity. This proportion design ensures that the tuning screw has sufficient mechanical stability in the cavity, avoiding deformation or vibration that may occur due to excessive length, while also ensuring that it has sufficient adjustment margin for fine tuning and calibration during production.

[0039] The broadband matching characteristics of the input / output ports are optimized, specifically using the direct tap coupling method. The implementation of this method is completed through the following detailed three-dimensional electromagnetic simulation process: In the HFSS software, create a rectangular cavity model representing the first resonant cavity of the filter. On the side wall of the cavity, draw a rectangular cross-section and cut out a rectangular window through the "Subtract" operation to simulate the interface of the connector extending into the cavity during actual assembly. At the center of the above window, create a vertical to the side wall of the cavity, thin and long rectangular sheet or cylinder to accurately represent the metal probe of the direct tap. One end of the probe is connected to the external port, and the other end is suspended inside the resonant cavity, with the length of its depth into the cavity, i.e., the coupling height (Port_Height), being set as a key design variable. At the side wall opening, use the "Wave Port" function of HFSS to define an excitation port. Set the integration line of this port to point from the probe to the cavity wall direction to accurately define the electric field polarization direction. At the same time, set the other surfaces of the cavity as ideal electric boundary (Perfect E) to simulate the closed metal cavity environment.

[0040] The "coupling height of the tap is set as a variable for parameter scanning optimization" in the present application is a key step of quantifying the relationship between coupling strength and geometric size through automation simulation. First, the variable is parameterized, and the coupling height Port_Height of the probe is defined as an optimization variable of HFSS; then in the analysis settings of the software, a parameter scanning task is added. The scanning range of Port_Height is set, for example, from 18.0 mm to 24.0 mm, and the scanning step is set to 0.2 mm. This fine step helps to accurately capture the best matching point. Finally, the "driven mode" solver is used to run the S parameter simulation. The software will automatically traverse all the set height values and calculate the return loss (S11 parameter) and insertion loss (S21 parameter) of the port at each height.

[0041] After the simulation is completed, a family of curves of S11 parameters with respect to frequency and coupling height is drawn. It is observed that near the center frequency 1250 MHz, the S11 parameter changes with Port_Height. The results show that the depth of S11 (i.e. the degree of port matching) is significantly improved with the increase of the coupling height. It is determined that the coupling height corresponding to the minimum S11 parameter at 1250 MHz (i.e. the maximum return loss) is 21.6 mm. At this optimized height, the external quality factor (Qe) is calculated by the following formula: ; wherein, (Stored energy) refers to the total electromagnetic field energy stored in the resonator (metal structure inside the cavity in the figure) in a complete oscillation period. This part of the energy is periodically converted between the electric field and the magnetic field. (Power consumed) refers to the energy radiated through the port and consumed due to conductor and dielectric loss in a oscillation period.

[0042] or the external quality factor (Qe) calculated directly from the S parameter fitting, is highly consistent with the target value (about 1980) calculated in advance by the filter synthesis theory. This result proves that by optimizing the coupling height (such as 21.6 mm), the Qe value can be accurately adjusted to any required value within the best performance interval of 1500-2500, thereby verifying at the circuit level that the structure size can ensure that the filter has good impedance matching characteristics within a wide band.

[0043] ​​​After the simulation verification of all sub-components (single cavity, port, coupling structure) is completed, the system-level collaborative design and optimization phase is entered. This phase aims to solve the performance deviation caused by complex electromagnetic coupling and ensure that the final design meets all preset indicators. The complete three-dimensional electromagnetic simulation model refers to the accurate digital reproduction of the physical structure of the filter, which is constructed based on the dimensions determined in the previous simulation. The following operations are performed in the HFSS software: First, the model is integrated, and ten independent resonant cavity models are arranged in three-dimensional space according to their topological structure (such as linear or folded). By using the "Unite" command in Boolean operation, the rectangular body representing the coupling window is accurately "Subtracted" from the metal partition between adjacent cavities to form a controllable energy channel. Then, the port and excitation are set. At both ends of the model, the optimized direct tap probe structure is associated with a "Wave Port" respectively. This port is defined as the input / output excitation end. Finally, full-wave electromagnetic simulation analysis is used, which is essentially the finite difference time domain method (FDTD) or the finite element method (FEM). Taking FEM as an example, the core is to solve the vector Helmholtz equation in the frequency domain: ; Where, is the electric field vector, is the relative magnetic permeability, is the relative permittivity, is the free space wave number. The HFSS software grids the entire model and solves the equation at each grid cell to accurately calculate the electromagnetic field distribution and resonance characteristics of the entire structure in the microwave frequency band. This "full-wave" analysis considers all possible radiation, coupling, and parasitic effects, so the simulation results are highly accurate, but there is also a deviation between the initial performance and the ideal synthesis theory.

[0044] The optimization algorithm used in the invention is an automatic process built into the simulation software based on mathematical optimization theory, mainly using gradient optimization algorithm and sequential nonlinear programming algorithm. The optimization process can be described as a multi-parameter constrained optimization problem. Let the optimization variable vector be: ; Where represents the tuning screw depth of the first resonant cavity, represents the coupling window width between the first and second resonant cavities, represents the rotation depth of the tuning screw of the second resonant cavity, represents the coupling window width between the second and third resonant cavities, represents the insertion depth of the input port (or output port) tap probe, This indicates the dimensions of a cross-coupled structure (e.g., the length of a coupling probe between non-adjacent cavities used to generate transmission zeros, or the offset of a coupling gap). This indicates the screw-in depth of the tuning screw in the third resonant cavity. This indicates the height of the coupling window between the third and fourth resonant cavities. It represents the local thickness compensation amount of a specific end wall or side wall of the cavity; This represents the Nth parameter, such as the depth and width of bosses or grooves added inside a specific resonant cavity to suppress unwanted modes; the layout and thickness of reinforcing ribs designed to enhance mechanical strength or heat dissipation; and the precise installation position of the input / output port grounding flange and its clearance from the cavity wall. Each component represents a geometric parameter (such as the depth of the tuning screw or the width of the coupling window).

