Plasma processing and controlling apparatus and method of controlling plasma processing apparatus

Abstract

To provide a technique capable of smoothly maintaining EtherCAT communication when a communication failure occurs at a plurality of places or a slave device is added in a communication network for control.SOLUTION: In a plasma processing / control apparatus of the present invention, a control communication network in which a master having a plurality of M communication ports and a plurality of slaves that operate by receiving a packet transmitted from the master are connected to each other has a ring-topology path formed by a plurality of slaves connected in a row from a first M communication port to a second M communication port and an M branch path formed by an M branch slave directly connected to a third M communication port, and when a communication failure is not detected by a control unit, the M branch path is closed and communication is performed in the ring-topology path. When a communication failure is detected, communication is performed in the redundant path in which the M-branch path is opened.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a plasma processing control device and a method for controlling a plasma processing device. [Background technology]

[0002] Known control systems for semiconductor manufacturing equipment such as plasma processing equipment include those that form a network by connecting multiple devices in a communicable manner to adjust processing conditions within appropriate ranges, such as the power supplied from a predetermined power source, the strength and distribution of the electric field used in sample processing, the temperature of the sample being processed, and the pressure within the processing chamber. Such control systems are composed of sensors, controllers, and actuators, and are configured such that the state of the controlled object is acquired by the sensors and input to the controller, the controller calculates control command values ​​to instruct the actuators, and the actuators act on the controlled object based on the command values. The desired control can be achieved by periodically repeating this series of processes.

[0003] In the control systems described above, multiple sensors, controllers, and actuators are connected via a network. The operation of semiconductor manufacturing equipment is controlled by exchanging data or command signals with devices via this network. To meet the requirements of control systems, such as time constraints, cost, reliability, and requirements specific to the application field, technology for such networks as control networks is evolving, and efforts are being made to increase communication capacity.

[0004] One form of such a control system is one that uses a control communication network in which a single control device controls one or more control objects via the network, and the network connection topology can be of various types, such as line, tree, star, or ring topology.

[0005] In a control communication network system, a control packet (data frame) sent from a control device is transmitted to multiple controlled devices in sequence. Examples of such control network systems include EtherCAT (registered trademark) and Ethernet (registered trademark).

[0006] In EtherCAT, a real-time Ethernet communication system, an EtherCAT master (also called a "master" or "master device"), which is a control device, and EtherCAT slaves (also called "slaves," "slave devices," "slave equipment," or "slave devices"), which are controlled devices (devices to be controlled), are connected so that they can communicate over the network to create a control communication network. The simplest network path is a line topology configuration. A line topology is a network formed by generally connecting an EtherCAT master, which is a control device, and multiple EtherCAT slaves, which are controlled devices, in a daisy chain (so-called daisy chain). In a network where the master has one port, which is an interface for Ethernet communication, a sent packet containing signal data is returned by the slave at the end of the topology and sent back to the master.

[0007] Technologies have been developed to ensure reliable control in network-based control systems. For example, a topology route has been considered that configures a network in which one or more master and slave devices are connected in a ring shape, ensuring redundancy by enabling communication even if a failure occurs at one point on the network. In a ring topology route, EtherCAT slaves generally have two or more slave communication ports, and the slave communication ports are connected sequentially to form a circular topology with the master as part of the ring communication route.

[0008] When a control network is configured using a ring topology with one master and multiple slave devices, the following conventional technologies are known to ensure the communication path or volume and prevent loss of reliability in the event of an abnormality in the communication path.

[0009] Patent Document 1 discloses a communication system with two network configurations for a single control device and multiple controlled devices. In addition to a network configuration in which a master device and slave devices are connected in a ring, the system also shows a method in which a single control device and multiple controlled devices are connected in a line topology via a separate route, with packets being returned by the terminal slave. The method in Patent Document 1 can selectively adopt either a line topology or a ring topology for communications during operation. Therefore, even if multiple failures occur on one route, communication between the master and slaves can be maintained via the other communication route.

[0010] Patent Document 2 discloses a configuration in which, in a ring topology consisting of a single control device (master) and multiple controlled devices (slaves), the master is equipped with three communication ports, and some slave devices are branched to connect the other slave devices to the communication ports of the master device for communication. This configuration essentially corresponds to a two-path loop topology, and when multiple failures occur in one of the loop paths from the branched slave devices to the master device, communication paths can be switched within the master to maintain communication on the other loop path. [Prior art documents] [Patent documents]

[0011] [Patent Document 1] Patent No. 6830608 [Patent Document 2] U.S. Patent No. 9,674,032 Summary of the Invention [Problem to be solved by the invention]

[0012] However, the following issues were found when maintaining EtherCAT communication when a failure occurred in a conventional network system. First, adding a controlled device (slave) requires significant changes to the hardware and software configuration of the master device. Communication from the master device to lower-level slave devices involves packets being sent from the master's built-in CPU via the MAC (Media Access Control) function, and then from the built-in PHY (Physical Layer), pulse transformer, and RJ-45 connector to the slave. When adding a slave device, it is necessary to change or add hardware configurations such as the master's built-in PHY and RJ-45 connector, which raises concerns about an increase in the amount of work required to add functions to the device.

[0013] Secondly, when communication becomes impossible due to a failure in a ring topology, possible causes of the failure include disconnection, disconnection, or slave device failure, but there is an issue in that the comprehensiveness of failure cases is low. For example, in both Patent Document 1 and Patent Document 2, it is difficult to maintain communication when two routes simultaneously experience one or more failures on each route.

