A radio frequency power supply system for plasma load and a design method thereof

Abstract

The application discloses a kind of for plasma load radio frequency power supply system and its design method, belong to plasma technical field;The method comprises: determining the limited reflection phase change region of plasma load;Based on the region and load traction data, the output matching network of amplifier is optimized and designed, so that its best load point is aligned with the high efficient center of gravity of load region;In the multi-stage power synthesis network thus constructed, the fixed phase difference between branches is optimized and set for the limited phase range, to balance the reflection stress.This application makes full use of the physical constraints of plasma load, changes the traditional "full-phase protection" into "precise regional optimization", while ensuring the robustness of the system, significantly improves the average efficiency, and is suitable for the radio frequency power supply of plasma generating equipment with extremely high reliability and energy efficiency requirements.

Description

Technical Field

[0001] This application relates to the field of plasma technology, specifically to a radio frequency power supply system for plasma loads and its design method. Background Technology

[0002] Radio frequency (RF) power supplies are used to excite and sustain plasma within the process chamber. The plasma load exhibits dynamic, nonlinear complex impedance characteristics, and its drastic changes can lead to severe mismatch at the RF power supply output, resulting in a high proportion of reflected power.

[0003] To address this issue, existing technologies mainly fall into two categories: First, backend protection, which uses an automatic matching network for real-time tuning, but its response speed is limited and it may fail during rapid load changes; second, robust design, which employs power combining topologies (such as Wilkinson combiners and quadrature combiners) and reserves sufficient voltage, current, and thermal design margins to withstand the worst standing wave conditions. These designs typically assume that the load reflection phase occurs with equal probability across the entire range of 0° to 360°, leading to an overly conservative design that sacrifices system efficiency under most real-world operating conditions.

[0004] The impedance of a plasma load is determined by its physical mechanism, and the phase of its reflected signal is actually confined within a finite sector of the Smith chart (usually a capacitive region).

[0005] However, existing technologies have failed to effectively utilize this key physical constraint for targeted optimization, resulting in an inability to achieve an optimal balance between performance and cost.

[0006] Furthermore, the introduction of automatic matching networks and transmission lines in practical systems further exacerbates the complexity of impedance variations at RF power supply ports. Automatic matching networks aim to dynamically tune and match varying load impedances to the system's characteristic impedance (e.g., 50Ω); transmission lines, on the other hand, perform phase rotation of the load impedance on the Smith chart. While both are necessary, they introduce a series of problems: the response speed of automatic matching networks is limited, and they may not be able to achieve complete matching in time during rapid changes in plasma impedance, resulting in transient mismatch at the amplifier end; while the fixed phase rotation of transmission lines widens the range of impedance variations observed at the amplifier end.

[0007] Therefore, even with the introduction of an automatic matching network and transmission line between the RF power supply and the plasma load, the actual observed impedance variation range at the RF power amplifier port is still significantly broadened due to the lag in the response of the automatic matching network to rapid impedance transitions and the fixed phase rotation effect of the transmission line on the load impedance. This results in a wide and dynamically unpredictable environment. Existing technologies have failed to address this problem at its root, forcing RF power supply designs to remain based on conservative assumptions of a full phase range, thus sacrificing system efficiency and cost optimization under most practical operating conditions. Summary of the Invention

[0008] To address the technical problem of low efficiency caused by the lack of full-phase protection in traditional designs, this application provides a radio frequency power supply system for plasma loads and its design method to achieve a balance between high efficiency and high reliability.

[0009] The technical solution is as follows: First, this application provides a radio frequency power supply system for a plasma load, comprising: Multiple power amplifier modules connected in parallel, each power amplifier module containing a power transistor and an output matching network; A multi-stage power combining network, whose input is connected to the output of the plurality of power amplifier modules, is used to combine the output power of each power amplifier module; The impedance characteristics of the output matching network are optimized and determined based on the finite variation range of the phase of the reflection coefficient of the plasma load impedance. A fixed phase difference is provided between at least one set of branches of the multi-stage power combining network. This phase difference is optimized and determined based on the finite variation range of the phase, and is used to balance the reflected electrical stress borne by the ports of each power amplifier module.

[0010] Optionally, the system is applied to a plasma excitation power source in a plasma generation process equipment.

