Power generator and control method
The RF power generator addresses inefficiencies in existing systems by using multiple amplifier blocks for discrete modulation and impedance conversion, achieving high efficiency and wide power range suitable for semiconductor processing and other industrial applications.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- MASSACHUSETTS INST OF TECH
- Filing Date
- 2026-02-19
- Publication Date
- 2026-05-11
AI Technical Summary
Existing power generators face inefficiencies when operating with variable load impedance, high frequency ranges, and high power levels, particularly in applications like plasma generation for semiconductor processing, leading to increased size, power rating, and low peak and average power efficiencies.
A radio frequency (RF) power generator with a unique system configuration and power control method, utilizing multiple amplifier blocks that perform discrete modulation by out-of-phase power signal generation and discrete drain modulation, combined with impedance conversion to match load impedance, enabling efficient power supply to varying loads.
The RF power generator achieves high efficiency and wide output power range with fast response, compensating for voltage and current losses, suitable for applications with high peak-to-average power ratios, including plasma generators and other industrial loads.
Smart Images

Figure 2026076357000006 
Figure 2026076357000007 
Figure 2026076357000008
Abstract
Description
[Technical Field]
[0001] Cross-reference of related applications
[0001] This application claims the interests of U.S. Provisional Patent Application No. 63 / 059,532, filed on 31 July 2020, and U.S. Provisional Patent Application No. 63 / 085,432, filed on 30 September 2020, under Section 119(e) of the U.S. Patent Act, the contents of those applications are incorporated herein by reference in their entirety.
[0002]
[0002] One or more embodiments described herein relate to power generation. [Background technology]
[0003]
[0003] In many industrial applications, power amplifiers operate with variable load impedance, high frequency ranges, and high power levels and peak-to-average power ratios. One example of such an application is plasma generation used in semiconductor processing. Existing power generators cannot function in a way that does not limit or otherwise adversely affect the operation of the plasma generator or the operation of the power generator itself. For example, existing power generators sacrifice efficiency to meet other metrics. This increases the size and power rating and results in particularly low peak and average power efficiencies among several limitations. [Overview of the project]
[0004]
[0004] One or more embodiments described herein provide a radio frequency (RF) power generator having a unique system configuration and power control method.
[0005]
[0005] According to one or more embodiments, the power generator includes a plurality of amplifier blocks, each block including one or more amplifiers and a combiner for combining the modulated power signals output from the plurality of amplifier blocks to generate an RF power signal for the load, wherein the plurality of amplifier blocks are configured to out-of-phase the modulated power signals based on the phase angle. Each of the plurality of amplifier blocks is configured to perform discrete modulation to generate each of the modulated power signals. Discrete modulation involves selecting different combinations of the plurality of power amplifiers to vary the RF power signal in discrete steps to correspond to changes in the power of the load. In embodiments, the amplifiers may be RF power amplifiers.
[0006]
[0006] According to one or more embodiments, a method for managing power includes the steps of: generating a first modulated power signal from a first amplifier block; generating a second modulated power signal from at least a second amplifier block; out-of-phase the first and second modulated power signals based on a phase angle; and generating an RF power signal for a load based on the out-of-phase first and second modulated power signals. The step of generating the first modulated power signal includes switching different combinations of a plurality of power amplifiers in the first amplifier block, and the step of generating the second modulated power signal includes switching different combinations of a plurality of power amplifiers in the second amplifier block. In embodiments, the amplifiers may be RF amplifiers.
[0007]
[0007] The aforementioned and other purposes, features, and advantages will become apparent from the following more detailed description of embodiments, as shown in the accompanying drawings, where similar reference letters refer to the same parts through different figures. The drawings are not necessarily to scale and are rather focused on illustrating the principles of the embodiments. [Brief explanation of the drawing]
[0008] [Figure 1]
[0008] This is a block diagram of a radio frequency (RF) power generator. [Figure 2]
[0009] This is a block diagram of an RF power generator. [Figure 3A]
[0010] This is a schematic diagram of an embodiment of an amplifier having an output section coupled via a single combiner. [Figure 3B] This is a schematic diagram of an embodiment of an amplifier having an output section coupled via a single combiner. [Figure 3C] This is a schematic diagram of an embodiment of an amplifier having an output section coupled via multiple combiners. [Figure 4]
[0011] This is a schematic diagram of a switched-mode power amplifier (PA). [Figure 5]
[0012] This figure shows a voltage-versus-time plot of the RF output voltage waveform for an example of an MIDB block, such as the Multi-Inverter Discrete Backoff (MIDB) block shown in Figure 2 or Figure 6. [Figure 6]
[0013] This is a block diagram of an RF power generator with two MIDB power amplifiers (PAs). [Figure 7]
[0014] This is a schematic diagram of an exemplary embodiment of an impedance transformer. [Figure 8]
[0015] This is a schematic diagram illustrating an exemplary embodiment of a discrete drain modulation circuit for PA systems. [Figure 9A]
[0016] This flowchart shows a method for generating an RF power signal. [Figure 9B] This flowchart shows a method for generating an RF power signal. [Figure 9C] This flowchart shows a method for generating an RF power signal. [Figure 10]
[0017] This is a graph plotting the MIDB block output voltage versus the load voltage (voltage vector graph). [Figure 11]
[0018] This figure shows a plot of uncompensated load admittance curves observed in MIDB blocks (YA and YB in Figure 6) as an example of admittance for evaluating performance and power management. [Figure 12]
[0019] This figure shows a plot of efficiency versus output power for an embodiment of the MIDB system. [Figure 13]
[0020] This figure shows a voltage-to-time plot illustrating the dynamic response to a step at an out-fading angle. [Figure 14]
[0021] This figure shows a voltage-to-time plot illustrating the dynamic response to a step in an MIDB configuration. [Figure 15]
[0022] This figure shows voltage-versus-time plots illustrating performance metrics for one or more exemplary embodiments. [Modes for carrying out the invention]
[0009]
[0023] One or more embodiments described herein provide a high-frequency (RF) power generator that meets the power requirements of a variety of applications. Across the various embodiments, one or more of the following features may be combined: (1) outfading power supply signals from multiple amplifier blocks for fast response (and, if necessary, continuous) power generation; (2) configuring each multiplex amplifier block to include multiple amplifiers; (3) configuring each multiplex amplifier to include multiple switch-mode power amplifiers; (4) modulating the power signal by switching (or selecting) different combinations of amplifiers within each block to generate discrete steps of the RF power signal voltage; and / or (5) performing discrete drain modulation of the supply voltage for the amplifier blocks to generate a high-efficiency operating power range.
[0010]
[0024] In one embodiment, the RF power generator may be used to power a load including a plasma generator of the type used in semiconductor chip fabrication processes. Other embodiments In this configuration, the RF power generator can power other types of loads, including those operating in various power ranges, such as those that do not have the same power range and performance requirements as the plasma generator. Furthermore, discrete step changes in the RF power signal can be performed dynamically to meet the power requirements of loads that change over time. In some embodiments, impedance converters may be included to address matching problems caused by changes in load impedance and / or impedance changes caused by switching different combinations of power amplifiers within an amplifier block during discrete modulation.
[0011]
[0025] Figure 1 shows one embodiment of RF power generator 1, including various stages for generating power for intended applications. One application could be an industrial application, such as power generation for plasma generators in semiconductor manufacturing processes, as described above. Other applications could be various types of communication and antenna systems, radar systems, and microwave cavity resonators, as well as others.
[0012]
[0026] Referring to Figure 1, the RF power generator 1 includes an amplifier stage 10, a power combiner 30, and an impedance converter 40. The amplifier stage includes N amplifier blocks (where N≧2) that output modulated power signals to be combined to generate an output RF power signal for a load. Each of the N amplifier blocks includes multiple power amplifiers that are selectively switched (e.g., selected or activated) to generate a given power block voltage. An example of the first amplifier block will be discussed under the understanding that the remaining N amplifier blocks may be the same or similarly configured. In other embodiments, the amplifier blocks may have different structural and / or functional configurations.
[0013]
[0027] The first amplifier block 201 includes a plurality of power amplifiers PA arranged in parallel. For convenience, only one power amplifier 21 is shown in detail in block 201, under the understanding that the remaining one or more power amplifiers in the block may be similarly configured. In one embodiment, the power amplifiers in each block may have an inverter configuration that operates in switch mode instead of linear mode.
[0014]
[0028] The power amplifier 21 may include power transistors, either alone or in conjunction with one or more other transistors and / or circuit elements. The power transistors may be, for example, MOSFETs having fixed or adjustable gain, a predetermined bandwidth, power efficiency, and impedance. In some embodiments, the RF amplifier may be a Class D amplifier, a Class E amplifier, a Class Φ amplifier, or another type of amplifier. In some embodiments, the power amplifier 21 may operate in switch mode. In other embodiments, the power amplifier 21 may operate in the linear region, for example, at relatively low power levels.
[0015]
[0029] In addition, the power amplifier 21 may include a first control input 22, a power input 23, an RF power output 24, and a second control input 25. The first control input 22 may receive a first control signal from a control driver system (or controller) 71. The first control signal may include a gate signal to the power transistors of the power amplifier 21. In one embodiment, the first control signal may include information indicating the amplitude A1 of the power signal (voltage) to be output, and / or timing information (e.g., phase φ1) for controlling the switching mode (and out-fading) of the power transistors. In switch-mode operation, the control information may omit the amplitude. In linear-mode operation, both amplitude and phase information may be included in the first control signal. The amplitude and phase information may also be received by other power amplifiers in the block.
[0016]
[0030] Power amplifier 23 receives one of a plurality of supply voltages V D1 ,..., V DN output from power supply 72, where N≥2. In one embodiment, the voltage signal from power supply 72 can correspond to a fixed voltage. In another embodiment, the voltage signal from power supply 72 can correspond to a variable voltage. Other power amplifiers within block 20 can receive the same supply voltage as power amplifier 23, or a different supply voltage. For purposes of illustration, it will be assumed that all power amplifiers within block 201 receive the same supply voltage.