[0014] Third, there is the issue of redundancy within grouped component groups. In plasma processing equipment (plasma etching equipment), slave devices are sometimes classified based on function and risk level, and power is supplied to each component group (slave group) from the same power source. Patent Documents 1 and 2 do not consider ensuring functional safety or a configuration that can maintain communication between each slave when equipment maintenance or a power supply system failure affects the component group. Therefore, redundancy measures are needed to ensure communication within the grouped slave group in the event of a communication failure or equipment maintenance.

[0015] To summarize the above, conventional methods have had issues with redundancy when failures occur in slave devices at multiple locations or when slave devices are added, resulting in the problem that operation of at least part of the semiconductor manufacturing equipment is stopped, resulting in a loss of operational efficiency. Therefore, an object of the present invention is to provide a technology that can smoothly maintain EtherCAT communication in a control communication network in which a master device and multiple slave devices are connected for communication, even when communication failures occur at multiple locations or when slave devices are added. [Means for solving the problem]

[0016] In order to solve the above-mentioned problems, one representative plasma processing control apparatus of the present invention is a plasma processing apparatus including a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, and a control unit that controls operation of a controlled object (slave) of the plasma processing apparatus, and the control unit includes a control communication network in which a master included in the control unit and having a plurality of communication ports (also referred to as "M communication ports") is connected via communication paths to a plurality of slaves that receive packets transmitted from the master and operate according to the packets, the control communication network including a ring topology path formed by the plurality of slaves connected in a daisy chain manner via communication paths from a first M communication port to a second M communication port, and a branch connection path (also referred to as an "M branch path") formed by one slave (also referred to as an "M branch slave") included in the plurality of slaves that is directly connected to a third M communication port via a communication path, and the control unit is configured to close the M branch path and perform communication via the ring topology path when no communication failure is detected in the control communication network, and to open the M branch path and perform communication via a redundant path when a communication failure is detected. [Effects of the Invention]

[0017] According to the present invention, in a control communication network in which a master device and multiple slave devices are communicatively connected, it is possible to smoothly maintain EtherCAT communication when communication failures occur at multiple locations or when slave devices are added. [Brief explanation of the drawings]

[0018] [Figure 1] FIG. 1 is a cross-sectional view schematically showing the configuration of a plasma processing apparatus according to an embodiment of the present invention. [Figure 2] FIG. 2 is a schematic diagram showing a conventional configuration example of a control communication network. [Figure 3] FIG. 3 is a schematic diagram showing an enlarged view of a conventional configuration example of a control communication network. [Figure 4] FIG. 4 is a schematic diagram of a situation where a communication failure occurs in the control communication network of FIG. [Figure 5] FIG. 5 is a schematic diagram showing an enlarged view of a configuration example of the control communication network according to the first embodiment. [Figure 6] FIG. 6 is a schematic diagram of a situation where a communication failure occurs in the control communication network of FIG. [Figure 7] FIG. 7 is a schematic diagram showing the structure of a packet (data frame) transmitted from the master device to the controlled device. [Figure 8] FIG. 8 is a schematic diagram comparing a conventional configuration example of a control communication network with a configuration example of the first embodiment. [Figure 9] FIG. 9 is a schematic diagram showing an example of the configuration of a control communication network according to the second embodiment. [Figure 10] FIG. 10 is a schematic diagram showing a case where a communication failure occurs in a conventional control communication network. [Figure 11] FIG. 11 is a schematic diagram showing a case where a communication failure occurs in the control communication network of embodiment 2-2. [Figure 12] FIG. 12 is a flowchart showing the operation procedure of the master device in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the present invention is not limited to this embodiment. In addition, in the description of the drawings, the same parts are designated by the same reference numerals.

[0020] <Configuration of plasma processing apparatus> Fig. 1 is a cross-sectional view schematically illustrating the configuration of a plasma processing apparatus according to an embodiment of the present invention. In particular, Fig. 1 illustrates a plasma etching apparatus that uses a microwave electric field as an electric field for generating plasma, generates ECR (Electron Cyclotron Resonance) between the microwave electric field and magnetic field, and uses the plasma to etch a substrate-like sample such as a semiconductor wafer.

[0021] A plasma etching apparatus (plasma processing apparatus) 100 shown in Fig. 1 will now be described. Plasma etching apparatus 100 has a vacuum vessel 101 equipped with a processing chamber 104 therein in which plasma is generated. Vacuum vessel 101 has a cylindrical shape and an open top, on which a dielectric window 103 (made of quartz, for example) for introducing microwaves is arranged as a cover member, forming processing chamber 104 with the inside and outside airtightly separated.

[0022] A vacuum exhaust port 110 is disposed at the bottom of the vacuum vessel 101 and is connected to a vacuum exhaust device (not shown) disposed below and connected to the vacuum vessel 101. Furthermore, a shower plate 102 that forms the ceiling surface of the processing chamber 104 is provided below the lower surface of a dielectric window 103 that forms a cover member at the top of the vacuum vessel 101.

[0023] The shower plate 102 has a plurality of gas inlet holes 102a arranged in the center, through which etching gas is introduced into the processing chamber 104. The shower plate 102 is a circular plate made of a dielectric material such as quartz.