[0045] objective function It is a function that measures the difference between the simulation results and the ideal S-parameter curve. For example, it can be defined as: ; in, These are the transmission coefficients obtained from simulation. It is the target transmission coefficient. They are different frequencies The weight, This refers to the number of sampling frequency points. The goal of optimization is to find a set of... , making Minimize. The execution flow of the algorithm is as follows: First, set the variables and objectives. In the "Optimetrics" module of HFSS, set about 10-15 key geometric parameters, such as the screw-in depth of the frequency tuning screw, the diameter of the coupling tuning screw, and the probe height of the input and output taps, as optimization variables. The algorithm is then selected and executed, with reasonable initial values ​​and a range of variation provided. The algorithm is either "Quasi-Newton" or "Sequential Nonlinear Programming". It automatically iterates: first, it runs an initial simulation to calculate the objective function. Then, by calculating the gradient or difference of the objective function with respect to each variable, the direction and step size of the adjustment for each variable in the next step are determined to reduce the error as quickly as possible. Then the variable values ​​are updated, and a new round of simulation begins. Finally, iteration and convergence checks are performed, and the above process is repeated cyclically. When the objective function... When the value is less than the preset tolerance (for example, the difference between the simulation curve and the target curve is within 0.1dB), or when the maximum number of iterations is reached, the optimization process automatically terminates and outputs the optimal combination of variables.

[0046] After the above-mentioned multi-round iterative optimization, the final three-dimensional model simulation result will stably meet all the preset indicators: insertion loss ≤ 0.3 dB at a center frequency of 1250 MHz, 0.3 dB bandwidth ≥ 400 MHz, and suppression ≥ 40 dB in the specified stopband. At this time, the electromagnetic design of the filter is completed, and the final model containing all the optimized sizes can be output to guide the processing and manufacturing.

[0047] The significant progress of the present application is to convert simulation high performance into physical high performance through the following specific process.

[0048] In order to minimize conductor loss, the cavity and internal resonator blank are made of 7075 series anti-rust aluminum alloy (in accordance with GB / T 3190-2020 standard), and the chemical composition contains 5.1-6.1% of zinc and 2.1-2.9% of magnesium. The material is detected by ultrasonic flaw detection to ensure that the internal defect size is not greater than 0.1 mm. The blank is subjected to solid solution treatment (heated to 470±5℃ for 2 hours, water quenching) and aging treatment (120±5℃ for 24 hours) before processing, so that the hardness reaches HRB≥85.

[0049] A five-axis linkage numerical control milling machine (such as DMG MORI CMX 1100 V) is used for precision machining, and the specific parameter settings are as follows: The specific content of tool path optimization: the spiral interpolation method is used for cavity inner wall processing, the tool axial cutting depth is 0.1 mm, and the radial cutting width is 0.5 mm; a ball end mill (diameter 6 mm, 4 blades) is used for curved surface finishing, and the line spacing is set to 8% of the tool diameter (0.48 mm); the tangent continuous feeding strategy is implemented to ensure smooth transition of the tool path and avoid vibration caused by sharp turns. The specific content of the cutting parameter precise control of the present application: spindle speed 18,000±100 rpm

[0050] , feed speed 800 mm / min, cutting depth: controlled at 0.05-0.1 mm in the finishing stage, micro-lubrication (MQL) technology is used, and the lubricant flow is controlled at 50 ml / h.

[0051] The surface finish is mainly ensured by the following measures: First, optimize the geometric parameters: rake angle: 12°, relief angle: 8°, and tool nose radius: 0.2 mm. In addition, control measures are set: set the dynamic stiffness coefficient of the machine tool to ≥80 N / μm; use an acceleration sensor to monitor the vibration amplitude in real time, control it within 0.5 m / s²; implement adaptive cutting force control, and limit the maximum cutting force to below 200 N.

[0052] After processing, the surface roughness instrument (Mitutoyo SJ-410) is used for detection, the Ra value is controlled to be below 0.4 microns, the surface profile is measured by a white light interferometer, it is ensured that the waviness is less than 0.1 microns, the surface micro morphology is checked by an electron microscope (500x magnification), and it is ensured that there is no microscopic crack. The process scheme makes the equivalent resistance of the current transmission path of the cavity surface reduce by about 35%, the surface conductivity is improved to 1.8 times of conventional processing at a frequency of 10 GHz, and reliable process guarantee is provided for realizing low insertion loss of <=0.3 dB.

[0053] The key parameters in the processing process are optimized in real time by the following formula:

[0054] Wherein, The cutting force (N) is, The material cutting coefficient (N / mm²) is, The cutting depth (mm) is, The feed per tooth (mm / z) is, The number of tool teeth is.

[0055] The surface roughness prediction adopts: ; Wherein The theoretical roughness (microns) is, The feed per tooth (mm) is, The tool nose radius (mm) is.

[0056] In order to effectively reduce the skin effect loss of the microwave frequency band, the following strictly controlled electrochemical silver plating process is carried out on the aluminum alloy cavity and resonator parts which have completed precision processing and cleaning treatment. The core of the process is to obtain a thick silver layer with dense crystal, uniform and low resistance by accurately controlling the plating solution system and electroplating parameters.

[0057] The application provides a high-reliability silver plating process for the inner surface of a cavity filter, which is realized through the following specific technical scheme: The plating solution formula mainly adopts a cyanide silver plating system, and specifically includes: silver potassium cyanide (KAg(CN)2): 40±2g / L, free potassium cyanide (KCN): 85±5g / L, potassium carbonate (K2CO3): 15±2g / L, brightener: 2-mercaptobenzothiazole 0.15-0.25mL / L.

[0058] The main process control parameters include the following: Temperature control: the plating solution temperature is maintained at 22±1℃ through a constant temperature system; Current density: a direct current power supply is adopted, and the cathode current density is controlled at 0.5±0.1A / dm²; Deposition time: Deposition time is calculated according to Faraday's law: ; Wherein is time (min), is target thickness (pm), is silver density (10.5 g / cm³), is area (dm²), is current density (A / dm²), is current efficiency (take 0.95), is electrochemical equivalent (4.025 g / Ah).