[0017]
[0031] Power supply 72 can be configured in various ways. In one embodiment, power supply 72 can generate regulated power supply voltages V D1 ,..., V DN from one or more input energy sources. In one embodiment, power supply 72 can generate independently controlled variable voltages as power supply voltages V D1 ,..., V DN which enables modulation of the RF output voltage V RF,j from the amplifier block using power supply (or drain) modulation. In the latter case, power supply 72 can enable rapid switching between available discrete voltage levels, thereby enabling rapid modulation of output power voltages V D1 ,..., V DN without the need to rapidly slew the power supply voltage. These functions will be described in more detail below.
[0018]
[0032] In the embodiment shown in Figure 1, the supply voltage output from the power supply 72 corresponds one-to-one with N amplifier blocks. Thus, the number of supply voltages may be equal to the number of amplifier blocks in this case. The same may or may not apply to the number of control signals output from the controller 71. For example, the control signals output from the controller may correspond one-to-one with amplifier blocks, or different controls (e.g., phase information) may be input to different power amplifiers of each block to generate the intended output power voltage from each block during outfading. In other embodiments, these one-to-one correspondences may not exist.
[0019]
[0033] The output voltage 24 of power amplifier 21 may be selectively combined with the output voltages of one or more other power amplifiers in block 201, or it may be output independently, for example, depending on the discrete modulation scheme (e.g., phase angle and / or power management method) of the implemented amplifier blocks. The voltages output from the power amplifiers (PAs) in block 201 may be combined, for example, by a combiner 29 further included in the block. As a result, the combiner, if amplifier 21 has a positive gain greater than 1, outputs a power signal (e.g., voltage V) with an amplitude equal to or different (e.g., greater) than the voltage supplied to the power input 23. RF1 Outputs ).
[0020]
[0034] The power signal 24 may be an intermediate voltage (for example, undergoing further processing before input to the power combiner), or it may be output directly to the power combiner. The gain of each power amplifier in block 201 is selectively combined with power signals from one or more other power amplifiers in block 201, for example, with respect to the load (e.g., Z Load The voltage output from the combiner 29 can be set to be within a predetermined range in order to meet the power requirements of ). The power signal output from the amplifier block 201 is the voltage V RF1 It is shown as having the following characteristics.
[0021]
[0035] The second control input 25 is coupled to receive a second control signal in the form of a switching control signal SCS. The switching control signal may be output from a switching control circuit 28 located within or coupled to the power amplifier. In one embodiment, the switching control signal SCS performs a switching augmentation function. When a supply voltage is received from the power supply 72, the power amplifier, operating in switch mode, converts the supply voltage into a desired RF power signal. Convert to.
[0022]
[0036] In addition to these features, the RF power amplifier 21 can be coupled to a reference potential or bias voltage source 27. The reference potential can be, for example, the ground potential, but in another embodiment it can be a different potential. Furthermore, before input to the power combiner 30, the power signal V from the amplifier block 201 is connected. RF1 This may pass through one or more active elements. Active elements may include reactance elements (e.g., capacitors, inductors, transmission lines, etc.) that can help set the output impedance of the block. If the active elements include a capacitor, the capacitor may also function, for example, as a noise filter and / or a smoothing capacitor. The output section of the first amplifier block 201 has a first impedance Z AL,1 It may have a value that is fixed or variable, for example, based on the state of the power amplifier in block 201.
[0023]
[0037] The remaining of the N amplifier blocks (block 20) N Up to this point, the first amplifier block can have a similar function to the first amplifier block. The output impedance of these blocks is Z ALj It is labeled as such, and here the Nth block is Z ALN It has an output impedance of . Therefore, according to one embodiment, the amplifier stage 10 has a plurality of amplifier blocks 201-20 N The j-th power block has a power input V Djand the power signal RF output V RFj which can be considered where 1 < j ≦ N.
[0024]
[0038] In one embodiment, the amplifier block can be controlled to be always on. In this case, different combinations of power amplifiers within each block can be selectively switched to control the respective power signal output of the block. This can be accomplished, for example, based on a predetermined and / or feedback-controlled outphasing pattern. Selective switching of the power amplifiers within the amplifier block can be used to perform discrete modulation of the power signals output from those blocks. As a result, this can vary the RF power signal output to the load in discrete steps.
[0025]
[0039] In linear mode operation, both the amplitude information and the phase information output from the controller 71 can control the power signals output from each block. Some or all of the amplitudes A1,..., A N can be the same or different from each other in this case. In switch mode operation, only the phase φ j can be used according to a predetermined outphasing scheme. The outphasing scheme can selectively activate a predetermined amplifier block at a predetermined time to generate the power signals to be combined by the power combiner to supply power to the load. By controlling the power amplifiers in each of the N amplifier blocks (e.g., by controlling the on / off state of the power amplifiers), the amplifier stage 10 can achieve an expandable power generator design sufficient to meet the requirements of many applications.
[0026]
[0040] In one embodiment, an amplifier block can be selected (or activated) based on control information from the controller 71, along with switching performed on the power amplifiers within each amplifier block. In this sense, the switching (or selection) of an amplifier block can be considered as performing coarse adjustment of the RF power signal to the load, while the switching of the power amplifiers within each amplifier block can be considered as performing fine adjustment of the RF power signal to the load. The control information that controls the switching of the amplifier blocks can be used, for example, to change the operating state of the block (e.g., activated / deactivated, or selected / deselected), thereby enabling different switching combinations of the block.
[0027]
[0041] In one embodiment, in order to improve efficiency and achieve a wide load range, N additional Each power transistor within a widener block may be operated to maintain zero-voltage switching (ZVS, or soft switching) over the entire or a predetermined portion of the block's operating range. Auxiliary circuitry may be included to facilitate ZVS operation under varying conditions.
[0028]
[0042] The power combiner 30 is used as the basis for supplying power to the load, and the combined power signal (V RFT ) generates 35. The combined power signal is based on a combination of power signals output from the amplifier stage 10, for example, a combination of power signals output from all or some of the N amplifier blocks. In one embodiment, the combination may be performed using a lossless (and therefore non-isolated) combination method. In another embodiment, the combination may be performed using an isolated combination method coupled with energy recovery. In these or other embodiments, the power signals may be combined based on weights or weighting coefficients assigned to each of the RF power signals. The output impedance of the power combiner 30 is Z T That is the case.
[0029]
[0043] In one embodiment, the power combiner 30 can perform one or more operations in addition to its combined operation. For example, the power combiner may, in addition to or in combination with, the impedance conversion performed by the impedance converter 40, perform a block impedance (e.g., Z AL1 ~Z ALN A power combiner can perform load modulation of a power amplifier block, including conversion. A power combiner can also perform further power control and / or narrow the operating range of a power amplifier block.
[0030]
[0044] The impedance converter 40 converts the output impedance of the power combiner 30 to the load Z Load The impedance is matched to at least within a predetermined tolerance. This impedance matching operation can improve the performance and efficiency of transferring RF power to the load. The characteristics of the load can control the impedance transformation to be performed. For example, the load impedance range relevant to a given application can determine the voltage and / or current levels corresponding to the load. These voltage and / or current levels can be scaled back, for example, to scale the load impedance to a more appropriate range, making them more suitable for efficient synthesis. In one embodiment, the impedance converter 40 can perform fixed or variable impedance transformation for this purpose. To accomplish this, the impedance converter 40 may include a matching variable network (TMN) or a resistive compression network. The output impedance of the converter 40 is the load Z Load It can match (effectively match) the impedance of the load. In addition, in one embodiment, the signal output from the impedance converter 40 can correspond to the RF power signal 90 supplied to the load.
[0031]
[0045] The RF power generator 10 may include several additional features. For example, the RF power generator may include, or be coupled to, a control system 80. The control system may include, for example, the controller 71 and power supply 72 discussed earlier. In addition, the control system 80 may include, but is not limited to, the following: V RFout The control system 80 may include one or more sensors 81 for measuring parameters, including corresponding output voltage, current, and power, ZVS detection and monitoring of power amplifier transistors, and / or system temperature. Accurate high-bandwidth measurements of voltage and current (or direct measurement of RF power) under variable load conditions may, in some embodiments, be advantageous for controlling output power more accurately with even faster response times. For example, the control system 80 may include sensors to dynamically change the power signal generation from the amplifier stage 10 to meet changing loads, impedances related to the load, and / or power requirements, or to compensate for various conditions of the RF power generator, for example, to maintain various components of one or more RF power generators operating within a predetermined range, from sensors to controllers. It should include a feedback loop to 71.
[0032]
[0046] Sensor 81 inputs the measured value to controller 71. Depending on the system, controller 71 can generate various types of control signals CS to output to power supply 72. In one embodiment, the control signal is a voltage V to be supplied to the amplifier block of stage 10. D1 ..., V DN All or part of it can be selected or changed. In some embodiments, this can be done entirely or partially, with respect to the reference voltage V D1,Ref ~V DN,RefThis can be performed by specifying the following. The controller 71 can also generate a signal (e.g., gate signal 22) to control the out-fading of the power signal from the amplifier block. In switch mode operation, the control signal can indicate the switching timing and relative phase between the amplifier blocks. In linear mode operation, the control signal is the voltage V of the output power signal output to the power combiner 30. RF,1 ..., V RF,N The amplitude for generating the signal may further be included. These and other features of the RF power generator 20 are discussed below.
[0033]
[0047] The amplifier stage 10 can control power generation based on the multi-inverter discrete backoff (MIDB) technique. Unlike other proposed power generators, the MIDB technique allows the RF power generator to combine the outputs of switch-mode power amplifier blocks arranged in parallel, current combination, and / or voltage combination groups without loss, while simultaneously performing outfading between groups of amplifier blocks for the purpose of generating an RF power signal. Such a technique allows high-bandwidth output power modulation to be performed via rapid changes in phase shift (e.g., the phase shift can be performed by setting gate-driver signal delays between power amplifier blocks, which can be changed at a very fast rate based on corresponding control signal commands), thereby improving reliability, performance, and power efficiency.