[0011] Secondly, this application provides a design method for an RF power supply system for a plasma load, comprising the following steps: Determine the finite range of phase variation of the reflection coefficient of the plasma load; Based on the limited range of phase variation and the load pulling data of the power transistor, the output matching network of each power amplifier module is optimized so that the load impedance region corresponding to the limited range of phase variation is mapped to the high-efficiency operating region of the power transistor, and the optimized output impedance of each power amplifier module is determined. A multi-stage power combining network is constructed based on the optimized output impedance, and an optimized phase difference is introduced between at least one set of branches of the multi-stage power combining network to balance the reflected electrical stress borne by each power amplifier module port within a finite range of phase variation.

[0012] Furthermore, the optimized phase difference is determined by minimizing the difference in the magnitude of the equivalent reflection coefficient of each power amplifier module port within a finite range of phase variation.

[0013] Preferably, the optimized phase difference is 90°.

[0014] Furthermore, the reflected electrical stress includes the peak voltage stress and / or peak current stress at the output terminals of each power amplifier module.

[0015] As a specific implementation architecture, the multi-stage power combining network adopts a hybrid architecture that includes orthogonal combiners and Wilkinson combiners. Specifically, in the multi-stage power combining network, the final or subsequent stage uses a Wilkinson combiner for power combining, while the preceding or first stage uses an orthogonal combiner, and the isolation port of the orthogonal combiner is connected to an absorption load.

[0016] Preferably, the optimized phase difference can be applied between branches connected to different quadrature synthesizer groups, or between a quadrature power synthesizer and a Wilkinson synthesizer.

[0017] Preferably, the optimized phase difference is achieved by adjusting the electrical length of the transmission line or by using a lumped parameter phase-shifting network.

[0018] The technical solution includes at least the following technical effects: End-to-end efficiency improvement: Traditional designs optimize amplifier efficiency or the combining network in isolation, without considering the limited area of ​​actual load variation. This application takes a holistic approach to the RF link, ensuring that the phase difference between the amplifier's optimal load point and the combining network is matched to the actual load variation characteristics, thereby minimizing power loss during link transmission and achieving a significant improvement in system-level average efficiency.

[0019] Link stress distribution optimization and enhanced reliability: Through targeted phase difference design, the uneven stress reflected from the load end is actively redistributed at multiple amplifier ports at the front end of the link; peak voltage, current stress and unevenness are significantly reduced, fundamentally improving the long-term reliability and robustness of the entire RF power link.

[0020] Cost optimization: By optimizing the phase difference of the synthesized network for a limited phase range, the electrical stress caused by reflection is evenly distributed among multiple amplifiers, avoiding single-transistor overload. There is no need to reserve excessive margins for the worst-case scenario across the entire phase, reducing device selection costs and thermal design complexity.

[0021] The methodology is systematic: it provides a complete approach from load characteristic analysis and transistor-level optimization to system network co-design. This application transforms the physical constraints of the load into deterministic design goals for the source system. The high efficiency and robustness of the source, combined with the fast and precise tuning capabilities of the terminal matching network, jointly achieve optimal performance of the entire RF power transmission link under dynamic loads. It is particularly suitable for the development of RF power supplies for high-power, high-reliability plasma generation equipment. Attached Figure Description

[0022] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0023] Figure 1 A flowchart illustrating a design method for a radio frequency power supply system for a plasma load, provided as an embodiment of this application; Figure 2 This is a schematic diagram of an optimized architecture for 13.56MHz 8-channel synthesized RF power supply in one embodiment of this application; Figure 3 This is a schematic diagram of the plasma load impedance on a Smith chart limited to a finite phase sector in one embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0026] Definitions: Wilkinson synthesizer: The Wilkinson synthesizer.

[0027] This application provides a radio frequency (RF) power supply system for a plasma load, and designs it using the RF power supply system design method of this application. This system is applied to the plasma excitation power supply in plasma generation process equipment.

[0028] This application provides a radio frequency power supply system for a plasma load, comprising: Multiple power amplifier modules connected in parallel, each power amplifier module containing a power transistor and an output matching network; A multi-stage power combining network, whose input is connected to the output of the plurality of power amplifier modules, is used to combine the output power of each power amplifier module; The impedance characteristics of the output matching network are optimized and determined based on the finite variation range of the phase of the reflection coefficient of the plasma load impedance. A fixed phase difference is provided between at least one set of branches of the multi-stage power combining network. This phase difference is optimized and determined based on the finite variation range of the phase, and is used to balance the reflected electrical stress borne by the ports of each power amplifier module.