[0024] An electric field and magnetic field generating unit 150 is disposed above the exterior of the vacuum vessel 101, generating an electric field and a magnetic field for generating the plasma 116. The electric field and magnetic field generating unit 150 is provided in the plasma etching apparatus 100 and includes the following components. Specifically, the electric field and magnetic field generating unit 150 includes a waveguide 105 disposed above the dielectric window 103, through which an electric field is transmitted to supply a high-frequency electric field of a predetermined frequency for generating the plasma 116 into the processing chamber 104. Furthermore, the electric field transmitted through the waveguide 105 is generated by oscillation in an electric field generating power supply 106. The frequency of the electric field is not particularly limited, but in this embodiment, microwaves of 2.45 GHz are used.

[0025] In addition, magnetic field generating coils 107 that form a magnetic field are arranged above the dielectric window 103 of the processing chamber 104, on the side wall of the vacuum vessel 101 that forms the cylindrical portion of the processing chamber 104, and on the outer periphery of the lower end of the waveguide 105, surrounding these.

[0026] In the above-described arrangement, the microwave electric field generated by the electric field generating power supply 106 propagates through the inside of the waveguide 105, passes through the dielectric window 103 and the shower plate 102, and is supplied from above to the processing chamber 104. Furthermore, the magnetic field generated by the magnetic field generating coil 107 interacts with the magnetic field supplied into the processing chamber 104, thereby generating electron cyclotron resonance (ECR). Then, atoms or molecules of the processing gas introduced into the processing chamber 104 through the gas introduction holes 102a of the shower plate 102 are excited and dissociated, thereby generating high-density plasma 116 in the processing chamber 104.

[0027] Furthermore, a wafer-mounting electrode (first electrode) 120 constituting a sample stage is provided in the lower part of the processing chamber 104, below the space where the plasma 116 is generated. The wafer-mounting electrode 120 has a mounting surface 120a on which a semiconductor wafer (also simply referred to as a wafer) 109, which is a sample (a processing object), is mounted.

[0028] The wafer mounting electrode 120 is disposed so that its mounting surface 120a faces the shower plate 102 or the dielectric window 103. As shown in Fig. 1, the wafer mounting electrode 120 has an upper surface 120b covered with a dielectric film 140 that forms the mounting surface 120a. Inside the dielectric film 140, there are disposed a plurality of conductive films (electrostatic attraction electrodes) 111 for electrostatic attraction, which are connected to a DC power supply 126 via a high-frequency filter 125 shown in Fig. 1.

[0029] Here, the conductive film 111 constitutes the mounting surface 120a of the sample stage and is a film-like electrode for electrostatic attraction to which DC power for electrostatically attracting a semiconductor wafer is supplied. In this case, the conductive film 111 may be bipolar, in which one of the multiple film-like electrodes is given a different polarity from the other, or may be monopolar, in which the same polarity is given, but in this embodiment it is shown as monopolar.

[0030] Further, a high frequency power supply 124 (first high frequency power supply) and a matching box 129 are arranged at a position closer to the electrostatic attraction electrode (conductive film 111) than the high frequency filter 125, and these high frequency power supply 124 and matching box 129 are connected to a conductive electrode substrate 108 having a circular or cylindrical shape and arranged inside the wafer mounting electrode 120. The high frequency power supply 124 is connected to the ground 112.

[0031] High frequency power of a predetermined frequency is supplied to the electrode substrate 108 from the high frequency power supply 124, and a bias potential is formed above the wafer 109, which is attracted and held on the upper surface of the wafer support electrode 120, during processing of the wafer 109. In other words, the sample stage has the wafer support electrode 120 to which high frequency power is supplied from the high frequency power supply 124 while the plasma 116 is being generated.

[0032] Inside the electrode base material 108, in order to remove the transferred heat and cool the wafer mounting electrode 120, multiple coolant flow paths 152 are arranged spirally or concentrically around the central axis in the vertical direction of the electrode base material 108 or the wafer mounting electrode 120. A cooling coolant for cooling the electrode base material 108 flows through these coolant flow paths 152.

[0033] Furthermore, on the outer periphery of the upper portion of the wafer-holding electrode 120, a recessed portion 120c is arranged on the outer periphery of the mounting surface 120a so as to surround the upper portion. A susceptor ring 113, which is a ring-shaped member made of a dielectric material such as quartz or ceramics such as alumina, is placed on the ring-shaped upper surface of the recess 120c, which is formed at a height lower than the mounting surface 120a of the sample stage.

[0034] When the upper surface of the susceptor ring 113 is placed on the recess 120c, the upper surface of the susceptor ring 113 is dimensioned to be higher than the mounting surface 120a of the wafer mounting electrode 120. The susceptor ring 113 is disposed on the outer periphery of the mounting surface 120a of the wafer mounting electrode (sample stage) 120, and covers the surface of the wafer mounting electrode 120. Specifically, the susceptor ring 113 is configured to cover the upper surface and cylindrical side wall surface of the recess 120c, as well as the cylindrical side wall surface of the wafer mounting electrode (sample stage) 120 below the recess 120c.

[0035] In such a plasma etching apparatus 100, an unprocessed wafer 109 is placed on the tip of the arm of a wafer transfer robot arranged in a vacuum transfer chamber that has been depressurized to the same pressure as that of a processing chamber 104 inside a vacuum transfer vessel, which is another vacuum vessel connected to the side wall of the vacuum vessel 101.

[0036] A gate, which is a passage connecting the vacuum transfer chamber and the processing chamber 104, is opened by the operation of a valve arranged in the vacuum transfer chamber, and the unprocessed wafer 109 is transferred into the processing chamber 104 while being placed on the tip of the arm of the robot.