[0034]
[0048] Combined with these advantages, one or more embodiments of RF power generator 1 can compensate for both voltage-related and current-related losses, which is a performance advantage not achievable with other RF power generators. As a result, higher efficiency can be achieved, particularly at relatively low power levels. For example, other proposed power amplifiers operate at full voltage. Consequently, loss components related to the supply voltage (e.g., device output capacitor losses, resonance losses, etc.) are not reduced and therefore not compensated. This lack of compensation negatively impacts efficiency, especially at lower power levels. In contrast, according to one or more embodiments, RF power generator 1 can compensate for voltage-related losses, as well as current-related losses, over a wide range including relatively low power levels. Furthermore, RF power generator 1 can achieve a wider range of output power backoff (e.g., including large ratios of 30 dB or more). For these and other reasons, RF power generator 1 can be suitable for a wide variety of applications, including, but not limited to, those with high peak-to-average power ratios. Embodiments of the MIDB method are discussed next.
[0035]
[0049] The MIDB method can be implemented to perform discrete modulation with an out-fading pattern. In this method, the power amplifiers within each amplifier block generate a fixed (or substantially fixed) number of RF voltage vectors. Discrete steps of the output RF power signal 90 voltage can be performed by varying (or switching) the number of active power amplifiers within one or more of the N amplifier blocks, and then by combining the power signals from the blocks. As a result, the out-fading angle can always be maintained within a given range, thereby enabling a higher level of efficiency (ideally, substantially higher level of efficiency) and, at the same time, a wider backoff range.
[0036]
[0050] To further extend the output power range while achieving high efficiency, in one or more embodiments, the RF power generator 1 can perform modulation in the power supply 71. For example, discrete drain modulation may be performed, in which the supply voltage input to one or more of the amplifier blocks is switched between several discrete levels. Outfacing is then used in a MIDB (implemented in all or part of the blocks) to provide continuous control over a predetermined output power range corresponding to the load. This can be achieved, for example, by interpolating between power levels obtained at various discrete supply voltage levels.
[0037]
[0051] With respect to output power control, the combination of discrete drain modulation and MIDB techniques can provide an additional basis for performing discrete steps of the RF voltage quantity so that it is out-of-phase from each block. This can allow the MIDB configuration to be highly modularized and reduce supply generation and modulation overhead, even at relatively low power levels. Moreover, in one or more embodiments, the generation of discrete supply voltages may be adapted continuously, and for example, the supply modulation operation may be controlled to occur on a timescale that may be much slower than, for example, that used in high-speed power control.
[0038]
[0052] Figure 2 shows an exemplary embodiment of the RF power generator 100. That is, the embodiment in Figure 2 may be one embodiment of the RF power generator of Figure 1. In this exemplary embodiment, the RF power generator 100 implements a MIDB technique combined with out-fading of the amplifier block.
[0039]
[0053] Referring to Figure 2, the RF power generator 100 includes a power supply 110, an amplifier stage 120, a power combiner 130, and an impedance converter 140. The power supply 110 may include a plurality of power supplies 111 and an optional modulator 112 which may include one or more switches. The power supplies output voltages at N discrete voltage levels, and the modulator 112 modulates the N discrete voltages and outputs them to the amplifier stage 120. In one embodiment, the modulator 112 performs discrete drain modulation on the N discrete voltages output from the power supplies 111. One embodiment of discrete drain modulation is discussed in more detail below.
[0040]
[0054] The amplifier stage 120 consists of multiple (M) MIDB amplifier blocks 1211-121 M This includes, where M>1. Each amplifier block may include multiple RF power amplifiers combined in parallel, resulting in the voltage V of each block. RF,1 ..., V RF,M The output RF power signal having this property can be discretely modulated.
[0041]
[0055] Figures 3A to 3C show MIDB blocks 1211 to 121. M Various embodiments of how each of these may be configured are shown. Figure 3A shows a block of one such configuration including two modular power amplifiers, namely power amplifier (PA1) 211 and power amplifier (PA2) 212, and a combiner 218. Power amplifiers 211 and 212 may have a structure as shown in Figure 1, for example, and may output voltages V1 and V2, respectively, as well as the same current I in this example. Power amplifiers 211 and 212 may be controlled to perform discrete modulation of the output voltage of the block. This may be done, for example, by controlling the on / off state of the power amplifiers to achieve a desired modulation. The on / off state of power amplifiers 211 and 212 may be controlled, for example, based on control information output from a controller, such as a gate signal, phase value φ, and / or other information output from controller 71.
[0042]
[0056] In one embodiment, the power amplifier keeps one or more of the transistors in a fixed-gating state (for example, a bias (or control) signal to the transistors). The power amplifier can be kept in the off state by applying a bias (or control) signal, in particular by applying a bias (or control) signal to one or more electrodes of the transistor to hold a particular switch "on". When operating (on), the power amplifier can have a fixed relative phase (e.g., (φ1-φ2)) based on the desired combiner type (e.g., in phase, perpendicular phase, 180-degree phase shift, or some other relative phase relationship in the illustrated combiner), and the average phase is controlled to provide out-fading between MIDB blocks. Alternatively, the relative phase may also be adjusted during operation. In this way, the respective output voltages V1 and V2 of the power amplifiers 211 and 212 can be selected (e.g., actuated or switched) based on different phases (φ1 and φ2, which may be the same in the embodiment shown in Figure 3A, for example), combined, and then produce the desired modulated power signal output of the block.
[0043]
[0057] In other words, PAs within a MIDB block can be either active (on) or inactive (off or AC-grounded) based on the control. Active PAs within the same MIDB block can be switched synchronously (or exactly 180° apart if an alternative combiner type is used), for example, as shown in Figures 3A-3C. Regarding out-fading with different phases, this is V rf1 ~V rfM This can occur during that period, i.e., after the MIDB block.
[0044]
[0058] As described above, in embodiments, the relative phase relationships may be fixed or selected in combination with a desired combiner structure (including, but not limited to, in-phase, 180-degree phase shift, right-angle phase, and / or relative phases of 45-degree or 135-degree phase shift) to allow their phases to be adjusted together to impart outfading to different blocks. In the combiner shown in Figure 3A, it is generally desirable to have the power amplifier controls in-phase. In embodiments, an option to allow adjustable relative phases within a block can be added, but this is less of a condition than usual. It should be noted that individual amplifiers are turned on or off depending on whether the gate signal is provided or held at a certain stage. In some cases, holding a particular power amplifier switch on so that that particular power amplifier switch appears as a short circuit is a desirable state for a PA that is "off".
[0045]
[0059] Thus, in one embodiment, the power amplifiers within each amplifier block may be active (on) or inactive (off or AC ground) based on control. Active power amplifiers within the same amplifier block may be switched synchronously (as shown, for example, in Figures 3A-3C), (or precisely 180 degrees offset, if an alternative combiner type is used). In another embodiment, the switching may be performed asynchronously. In one embodiment, the phases may differ in the out-fading portion. This may occur, for example, after the MIDB block. In one embodiment, the relative phase relationship may be fixed by the desired combiner structure (e.g., in phase, 180-degree phase shift, perpendicular phase), but their phases may be adjusted together to impart out-fading for different blocks. In the illustrated illustrative combiner, the power amplifiers may be controlled in phase. In one embodiment, adjustable relative phase may be performed within one or more of the amplifier blocks. Whether or not a gate signal is received can control whether the corresponding power amplifier is on or off, or held in a certain position. (In some cases, holding a particular PA switch ON so that it appears as a short circuit may be a desirable state for a PA that is "OFF".)
[0046]
[0060] The combiner 218 can combine the output voltages of power amplifiers 211 and 212 in common mode. Although the combiner 218 is shown as a coupled inductor to better illustrate the voltage combination characteristics within the MIDB block, in practice... It should be noted that the combiner does not necessarily have to be based on the inductor of a conventional AC transformer. The combiner 218 can combine the output voltages of power amplifiers 211 and 212 in common mode using a transformer base structure such as an interphase transformer, or an RF power combiner or RF coupler (e.g., an RF combiner or RF coupler including a transmission line section) including a transmission line section, as in a transmission line transformer. For example, as shown in Figure 3A, the combiner 218 includes two windings 215 and 216 coupled to effectively form a transformer with a predetermined winding ratio. In this example, the combiner includes a node N1 that generates the output power signal voltage of the block. In one embodiment, the output power signal voltage of the block (e.g., V RF1 ..., V RFM One of these can correspond to the weighted sum (e.g., average) of the output voltages V1 and V2 of the power amplifiers within the block. The output current from the amplifier block can correspond to the sum of the output currents of power amplifiers 211 and 212, which in this example is 2I.
[0047]
[0061] Figure 3B shows a stage 220 representing the configuration of each amplifier block. In this exemplary embodiment, the stage 220 includes three modular power amplifiers 221, 222, and 223 and a combiner 228. Power amplifiers 231-223 may have a structure as shown in Figure 1, for example, and may output voltages V1, V2, and V3, as well as the same current I in this example. The power amplifiers may be controlled to perform discrete modulation of the output voltage of the block. This may be done, for example, by controlling the on / off state of the individual power amplifiers to achieve a predetermined modulation. The on / off states of power amplifiers 221, 222, and 223 may be controlled, for example, based on control information output from a controller, for example, a gate signal output from controller 71. In this way, the respective output voltages V1, V2, and V3 of the power amplifiers can be selected (or actuated) with appropriate phase (φ1, φ2, and φ3) and gate control, combined, and then generate a signal voltage of the desired modulated output power of the block.
[0048]
[0062] One difference between the examples shown in Figures 3A and 3C is the number of PAs that make up the block, and furthermore, the number of combiners required to achieve the common-mode voltage combination (for example, Figures 3A and 3C show the use of a 2-way common-mode combiner / phase transformer, while Figure 3B shows the use of a 3-way common-mode combiner / phase transformer).