[0059] The main process flow includes pretreatment process, electroplating process control and post-treatment. The detailed steps will be introduced as follows: In the pretreatment process stage, the main purpose is to provide a clean and activated plating surface for the aluminum cavity parts. First, alkaline degreasing is performed, and the machined aluminum alloy cavity parts are immersed in an alkaline degreasing solution with a temperature of 60±5℃ and a sodium hydroxide concentration of 50±5g / L for 5±1 minutes to remove surface oil. After completion, the parts are rinsed with deionized water. Subsequently, acid pickling and activation are performed, and the parts after oil removal and rinsing are immersed in a 10% nitric acid aqueous solution at room temperature for 60±10 seconds to remove the surface oxide film. After treatment, deionized water is used for spraying and rinsing again. Finally, zinc immersion treatment is performed, and a dense transition layer is prepared on the aluminum alloy substrate. The parts are immersed in a zincate solution (formula: zinc oxide 100±10g / L, sodium hydroxide 500±20g / L) at room temperature for 60±10 seconds, and then rinsed after taking out; the zinc immersion-rinsing process is repeated once, i.e. twice zinc immersion, to ensure uniform coverage of the zinc layer.

[0060] The main core deposition process is carried out in the automatic control electroplating tank when the electroplating process control is carried out. First, hanging is carried out, and the parts after the completion of the pretreatment are used as cathodes, and the hanger is made of titanium metal to ensure the electrical conductivity and corrosion resistance. The anode is made of pure silver plate (silver content ≥ 99.99%), and the area ratio of the cathode to the anode is controlled to be 1:2. Then, the plating solution is circulated and filtered, and the plating solution circulation and filtration system is started, so that the plating solution continuously passes through the filter core with a precision of 5 μm at a rate of 2 times per hour, so as to maintain the cleanliness of the plating solution. After the above step is completed, electro-deposition is carried out, the direct current source is started, the cathode current density is set to 0.5±0.1 A / dm², and the electro-deposition process is started. The plating solution temperature is maintained at 22±1℃ by a constant temperature system. The deposition time is determined according to the target thickness by Faraday's law, and the silver layer thickness is ensured to be not less than 5 μm. Finally, composition monitoring is carried out, and the silver ion concentration in the plating solution is detected by atomic absorption spectrometry during the electroplating process, and the concentration is stabilized in the range of 28-32 g / L by adding silver potassium cyanide mother liquor.

[0061] Finally, the post-treatment process is carried out. First, water washing is carried out, and after the electroplating is completed, the parts are immersed in three countercurrent rinsing tanks in series, each tank uses flowing deionized water for rinsing, and the total water washing time is not less than 3 minutes, so as to completely remove the residual plating solution on the surface of the parts. Then, drying is carried out, and the washed parts are placed in a hot air drying oven at a temperature of 80±5℃ for 30±5 minutes to obtain dry and clean final products.

[0062] All the parts completed by the above processes are subjected to quality detection according to the following standards: Thickness uniformity detection is carried out by using an eddy current thickness gauge, and at least five points on the main surface of the part (such as the inner wall of the resonant cavity and the surface of the resonant rod) are randomly selected for thickness measurement, and the thickness difference between any two points is required to be not more than 0.5 μm. The bonding force test is carried out according to the standard GB / T 5270-2005, and the sampled parts are subjected to thermal shock test. The specific method is to place the parts in an oven at 200±5℃ for 30 minutes, and then quickly put them into deionized water at 25±5℃, and this process is repeated for 3 times. After the test, the plating layer is observed using a 10 times magnifying lens, and no blistering or peeling phenomenon is considered to be qualified. Surface quality inspection is carried out by using a surface roughness gauge, and the surface roughness Ra value after plating is required to be not more than 0.2 μm. Visual inspection under a white fluorescent lamp shows that the surface should be uniform in color, without defects such as pinholes, cracks and peeling.

[0063] The cavity filter part processed by the above process is detected, and the average thickness of the plated layer can reach 5.2±0.3 microns, the volume resistivity of the plated layer is not greater than 2.5 mu omega*cm, compared with the surface of the silver-plated aluminum alloy, the surface conductivity is increased by about 8 times at the working frequency of 10 GHz, and the skin effect loss is significantly reduced, which provides a key process guarantee for realizing the low insertion loss of the filter of ≤0.3 dB.

[0064] The present application determines the optimized size according to electromagnetic simulation to process and manufacture the resonator, so that the shape and area of the cross section of the resonator can support the lowest current distribution resistivity per unit area. This design cooperates with the high-finish substrate and the thick silver-plated layer to minimize the energy loss from the physical root.

[0065] Compared with the prior art, the technical scheme provided by the present application has the following advantages: Through the above-mentioned systematic scheme from comprehensive design, simulation optimization to precision manufacturing, the excellent performance combination of wideband, low insertion loss and high suppression is successfully realized, and the technical contradiction between bandwidth and insertion loss in the traditional cavity filter design is effectively solved.

[0066] The present application provides a complete and repeatable design and manufacturing method from theoretical calculation, simulation optimization to process landing. The method has clear process flow and clear key process parameters, significantly reduces the discreteness of product performance, ensures the consistency of product performance and high yield rate under batch production conditions, and has high industrial practicability.

[0067] The present application provides a complete and repeatable design and manufacturing method from theoretical calculation, simulation optimization to process landing. The method has clear process flow and clear key process parameters, significantly reduces the discreteness of product performance, ensures the consistency of product performance and high yield rate under batch production conditions, and has high industrial practicability.

[0067] The present application provides a complete and repeatable design and manufacturing method from theoretical calculation, simulation optimization to process landing. The method has clear process flow and clear key process parameters, significantly reduces the discreteness of product performance, ensures the consistency of product performance and high yield rate under batch production conditions, and has high industrial practicability.

[0068] The cavity is precisely divided into multiple resonant cavities by a metal partition plate. The internal width size (26.1 mm) shown in the cross-sectional view and the fixing mode of the resonant column jointly determine the fundamental mode frequency and high mode suppression capability of the resonator, which are the physical core of realizing high no-load Q value (low insertion loss) and wide bandwidth.