[0037] In addition to the above, the wafer 109 transported to above the mounting surface 120a of the wafer mounting electrode 120 in the processing chamber 104 is transferred onto the lift pins by the up and down movement of the lift pins, and then placed on the mounting surface. After that, the wafer 109 is attracted to and held on the mounting surface 120a of the wafer mounting electrode 120 by the electrostatic force generated by DC power applied from the DC power supply 126.

[0038] In this state, the flow rate or speed of the etching gas is adjusted by the mass flow controller 132 and introduced into the space between the dielectric window 103 and the quartz shower plate 102, and after diffusing in this space, the gas is introduced into the processing chamber 104 through the gas inlet holes 102a of the shower plate 102. Thereafter, the gas and particles in the processing chamber 104 are exhausted through the vacuum exhaust port 110 by operating the vacuum exhaust device. The pressure inside the processing chamber 104 is adjusted to a predetermined value within a range suitable for processing the wafer 109 depending on the balance between the amount of gas supplied from the gas introduction holes 102a of the shower plate 102 and the amount of gas exhausted from the vacuum exhaust port 110.

[0039] Furthermore, while the wafer 109 is being held by suction, a heat-conductive gas such as He (helium) is supplied to the gap between the wafer 109 and the upper surface of the dielectric film 140, which is the mounting surface 120a of the wafer mounting electrode 120, from an opening (not shown) in the upper surface of the dielectric film 140, thereby promoting heat transfer between the wafer 109 and the wafer mounting electrode 120.

[0040] The temperature of the wafer supporting electrode 120 or the electrode base material 108 is adjusted in advance before the wafer 109 is placed on it by circulating a coolant adjusted to a predetermined temperature range through the coolant flow path 152 arranged in the electrode base material 108 of the wafer supporting electrode 120.

[0041] Therefore, by heat transfer between the wafer support electrode 120 or the electrode substrate 108, which have a large heat capacity, the temperature of the wafer 109 is adjusted to be close to these temperatures before processing, and even after processing begins, heat is transferred from the wafer 109 to adjust the temperature of the wafer 109.

[0042] In this state, a microwave electric field and magnetic field are supplied into the processing chamber 104, and plasma 116 is generated using the gas. When plasma 116 is formed, radio frequency (RF) bias power is supplied to the electrode substrate 108 from the radio frequency power supply 124, a bias potential is formed above the upper surface of the wafer 109, and charged particles such as ions in the plasma 116 are attracted to the upper surface of the wafer 109 in accordance with the potential difference between the potential and the potential of the plasma 116.

[0043] In addition, the etching process is performed by colliding the charged particles with the surface of the film layer to be processed of the film structure including the mask and the film layer to be processed that are placed in advance on the upper surface of the wafer 109. During the etching process, the processing gas introduced into the processing chamber 104 and particles of reaction products generated during the process are exhausted from the vacuum exhaust port 110.

[0044] In the plasma etching apparatus 100 of this embodiment, during plasma processing, a second high-frequency power is supplied from a high-frequency power supply (second high-frequency power supply) 127 to a conductor ring (second electrode) 131 arranged on the outer periphery of the sample stage via a power supply connector 161 (described later) that is provided on the sample stage and has an elastic conductive member.

[0045] In the wafer-supporting electrode 120 of this embodiment, the AC high voltage generated from the high-frequency power supply (second high-frequency power supply) 127 is introduced into a conductor ring (second electrode) 131 made of a conductive material and arranged within the susceptor ring 113 via a load matching box 128 and a load impedance variable box 130.

[0046] With the above configuration, the load impedance variable box 130, which is adjusted to a suitable impedance value, is combined with a relatively high impedance portion disposed above the susceptor ring 113 to relatively reduce the impedance value for the high frequency power from the high frequency power source 127 through the electrode substrate 108 to the outer periphery of the wafer 109. This allows the high frequency power to be effectively supplied to the outer periphery and outer periphery of the wafer 109, and reduces the concentration of the electric field at the outer periphery or outer periphery, thereby attracting charged particles such as ions in the plasma to the upper surface of the wafer 109 in the desired direction.

[0047] The high frequency power supply 127 is connected to the ground 112. In this embodiment, the frequency of the high frequency power supply 127 is preferably set to the same value as that of the high frequency power supply 124 or to a value that is a constant multiple of that.

[0048] Each part of the plasma etching apparatus 100 receives a command signal from a control unit 170 disposed outside the vacuum chamber 101, and operates based on the command signal, and the operation is controlled by the control unit 170. An operator of the plasma etching apparatus 100 can set the operation of the plasma etching apparatus 100 to be adjusted using a PC 160, which is a setting terminal communicably connected to the control unit 170. In this embodiment, the controlled devices (slaves) include the high-frequency power supplies 124 and 127, mass flow controller 132, electric field generating power supply (magnetron power supply) 106, heater power supply, autotuner, and other sensors, components, and power supplies that control the etching apparatus, and are connected to control unit 170. These controlled devices are connected to control unit 170 via wired communication paths so that they can communicate with each other, constituting a control communication network, which will be described later. Note that the configuration in which plasma processing apparatus 100 and control unit 70 are connected via a control communication network is referred to as a plasma processing control apparatus.

[0049] <Control communication network> [Example of conventional configuration] Fig. 2 is a schematic diagram showing a conventional configuration example of a control communication network. In the example of the control communication network 220 shown in Fig. 2, a plurality of slaves 210 (slaves such as slv1 to 5 are collectively referred to as slaves 210) which are devices to be controlled are connected in sequence via communication paths, and one end and the other end of the plurality of devices are connected via communication paths to communication ports provided in a master device (master) 200 arranged in the control unit 170, forming a loop path of the network in which the master 200 and each of the plurality of devices are connected in a circular (ring) shape via the communication paths (also referred to as a "cyclic topology path," "ring topology path," "loop topology path," etc.). In the example shown in Fig. 2, the master 200 is connected to two cyclic topology paths.