[0049]
[0063] The combiner 228 can combine voltages 221, 222, and 223 in common mode using an interphase transformer base structure (e.g., a coupled conductor base structure). For example, as shown in Figure 3B, the combiner 228 includes three windings (conductors) 224, 225, and 226, each winding (conductor) located on the legs of a tripod transformer core, coupled at the first end to the output of each of the power amplifiers among the multiple power amplifiers, and coupled at the second end in common to node N2. This results in proportional currents in the individual power amplifiers, and the output voltage is a weighted sum of the three power amplifier voltages. The outputs of the power amplifiers are combined at node N2, which generates the output power signal voltage of the block. In one embodiment, the output voltage of the block (e.g., V RF,1 ..., V RF,M The corresponding values among them can correspond to the weighted sum (e.g., average) of the output voltages V1, V2, and V3 of the power amplifiers within the block. The output current from the power amplifier block can correspond to the sum of the output currents of power amplifiers 321, 322, and 323, which in this example is 3I.
[0050]
[0064] Figure 3C shows block 230 which may represent the configuration of each block of the amplifier stage. In this exemplary embodiment, block 230 is two subblocks of the power amplifier 2 This includes 40 and 250. Each of the subblocks may correspond to, for example, the block in Figure 3A, but the output of each stage is coupled to the combiner 260. The first subblock 240 includes power amplifiers 241 and 241 having output voltages coupled to the combiner 243. The second subblock 250 includes power amplifiers 251 and 252 having output voltages coupled to the combiner 253.
[0051]
[0065] The power amplifiers within block 230 can be controlled to perform discrete modulation of the block's output power signal voltage. This can be done, for example, by controlling the on / off states of power amplifiers 241, 242, 251, and 252 to achieve a predetermined modulation. The on / off states can be controlled, for example, based on control information output from a controller, such as a switch-gate signal output from controller 71. In this way, the respective output voltages V1, V2, V3, and V4 of the power amplifiers are selected (or activated) and combined to generate the desired modulated output voltage of the block. (In this regard, it should be noted that in embodiments, the relative phase relationships of the PAs within the block may be fixed, and the rf output is modulated in discrete steps by turning one or more amplifiers on or off. The phases of the "on" amplifiers can be modulated together to be out-of-phase with respect to different blocks.) The combiner 260 controls the output voltage V of the first subblock 240. 12 and the output voltage V of the second subblock 250 34 The two are combined. In one embodiment, combiner 260 can be similar to combiner 213 in Figure 3A. The output of combiner 260 from node N3 is, for example, the weighted sum (e.g., average) of the output voltages of subblocks 240 and 250 (V RF ) can be the sum of the currents in those stages, which in this example is 4I.
[0052]
[0066] In one embodiment, the combiner in Figures 3A-3C may be implemented using phase-to-phase transformers or transmission line transformers. For example, each combiner in the MIDB amplifier block may be implemented using an n-way phase-to-phase transformer or a "wiffle tree" of two-way phase-to-phase transformers with a quarter-wavelength line power combiner. In such a case, the output power signal voltage of each block can correspond to the direct average of the output voltages of the power amplifiers in that block. In the illustrated example, the power amplifiers in each block have the same output current. In another embodiment, a double of this configuration may be implemented, where all the power amplifiers in each block have the same voltage and output current, and they are summed, for example, weighted averaged.
[0053]
[0067] For each block, the peak output voltage and power can be achieved by turning on all power amplifiers within the block. (Similarly, the peak output voltage and power from the amplifier stage can be achieved by turning on all power amplifiers within the block.) According to one embodiment, the peak output voltage and power of any one block can be reduced in discrete steps by selectively turning off one or more power amplifiers within that block (e.g., deselecting them based on phase φ or other control information). This can be performed, for example, based on a predetermined pattern of the on / off states of the power amplifiers, resulting in the achievement of a desired modulated (or changing) block output voltage over time that satisfies the load requirements. The predetermined pattern can be a symmetrically switched pattern in one embodiment, or an asymmetrically switched pattern in another embodiment. Power amplifiers can be turned off in a variety of ways. For example, each power amplifier can be turned off using an AC grounding technique.
[0054]
[0068] In one embodiment, the power amplifiers (PAs) within each block output the same current (I). In another embodiment, one or more power amplifiers within each block may output a different current from one or more of the other power amplifiers. Furthermore, the combiners of the amplifier blocks have been described as common-mode combiners. In one embodiment, all or some of the combiners within each block may be other types of combiners, e.g. For example, a differential combiner can be used, where the output voltage of each MIDB block is based on the difference between the individual voltages of the block, and the output current is the same as the current of the individual power amplifiers, e.g., current I. In one embodiment, output voltage modulation using a differential combiner can be achieved by implementing on / off control of the power amplifiers within each MIDB block.
[0055]
[0069] MIDBamp Blocks 1211-121 M It has been previously shown that all of them can have the same configuration, for example, one of the configurations shown in Figures 3A to 3C. In another embodiment, one or more of the MIDB amplifier blocks can have a different configuration from the others, for example, in one embodiment, the MIDB amplifier blocks can have different configurations from those shown in Figures 3A to 3C.
[0056]
[0070] Figure 4 shows one embodiment of a power amplifier 400 that may be used to implement a power amplifier (PA) within an amplifier block. In one embodiment, the power amplifier 21 in Figure 1 and / or the power amplifier (labeled PA) in each of the MIDB blocks in Figures 3A-3C may include a power amplifier 400. While the configuration in Figure 4 may be used in several applications, in other embodiments or applications, the power amplifier 400 may have a different configuration. In plasma generation applications, the power amplifier 400 may have all or some of the following features, for example: (1) high (or desired) efficiency in a given switching frequency range (e.g., tens of MHz), (2) efficient AC grounding for MIDB control (and / or other connections for turning off the power amplifier in the MIDB block and the combiner), (3) maintaining efficiency against load impedance variations (e.g., for out-fading and requirement relaxation in impedance transformation), and / or (4) fast dynamic response to discrete step changes in output power, e.g., out-fading angle, on / off state, and step changes in supply voltage level.
[0057]
[0071] Referring to Figure 4, the power amplifier 400 consists of an example of a ZVS Class D inverter circuit including a filter 410, a shunt leg 420, a switch 430, and a switch 440. The filter 410 is coupled in series with an inductor L to the output terminal 401, which is coupled to the associated block combiner in the case of Figures 3A-3C. r and capacitor C r L r and C r Based on the value, the filter is adjusted to resonate at the relevant switching frequency in order to filter out the fundamental components of the switching node voltage.
[0058]
[0072] The shunt leg 420 provides an additional inductive load to the switching node, and therefore helps the switch achieve zero-voltage switching even when the load of the power amplifier changes, for example, within a predetermined range. The shunt leg is connected in series with an inductor L, as previously mentioned, which has a value sufficient to achieve zero-voltage switching operation. ZVS and capacitor C ZVS Includes.
[0059]
[0073] Switches 430 and 440 operate in switch mode and can each be implemented by a transistor having a predetermined conductivity. Switch 430 (Q1) is located at node N ZVS and power supply V dd Coupled between them, for example, can be an N-channel transistor having an intrinsic body diode or equivalent 435 for zero-voltage switching. Switch 440(Q2) is at node N ZVS Coupled between the power amplifier 400 and a reference potential 450 (e.g., ground), the power amplifier 440 can also be an N-channel transistor having an intrinsic body diode or equivalent 445 for zero-voltage switching. During operation, when the power amplifier 400 is on (e.g., selected or activated), both transistors 430 and 440 switch by conducting, each for, for example, about half of the switching period, preferably coinciding with the achievement of zero-voltage switching for each transistor. When the power amplifier 400 is off (e.g., not selected or activated) When this happens, switch 430 is turned off and switch 440 remains constantly on. This effectively grounds the output to AC ground and alters the output voltage of the MIDB block. Note that, when the power amplifier is off, you can, if necessary, reverse which switches remain on and which remain off.
[0060]
[0074] In other embodiments, the power amplifier 400 may have a different configuration. For example, the power amplifier may have a Class E inverter structure or a Class φ2 type structure. In these structures, an additional switch may be used to disconnect the power amplifier from the input power supply when it is in the off state. Furthermore, one or more of the transistors of the power amplifier (PA) may be kept on to achieve the desired AC ground. In other MIDBs, different switch state settings may be used for the power amplifier in the off state. For example, in an MIDB configuration that forces an equal voltage across the amplifier outputs and averages the current between power amplifiers in a block or across blocks, the off state may be achieved, for example, by using a switch to open the power amplifier output or by keeping the amplifier transistors in the off state.
[0061]
[0075] Figure 5 shows power amplifier blocks 1211-121 used to generate the RF output voltage (Vrf) of a MIDB block with four PAs, for example, as shown in Figure 3C. M An example of a switching waveform 500 that can be used to selectively turn on and off four power amplifiers in one of the M amplifier blocks is shown. The same or different waveforms can be used for power amplifiers in other of the M amplifier blocks. In this example, the switching waveform shows a symmetrical pattern of the on / off states of the power amplifiers. In section 510, the switching waveform turns on all the power amplifiers in the MIDB block, thereby causing the block to output a peak voltage. In section 520, the switching waveform turns on three of the power amplifiers and off the remaining power amplifiers. In section 530, the switching waveform turns on two of the power amplifiers and off the remaining two power amplifiers. In section 540, the switching waveform turns on one of the power amplifiers and off the remaining three power amplifiers. The waveform can then be repeated, and may or may not be in the amplitude order shown here, depending on the desired output power and load.
[0062]
[0076] Therefore, in this example waveform, each MIDB block is always on within the cycle and outputs a voltage. Furthermore, the modulated output voltage of the blocks illustrated by waveform 500 is reduced at a discrete level (i.e., waveform 500 is an example of a discretely reduced modulated output voltage Vrf).
[0063]
[0077] Next, the modulated voltages from each block can be out-phased into the combiner along with the modulated output voltages of the remaining blocks. Out-fading of the modulated voltages from the MIDB blocks can be controlled based on phase information φ (or phase angle α) that selectively controls which amplifier blocks should be turned on or selected. In one embodiment, all MIDB amplifier blocks can always be turned on. In another embodiment, different combinations of MIDB amplifier blocks (all or less than all) can be turned on over time for the purpose of generating an RF power signal for the load, which can be particularly advantageous when the load power requirements change as expected or unexpectedly as a result of information fed back from one or more sensors 81.