[0029] like Figure 1 As shown, this application provides a design method for an RF power supply system for a plasma load, applied to the aforementioned RF power supply system for a plasma load, comprising the following steps: Determine the finite range of phase variation of the reflection coefficient of the plasma load; Based on the limited range of phase variation and the load pulling data of the power transistor, the output matching network of each power amplifier module is optimized so that the load impedance region corresponding to the limited range of phase variation is mapped to the high-efficiency operating region of the power transistor, and the optimized output impedance of each power amplifier module is determined. A multi-stage power combining network is constructed based on the optimized output impedance, and an optimized phase difference is introduced between at least one set of branches of the multi-stage power combining network to balance the reflected electrical stress borne by each power amplifier module port within a finite range of phase variation.

[0030] Furthermore, the optimized phase difference is determined by minimizing the difference in the equivalent reflection coefficient magnitude of each power amplifier module within a finite range of phase variation.

[0031] Furthermore, reflected electrical stress includes peak voltage stress and / or peak current stress at the output terminals of each power amplifier module.

[0032] Furthermore, the multi-stage power combining network adopts a hybrid architecture of quadrature combiners and Wilkinson combiners. The final or subsequent stage uses Wilkinson combiners for power combining, while the preceding or first stage uses quadrature combiners, and the isolation ports of the quadrature combiners are connected to absorption loads.

[0033] Furthermore, the optimized phase difference is applied between branches of different quadrature synthesizer groups, or between branches of the quadrature power synthesizer and the Wilkinson synthesizer.

[0034] Furthermore, the optimized phase difference can be achieved by adjusting the transmission line electrical length or by using a lumped parameter phase-shifting network.

[0035] This application provides a radio frequency power supply system for a plasma load, comprising multiple parallel power amplifier modules and a multi-stage power combining network. The output matching network of each power amplifier module is optimized to ensure that its nominal output impedance is not 50 Ω, and the multi-stage power combining network contains at least one controllable phase difference, which varies according to the finite range of the load's phase. i min , i max ]Optimization determined.

[0036] The core advantage of the technical solution provided in this application lies in its synergy with downstream links: To reduce the burden on the automatic matching network. Since the source system has already addressed the limited range of phase variations... i min , i max The phase change region is optimized, significantly narrowing the range of equivalent impedance changes from the automatic matching network to the source and improving its characteristics. This eliminates the need for the automatic matching network to handle extreme full-phase conditions, allowing for a more relaxed matching speed requirement, easier achievement of matching accuracy and dynamic range, and reduced design complexity and cost.

[0037] Compensation for transmission line effects. The fixed phase rotation caused by the transmission line changes the impedance phase viewed from the source to the load. This application incorporates the known electrical length of the transmission line segment as a fixed offset into the calculation model when determining the optimization parameters (phase difference), thus ensuring that the optimization results remain accurate and effective in actual physical links.

[0038] Improved system-level stability: When the plasma undergoes drastic changes (such as ignition), the uniform stress distribution capability of the source-end system in this application can prevent damage from overload of a single amplifier; at the same time, the power reflected back to the automatic matching network is more balanced, reducing the risk of overvoltage breakdown of the automatic matching network components, thereby improving the robustness of the entire link from source to load.

[0039] Example 1: Design of an 8kW / 13.56MHz RF Power Supply System 1. System Structure like Figure 2 As shown, an embodiment of this radio frequency power supply system for a plasma load includes: Eight power amplifier modules are included, each consisting of an LDMOS power transistor and its output matching network. The power amplifier modules are composed of eight radio frequency (RF) amplifiers (PA1-PA8), each with its input connected to one of eight excitation signal sources.

[0040] The first-stage power combining network consists of four quadrature couplers. The two inputs of each quadrature coupler are connected to the outputs of two power amplifier modules, respectively. Each quadrature coupler receives two output signals from the RF amplifier. Each quadrature coupler's isolation port is connected to absorption resistors (R1-R4) to absorb reflected power and improve the system's port isolation and stability.