[0050] These multiple slaves 210 (slave groups) may be divided into groups (slave groups) according to a predetermined criterion, for example, according to function. In such a control communication network 220, command signals for controlling each of the multiple devices (controlled devices: slaves) 210 that are the control targets, and signals indicating data including information from each slave 210, are transmitted to the master 200 through communication paths. The master 200 that receives the signals transmits the signals to the CPU 171 arranged therein, and is further configured to be able to transmit and receive communication frames including slave information from the control unit 170 to the higher-level PC 160.

[0051] In this example, each loop is connected to two communication ports of the master 200, one port is connected to a slave 210 at one end of a slave group via a communication path, and the other port is connected to a slave 210 at the end of the slave group. Each loop includes the control unit 170 or the master 200, and forms a circular ring topology path. A packet containing a control signal transmitted from one port of the master 200 is transmitted sequentially to each slave 210 in the slave group of the ring topology via the communication path, and when it reaches the slave 210 at the end, it is returned to the master 200 from the other port again, and communication between the master 200 and the slave 210 is carried out.

[0052] The slaves 210 correspond to sensor components and power supplies such as the high frequency power supplies 124 and 127, the mass flow controller 132, and the electric field generating power supply (magnetron power supply) 106 in FIG. 1, and the slave devices are connected in a daisy chain.

[0053] The master 200 is equipped with a CPU 171 that handles overall IO control for communicating with the upper main PC 160, an EC master controller for EtherCAT communication built into the master, and a communication port for sending and receiving data with the slave 210 (a communication port is sometimes simply called a "port." The master and slave communication ports are sometimes distinguished by being called an M communication port, an S communication port, etc.). The EC master controller has a MAC (Media Access Control) 172, and communication with the lower slave 210 is performed via a PHY (Physical Layer), pulse transformer, and RJ-45 connector, which are also built in. The MAC 172 controls the transmission of data frames to the lower slave 210, such as defining MAC addresses and controlling transmission timing, in response to commands from the upper PC 160 and CPU 171. The communication bus is a communication standard that connects the MAC layer and PHY layer of Ethernet, and examples of such standards include MII (Media Independent Interface), GMII (Gigabit MII), and RMII (Reduced MII).

[0054] Networks in EtherCAT communication can have a variety of configurations, including line, tree, star, and ring topologies, but the most common configuration, which includes redundancy, is the ring topology, in which controlled devices (slave devices) including a main control device (master) are connected in a ring shape.

[0055] FIG. 3 is a partially enlarged schematic diagram showing an example of a conventional configuration of a control communication network. The master 200 has EtherCAT communication ports 300 and 301, which are connected to the two end devices of a slave group via communication paths, forming a ring topology with the chained slave group. Communication between the slave 210 and the master 200 on the control communication network is performed between the slave 210 and the master 200 via the CPU 171 and MAC 172 of the master 200. In this example, the EtherCAT M communication ports 300 and 301 inside the master 200 are connected via communication paths to the pair of transmitter TX and receiver RX of the S communication port of each of slave A 211 and slave E 215, which constitute one end of the slave group. Similarly, the pair of transmitter TX and receiver RX of adjacent slaves 210 are connected to each other, forming a network loop.

[0056] When a packet is transmitted from a higher-level device 160 or master 200 on the network, a data frame 230 contained in the packet is sent from communication ports 300, 301 of master 200 to slave A211 or slave E215 connected via a communication path, and is then transmitted sequentially through slaves B212, C213, and D214 connected in a daisy chain (connected in a daisy chain), and returned from either slave A211 or slave E215 to master 200. Master 200 is configured to communicate control information and status information on the fly among slaves A211 to E215 connected in a daisy chain on a single ring topology to which it is also connected.

[0057] FIG. 4 is a schematic diagram of a communication failure occurring in the control communication network of FIG. 3. In this example, we consider a case where failures occur in slave B212 and slave D214. When data frames are transmitted using the communication port 300 of the master device as the primary port, a failure occurs in slave B212, rendering communication impossible. This causes a loopback 400 in slave A211, where signals are transmitted and received between the transmitter TX and receiver RX of the S communication port, preventing the signal from propagating to the downstream slave B212. Furthermore, when the M communication port 300 receives a packet, a redundant path switch occurs, and the data frame is transmitted from the M communication port 301 to slave E215. However, in this case, as with the path from the M communication port 300, a loopback 401 occurs in slave E215.

[0058] For this reason, if a communication failure occurs in the communication path between slave B212 or slave A211, or between slave D214 or slave E215, a problem occurs in that even if slave C213 is normally powered and functioning normally, the communication path is physically interrupted and it is not possible to send a command signal to slave C213 or to send data indicating its status from slave C213 to master 200.

[0059] Each slave device (particularly an EtherCAT slave) is equipped with an EtherCAT Processing Unit (EPU) and multiple S communication ports each having a transmitter TX and a receiver RX, and each S communication port has an Auto-Forward function and a LoopBack function, and within each slave, the multiple S communication ports equipped with these functions and the EPU are connected in a circular fashion.

[0060] The EPU is an IC that conforms to the EtherCAT specification and executes communication processing transmitted inside the slave device. The EPU notifies the CPU 171 built into the master device 200 of EtherCAT information via the control communication network 220, and the CPU 171 sets the EPU built into the slave device.