[0069] In order to ensure the performance consistency, the key size needs to be strictly controlled. For example, the positioning size tolerance between the resonant cavities should be controlled within ±0.05 mm, and the uniformity tolerance of the cavity side wall thickness (7.45 mm) should be controlled within ±0.1 mm, so as to ensure the accuracy of the electromagnetic field distribution.

[0070] The top cover of the cavity is provided with a threaded hole for mounting a tuning screw at the position corresponding to the center of each resonant cavity. By rotating the screw, the resonant frequency of each resonant cavity can be fine-tuned to ensure that the ripple in the passband is ≤0.2 dB. The cover and the cavity are fastened by peripheral screws to form a complete electromagnetic shield, ensuring that the out-of-band rejection is ≥40 dBc.

[0071] To clearly show the specific technical solutions of the present application and facilitate understanding and implementation by those skilled in the art, the design process, key structure, simulation verification and process implementation of the low-insertion-loss wideband cavity filter of the present application will be described in detail below in conjunction with the drawings of the specification. The drawings are intended to visually demonstrate the entire process from theoretical synthesis, structural optimization to performance verification.

[0072] Figure 1 The frequency response simulation curve of the ten-order Chebyshev bandpass filter of the present application is a theoretical basis and optimization target for the comprehensive design of the physical structure of the filter (such as cavity size, coupling coefficient). In subsequent embodiments, all three-dimensional electromagnetic simulations and iterative optimization of physical structures aim to approach the response characteristics of this curve as the final goal, thereby ensuring that the manufactured physical filter can achieve the expected low insertion loss, high selectivity and excellent signal fidelity.

[0073] Figure 2 The ideal frequency response curve obtained based on the filter synthesis theory of the present application is a theoretical blueprint for filter design. It determines the final performance indicators that the filter should achieve, including center frequency, bandwidth, in-band insertion loss, out-of-band rejection and port matching degree, through mathematical synthesis method. In subsequent three-dimensional electromagnetic simulations and physical structure optimization, all designs aim to approach this target curve as the final goal.

[0074] Figure 3 The isometric line frame model of the filter resonant cavity assembly of the present application is shown in Figure 1, Figure 3 which shows the three-dimensional structure model of the filter resonant cavity assembly in the embodiment of the present application. As shown in Figure 1, Figure 3 the main body of the assembly is cylindrical and adopts a multi-layer sleeve structure design, which is placed in a hexagonal boundary space, showing the mounting or heat dissipation interface. Figure 3 A scale of 0 to 20 millimeters is provided in the lower left corner to clearly indicate the relative size of the structure. The coordinate system is marked with X-axis (red), Y-axis (green) and Z-axis (blue), and the Z-axis direction is the main axis direction of the assembly. This three-dimensional line frame model clearly expresses the spatial geometric shape, proportional relationship and basic structural characteristics of the resonant cavity, providing a basis for subsequent detailed engineering design and electromagnetic simulation.

[0075] Figure 4 The three-dimensional line frame model with cross-section of the filter resonant cavity assembly of the present application is shown in Figure 2, Figure 4The internal structure of the resonant cavity is further shown in cross-section. Figure 4 The middle cylindrical body is composed of solid lines and dashed lines, which reveals the three-dimensional level and internal profile of the component. The hexagonal outline defines the installation boundary of the assembly. The millimeter scale at the bottom provides a reference for size measurement. The X, Y, and Z three-dimensional coordinate axes clearly indicate the orientation of the component in space. Figure 4 The internal structure, wall thickness, and relative position between the components of the resonant cavity are clearly expressed, which is crucial for understanding its working principle and manufacturing process.

[0076] Figure 5 The three coupling structures used for the filter input / output port of the invention are shown in the schematic diagram, Figure 5 Three core coupling structures are shown to achieve wideband matching of the filter: direct tap coupling, probe coupling, and ring coupling. The core is to efficiently feed the signal into the resonant cavity system through different electromagnetic field interaction mechanisms. The direct tap coupling (left) structure extends the metal probe directly into the cavity, and adjusts the insertion depth (key dimensions such as 21.6mm) to accurately control the coupling strength. Its external quality factor (Qe =1980) is the design goal to achieve a specific bandwidth. The probe coupling (middle) structure optimizes the geometry of the probe and the matching network to achieve good impedance matching in a wider frequency range (such as 18-24mm length adjustment range), suitable for wideband applications. The ring coupling (right) structure uses the magnetic field generated by the metal ring to inductively couple with the resonant cavity. This mechanism can provide higher port isolation and structural stability, and its performance is determined by the size of the ring (such as outer diameter 44.0mm).

[0077] Figure 6 The S-parameter simulation results of the filter circuit of the invention are shown in the figure. The S21 transmission curve (purple) accurately outlines the passband characteristics of the filter at a center frequency of about 1.25GHz. Its flat top and steep edges indicate that the design ensures low-loss (close to 0dB) transmission of signals within a bandwidth of about 400MHz, while achieving sharp suppression (more than -60dB) of out-of-band signals. The S11 return loss curve (cyan) is deeply below -20dB in the passband, with a minimum of -40dB, demonstrating good port impedance matching performance, ensuring efficient transmission of signal energy rather than reflection.

[0078] Figure 7 The structural diagram of the invention is shown in the figure, which uses front view, top view and left view Figure ThreeThe basic projection view and the two partial sectional views of A-A' and B-B' are provided to fully show the internal and external structures of the filter cavity. The overall external dimensions of the cavity (length 71.66 mm, width 52 mm, height 31 mm), the layout and key dimensions of the internal resonant cavity, the mounting positions of the input and output ports (the center distance of the mounting holes is 4 mm, the center distance of the two holes is 23 mm, and the hole diameter is Φ2.8 mm), and the wall thickness and mounting structure for ensuring mechanical strength and electromagnetic shielding are clearly marked in the drawing. The drawing provides accurate engineering basis for the specific physical implementation of the filter of the application.