[0064]
[0078] Which MIDB blocks are on or off can be determined by phase control information output from the controller 71, as previously described. In one embodiment, a block-off state may include the controller 71 outputting phase information that turns off all power amplifiers within that block. Since the power signal voltage output from the device changes in discrete steps, the total output power signal voltage from the amplifier stage 120 can be changed in discrete steps to satisfy the power requirements of the load.
[0065]
[0079] After a step change occurs, the output voltage of the power amplifier (or a given block) can be settled very quickly, for example, within a few RF cycles. This structure is highly modular, with each block containing two or more power amplifiers to obtain a predetermined discrete number of voltage levels to be output from each block. This modular configuration can also achieve higher peak power and a wider output power range compared to other proposed designs.
[0066]
[0080] Figure 6 shows one embodiment of a power combiner 130 in the form of an M-way combiner. The combiner 130 takes out-of-phase voltages from the MIDB block of the amplifier stage 120, for example, V in Figure 2. RF1 ..., V RFM A lossless combiner can be formed by combining corresponding components from among them. For illustrative purposes, combiner 130 is a voltage V output from a power amplifier stage 120 containing two MIDB amplifier blocks 610 and 620. X and V Y The combination is shown as follows: In this case, power supply 71 includes two power supplies 601 and 602 corresponding to 100V and 400V respectively, which are switched (using switching logic 603) to generate a modulated supply voltage for input to power amplifier 120. In one exemplary embodiment, the high voltage (e.g., 400V) may come from the main DC power supply and the low voltage (e.g., 100V) may come from the auxiliary power supply. In this embodiment, the number of power supplies is equal to the number of MIDB blocks in this non-limiting embodiment. In one embodiment, the number of power supplies and MIDB blocks may differ.
[0067]
[0081] In the exemplary embodiment shown in Figure 6, the RF power combiner 130 includes a Chireix-type combiner with asymmetric compensation. In Figure 6, the RF combiner is shown as a transformer 650 that performs m:n voltage and impedance conversion (where generally m can be greater than or equal to n, or less than n). For example, the combined output voltage VL To generate the desired output, if a rise in impedance transformation from the load to the output of the PA block is required based on the structure of the coupling conductor, then m > n. The combiner 130 is a shunt reactance compensation component jX A and -jX B The combination further includes, which may be provided to adjust the impedance output of the MIDB within a predetermined range. Output current I of combiner 130 L This refers to the current I output from MIDB power amplifier blocks 610 and 620, respectively. X and I Y Based on the combiner's output voltage V L (or V in Figure 1) RFT ) is connected to the load Z via the impedance converter 140. L It interfaces with and one embodiment corresponds to box 680.
[0068]
[0082] Load Z L It can be fixed or variable. For plasma generation applications, Z L This can be a variable load. In this case, the RF power generator can vary the output voltage level to meet the fluctuating power requirements of the load. This can be accomplished, at least in part, by selectively operating the MIDB power amplifier block.
[0069]
[0083] The impedance converter 140 (see Figure 1) adjusts the impedance of the RF power generator to match the load. In one embodiment, the impedance converter adjusts the output impedance of the RF power generator to at least within a predetermined tolerance to match the variable RF load Z L It converts the impedance to a predetermined impedance (or range of impedances) that matches the impedance of the load. Performing impedance matching in this way can improve efficiency and reduce the over-rated power amplifier hardware to adapt to the load range.
[0070]
[0084] More specifically, the switch-mode MIDB block can have a variable impedance based on which combination of individual power amplifiers is selected. The impedance converter 140 can map the output impedance of the amplifier stage 120 to match the impedance of the load. In one embodiment, the impedance converter 140 allows the RF power generator to interface directly with a variable load (e.g., a plasma chamber), thereby mitigating the need for an external system to impedance match the load to a specific intermediate impedance value, e.g., 50Ω. Furthermore, by including the impedance converter in the RF power generator, the impedance converter can perform less extreme impedance compression, which results in a narrower range of load impedances that can be presented to the power amplifier in an acceptable manner.
[0071]
[0085] The impedance converter 580 can be implemented in various ways. In one embodiment, the impedance matching section uses one or more of the following matching variable networks (tunable): (a) phase-switched impedance modulation, (b) switched capacitor, (c) dynamic frequency adjustment, and / or (d) resistive compression networks. It may include a matching network. One embodiment of the matching variable network (TMN) 580 (see Figure 6) may have all or some of the following features.
[0072] • Fixed impedance transformation ratio, where the load impedance range is expanded or contracted by a specific transformation ratio k. • Fixed matching network, where fixed-value passive components are used to perform some load transformation, and in some cases, also to perform some compression if the frequency changes within a predetermined (e.g., relatively small) range. • A matching network having discrete switching passive components, where several binarization components are optionally connected to the system via switches to provide a range of discrete matching reactances. • Dynamic frequency tuning (DFT), where the frequency is dynamically changed within a predetermined (e.g., relatively small) range, and combined with a passive component (e.g., a high-Q resonant tank), provides a continuously matchable variable reactance within a range. Phase-switched impedance modulation (PSIM) can be performed, for example, by controlling the duration for which a fixed capacitor is connected to the system for each RF cycle. This technique allows for a range of continuously matchable and variable reactance at a given fundamental frequency. • One or more resistive compression networks (RCNs), which may be particularly suitable for situations where there is a pair (or many) of loads with similar fluctuation patterns. This technique can then compress the range of real impedance changes of the two (or many) loads, individually or together as a combination, through a passive network.
[0073]
[0086] As mentioned above, one example of a variable load is a plasma generator used in semiconductor processing applications. Plasma generators have a wide load impedance range. Other embodiments of the TMN580 or impedance converter 140 can be tuned or remapped to match the fluctuating load impedance to ensure acceptable operation of the switch-mode power amplifier.
[0074]
[0087] Figure 7 shows one embodiment of a variable matching network 900 that can be used to implement the impedance converter 680 of Figure 6 or the impedance converter 30 of Figure 1. The variable matching network is dynamically adjustable between the output of the power combiner and the load. It can operate to perform capable impedance matching. This may include performing dynamically variable voltage conversion and reactance impedance adjustment, and may include a dynamic frequency tuner (DFT) 710 coupled to the input of a discrete switching passive network 720. In one embodiment, the matching variable network can operate in conjunction with a power combiner that provides an additional impedance conversion ratio, for example, a fixed ratio based on the turns ratio of a transmission line-based power combiner.
[0075]
[0088] The output impedance observed by the MIDB block is, for example, the square of the combiner's turns ratio (m / n). 2 This allows for matching load impedances that have been multiplied by this ratio. Depending on the power amplifier configuration, this ratio may be set so that the power amplifier operates within a predetermined range, for example, near the optimal load impedance range in which efficiency is optimized.
[0076]
[0089] The DFT710 can receive the output of the power combiner 130 and generate a continuously changing series reactance. The DFT unit can be implemented, for example, using a high-Q series inductor (L) and capacitor (C) tuned to resonate at the center of the operating frequency. Thus, at the center operating frequency (ideally, precisely at the center operating frequency), the DFT tank presents near zero impedance (ideally zero impedance). However, for continuously controlled frequency fluctuations above and / or below the center frequency (and, in at least some embodiments, small, continuously controlled frequency fluctuations above and / or below the center frequency), the DFT tank correspondingly exhibits a load impedance Z L The desired impedance Z in A series reactance, which may be inductive and / or capacitive, can be presented to function as part of the impedance matching network when matching to the load Z. The discrete switching passive network 720 connects the DFT 710 to the load Z via switch 724. LThe system includes multiple capacitors 722 connected in parallel between the two components. In one exemplary embodiment, the capacitors 722 may be binarized capacitors that are selectively switched to provide shunt reactance. By switching different combinations of capacitors, it becomes possible to vary the shunt reactance in discrete steps based on the capacitance values. In this embodiment, three capacitors are represented by predetermined capacitances C0, 2C0, and 4C0 that increase progressively. In another embodiment, different numbers of capacitors and / or capacitors with different progressive capacitances may be used.
[0077]
[0090] Switch 724 can be selectively opened to vary the applied shunt reactance in discrete steps, as described above. The switch can be selectively opened and closed in different combinations, for example, based on a switching signal from controller 71. During operation, the small frequency fluctuations corresponding to the selected discrete steps applied, in addition to the top of the series reactance from DFT710, can enable the generation of a continuously controlled series reactance to satisfy the changing conditions of the load. Thus, when combined with the series reactance output from DFT710, the matching variable network 680 achieves impedance matching that matches the impedance of the variable load to at least within a given tolerance or range. Moreover, this design allows the matching variable network 680 to compress load fluctuations into an impedance range, keeping the overall system complexity low while ensuring that the efficiency of the power amplifier is not adversely affected. It will be recognized that other matching variable network designs, including those based on varactors, phase-switched impedance modulation, and other techniques, may be used in a similar manner.
[0078]
[0091] Figure 8 can correspond to one example of switching logic 603 (see Figure 6), and one actual modulator 800 that can be used to perform discrete drain modulation in power supply 72. The implementation details are shown. Discrete drain modulation can extend the achievable output power range of an RF power generator. It can also reduce voltage-related losses in the power amplifier that may become dominant at lower power levels (e.g., transistor cross-related loss, ZVS resonance loss, etc.).
[0079]
[0092] Referring to Figure 8, the modulator 800 receives the first supply voltage V through the first input section 810. High The second supply voltage V is received and passed through the second input section 820. Low It is designed as a two-level power modulator that receives the following. The supply voltage can come from different types of power sources or the same type of light source. In one embodiment, the supply voltage can be obtained from power sources with different power ratings, for example, a high voltage V High This can be obtained from a mains power supply rated in the peak to intermediate power range, and low voltage V Low This can be obtained from an auxiliary power supply rated only at low power levels. For plasma generation applications, a second supply voltage V Low This can have a value at the lowest voltage level before the capacitance nonlinearity of the semiconductor device adversely affects the performance of the power amplifier, for example, the loss of zero-voltage switching (ZVS), or a value just above the lowest voltage level.
