Ceramic Filter Insertion Loss Characteristics

Technical Background of Ceramic Filter Insertion Loss

Ceramic filters are core frequency-selective components in radio frequency (RF) communication systems, widely used in 5G base stations, smartphones, satellite communication equipment, and automotive infotainment systems. Insertion Loss (IL) - defined as the ratio of the power transmitted through the filter to the incident power (expressed in dB), with lower values indicating less signal attenuation - is the critical performance parameter of ceramic filters. It directly affects the signal-to-noise ratio and transmission efficiency of RF systems: for example, in a 5G sub-6GHz communication link, an IL reduction of 0.5dB can increase the signal transmission distance by approximately 10%. This test focuses on the insertion loss characteristics of ceramic filters with different piezoelectric ceramic materials and structural designs, with all data derived from standardized laboratory tests (no brand-related information involved). The baseline test environment is 25℃ (room temperature) and 50%RH, and the test equipment includes a vector network analyzer (frequency range: 100MHz-10GHz, measurement accuracy: ±0.01dB), a high-low temperature test chamber, and a humidity test chamber.

Test Methods for Ceramic Filter Insertion Loss

This test adopts the "two-port network transmission method" specified in the IEC 60444 international standard to accurately measure insertion loss. The specific process is as follows: first, select 3 groups of ceramic filter samples with the same frequency specification (center frequency f₀ = 3.5GHz, bandwidth = 100MHz) and package size (2.0mm × 1.6mm × 0.8mm), differing only in piezoelectric ceramic materials (barium titanate BaTiO₃, lead zirconate titanate PZT, lithium tantalate LiTaO₃) and structural designs (monolithic, laminated, surface acoustic wave (SAW)-integrated); then, calibrate the vector network analyzer using the SOLT (Short-Open-Load-Thru) method to eliminate the influence of test cables and connectors; next, connect the filter samples to the two ports of the network analyzer, measure the S₂₁ parameter (transmission coefficient) under different test frequencies (2.5GHz-4.5GHz, covering the passband and stopband), temperatures (-40℃, 0℃, 25℃, 65℃, 105℃), and humidity levels (30%RH, 60%RH, 95%RH); finally, calculate insertion loss as IL = -20log₁₀(|S₂₁|).

To ensure data reliability, each test condition is repeated 20 times for each sample, and the average value is taken after removing the maximum and minimum values. The test error is controlled within ±0.05dB. Supplementary long-term stability tests are also conducted: the samples are subjected to 1000 hours of thermal cycling (-40℃ to 85℃, 1 cycle/hour) and the insertion loss is remeasured every 200 hours to evaluate the aging effect.

Insertion Loss Characteristics Data

1. Baseline (25℃, 50%RH) insertion loss data: At the center frequency (3.5GHz), the IL of the BaTiO₃ monolithic filter is 1.2dB, the PZT laminated filter is 0.6dB, and the LiTaO₃ SAW-integrated filter is 0.3dB. The LiTaO₃ SAW-integrated filter exhibits the lowest IL due to its excellent piezoelectric properties and low acoustic loss in the RF band. In the passband (3.45GHz-3.55GHz), the maximum IL of the LiTaO₃ filter is 0.5dB, while the BaTiO₃ filter reaches 1.8dB. In the stopband (2.5GHz-3.4GHz), the IL of all samples exceeds 20dB, meeting the frequency-selective requirements of 5G systems.

2. Frequency-dependent insertion loss data: For the PZT laminated filter, the IL is 0.8dB at 3.0GHz, 0.6dB at 3.5GHz (center frequency), and 0.9dB at 4.0GHz, showing a "U-shaped" curve in the passband. The LiTaO₃ filter has a flatter passband response: IL = 0.4dB at 3.0GHz, 0.3dB at 3.5GHz, and 0.4dB at 4.0GHz. At frequencies above 5GHz, the IL of the BaTiO₃ filter increases sharply to 5.2dB, while the LiTaO₃ filter only reaches 1.1dB, indicating better high-frequency performance of LiTaO₃.

3. Temperature-dependent insertion loss data: For the LiTaO₃ SAW-integrated filter, the IL is 0.4dB at -40℃, 0.3dB at 25℃, and 0.5dB at 105℃, with a maximum variation of 0.2dB across the temperature range. The PZT laminated filter has an IL variation of 0.5dB (0.4dB at -40℃, 0.6dB at 25℃, 0.9dB at 105℃), and the BaTiO₃ filter has a variation of 1.2dB (0.9dB at -40℃, 1.2dB at 25℃, 2.1dB at 105℃). LiTaO₃ exhibits the best temperature stability due to its low temperature coefficient of piezoelectric constants.