[0061] The settings include Open mode, which always opens the transmitter TX of the slave device and transmits communication frames to the connected slave 210 through the communication path; Close mode, which always transfers (loopbacks) communication frames from the RX part of the slave port that received the frame to the TX part of the same port; Auto mode, which automatically switches the data frame transmission path between Open and Close states according to the setting information of the connected communication port; and Auto Close mode, which must be explicitly set and specified by the master device 200 when switching the connected communication port and re-establishing the communication path. A communication frame is transmitted from an Open port to the adjacent slave 210 (or master 200) connected to that port. Each port is basically set to Auto mode. Depending on the settings from the EPU, each port selects whether to forward the communication frame to an adjacent port or perform a LoopBack depending on the connection status with the communication path.

[0062] The Auto-Forward function operates in Open mode and forwards communication frames containing data transmitted to each port in the EtherCAT slave to the function corresponding to each port. A communication frame sent from an adjacent slave 210 in the upper row among multiple ports is forwarded from the port of the receiving slave to the EPU of that slave, and then forwarded to the port of the adjacent slave. The LoopBack function operates in Close mode, is set by the EPU based on information (command values) in a data frame transmitted from the master device 200, and has different communication paths and methods.

[0063] 4, when any of slaves A 211 to E 215 fails, the LoopBack function of the slave preceding the failed slave causes the communication frame to be looped back to the original communication path on which the preceding slave received the communication frame, and the communication frame is returned to master 200. This allows CPU 171 of master 200 to detect information contained in the returned communication frame. CPU 171 compares the information on the total number of slaves 210 and the WKC value in all communications between master 200 and connected slaves 210 during normal operation with the information contained in the returned communication frame, and the difference between the information is used to detect and identify a slave 210 that is assumed to have experienced a communication failure or failure.

[0064] [Embodiment 1] The redundant path switching configuration of the first embodiment described below is a function that, when a failure occurs in the communication network of the ring-shaped master 200-slave 210 that constitutes the ring topology, the CPU 171 detects the failure and creates a new communication path (redundant path) and switches to it.

[0065] Fig. 5 is a schematic diagram showing an enlarged portion of an example configuration of the control communication network of embodiment 1. In addition to the network configuration of the ring topology path shown in Fig. 3, branch connection paths 501 and 502 that connect the third M communication port 302 of the master 200 and the S communication port 503 of the slave C213 are provided, thereby branching the ring topology path. A branch connection path in which the master 200 and the slaves are directly connected in this manner is called an M-branch path, and the slave (slave C213) in this case is sometimes called an M-branch slave.

[0066] During operation of the plasma etching apparatus 100 shown in Fig. 1, a signal including a communication frame (data frame) 230 for controlling slaves A211 to E215 is transmitted from CPU 171 of master 200 of the control communication network shown in Fig. 5 via port 300. Data frame 230 received at one port of slave A211 is transmitted sequentially from slave A211 to slave B212 and slave C213, and is used to control each device in each slave. Furthermore, data frame 230 is transmitted from the port at which it was received in slave C213 to port 503, and is transmitted from the TX unit to port 302 of master 200 via path 501 connected thereto.

[0067] At port 302 of master 200, the data frame from slave C 213 is looped back (Close mode) and transmitted from port 302 to the RX unit of port 503 of slave C via communication path 502. At slave C, data frame 230 is transferred from port 503 to another port, and transmitted from that port to slave D 214. In the same manner, data frame 230 is transmitted sequentially through slave D 214 and slave E 215, and returned from slave E 215 to port 301 of master 200 and CPU 171. In this way, the data frame is transmitted sequentially through the network of the ring topology path from master 200 and slaves A 211 to E 215.

[0068] Fig. 6 is a schematic diagram showing when a communication failure occurs in the control communication network of Fig. 5. Similar to the example shown in Fig. 4, a case where a failure occurs in slave B 212 and slave D 214, impairing communication, will be described. In this state, when a data frame 230 is transmitted from the TX section of port 300 of master 200, a loopback 600 occurs in slave A 211, and the data frame 230 is not transmitted to slave B 212 but is returned to the RX section of port 300 of master 200. Similarly, a loopback 601 occurs when a data frame is transmitted from port 301 to slave E 215. Meanwhile, in accordance with the settings inside the master 200, port 302 is switched to an open (Open state) redundant path, and transfer of the data frame 230 becomes possible between ports 300 and 302, so the data frame 230 returned to the RX unit of port 300 is transferred 610 to port 302, and further transferred 611 from port 302 to port 301. Therefore, communication of the data frame 230 is performed between port 301 and slave E215, and between port 302 and slave C213.

[0069] In this way, in the control communication network of embodiment 1, even if a failure occurs in communication between slave B212, slave D214 or their adjacent slaves A211, C213, and E215 on the loop topology network, communication between slave A211, C213, and E215 and master 200 can be performed.

[0070] (Data Frame) Fig. 7 is a schematic diagram showing the structure of a packet (data frame) transmitted from a master device to a controlled device. As shown in Fig. 7, an EtherCAT data frame 700 is made up of an Ethernet header 701, an Ethernet data area 702 that stores EtherCAT communication data, padding 703 that is dummy data used to adjust the data length to the minimum, and a frame check sequence (FCS) 704 that checks whether the data in the Ethernet data area 702 is correct.

[0071] The Ethernet data area 702 consists of an EtherCAT header and one or more datagrams. An EtherCAT datagram is further composed of a datagram header, a data area 711 in which slave operation information and status information read and written by each slave is stored, and a working counter (WKC) 712. The WKC is an area incremented by a predetermined number by the slave each time the slave that is supposed to process the internal datagram processes it correctly.