[0079] Figure 8 The machining drawing of the filter cavity structure of the application uses the front view, top view and side view to jointly present the external contour and layout of the part, and accurately reveals the internal structural features (such as wall thickness, stepped groove, etc.) of the part through the two partial sectional views of A-A' and B-B'. The letters A / A', B / B' and the like marked in the drawing are sectional markers for clearly indicating the position and direction of the section plane, ensuring that the view correspondence is accurate and reliable.

[0080] Embodiment: The embodiment relates to a wideband cavity filter working in a 1050-1450 MHz frequency band, the center frequency of which is 1250 MHz, and the 0.3 dB bandwidth is 400 MHz. The design of the filter starts from a strict filter synthesis process, and through professional software, indexes are systematically analyzed to determine that a ten-order resonant cavity structure is the optimal scheme for realizing excellent out-of-band suppression performance. This decision is based on detailed iterative analysis: 8-order, 9-order and 10-order topologies are sequentially synthesized and simulated, and the results show that the 10-order structure can provide more than 40 dB of suppression capability at 900 MHz and 1800 MHz, while maintaining a passband ripple of less than 0.2 dB, fully meeting the requirements of a high-performance communication system.

[0081] After determining the basic topology, the application designs key components through three-dimensional electromagnetic simulation software. The simulation of the single-cavity resonator adopts the eigenmode analysis method, and the accurate relationship between the tuning screw length and the resonant frequency is determined through parameter scanning, and the optimal screw length at the center frequency of 1250 MHz is optimized to be 24 mm. This size not only ensures the accuracy of the electrical performance, but also ensures the stability and adjustability of the mechanical structure. At the same time, the port coupling structure is realized through a direct tap method, and through parameter scanning optimization, the probe insertion depth is determined to be 21.6 mm, so that the external quality factor of the port is well matched with the theoretical value 1980, providing ideal interface conditions for wideband signal transmission.

[0082] After the simulation verification of all sub-components is completed, a complete three-dimensional model containing ten resonant cavities is established to perform full-wave electromagnetic simulation analysis. The initial simulation result shows that due to the complex electromagnetic coupling between the cavities, the performance deviates from the ideal target. Therefore, the present application sets the frequency tuning screw, the coupling tuning screw and the input / output probe height as the optimization variables, and uses the optimization algorithm of the simulation software to perform multiple rounds of iterative simulation with the ideal scattering parameter curve as the target. This process continuously adjusts the parameters until the simulation result meets all the preset indicators such as the center insertion loss not greater than 0.3 dB, the bandwidth not less than 400 MHz, and the out-of-band suppression not less than 40 dB.

[0083] In terms of manufacturing process, the present application ensures the performance realization through a series of precision machining and processing techniques. The cavity is made of 7075 series rust-resistant aluminum alloy material and is processed by a high-precision numerical control milling machine, with optimized tool path and cutting parameters to ensure that the smoothness of all microwave transmission surfaces reaches an extremely high standard, fundamentally reducing the conductor loss. To further suppress the skin effect in the microwave frequency band, all internal surfaces of the parts are subjected to strict electrochemical silver plating treatment by controlling parameters such as plating solution composition, temperature, current density and deposition time to ensure that the silver layer thickness is not less than 5 microns, providing a low-resistance channel for surface current. Finally, the manufactured filter based on the optimized design model is verified to meet or exceed the design requirements in terms of performance indicators, fully proving the effectiveness and reliability of the technical solution.

[0084] The filter has an outer dimension of 71.66mm x 52mm x 31mm (excluding the size of the radio frequency connector and the tuning screw), and this compact design is due to the optimization of the single-cavity resonator structure, which effectively controls the product size while meeting the electrical performance indicators. Specifically, when the frequency is low and the size requirement is strict, the single-cavity size is reduced by optimizing the resonator structure (such as using a resonator rod with a loading structure), thereby realizing overall miniaturization.

[0085] The filter topology structure of the present application is determined systematically by professional synthesis software (such as Ansoft Filter Synthesis). The specific implementation process is as follows: In the software interface, a new filter synthesis project is created. First, set the basic parameters: select the bandpass filter for the filter type and the Chebyshev response for the response type. In the parameter input interface, accurately input the design indicators: center frequency 1250MHz, bandwidth 400MHz, passband ripple 0.1dB. To achieve the out-of-band suppression not less than 40dB at 900MHz and 1800MHz, the software introduces two transmission zeros in the transfer function, and the mathematical expression is: ; wherein, and are polynomials, s is a complex frequency variable, is an index variable, here the summation / integration index, representing the transmission zero number. is the transmission zero position, is the conjugate complex. The software performs order optimization analysis through an iterative algorithm, and sequentially calculates the performance indicators of 8th, 9th and 10th order topologies. When the 10th order Chebyshev response is adopted, the coupling matrix generated by the software shows that there is appropriate cross coupling between non-adjacent cavities , and this structure can generate the required transmission zero in the stop band. At the same time, the software calculates the external quality factor = 1980, which is verified by the following formula: ; wherein, , is the low-pass prototype parameter (i.e. the first two element values of the normalized low-pass prototype filter, which are determined by the passband ripple), is the relative bandwidth.

[0086] In the actual design process, in order to ensure the performance margin of the product, the set indicators are improved compared with the final requirements: the bandwidth is set to 420MHz, the out-of-band suppression requirement is increased to 45dB, and the VSWR requirement is increased to 1.25:1. This method of reserving design margin ensures that the final product can meet the performance requirements stably under various working conditions.

[0087] Through the above systematic synthesis analysis method, the ten-order Chebyshev topology structure is determined as the optimal scheme to achieve high performance indicators, which provides a reliable theoretical basis for subsequent detailed design and optimization.

[0088] The single-cavity resonator adopts a coaxial cavity structure, the inner conductor is a cylindrical resonant rod, and the frequency tuning is realized by a metal tuning screw. In the initial design, the single-cavity length is set to one-eighth of the working wavelength (about 1 / 8λ) to balance the size and performance. The bottom of the resonant rod is kept a certain distance from the bottom of the cavity, and this distance together with the tuning screw determines the resonant frequency; in the simulation model, the tuning screw is set to the same diameter as the actual processing to improve the simulation accuracy.

