View 3D results Used Tools:

Introduction

The demand for 5G applications increases the need for radio frequency (RF) and microwave filter components in the mm-Wave frequency bands. 5G Network systems require filters with high percent bandwidth, good selectivity, and excellent temperature stability in compact packages.

HFWorks offers RF and microwave design, analysis and optimization platform for 5G and mm-Wave components. It supports a complete line of structures including cavity, waveguide, coaxial, ceramic and microstrips. This analysis is studying a mmWave Waveguide band-pass filter design with Tuning screws operating between 27GHz and 29GHz. A parametric analysis is used to investigate the effect of screws depth on the filter performance.
 

Problem Description

The optimized mmWave WG filter design presented by ^[1] is made of 7 standard WR-34 wavegide resonators separated  by 6 cylindrical inductive irises and mounted by 7 tuning screws. These screws are added to the center of each resonator including one before and one after the iris from the wave port. They are added to the waveguide filter to tune toward the desired filter responses and each screw acts like an inductive iris, but it does not have an end to-end touch of the guide. The 3D design is presented by the next figure Fig1.

Simulation and Results

The S-parameters solver of HFWorks is used for a working frequency range of [275GHz-29GHz]. The simulation revealed the next S-parameter results of the studied Band-pass filter presented by the 2D plot of the return and insertion loss data versus frequency in Fig2.
The simulated return loss level depicts a satisfactory performance for the entire passband. The S-parameter simulation results also indicate that the filter has a lower and upper stopband insertion loss values of -103.42 dB and -84.75 dB at 26 GHz and 30 GHz, respectively.

Fig. 2. Return and Insertion Loss Results vs Frequency

Breakdown Level Analysis

All dielectrics, including air, have a maximum electric field strength above which ionization occurs resulting in corona and sparking.  It is the well-known by dielectric breakdown.  This phenomenon happens if the applied power is too high or/and if the spacing, i.e. air gap, between components is too small.   In our case, the penetration depth of the tuning screws which defines the air gap to the waveguide bottom can cause higher E-field strength levels. A cross-section presentation of the E-field safety factor distribution across the studied filter is shown by the following plot for an incident power of 1kW. As noticed, a maximum factor of 0.54 is achieved below the central tuning screw for the used phase.

Parametric Analysis: Filter S-parameter Sensitivity to Tuning Screws Depth

At a 5G frequency of 28 GHz, the wavelength is around 10.71 millimeters. 1 millimeter is basically lambda over 10. As a result, the passing properties of the filter are expected to be sensitive to the screw's mechanical penetration inside the filter waveguide.

A final parametric analysis is performed to study the effect of this penetration depth on the obtained S-parameter performance of the bandpass filter. Since the filter design presents a symmetry compared to the central screw, only 4 geometrical variables are optimized like shown in the table below.

The obtained results are shown by the next 2D plot of the return loss data versus frequency for four different scenarios. A significant variation in S11 results is noticed from one scenario to another which confirms the expected sensitivity of the studied filter to the tuning element`s depth.

Fig. 5. Corresponding Screw Depth Combinations for Studied Scenarios

Fig 6. Return Loss Results vs Frequency and Scenarios

Conclusion

Analyzing and optimizing mmWave RF Filters, with extra-small sizes dedicated to 5G network, represent a challenging task for FEM simulation. HFWorks presents a powerful tool helping to investigate and optimize extremely sensitive mmWave filter designs and boost their stability.

References