[0072] The datagram header contains commands for read / write access to the ESC (EtherCAT Slave Controller) register of each slave 210, address (position / offset) information, etc. The ESC is a part of the slave 210 that controls the communication of each slave 210, and provides the function of taking data addressed to that slave from the datagram in the packet, writing data addressed to the master 200 into the packet, or passing the packet to the next slave, and the ESC processes the data frame 700 based on the information in the datagram header.

[0073] The address area where the ESC reads / writes data frame 700 is the data area 711, and a fixed address area is set for each ESC. Therefore, an address area is set for each slave for multiple datagrams in the Ethernet data area 702, and when the data frame 700 communicates with each slave, the slave side processes the data frame 700 on the fly by reading / writing data stored in the slave's unique address area.

[0074] Furthermore, when a read / write is executed on the built-in register in the ESC, a process of incrementing the WKC 712 is performed. When a communication failure occurs due to a failure of one of the slaves 210 on the control network, the WKC increment value is compared with the expected WKC increment value when communication with all the slaves 210 and EtherCAT is normal, and the difference between these values ​​enables the master 200 and the higher-level device 160 on the network to detect the occurrence and location of a communication failure or failure on the network.

[0075] A case will be described in which the slaves A211 to E215 of the first embodiment are further classified by function etc. and divided into slave sets. Fig. 8 is a schematic diagram comparing a conventional configuration example of a control communication network with a configuration example of the first embodiment. Fig. 8(a) shows a control communication network of the conventional technology, and Fig. 8(b) shows a control communication network according to the first embodiment. Both Figures 8(a) and 8(b) show networks that have multiple slaves connected in a daisy chain to a master 200, forming a ring topology (or loop topology). The slaves are grouped according to their functionality and reliability, and are collections of devices (slave collections) that have at least one element, with each slave collection receiving power from a different power source. In the example of Figure 8, there are three power sources (power sources A, B, and C), each electrically connected to and supplying power to slave collection 801 (slv1, 2), slave collection 802 (slv3, 4), and slave collection 803 (slv5).

[0076] In the first embodiment, the modes of the communication ports, such as the Auto-Forward function and LoopBack function, of the slave devices described above are set, and as shown in Fig. 8(b), either slv3 or slv4 belonging to a slave set 802 that receives power from power supply B is set as an M-branch slave to form an M-branch path 806, which is a branch connection path, between the master 200 and the slave device. The LoopBack setting of either slv3, which is the M-branch slave, or the port of the master 200 is set to Close (in the example of Fig. 8, slv3 is set to Close). By configuring in this way, even if power supplies A and C are cut off and a communication failure occurs between slv1, 2, and slv5, it is possible to maintain communication between the master and slv3 and 4 by opening M branch path 806 and switching to the redundant path.

[0077] When grouping the controlled devices (slave devices) in the plasma processing apparatus according to their functions, they can be divided into groups based on the power supplies, sensors, components, etc. for each function, such as the control board that controls the plasma pulse in the chamber, the mass flow controller 132 that controls the gas flow rate, and the valves that control the vacuum in the chamber processing chamber 104 and the vacuum transfer chamber.

[0078] [Embodiment 2] When adding a slave device, it is possible to form a new network topology by devising branch wiring, as will be explained below. Fig. 9 is a schematic diagram showing an example of the configuration of a control communication network of embodiment 2. Fig. 9 shows a schematic diagram of changes in the network topology and operation that accompany the addition of a slave in the network of embodiment 1 shown in Fig. 8. That is, Fig. 9(a) shows the state (step 0) before the slave is added, and shows the network according to embodiment 1 (see Fig. 8(b)).

[0079] (Embodiment 2-1) 9(b) shows a state (step 1) in which one slave (slv6) has been added to FIG. 9(a), and shows a network according to embodiment 2-1. That is, in embodiment 2-1, new slv6, which is supplied with power from the same power supply B as slv3 and 4, is added after slave 5 on the ring topology and on the communication path between master 200. slv6 belongs to the same slave set 802 as slv3 and 4. In the example shown in Figure 9(b), the ring topology route is branched by forming a communication route (branch connection route 904) that connects slv4 of the slave set 802 with the added slv6. A branch connection route that connects slaves in the same slave set, such as branch connection route 904, is called an S-branch route, and the slaves (slv4, 6) in this case are sometimes called S-branch slaves. In addition, the LoopBack setting for either the slv4 or slv6 port that connects the branch connection route 904 is set to Close (in the example of Figure 9(b), slv4 is set to Close).

[0080] (Embodiment 2-2) 9(c) shows a state in which two more slaves have been added (step 2) to FIG. 9(b), illustrating a network according to embodiment 2-2. That is, in embodiment 2-2, slv7, which is supplied with power from the same power source A as slv1 and 2, and slv8, which is supplied with power from the same power source C as slv5, are added after slave 6 on the ring topology and on the communication path between master 200. Slv7 belongs to the same slave set 801 as slv1 and 2, and slv8 belongs to the same slave set 803 as slv5. 9(c), the loop network path is branched by forming S branch paths 905 and 906 between slv2 and the added slv7 in the slave set 801, and between slv5 and the added slv8 in the slave set 803. The LoopBack setting is set to Close for either the port of the S branch slave slv2 or slv7 that connects to the S branch path 905, and for either the port of the S branch slave slv5 or slv8 that connects to the S branch path 906 (in the example of FIG. 9(c), slv2 and 5 are set to Close).