[0080]
[0093] Both supply voltages are coupled to multiple different inputs of transistor 830, which can be, for example, an N-channel transistor. First supply voltage V High This may be input to the drain of the transistor, and the second supply voltage V Low The transistor's source is coupled through a blocking diode 840. The transistor's gate may be coupled to receive a control signal to control the transistor's switching, for example, the control signal may be a CS generated by a controller 71 as shown in Figure 1.
[0081]
[0094] In one embodiment, V Highand V Low is received continuously, and V Low is assumed to be greater than the forward bias voltage of blocking diode 840. When a gate signal is input to turn on transistor 830, the first supply voltage V High passes through the transistor and node N out (i.e., in this exemplary embodiment, the voltage V Low is blocked by blocking diode 840, and as a result, the voltage output from node N out is V High ). Then, the combined voltage is sent to one or more selected MIDB power amplifier blocks of amplifier stage 120 (e.g., of FIG. 2). When the gate signal is not received by transistor 830, the diode conducts and the second supply voltage V Low is output to the power amplifier at output node N out . Blocking diode 840 serves to block the input of the first supply voltage to input terminal 820 when the transistor is turned on. In this way, modulator 800 generates (or modulates between) two discrete levels of voltage (e.g., two discrete DC voltage levels V Low or V High ) for input to the MIDB block of the power amplifier.
[0082]
[0095] Modulator 800 may be beneficial for some applications, but for other applications, more than three levels of supply voltage may be beneficial. Thus, in one embodiment, the modulator of power supply 72 can output more than three levels of voltage. For efficiency, the number of discrete voltages output from power supply 72 may vary between embodiments to achieve the desired power range, response speed, and / or efficiency.
[0083]
[0096] FIG. 9A shows an embodiment of a method for generating RF power for a load, which may be, for example, a plasma generator or another type of load. This method may be performed based on one or more embodiments of the RF power generator described herein.
[0084]
[0097] Referring to Figure 9A, in 901, this method first involves amplifier stage 1 This includes generating (or receiving) one or more modulated power supply voltages to output to 0. The same modulated power supply voltage may be output to all amplifier blocks in the amplifier stage at the same or different times, or different modulated power supply voltages may be output to the amplifier blocks at the same or different times. The modulation can be discrete drain modulation of the voltage supplied by one or more power supplies. Which power supply is selected and the modulation to be performed can be based on one or more control signals output from, for example, a controller 71. The control signals can correspond to a predetermined pattern or scheme and / or can be generated adaptively based on feedback from, for example, one or more sensors 81 (see, for example, Figure 1).
[0085]
[0098] In 903, the out-fading pattern is determined, for example, by a controller. The out-fading pattern can correspond to pre-stored control information 75 stored in a non-temporary computer-readable medium (e.g., memory) 75. The pre-stored control information can be, for example, in the form of instructions, or in other forms of firmware or software. The pre-stored control information can control the operation of the MIDB block to generate an RF power signal to the load. The out-fading pattern, and / or other control information embodied in the control information (e.g., instructions) for the controller 71, can determine the discrete modulation to be performed for the power amplifier in each of the MIDB amplifier blocks.
[0086]
[0099] In the 905, each MIDB block is controlled to generate its own power signal based on out-fading and / or other control information from the controller. When all MIDB blocks are always on, discrete modulation of the out-fading power signals from all MIDB amplifier blocks is used to generate the RF power signal for the load. In this case, the amplifier blocks are not switched (or selected), and the power amplifiers within each block are switched, for example, according to a symmetric or asymmetric sequence. When the MIDB amplifier blocks are to be switched along with the power amplifiers within each block, discrete modulation of the out-fading power signals from the blocks is used to generate the RF power signal for the load. To perform the switching for discrete modulation, some of the power amplifiers in one or more blocks must be deactivated according to an out-fading pattern. This can be done, for example, by AC grounding.
[0087]
[0100] In 907, the power signals output from the power amplifier block are combined. This forms an RF power signal for the load. Combining power signals can be done in various ways, for example, using a sieve combiner with or without weights. As previously described, discrete modulation performed in the amplifier block can generate discrete steps in the RF power signal to satisfy the fluctuating power requirements of the load.
[0088]
[0101] In 909, the output impedance of the RF power generator is at least predetermined The impedance is adjusted to match the load impedance within the tolerance. This adjustment can be performed by a matching variable network or any of the other types of impedance transducers described herein.
[0089]
[0102] In 911, over time, the power requirements and / or impedance of the load change. The load may change, and / or the RF power generator may need to be modified. To meet the changing power requirements of the load and match the changes in load impedance, a balanced change is made to the RF output power and / or impedance. This is achieved by different combinations of MIDB blocks, and / or selected MIDB blocks, to achieve a change in the voltage of the output RF power signal 70 (e.g., discrete steps). This can be accomplished by selecting different power amplifiers within the system. This selection can be performed, for example, by changing the on / off state of the corresponding power amplifier (PA) based on a control signal from the controller 71.
[0090]
[0103] Figure 9B shows, for example, one of the embodiments of an RF power generator disclosed herein. An embodiment of a method for generating RF power for a load, which may be performed by one or more, is shown. Referring to Figure 9B, this method includes selecting a first number of power amplifiers in a first amplifier block (921), selecting a second number of power amplifiers in at least a second amplifier block (922), combining the voltages from the first number of power amplifiers to generate the output voltage of the first amplifier block (923), combining the voltages from the second number of power amplifiers to generate the output voltage of the second amplifier block (924), outfacing the output voltages of the first and second amplifier blocks (925), and combining the out-phased voltages to generate an RF power signal for the load (926). As in other embodiments, this method may be supplemented by an operation to select different amplifier blocks. Otherwise, the amplifier blocks may be fixed to remain continuously on.
[0091]
[0104] Figure 9C shows, for example, one of the embodiments of an RF power generator disclosed herein. An embodiment of a method for generating RF power for a load, which may be performed by one or more units, is shown. Referring to Figure 9C, the method includes generating a first modulated power signal from a first amplifier block (931), generating a second modulated power signal from at least a second amplifier block (932), out-of-phase the first and second modulated power signals based on their phase angles (933), and generating an RF power signal for a load based on the out-of-phase first and second modulated power signals (934). Generating the first modulated power signal may include switching one or more of a plurality of power amplifiers in the first amplifier block, and generating the second modulated power signal may include switching one or more of a plurality of power amplifiers in the second amplifier block.
[0092] <Example of power control management>
[0105] One of the many applications of the embodiments of RF power generators described herein is semi The objective is to power a plasma generator in a conductor manufacturing process. Plasma generators sometimes require a very wide output power modulation range compared to other applications. Embodiments of RF power generators can achieve this range by implementing a wide range of discrete steps in the RF output voltage using discrete drain modulation and on / off control of MIDB power amplifier blocks, in addition to fine continuous adjustment of the RF output voltage (with outfading between MIDB blocks). Non-limiting embodiments of such power control management techniques are described below.
[0093]
[0106] Figure 10 shows an example of a voltage vector graph for a power management control method of an RF power generator. This example corresponds to the embodiment of the RF power generator in Figure 6, which has two MIDB phase amplifier blocks.
[0094]
[0107] Referring to Figure 10, the RF voltage output from the first MIDB block 510. The voltage is V XCorresponding to this, the RF voltage output from the second MIDB block 520 is voltage V Y These correspond to the following. These voltages can be given by the following equations.
[0095]
number
[0096]
[0108] As shown by these equations, the load voltage at the output of combiner 230 ( VL), therefore, the power output from the RF power generator is controlled via both the out-fading angle and the output voltage amount of the MIDB block. The out-fading angle (α) is based on the phase shift between the two RF voltages output from the MIDB block.
[0097]
[0109] The output voltage of the selected MIDB block (in this example, voltage| V| is obtained by drain modulation and / or on / off control of the power amplifier in each of the blocks. In the two-level power supply modulation and MIDB block configuration of Figure 6, two groups of four non-zero voltage quantities are available at the same out-fading angle (α). The first group corresponds to vectors 1010, 1020, 1030, and 1040, and the second group of vectors corresponds to vectors 1060, 1090, 1080, and 1090.
[0098]
[0110] More specifically, the load voltage (VL) is directly correlated with the output power (for example, | VL| 2It can be modulated by a combination of outfacing and discrete drain / MIDB modulation (directly proportional to the phase shift). Outfacing provides continuous control, as the phase shift can be continuously changed depending only on the resolution of the implementation hardware in one embodiment. Discrete voltage modulation (achieved, e.g., by switching to different power rails or MIDB on / off configuration) can cover a very wide output power range (e.g., 1000:1 or 30 dB) and may be implemented, for example, when the RF power generator can achieve higher efficiency in different voltage domains or when subject to large power fluctuations.
[0099]
[0111] The discrete voltage step at the output of the MIDB block is controlled by power supply modulation and MIDB control. This can be achieved based on the combination of the following. For example, the two-level supply voltage in Figure 8 is V High When set to =V and V=V / 3, and in a 2MIDB power amplifier block configuration, the available voltage levels can be output from the power amplifier 120 shown below, where PA refers to the individual power amplifiers within the MIDB block. Using this power management method, a single out-fading power range can be obtained (for example, by expanding or contracting according to the squared voltage ratio) to cover a very wide power range while maintaining high efficiency.
[0100] [Table 1]
[0101]
[0112] Figure 11 shows how performance is determined and power management is performed according to one or more embodiments. A graph of admittance values that can be used to perform the control function is shown. This graph can be useful for evaluating the load modulation effect on the power amplifier in the MIDB block during out-fading and for designing the corresponding compensation component and selecting the out-fading angle. For illustrative purposes, the graph in Figure 11 can correspond to the 2-MIDB block phase amplifier in Figure 6, but it may be extrapolated to embodiments with three or more MIDB blocks.