[0089] To obtain high Q value and reduce insertion loss, the characteristic impedance of the single-cavity coaxial line is preferably set to 76Ω, at which the theoretical Q value is the highest. The cavity material is selected from the 7075 series of rust-resistant aluminum materials, which has high density and good hardness, and the surface finish after processing is excellent, which is beneficial to reducing microwave transmission loss. The selection of this material is based on its dense texture, fewer bubbles and holes, and high smoothness surface can be obtained directly through CNC processing, which is one of the key technologies to reduce insertion loss.

[0090] The simulation modeling of the single cavity adopts the eigenmode solver of the HFSS software. The influence of the length of the tuning screw on the resonant frequency is analyzed through parameter scanning, and it is determined that the initial length of the tuning screw is about 24 mm at the center frequency of 1250 MHz, which is about half of the height of the cavity, facilitating subsequent production debugging. The simulation results show that the longer the tuning screw, the lower the resonant frequency, and the relationship curve provides an important reference for subsequent debugging.

[0091] In the present application, the fine adjustment of the inter-cavity coupling is a systematic engineering based on three-dimensional electromagnetic field simulation and parameterized scanning, and the core is to accurately map the electrical parameters (coupling coefficients) obtained from filter synthesis theory to the geometric dimensions of the physical structure. The specific technical means are as follows: First, a simulation model containing two adjacent resonant cavities is established in the HFSS software. The model accurately reproduces all the key dimensions of a single cavity, and a rectangular opening, i.e. a coupling window, is created on the common wall between the two cavities. The width (W) and height (H) of the window, as well as the position of the window center relative to the cavity wall (Offset), are set as key design variables. The solver type is set to driven modal (Driven Modal). Two wave ports (Wave Port) are set at the ends of the two cavities of the model away from the coupling window for excitation and energy extraction. The simulation analysis type is terminal S parameter (Terminal S-Parameter).

[0092] Coupling coefficient The extraction is based on the frequency splitting principle of two resonant cavities under weak coupling conditions. Eigenmode (Eigenmode) simulation is performed on the established double-cavity model. In this simulation mode, the software directly calculates all the eigenmodes of the coupling structure and their corresponding resonant frequencies by solving the Helmholtz equation. The two lowest order eigenmode frequencies are the even mode frequency and the odd mode frequency . The even mode frequency : In this mode, the electric field vector directions of the two cavities are the same, and the symmetry plane (i.e. the midplane of the coupling window) is approximately a magnetic wall (tangential magnetic field is zero, normal electric field is zero). The odd mode frequency : In this mode, the electric field vector directions of the two cavities are opposite, and the symmetry plane is approximately an electric wall (tangential electric field is zero, normal magnetic field is zero). Energy is strongly exchanged through the coupling window, and the coupling effect is significant. Energy is constrained within the respective cavities, and the coupling effect is weak. The coupling coefficient is calculated by the following formula: ; By this method, the coupling coefficient value corresponding to the current coupling window size can be directly obtained from electromagnetic field simulation.

[0093] The "fine tuning" of the present application is an iterative process realized through systematic parameter scanning and data analysis. First, fix the height H and position Offset of the coupling window, set the width W as the scanning variable, and set a reasonable range (for example, from 2.0 mm to 5.0 mm, with a step of 0.2 mm). After running the simulation, HFSS will output a set of coupling coefficient k values corresponding to different widths W. Plot the curve of k versus W. According to the target coupling coefficient obtained from the filter synthesis theory (for example, for a certain pair of cavities, =0.05), find the corresponding target width W_target on the above curve. If the data point on the curve does not pass through the target value directly, the exact W_target can be calculated by interpolation. After completing the width optimization, further set the height H as a variable and repeat the above scanning process to fine-tune the coupling strength or optimize the frequency dependence of the coupling (mainly for achieving specific zero-point positions). For designs requiring extremely high precision, it may be necessary to set both W and H as variables and use the optimization toolbox (such as Optimetrics) of HFSS to automatically iterate to converge to the best size combination as quickly as possible. During the adjustment process, the sensitivity of the coupling coefficient to changes in window size should also be evaluated. The higher the sensitivity, the stricter the processing tolerance requirements. Through sensitivity analysis, a design point that is relatively insensitive to size changes can be selected to improve product yield and consistency in mass production while meeting performance indicators.

[0094] Through the above systematic parameter scanning, data analysis and target-driven optimization, the present application realizes fine tuning of the coupling window size, ensuring that the physically realized coupling coefficient matches the theoretical synthesis result highly, thereby accurately controlling the passband bandwidth and shape of the filter.

[0095] The input-output coupling adopts a direct tap (barrel tap) method, which is simple and reliable, and easy to debug. The optimal insertion depth of the tap is determined by simulating the port impedance matching, and the external Q value is designed as 1980, based on which the initial tap height is calculated to be about 21.6 mm. This tap method belongs to inductive coupling, with moderate coupling degree, which is particularly suitable for the wideband filter application of the present embodiment.

[0096] The port coupling simulation of the present application is completed in ANSYS HFSS software, and the specific implementation process is as follows: First, an accurate port simulation model is established, a rectangular window is opened on the side wall of the resonant cavity, and the depth parameter of the probe into the cavity is defined as a variable Port_Height. The wave port excitation is set, and the integration line direction is from the probe to the cavity wall to accurately define the electric field polarization direction. The simulation uses the driving mode solver, and the tap height is systematically scanned and analyzed. The scanning range is set from 18.0 mm to 24.0 mm, the scanning step is 0.2 mm, and there are 31 sampling points in total. For each height value, the software solves the Maxwell equations by the finite element method to calculate the complete S parameter matrix.

[0097] The simulation results are analyzed by extracting the S11 parameter curve corresponding to each height and observing the return loss value at the center frequency 1250 MHz. Data analysis shows that as the tap height increases from 18.0 mm to 24.0 mm, the return loss improves from -8.2 dB to -26.5 dB, indicating that the port matching degree is significantly improved. The quantitative relationship between coupling strength and tap height is characterized by the external quality factor . The calculation formula of the value is: ; Wherein, is the frequency bandwidth corresponding to the change of S11 phase by 180°. Simulation data shows that when the tap height increases from 18.0 mm to 24.0 mm, the value decreases from 2850 to 1650, indicating that the coupling strength increases with the increase of height.