[0041] Phase adjustment unit: Located between the output of the power amplifier module and the first-stage power combining network. In this embodiment, it is specifically located on the two input branches of the second-stage Wilkinson combiner to introduce an optimized phase difference ΔΦ.

[0042] The second-stage power combining network consists of two Wilkinson combiners, each of which combines the power outputs of the two quadrature combiners from the first-stage power combining network.

[0043] The third-stage power combining network consists of a Wilkinson synthesizer, which performs final combining of the two signals from the second-stage power combining network to obtain the radio frequency output signal, which is then output to the load, and connected to the plasma chamber via a transmission line and an automatic matching network.

[0044] The 8kW / 13.56MHz RF power supply system achieves high-efficiency power synthesis and robust operation over a wide range of load impedance variations through the cascaded combination of quadrature synthesizers and Wilkinson synthesizers, along with a phase adjustment unit and absorption resistor. It is particularly suitable for plasma loads with rapidly changing impedance characteristics.

[0045] This embodiment of a radio frequency power supply system design method for a plasma load includes the following steps: Step S1: Determine the finite range of phase variation of the plasma load reflection coefficient.

[0046] In the actual process chamber, a vector network analyzer was used to measure the impedance of Ar / CF4 plasma within a typical etching process window. The measured complex impedance was converted into a reflection coefficient, and its phase distribution was statistically analyzed. The reflection phase θ was determined to be mainly distributed within the [-60°, +45°] sector. This range is much smaller than the full circumference of 0°-360°, confirming the phase-constrained characteristics of the plasma load.

[0047] Step S2: Optimize the output matching network of the power amplifier module S2.1. Obtain the load pull data of the power transistor at the target frequency, and obtain a contour plot of its efficiency and output power as a function of load impedance.

[0048] S2.2. Map the load impedance variation region determined in step S1 onto the load traction contour map.

[0049] S2.3. Design the amplifier's output matching network so that the entire load impedance variation region falls within the transistor's high-efficiency region; this optimized matching makes the amplifier module's nominal output impedance... Its conjugate value corresponds to the "high-efficiency centroid" of the load region (based on the limited variation range of the phase and the load pull data of the power transistor, the output matching network of each power amplifier module is optimized so that the load impedance region corresponding to the limited variation range of the phase is mapped to the high-efficiency operating region of the power transistor).

[0050] Specifically: An LDMOS transistor, BLF188XR, was selected, and a load pull test was performed at 13.56MHz to obtain a contour plot of power-added efficiency (PAE) as a function of load impedance. The load impedance region determined in step S1 (considering the impedance transformation between the amplifier output and the actual load) was mapped onto the load pull contour plot. An output matching network for the power amplifier module was designed so that the entire load impedance region falls within the high-efficiency region where PAE > 70%. After optimization, the optimized output impedance of the power amplifier module was obtained. (Not 50Ω). This The conjugate value corresponds to the "efficient centroid" of the load region.

[0051] Step S3: Construct a multi-stage power combining network and introduce an optimized phase difference.

[0052] The multi-stage power combining network adopts a three-stage combining architecture: the first stage consists of four quadrature combiners, and the second and third stages use Wilkinson combiners. The specific steps are as follows: S3.1. Based on the power amplifier module obtained in step S2 (output impedance is...) ), construct a multi-level power combining network.

[0053] S3.2. Introduce a phase difference between different branches of a multi-stage power combining network. The optimization objective is: the finite range of phase variation determined in step S1 [ i min , i max Within this scope, minimize the difference in peak voltage stress at the ports of each power amplifier module.

[0054] S3.3. The phase difference ΔΦ is achieved by adjusting the electrical length of the transmission line or by using a lumped parameter phase-shifting network.

[0055] Preferably, in step S3.2, the optimized value of the phase difference ΔΦ is determined by searching or simulating a function with the goal of minimizing the function: ; in, i ∈[ i min , i max ],|Γ i | represents the equivalent reflection coefficient magnitude seen at the i-th amplifier port.

[0056] Optimization goal: In [ i min , i max [Internal equalization of voltage stress at each power amplifier port]

[0057] Step S3.4 Implementation: Adjust the transmission line length / lumped parameter phase-shifting network; such as... DF =90° through λ / 4 delay line. Preferably, the multi-stage power combining network adopts a hybrid architecture using orthogonal combiners and Wilkinson combiners, wherein the final or subsequent stage uses Wilkinson combiners for power combining, the preceding or first stage uses orthogonal combiners, and the isolation port of the orthogonal combiners is connected to a matching absorption load.