[0102]
[0113] Referring to Figure 11, the uncompensated load admittor of MIDB block 610 in Figure 6 The upper semicircular curve 1110 indicates the uncompensated load admittance of MIDB block 520, and the lower semicircular curve 1120 indicates the outfading (for example, Y in Figure 6). A and Y B See also). Uncompensated load admittance (Y) of MIDB block 610 A ) and MIDB block 620(Y B The uncompensated load admittance of ) can be expressed as follows:
number
[0103]
[0114] The axis is a matching variable network 680 and combiner 130 that purely loads Stylish resistive effective impedance R L Based on the assumption that it will match, 1 / R L It is normalized to . Each MIDB block has a load admittance point determined by the outfading angle and compensation component selection mechanism on the corresponding curve. For a fixed MIDB output voltage quantity, the output power is proportional to the real part of the load admittance, for example, Re{Y A or Y B The larger the value of}, the greater the output power.
[0104]
[0115] In a power amplifier (PA), the closer the load is to the real axis (dashed line in Figure 11), the better the PA performs. It operates efficiently. However, when the output power is changed by changing the out-fading angle, both the conductance and susceptance loads observed in the MIDB block (and therefore the PA) change, which affects the system efficiency during the process. This can be seen from the graph. Therefore, in some cases, outfading alone may limit the range of power modulation for efficient operation. In some cases, the compensating reactance may be selected to shift the curve so that the admittance remains close to the real axis over the widest possible power modulation range, for example, so that the portion of the admittance curve enclosed by box RB in Figure 11 is close to the real axis.
[0105]
[0116] The load admittance characteristic ensures the continuity of power control across voltage steps. Therefore, it may be independent of voltage. In some applications, a specific minimum span may exist within the out-fading angle range. In one embodiment, the modulation voltage level of all MIDB blocks
number
[0106]
[0117] In some embodiments, for a given output power profile, RF power generation Efficiency optimization of the combiner may be possible with respect to the optimal number of power amplifiers per MIDB block, supply voltage level, device area, compensating reactance, and / or target load impedance of a template-matching combiner (e.g., 680 in Figure 6).
[0107]
[0118] Regarding dynamic response, there are three control methods, namely (1) Outf The commanded step changes in output power can be satisfied by (2) changing the fading angle, (3) changing the MIDB on / off configuration, and (4) changing the discrete power supply modulation. Changing the outfading angle can perform a step change in the effective load impedance observed in the power amplifier within each MIDB block. Changing the MIDB on / off configuration and changing the discrete power supply modulation can perform a step change in the common-mode voltage across one or more DC blocking components in the block power amplifier. By distributing the resonant tanks across each of the block power amplifiers, the RF power generator can achieve very fast settling times by reducing or minimizing the DC blocking capacitor values in all three scenarios. This can also have the additional advantage of fully modularizing at least the power amplifier 120 of the RF power generator, and the MIDB blocks can be easily reconfigured to meet new specifications for the intended application.
[0108] <Example simulation results>
[0119] To evaluate the performance of a simulated version of an RF power generator, an example The circuit model shown was implemented in LTSpice for a two-level MIDB-2 configuration power generator in Figure 6, using ZVS Class D amplifiers in each MIDB block and a lossless out-fading combiner in power amplifier 220. The power generator was configured for an output power profile corresponding to a real plasma load application, with the following parameters: 0.2kW for 1ms, 1kW for 1ms, and then 5kW for 20μs. Furthermore, a GaN FET PGA26E19BA switch was specified at an ISM band frequency of 13.56MHz. A matching variable network 580 was used to match the variable plasma load to a fixed real impedance R L It was assumed that it would match. The following simulation results were obtained.
[0109]
[0120] Simulation results for efficiency and power range. The efficiency plot shows the cross-loss. Peak dV / dt loss fitting was employed for the calculations, applying a 5.5 × scalar to the nominal Rds,on of the active block PA device and a 2 × scalar to the conductive switch when the PA is turned on. Additionally, a quality factor (Q) of 500 was set for the inductor and the ideal common-mode combiner, and the supply voltage levels were 300V and 100V based on considerations of switch loss characteristics and power range. A 99% scalar was applied for the lower supply voltage to account for modulator losses (e.g., losses across diode 740 in Figure 7). Several key system design parameters are listed in the table below.
[0110] [Table 2]
[0111]
[0121] Figure 12 shows the MIDB power amplifier block, out-fading, and slaying. The efficiency versus output power backoff curves for a simulated RF power generator using a combiner are shown. The main DC supply voltage level was 300V and rated to 5kW peak power, and the auxiliary supply voltage level was 100V and rated to less than 500W peak power. The output power was modulated with outfading every 1210, 1220, 1230, and 1240 consecutive segments, and discrete RF voltage modulation (by MIDB on / off control and / or power supply modulation) was used at the crossovers to maintain high efficiency over a wide power range.
[0112]
[0122] Furthermore, in Figure 12, the vertical lines represent the power levels in the load profile. As can be seen, high efficiencies exceeding 90% are achieved across the entire power range (200W to 5kW), and at least 30dB is achieved in the output power range (5W to 5kW). Although rated to several kW peak power, the power generator can reliably deliver power at the lowest design power level of >20% efficiency, even at the very low power of 5W.
[0113]
[0123] Figure 13 shows the dynamic response of the power amplifier's output voltage to out-fading. The waveforms in Figure 13 demonstrate that very fast dynamic behavior is achieved, and the new command voltage level is set without a few RF cycles. In particular, waveform 1310 corresponds to the VLoad voltage and shows the output voltage of a 1:1 turns ratio Shiley combiner (e.g., VL in Figure 5). Waveforms 1320 and 1330 correspond to VPAO1 and VPAO3, respectively, and show the unfiltered switching mode voltages of two power amplifiers located in different MIDB blocks. In this simulation, the phase angle α advanced from 30° to 60° in time = 3.69 μs and returned to 30° in time = 4.42 μs only after, for example, 10 RF cycles.
[0114]
[0124] Figure 14 shows that the dynamic response of the output voltage of the MIDB power amplifier was very fast. An example is shown. In Figure 14, waveform 1410 corresponds to the output voltage of a 1:1 turns ratio combiner (VL in Figure 6). Waveforms 1420 for VPAO2 and 1330 for VPAO1 are the switching modes of the two PAs in the same MIDB block. This represents voltage. One PA is turned off at time = 3.69 μs (e.g., AC grounded) and turned back on at time = 4.42 μs (e.g., switched).
[0115]
[0125] Figure 15 shows an example of the dynamic step response of the output voltage to discrete power supply modulation. High-speed setting at the microsecond level can be achieved. For example, the dynamic response to a step in the supply voltage (performed, e.g., by the modulator of power supply 72) is represented by the following waveforms. Waveform 1510 corresponds to the output voltage of a 1:1 turns ratio Shiley combiner (corresponding to Vvload in Figure 15 and VL in Figure 6). Waveform 1520 (marked V(in1)) corresponds to the supply voltage of a power amplifier in a selected MIDB block, and waveform 1530 corresponds to the switching node voltage of that power amplifier. The supply voltage dropped from 300V to 100V in time = 3.69μs and returned to 300V in time = 4:42μs.
[0116]
[0126] According to one or more embodiments, the RF power generator has a unique system configuration and a power control method are provided. In some embodiments, the RF power generator may have various combinations of the following features: (1) outfading of the power signal for fast response (and, if necessary, continuous) power generation; (2) discrete voltage modulation using a switch-mode (e.g., on / off controlled) power amplifier in a MIDB block; and / or (3) discrete drain modulation of the supply voltage to extend the high-efficiency operating power range of the load in various applications. In some embodiments, the load may be a plasma generator used in a semiconductor chip fabrication process. In other embodiments, the load may be different and may operate in different (higher or lower) power ranges, e.g., the high power range of the plasma generator and power ranges that do not have performance requirements.
[0117]
[0127] Discrete drain modulation can be performed in various ways. Two MIDB blocks In a power generator, for example, two power sources may be used, such as a main DC power supply (e.g., for peak to intermediate power levels) and an auxiliary DC power supply (e.g., for lower power levels). Thus, the RF power generator can use low-overhead two-level discrete-drain modulation. In doing so, the power generator can maintain high efficiency and perform fast RF power control over a very wide backoff range. In other embodiments, the power generator may have three or more MIDB power amplifier blocks.
[0118]
[0128] In one embodiment, the RF power supply is matched to the fluctuating load impedance. To enable this and to provide acceptable operation to the switch-mode power amplifier, an impedance converter that performs adjustment (or remapping) may be included. Additionally, the switch-mode power amplifier itself may have a variable impedance that the impedance converter can map to the impedance of a load, such as a plasma generator or another load.
[0119]
[0129] In addition to plasma generation applications, various embodiments of RF power generators are used in other industrial applications. It can meet various requirements, such as high frequencies (e.g., tens of MHz) and power levels (e.g., peak power in kW units), as well as the requirement to operate with variable load impedance over a wide overall power range (e.g., 30 dB) and a high peak-to-average power ratio.
[0120]
[0130] The methods, processes, and / or operations described herein are computer-controlled. It may be executed by code or instructions that can be executed by a processor, controller, or other signal processing device. The computer, processor, controller, or other signal processing device may be one of those described herein, or in addition to the elements described herein. Method (or computer, processor, controller, or Since algorithms that form the basis of the operation of other signal processing devices are described in detail, the code or instructions for performing the operation of embodiments of the method can be translated into a computer, processor, controller, or other signal processing device into a dedicated processor for performing the method herein.
[0121]
[0131] Furthermore, another embodiment involves a computer for storing the aforementioned code or instructions. This may include a computer-readable medium, such as a non-transient computer-readable medium. The computer-readable medium may be a volatile or non-volatile memory or other storage device, which may be detachably or fixedly coupled to a computer, processor, controller, or other signal processing device capable of executing code or instructions for performing the operation of embodiments of the method or apparatus described herein.
[0122]
[0132] Controllers, processors, and generators of embodiments disclosed herein Controllers, processors, generators, logic gates, modulators, combiners, transformers, matching networks, drivers, and other signal generation and processing functions may be implemented in non-transient logic, for example, in hardware, software, or both. When implemented at least partially in hardware, controllers, processors, generators, logic gates, modulators, combiners, transformers, matching networks, drivers, and other signal generation and processing functions may be any of a variety of integrated circuits, including, but not limited to, application-specific integrated circuits, field-programmable gate arrays, combinations of logic gates, systems on a chip, microprocessors, or other types of processing or control circuits.