[0098] Finally, the optimal tap height is determined to be 21.6 mm, at which the return loss measured at 1250 MHz is -25.8 dB, the value is 1980, which is completely consistent with the theoretical value of filter synthesis. At this height, the port voltage standing wave ratio reaches 1.15, completely meeting the design requirement of VSWR≤1.3.

[0099] Through the above systematic parameter scanning simulation, the present application establishes an accurate corresponding relationship between the tap height and the coupling strength, providing a reliable optimization basis for port design.

[0100] After completing the simulation of all sub-components (including single cavity, coupling structure, input and output port), the overall three-dimensional model of the filter is established. The model includes ten resonant cavities, all inter-cavity coupling structures, input and output taps, and frequency and coupling tuning screws, laying a foundation for accurate simulation. The overall model integrates all adjustable variables and is the basis for final performance verification and optimization.

[0101] The initial simulation result of the whole model shows that the center frequency is shifted, which is caused by the mutual influence between cavities after the whole modeling. By reducing the size of all resonant cavities or shortening the length of the tuning screw, the center frequency is raised to the target value. This is because in the whole model, the coupling between cavities introduces additional parasitic capacitance, resulting in a decrease in the overall resonant frequency, which is a common phenomenon in modeling and needs to be corrected by systematic adjustment.

[0102] The further optimization of this embodiment is assisted by the CST Filter Designer tool. The specific implementation steps are as follows: First, the complete model S parameter results obtained by HFSS simulation are exported as standard.s2p file format. This file contains the complete scattering parameter data of the filter in the frequency band. In the CST Filter Designer, create a new project, select the coupling matrix synthesis type, and import the exported.s2p file into the software for data fitting. The software fits the response by the following algorithm: ; Wherein, is the simulation value, is the target response, and N is the number of sampling points.

[0103] After fitting, the software displays the deviation of each design parameter in a graphical way. The tool interface clearly shows the comparison curve of the actual response and the ideal response, and lists the key parameters such as the deviation of the resonant frequency and the error of the coupling coefficient in numerical form. Based on the deviation analysis results, the system automatically generates optimization suggestions. For the resonant cavity with a large frequency deviation, adjust the length of its tuning screw. The adjustment amount ΔL is calculated by the following formula: ;

[0104] Wherein, is the proportionality coefficient, is the frequency deviation, is the current screw length. For the coupling coefficient deviation, adjust the size of the coupling window accordingly. The width adjustment amount is related to the coupling coefficient deviation : ; Wherein, is the sensitivity coefficient (determined by the previous simulation data), is the current design width of the coupling window, is the coupling coefficient deviation. After 3-5 iterations of optimization, the deviation between the simulation result and the ideal response is significantly reduced, and all parameters meet the design requirements. This method greatly improves the optimization efficiency, shortening the traditional optimization process of several days to a few hours.

[0105] After several rounds of iterative optimization, the simulation results meet the requirements: the in-band insertion loss is less than the specified value, the out-of-band suppression meets the requirements, and the in-band VSWR is better than the specified value. At this time, the filter design is finalized, and the processing drawing can be drawn. The final goal of optimization is to make the S parameter curve (including S11 and S21) obtained by simulation completely fall within the template range determined by the index requirements.

[0106] To further reduce the loss caused by the skin effect, all aluminum cavity surfaces are plated with silver. The thickness of the silver plating is optimized and inversely proportional to the insertion loss performance. In the embodiment, a thick silver plating layer is used to improve performance. That is, by increasing the thickness of the silver plating on the surface of the conductor, the high-frequency skin effect is directly suppressed, thereby significantly improving the insertion loss.

[0107] The present application adopts a systematic design method for the optimization of the resonant rod, and the specific implementation process is as follows: First, material selection optimization. By comparing and analyzing the electrical conductivity, processability and cost of various metal materials, it is determined to use H62 brass as the base material, and its electrical conductivity reaches 1.6 x 10 7 S / m. To further improve the surface conductivity, the brass base material is plated with silver, and the silver plating layer thickness is controlled at 8-12 μm, so that the surface conductivity is improved to In terms of geometric shape optimization, the present application adopts a parameterized modeling method. A three-dimensional model of the resonant rod is established, and its diameter D and height H are set as design variables. Through electromagnetic field simulation analysis of current distribution under different size combinations, it is found that there is obvious skin effect of current on the surface of the resonant rod. To optimize the current distribution, the traditional cylindrical cross section is changed to a special shape with a round corner, and the round corner radius R is set to 0.5 mm.

[0108] The following evaluation indicators are used in the optimization process: ;

[0109] Among them, is the AC resistance, is the electrical conductivity, is the skin depth, is the thickness of the conductor, is the current path length.

[0110] The optimal size combination is obtained by finite element analysis: diameter D = 8 mm, height H = 25 mm, and round corner radius R = 0.5 mm. This shape makes the current distribution more uniform, effectively reducing the eddy current loss.

[0111] In terms of processing technology, a combination of precision turning and grinding process is adopted. First, a CNC lathe is used for rough machining, leaving a 0.1mm allowance; then, fine grinding is performed to ensure that the diameter tolerance is controlled within ±0.01mm, and the corner profile error is not more than 0.005mm. The final optimized unit area resistivity of the resonant rod: ≤3.2×10 -8 Ω·m, surface roughness: Ra≤0.4μm, Corner consistency: ±0.003mm.

[0112] Through the above systematic optimization method, the performance of the resonant rod is significantly improved, laying a solid foundation for the optimization of the overall performance of the filter.