[0058] Preferably, the optimized phase difference ΔΦ is applied between branches connected to different quadrature combiner groups. For example, between the quadrature power combiner and the Wilkinson combiner, the optimal phase difference ΔΦ is approximately 90°. Optimizing the phase difference between the two input branches of the second-stage Wilkinson combiner to 90° achieves orthogonal stress distribution and peak shifting.

[0059] like Figure 2 As shown, in the hybrid architecture consisting of quadrature synthesizers and Wilkinson synthesizers, a fixed 90° phase difference ensures that when a load reflection phase change causes one set of amplifiers to approach the peak voltage stress, the other set of amplifiers remains under relatively low stress due to the quadrature of the signal phases. This design is equivalent to "quadrature sampling" of the reflected stress in the load phase domain, thereby distributing instantaneous peak stress across different subsets of amplifiers at the system level, preventing all amplifiers from simultaneously experiencing the worst conditions.

[0060] For a single amplifier, under a specific load, phase Down: (1) (2) in, It is the amplitude of the positive wave voltage output by the amplifier (calibrated to 1 in the matched state). It is the equivalent reflection coefficient seen at the output of the amplifier.

[0061] Voltage uniformity can be calculated as follows: (3) in, It is the peak voltage of the i-th amplifier when the load phase is θ; It represents the standard deviation of 8 values, reflecting the degree of dispersion; It is the average of 8 values, used for normalization.

[0062] Similarly, the current uniformity can be calculated.

[0063] On the two output paths of the second-stage Wilkinson synthesizer, 0° and 0° are introduced respectively. l A phase delay unit of 4 (90°) is used, i.e., ΔΦ=90° is set. In the range of θ∈[-60°,+45°], the peak voltage non-uniformity of the 8 amplifiers is reduced by more than 60% compared with the conventional design (ΔΦ=0°).

[0064] Simulation results show that F(ΔΦ) reaches its minimum value when ΔΦ=90°. Therefore, in this embodiment, ΔΦ=90° is set, specifically by adding a λ / 4 (90°) transmission line to one of the two branches.

[0065] Appendix Figure 3 This illustrates the possible range of load impedance, which means that overall impedance matching is not required on the complete Smith chart, nor is full-phase operation necessary.

[0066] 3. Effect Verification On simulation and experimental platforms, the optimized system of this application was compared with the traditional design (ΔΦ=0°, and the amplifier matched to 50Ω). With a load voltage standing wave ratio (VSWR) of 3:1 and the reflection phase dynamically changing within the range of [-60°, +45°], the peak voltage, peak current, and non-uniformity at the output of each amplifier were measured. The results are shown in Table 1. Table 1 is a comparative schematic diagram showing the peak voltage stress borne by the two amplifiers as a function of the load phase θ after introducing a phase difference ΔΦ between the two main branches of the synthetic network. The optimized curve is flatter and has smaller differences.

[0067] Table 1 Note that the smaller the non-uniformity index value, the smaller the difference between each amplification unit, and the better the system uniformity.

[0068] As shown in Table 1, the system of this application exhibits significant improvements in voltage, peak current stress, and non-uniformity. Furthermore, the average efficiency of the system of this application is more than 8% higher than that of traditional designs, and the temperature rise difference among the amplifiers is significantly reduced, demonstrating the effectiveness of the system and method of this application.

[0069] 4. Synergistic effects on downstream links In the actual system, the RF power supply output of this embodiment is connected to a 50Ω coaxial cable approximately 2m long (electrical length approximately 20°), which is then connected to the automatic matching network and the plasma chamber. Due to the fixed phase rotation introduced by the transmission line, without compensation, the load phase range observed from the power supply end will widen to [-80°, +65°]. In designing and optimizing the phase difference ΔΦ, this application incorporates the electrical length of the transmission line as a known offset into the simulation model of step S3. Therefore, the optimized ΔΦ=90° remains optimal in the actual physical link. Experiments show that after adding the transmission line, the system efficiency and stress uniformity remain essentially unchanged, demonstrating the robustness of the method presented in this application.