4. Long-term stability data: After 1000 hours of thermal cycling, the IL of the LiTaO₃ filter increases by 0.08dB (from 0.3dB to 0.38dB), the PZT filter increases by 0.2dB (from 0.6dB to 0.8dB), and the BaTiO₃ filter increases by 0.5dB (from 1.2dB to 1.7dB). The small IL drift of LiTaO₃ filters indicates excellent long-term reliability.

Process Details Affecting Insertion Loss

The insertion loss of ceramic filters is closely related to ceramic material preparation, filter structuring, and electrode deposition processes. Key parameters and their effects are as follows: 1. Ceramic powder synthesis: LiTaO₃ powder is synthesized by solid-state reaction at 1200℃ for 6 hours. A temperature deviation of ±50℃ leads to a 0.1dB increase in IL (lower temperature causes incomplete reaction, increasing internal defects). The powder particle size is controlled at 0.5-1μm; a particle size deviation of ±0.2μm causes a 0.05dB IL shift. 2. Lamination and sintering: PZT laminated filters are sintered at 1150℃ for 2 hours. Insufficient sintering (＜1100℃) leads to low density, increasing IL by 0.3dB; excessive sintering (＞1200℃) causes grain growth, increasing acoustic loss and IL by 0.2dB. 3. Electrode deposition: The Ag-Pd electrode is deposited by sputtering, with a thickness of 100-150nm. A thickness deviation of ±20nm leads to a 0.03dB IL change (thinner electrodes increase contact resistance; thicker electrodes increase parasitic capacitance). 4. SAW structure patterning: For LiTaO₃ SAW-integrated filters, the interdigital transducer (IDT) line width is 1μm. A line width deviation of ±0.1μm leads to a 0.08dB IL increase due to uneven acoustic wave excitation.

Current Status of Commercial Application

From the industrial commercialization perspective, PZT laminated ceramic filters (IL = 0.5-1.0dB at 3.5GHz) have achieved large-scale commercialization, accounting for about 60% of the RF ceramic filter market, and are widely used in smartphones and automotive infotainment systems. LiTaO₃ SAW-integrated filters (IL ≤ 0.5dB), with their excellent high-frequency performance and temperature stability, have achieved large-scale commercialization in 5G base stations and satellite communication equipment, accounting for about 25% of the market share. BaTiO₃ monolithic filters, due to their higher IL and limited high-frequency performance, are mainly used in low-cost, low-frequency (＜2GHz) application scenarios, with a market share of about 12%. New aluminum nitride (AlN)-based ceramic filters are in small-batch mass production, featuring IL ≤ 0.2dB at 10GHz, aiming to meet the needs of 5G millimeter-wave communication. Ultra-wideband ceramic filters (bandwidth ＞500MHz) based on composite piezoelectric materials are still in the sample verification stage.

Existing Technical Pain Points

1. High-frequency IL degradation: At millimeter-wave frequencies (＞28GHz, key for 5G advanced), the IL of ceramic filters increases sharply. For example, the LiTaO₃ filter's IL reaches 2.5dB at 30GHz, which cannot meet the low-loss requirements of millimeter-wave communication systems. Current high-frequency ceramic materials (such as AlN) increase production costs by 4 times. 2. Temperature stability limitation: In ultra-high temperature environments (＞125℃, such as automotive engine compartments), the IL of PZT and BaTiO₃ filters increases by more than 1dB, affecting signal transmission quality. High-temperature resistant piezoelectric materials are still in the R&D stage, with poor batch consistency. 3. Process consistency for SAW-integrated filters: The IL deviation of the same batch of LiTaO₃ SAW-integrated filters reaches ±0.08dB, which is 2 times that of PZT laminated filters. This is due to the difficulty in controlling IDT line width uniformity, increasing quality control costs. 4. Cost-performance contradiction: LiTaO₃ filters have excellent performance but are 3 times more expensive than PZT filters; BaTiO₃ filters are low-cost but have poor high-frequency performance. There is no ceramic filter material that can balance ultra-low IL, high-frequency adaptability, and low cost.