[0113] The tuning system in the application includes two types of frequency tuning screws and coupling tuning screws, both of which adopt specific structural designs to ensure optimal performance and production feasibility. All tuning screws are made of H62 brass material, which has excellent non-magnetic properties, effectively avoiding the introduction of additional magnetic loss. The screws adopt M3×0.5 fine thread structure, with a pitch accuracy controlled within ±0.01mm, ensuring precision and stability during adjustment. The total length of the screws is designed differently according to their position in the filter, with the standard length of the frequency tuning screw being 25mm and the standard length of the coupling tuning screw being 15mm. The end of the frequency tuning screw is designed in a conical shape with a taper angle of 60° and a top corner radius of 0.1mm. This design allows the contact area between the screw and the resonant rod to change gradually during screwing, which is beneficial for linear adjustment of the frequency. The screw head is designed as a cross slot structure with a slot depth of 0.8mm and a slot width of 0.6mm, which perfectly matches the standard debugging tool. The end of the coupling tuning screw is designed in a hemispherical shape with a spherical radius of 1.5mm. This shape allows for smoother coupling changes when adjusting the coupling window. The screw head adopts an internal hexagonal design with an angular size of 2.5mm and a depth of 3mm, which provides greater torque transmission capacity to prevent slipping during frequent debugging.

[0114] Both types of tuning screws are designed with anti-loosening structures at the thread tail, specifically by machining a 0.1mm chamfer at the last two threads, and setting corresponding elastic washers in the screw seat. This double anti-loosening design ensures that the filter can maintain stable electrical performance in a vibrating environment. The surface of all tuning screws is plated with silver, with a plating thickness of 5-8μm. After silver plating, passivation treatment is required to form a protective film, which not only ensures good electrical conductivity but also improves corrosion resistance. Tests have shown that this surface treatment can reduce the contact resistance of the screw to below 0.5mΩ.

[0115] Through the above special structure design, the tuning screw of the application not only meets the electrical performance requirement, but also greatly improves the production debugging efficiency and reliability, and provides a strong guarantee for mass production of the filter.

[0116] The cover plate design ensures good contact with the cavity, providing sufficient electromagnetic shielding. The cover plate and the cavity joint surface are designed with a flat flange, and conductive gaskets or soldering process can be used to ensure electrical continuity and prevent energy leakage. Good shielding is a key mechanical structure requirement to ensure that the out-of-band suppression performance is not deteriorated.

[0117] The radio frequency connector is preferably a standard SMA type joint, which is fixed on the filter input and output port by welding or screwing, to ensure the reliability of the connection and the stability of the impedance matching. The selection of the connector needs to consider the working frequency range, power capacity and cost, and the SMA joint is an excellent standard choice in terms of cost performance in this frequency range.

Claims

1. A broadband cavity filter with low insertion loss, characterized in that, It includes a cavity (1), a cover plate (2) covering the cavity (1), an input interface (3) and an output interface (4) disposed on the cavity (1), and several resonators (5) disposed in the cavity (1); The resonator (5) adopts a 10th-order resonant cavity topology. The adjacent resonators (5) exchange energy through the coupling structure (6), so that the insertion loss of the filter in the passband frequency range of 1050MHz to 1450MHz is not greater than 0.3dB, the 0.3dB bandwidth is not less than 400MHz, the ripple in the passband is not greater than 0.2dB, and the port voltage standing wave ratio is not greater than 1.

3.

2. The low insertion loss broadband cavity filter according to claim 1, characterized in that, The coupling structure (6) is a coupling window formed between adjacent resonators (5).

3. The low insertion loss broadband cavity filter according to claim 1, characterized in that, The input interface (3) and output interface (4) are connected to the resonator (5) via direct tap coupling. The insertion depth of the tap probes is configured to achieve an external quality factor. In the range of 1500 to 2500.

4. The low insertion loss broadband cavity filter according to claim 1, characterized in that, The sidewall thickness of the cavity (1) is 7.45mm ± 0.1mm, the endwall thickness is 4mm to 6mm, and the bottom of the cavity (1) is provided with an installation through hole with a diameter of 2.5mm to 3.2mm.

5. The low insertion loss broadband cavity filter according to claim 1, characterized in that, The joint surface between the cover plate (2) and the cavity (1) is provided with an electromagnetic shielding structure.

6. A design method for a low-insertion-loss broadband cavity filter, the design method being used to design the low-insertion-loss broadband cavity filter according to any one of claims 1-5, characterized in that, The design method includes the following steps: S1. Filter Synthesis Design: Based on the generalized Chebyshev function approximation method, the center frequency, bandwidth, passband ripple and stopband suppression index are input. The resonator topology is determined through iterative analysis and a coupling matrix is ​​generated. Cross coupling is set between non-adjacent resonators to introduce transmission zeros. S2. Three-dimensional electromagnetic simulation optimization: The resonator model is established using electromagnetic simulation, the tuning structure parameters are optimized through eigenmode analysis, and the coupling structure dimensions are adjusted by parameter scanning. S3. Input-output coupling design: Tap coupling method is adopted to optimize probe insertion depth to achieve the target external quality factor; S4. Overall Model Integration and Optimization: Integrate the resonant cavity, coupling structure and port into a complete model, and use the optimization algorithm to perform multivariate iteration with the S-parameter curve as the target until the simulation results meet the requirements of insertion loss not greater than 0.3dB and bandwidth not less than 400MHz.

7. The design method for a low insertion loss broadband cavity filter according to claim 6, characterized in that, In step S1, the order of the resonator topology is determined through iterative analysis to achieve a stopband suppression of not less than 40 dBc, and the coupling matrix contains cross-coupling elements to generate transmission zeros.

8. The design method for a low insertion loss broadband cavity filter according to claim 6, characterized in that, In step S2, the formula for calculating the coupling coefficient is:

9. Among them, It is an even-mode frequency. It is the odd-mode frequency.

10. The design method for a low insertion loss broadband cavity filter according to claim 6, characterized in that, The optimization algorithm is an automated process based on gradient optimization or sequential nonlinear programming, and the objective function is defined as minimizing the difference between the simulated S-parameters and the ideal S-parameters.

11. The design method for a low insertion loss broadband cavity filter according to claim 6, characterized in that, The filter cavity undergoes an electroplating process to reduce conductor loss. The electroplating process is silver plating, with a silver layer thickness of not less than 5μm, and the plating solution temperature is controlled within the range of 20°C to 25°C.

Images