[0070] This application provides a complete methodology from load characteristic analysis and transistor-level optimization to system network co-design. It transforms the physical constraints of the load into deterministic design objectives for the source-end system. The high efficiency and robustness of the source end, combined with the fast and precise tuning capabilities of the terminal matching network, achieve optimal performance of the entire RF power transmission link under dynamic loads. This is particularly suitable for the development of RF power supplies for high-power, high-reliability plasma generation equipment.

[0071] Other implementation methods The above embodiments are illustrated using an 8-channel combining, 13.56MHz, orthogonal + Wilkinson hybrid architecture as an example, but this application is not limited thereto. Those skilled in the art can select other combining network topologies (such as full Wilkinson tree, chain combining, etc.) based on the actual power level, frequency, and number of amplifiers, and determine the optimized output matching network and branch phase difference according to the method of this application. Any scheme that uses a finite phase range of the plasma load to collaboratively optimize amplifier matching and combining network phase falls within the protection scope of this application.

[0072] The RF power supply system and design method provided in this application can be widely used in process equipment that requires plasma generation, such as semiconductor etching, deposition, ion implantation, and surface treatment. It is particularly suitable for the development of high-power, high-reliability RF power supply products with a wide load range, and has good industrial practical value and economic prospects.

[0073] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

Claims

1. A radio frequency power supply system for a plasma load, characterized in that, include: Multiple power amplifier modules connected in parallel, each power amplifier module containing a power transistor and an output matching network; A multi-stage power combining network, whose input is connected to the output of multiple power amplifier modules, is used to combine the output power of each power amplifier module; The impedance characteristics of the output matching network are optimized and determined based on the finite variation range of the phase of the reflection coefficient of the plasma load impedance. At least one set of branches in the multi-stage power combining network is provided with a fixed phase difference, which is optimized and determined based on the finite variation range of the phase, and is used to balance the reflected electrical stress borne by the ports of each power amplifier module.

2. The radio frequency power supply system for a plasma load according to claim 1, characterized in that, The system is used as a plasma excitation power source in plasma generation process equipment.

3. A design method for an RF power supply system for a plasma load, applied to the RF power supply system for a plasma load as described in claim 1, characterized in that, include: Determine the finite range of phase variation of the reflection coefficient of the plasma load; Based on the limited range of phase variation and the load pulling data of the power transistor, the output matching network of each power amplifier module is optimized so that the load impedance region corresponding to the limited range of phase variation is mapped to the high-efficiency operating region of the power transistor, and the optimized output impedance of each power amplifier module is determined. A multi-stage power combining network is constructed based on the optimized output impedance, and an optimized phase difference is introduced between at least one set of branches of the multi-stage power combining network to balance the reflected electrical stress borne by each power amplifier module port within a finite range of phase variation.

4. The design method for a radio frequency power supply system for a plasma load according to claim 3, characterized in that, The optimized phase difference is determined by minimizing the difference in the magnitude of the equivalent reflection coefficient of each power amplifier module within a finite range of phase variation.

5. The design method for a radio frequency power supply system for a plasma load according to claim 3 or 4, characterized in that, The optimized phase difference is 90°.

6. The design method for a radio frequency power supply system for a plasma load according to claim 3, characterized in that, The reflected electrical stress includes the peak voltage stress and / or peak current stress at the output terminals of each power amplifier module.

7. The design method for a radio frequency power supply system for a plasma load according to claim 3, characterized in that, The multi-stage power combining network adopts a hybrid architecture of orthogonal synthesizer and Wilkinson synthesizer.

8. The design method for a radio frequency power supply system for a plasma load according to claim 7, characterized in that, In the multi-stage power combining network, the final or subsequent stage uses a Wilkinson combiner for power combining, while the preceding or first stage uses an orthogonal combiner, and the isolation port of the orthogonal combiner is connected to an absorption load.

9. The design method for a radio frequency power supply system for a plasma load according to claim 7 or 8, characterized in that, The optimized phase difference is applied between branches of different quadrature combiner groups, or between branches of a quadrature power combiner and a Wilkinson combiner.

10. The design method for a radio frequency power supply system for a plasma load according to claim 3, characterized in that, The optimized phase difference is achieved by adjusting the transmission line electrical length or by using a lumped parameter phase-shifting network.

Images