[0123]
[0133] If implemented at least partially in software, the controller, processor The generators, logic controllers, modulators, combiners, transformers, matching networks, drivers, and other signal generation and processing functions may include, for example, memory or other storage devices for storing code or instructions that can be executed by a computer, processor, microprocessor, controller, or other signal processing device. The computer, processor, microprocessor, controller, or other signal processing device may be one of those described herein or an additional element thereof. Since the algorithms forming the basis of the method (or the operation of the computer, processor, microprocessor, controller, or other signal processing device) are described in detail, the code or instructions for performing the operation of embodiments of the method can be translated into a computer, processor, controller, or other signal processing device that is a dedicated processor for performing the method herein.
[0124]
[0134] Concepts, systems, devices, structures, and techniques that are required to be protected Various embodiments are described herein with reference to the relevant drawings. Alternative embodiments may be devised without departing from the scope of the concepts, systems, devices, structures, and techniques described herein. Note that various connections and positional relationships between elements (e.g., above, below, adjacent, etc.) are described in the following description and drawings. These connections and / or positional relationships may be direct or indirect unless otherwise specified, and the concepts, systems, devices, structures, and techniques described herein are not intended to be limited in this respect. Consequently, the joining of entities may refer to either direct or indirect joining, and the positional relationships between entities may be direct or indirect.
[0125]
[0135] As an example of an indirect positional relationship, layer "A" is formed on top of layer "B" in this explanation. The reference to accomplishing something means that unless the relevant properties and functions of layers "A" and "B" are substantially altered by the intermediate layers, one or more intermediate layers (e.g., layer "C") may be layer "A" This includes the situation existing between " and layer "B". The following definitions and abbreviations may be used for the purposes of the claims and interpretation of this specification. In this specification, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having,” “contains,” or “containing,” or other variations thereof, are intended to cover non-exclusive inclusion. For example, a composition, mixture, process, method, article, or apparatus consisting of a list of elements is not necessarily limited to those elements alone, and may include other elements not explicitly listed or that are specific to such composition, mixture, process, method, article, or apparatus.
[0126]
[0136] In addition, the term “exemplary” is used herein to mean “example, case, or example.” It is used to mean "serving as evidence." Any embodiment or design described herein as “exemplary” should not necessarily be construed as being preferable or advantageous to other embodiments or designs. The terms “one or more” and “one or more” are understood to include one or more any integers, i.e., 1, 2, 3, 4, etc. The term “multiple” is understood to include two or more any integers, i.e., 2, 3, 4, 5, etc. The term “connection” may include indirect “connections” and direct “connections.”
[0127]
[0137] In this specification, "one embodiment," "one example embodiment," and "exemplary embodiment" are used interchangeably. References to such terms indicate that while the described embodiments may include certain features, structures, or characteristics, all embodiments may include certain features, structures, or characteristics. Furthermore, such phrases do not necessarily refer to the same embodiments. In addition, if certain features, structures, or characteristics are described in relation to one embodiment, it is presented that, whether explicitly stated or not, it is in the knowledge of those skilled in the art that such features, structures, or characteristics may be affected in relation to other embodiments.
[0128]
[0138] For the following explanation, "top", "bottom", "right", "left", "vertical", "water" The terms “flat,” “top,” and “lowest,” and their derivatives, shall be used in relation to the described structure and method as an orientation in the drawing. The terms “overlying,” “atop,” “on top,” “positioned on,” or “positioned atop” mean that a first element, such as a first structure, lies on a second element, such as a second structure, and that intervening elements, such as an interface structure, may be present between the first and second elements. The term “direct contact” means that a first element, such as a first structure, and a second element, such as a second structure, are connected without any intermediate elements.
[0129]
[0139] In order to modify the claim elements, the claim may contain "first," "second," The use of sequential terms such as "third" does not in itself imply a temporal order in which the priority, precedence, or sequence or method of action of one claim element over another is performed, but is simply used as a label to distinguish claim elements, to differentiate one claim element having a certain name from another element having the same name (except for the use of sequential terms).
[0130]
[0140] "Approximately" and "about" The terms may be used to mean within ±20% of the target value in some embodiments, within ±10% of the target value in some embodiments, within ±5% of the target value in some embodiments, and within ±2% of the target value in some embodiments. The terms “approximately” and “about” may include the target value. The term “substantially equal” may include several actual values. In some embodiments, these values can be used to indicate values within ±20% of each other, within ±10% of each other, within ±5% of each other, and within ±2% of each other in some embodiments.
[0131]
[0141] The term "substantially" means, in some embodiments, within ±20% of the comparison scale. In some embodiments, this can be used to refer to values within ±10% of the comparison scale, in some embodiments, within ±5% of the comparison scale, and in some embodiments, within ±2% of the comparison scale. For example, a first direction that is "substantially" perpendicular to a second direction may refer to a first direction within ±20% of the 90° angle it makes with the second direction, in some embodiments, within ±10% of the 90° angle it makes with the second direction, in some embodiments, within ±5% of the 90° angle it makes with the second direction, and in some embodiments, within ±2% of the 90° angle it makes with the second direction.
[0132]
[0142] The subject matter of this disclosure, in its application, is described in the following description or in the drawings. It should be understood that the disclosed subject matter is not limited to the details of the structure and the arrangement of its components. Other embodiments are possible and can be practiced and implemented in various ways.
[0133]
[0143] Furthermore, the expressions and terms used in this specification are for illustrative purposes only. It should be understood that this should not be considered limiting. Therefore, those skilled in the art will recognize that the underlying concepts of this disclosure can be readily used as a basis for designing other structures, methods, and systems to accomplish some of the objectives of the subject matter of this disclosure. Accordingly, the claims should be deemed to include such equivalent structures, as long as they do not deviate from the spirit and scope of the subject matter of this disclosure.
[0134]
[0144] The subject matter of this disclosure has been described and illustrated in the exemplary embodiments described above, Please understand that this disclosure is provided merely as an example, and that a great many modifications to the details of embodiments of the subject matter of this disclosure may be made without departing from the spirit and scope of the subject matter of this disclosure.
Claims
1. Each of the amplifier blocks contains multiple amplifiers, The system includes a combiner for generating a high-frequency (RF) power signal for a load by combining the modulated signals output from the plurality of amplifier blocks, A power generator comprising a plurality of amplifier blocks configured to out-of-phase the modulated power signal based on at least one phase angle, each of the plurality of amplifier blocks configured to perform discrete modulation to generate the respective signals of the modulated power signal, wherein the discrete modulation includes selecting different combinations of the plurality of amplifiers and varying the RF power signal in discrete steps.
2. The power generator according to claim 1, wherein each of the plurality of amplifiers is configured to generate a fixed voltage.
3. The power generator according to claim 1, wherein each of the plurality of amplifiers is configured to operate in switch mode.
4. The power generator according to claim 1, wherein the different combinations of the plurality of amplifiers are configured to perform a predetermined sequence of discrete step changes of the RF power signal.
5. The power generator according to claim 1, wherein each of the plurality of amplifiers is configured to generate a corresponding of the modulated power signals based on at least one modulated power supply voltage.
6. The power generator according to claim 5, further comprising a discrete drain modulator configured to modulate the power supply voltage via discrete drain modulation.
7. The power generator according to claim 1, further comprising an impedance converter configured to change the output impedance of the combiner to match the impedance of the load.
8. The impedance converter includes a matching variable network, The power generator according to claim 7, wherein the matching variable network is configured to change the output impedance of the combiner in discrete steps to match the change in the impedance of the load.
9. The power generator according to claim 8, wherein the matching variable network is configured to change one or more shunt reactances in discrete steps to match the change in impedance of the load by changing the output impedance of the combiner.
10. The output impedance of the combiner changes when different combinations of the plurality of power amplifiers are selected for each of the plurality of amplifier blocks. The power generator according to claim 7, wherein the impedance converter converts the changing output impedance of the combiner to match the impedance of the load.
11. The power generator according to claim 1, wherein each of the plurality of amplifier blocks is configured to operate in common mode.
12. At least one of the plurality of amplifier blocks is configured to receive a signal. The power generator according to claim 1, wherein the signal controls the on or off state of at least one of the plurality of amplifier blocks to change the RF power signal in discrete steps.
13. The power generator according to claim 12, wherein the discrete step change of the RF power signal caused by selecting different combinations of the plurality of amplifiers is different from the discrete step change of the RF power signal caused by controlling the on state or the off state of at least one of the plurality of amplifier blocks.
14. The power generator according to claim 1, wherein the load includes a plasma generator.
15. A method for managing electricity, The steps include generating a first modulated power signal from a first amplifier block, The steps include generating a second modulated power signal from at least a second amplifier block, Steps include out-of-phase steps for the first and second modulated power signals based on the phase angle, The process includes the step of generating an RF power signal for a load based on the out-phased first and second modulated power signals, A method comprising the step of generating the first modulated power signal, which includes the step of switching between different combinations of a plurality of power amplifiers in the first amplifier block, and the step of generating the second modulated power signal, which includes the step of switching between different combinations of a plurality of power amplifiers in the second amplifier block.
16. The method according to claim 15, wherein the different combinations include different numbers of switching amplifiers in the first amplifier block and different numbers of switching amplifiers in the second amplifier block.
17. The method according to claim 16, wherein the different combinations of the plurality of amplifiers are configured to perform a predetermined sequence of discrete step changes of the RF power signal.
18. The steps include inputting at least a first supply voltage to the first amplifier block, The steps include inputting at least a second supply voltage to the second amplifier block and Furthermore, The method according to claim 15, wherein the first and second supply voltages are discrete drain modulated voltages.
19. The further step involves changing the output impedance of the combiner to match the impedance of the load, The method according to claim 15, wherein the combiner generates the RF power signal based on the out-phased first and second modulated power signals.
20. The further step includes changing the output impedance of the combiner to match it to an impedance load, The method according to claim 15, wherein the combiner generates the RF power signal based on the out-phased first and second modulated power signals and has an impedance that changes based on a change in the number of switching power amplifiers in the first amplifier block.