[0081] This type of control communication network configuration can accommodate changes or additions to the power supply configuration, sensors, components, etc., when the equipment specifications required by each customer differ, or when the specifications of the plasma etching equipment differ depending on the semiconductor wafer configuration or type of processing. In particular, in the plasma etching equipment shown in Figure 1, the power supply system may need to be increased or decreased due to the addition or deletion of vacuum vessel 101, which houses processing chamber 104 for etching substrate-like samples such as semiconductor wafers, or when valves, mass flow controllers, etc. need to be replaced or added in order to achieve stricter control than the specified equipment specifications.

[0082] Next, the operation when a failure occurs in embodiment 2 will be described. First, we will consider the case where the network has a conventional configuration example. Figure 10 is a schematic diagram of when a communication failure occurs in a conventional control communication network. As shown in Figure 10(a), the network configuration is the same as in embodiment 2-2, with slv6 to slv8 added and a ring topology route formed, but it differs in that branch connection routes 806, 904 to 906 are not formed. In this example, we consider a case in which power supply from power supply A is interrupted when a slave device is added. At this time, as shown in FIG. 10(b), slv1, 2, and 7 (slave set 801), which are powered by power supply A, are shut down. In a conventional network, communication failures occur not only in these slaves but also in slv3 to 6 located between them, even though there is no problem with the power supply. Therefore, in the conventional network, if a failure occurs in a device that affects the operation of some of the slaves or the power supply that supplies power to them, the functionality of the control communication network for plasma etching apparatus 100 is impaired, which can force the apparatus to shut down, resulting in a loss of operational efficiency.

[0083] In contrast to this, the case of the control communication network of embodiment 2 will be described. Figure 11 is a schematic diagram of when a communication failure occurs in the control communication network of embodiment 2-2. Figure 11(a) shows an example of the operating state in normal times (operation mode). One port of the M branch slave and one port of the S branch slave are set to Close, and data frames are transmitted and received in sequence on the ring topology route formed between the master 200 and slv1 to slv8.

[0084] FIG. 11(b) shows a case where power supply A is shut off or a power supply failure occurs, causing a failure in the operation or communication of slv1, 2, and 7, which are powered by power supply A. Immediately after power supply A is shut off, the communication ports of slv2 to 5 connected to branch connection paths 806, 904, 905, and 906 remain set to Close, and slv3 to 6 remain unable to communicate. For example, slv5 and slv8 form an S branch path 906, but immediately after power supply A is shut off, only slv8 can communicate because the communication port of slv5 is set to Close. Meanwhile, the ESC in each slave can set the Auto-Forward function and LoopBack function for each slave port, and the port operation mode can be set to Open, Close, or Auto. In this embodiment, when initially setting up EtherCAT communication, based on a command signal from the master 200 or the upper device 160, the communication port setting in each slave is selected and set to "Auto," which opens the communication port in the slave after a predetermined time (port open time) during which communication is not possible has elapsed.

[0085] For this reason, in this embodiment, the operation mode of the slv5 port is switched from Close to Open after a predetermined time during which communication is disabled has elapsed since power supply A was shut off or a failure occurred in communication of a slave receiving power from power supply A. Similarly, if the ports in the ESCs of the other slaves are set to be switched from Close to Open at the same time after the same predetermined time has elapsed, the operation mode of the slv3 to 5 ports will change from Close to Open at the same time, and communication will be maintained between slv4, 5, and 6 and the master 200, as well as slv3 and 8, which are slaves other than slv1, 2, and 7 that have had their power supply stopped and are disabled to operate and communicate, and whose ports are directly connected to the master 200 by a communication path.

[0086] Although this example only mentions the case where power supply A is shut off, similarly, even if a failure occurs in power supply B or power supply C, multiple power supply systems, or multiple individual slaves, not limited to the power supply source, communication can be maintained between master 200 and slaves other than the slave in whose operation the failure has occurred.

[0087] (flowchart) 12 is a flowchart showing the operation procedure of the master device in an embodiment of the present invention. In EtherCAT communication in operation mode, the master device recognizes and counts the total number of slaves connected to the control communication network using the working counter (WKC) in the data frame information. The total number of slaves connected to the network is defined when the system is configured.

[0088] The operation mode is a state in which there are no abnormalities in the control communication network, and in a ring topology network configuration in which at least one slave is connected in a daisy chain to the master 200 as shown in Figure 5, packets containing data frames from the master 200 are transmitted sequentially from port 300 to each slave, and are returned to port 301 of the master 200 from the communication port of the last (terminal) slave. In operation mode, for each data frame packet transmitted from the master 200 to a slave, the read / write information is updated in the register of the EPU of each slave and returned to the master 200. At this time, the master 200 detects the connected slaves and their total number as initial values ​​using the WKC (step 1201).

[0089] Next, a predetermined time (port open time) from the detection of a communication failure to the switching of the redundant path from Close to Open of the communication port is set for all the recognized slave devices (step 1202). After the settings are complete, EtherCAT communication is executed (step 1203). The data frame information looped back to the master device is acquired, and the total number of slaves and the WKC value are confirmed (step 1204). It is determined whether or not the value matches the initial value of the total number of slaves defined in 1201, and if it matches, it is determined that it is an operation mode (step 1205).

[0090] If the device is operating in the operation mode, it is determined from the total number of slaves and the WKC value whether an abnormality has occurred in the communication path (step 1206). A communication failure is determined by checking the total number of slaves and the WKC, which is counted up by read / write access to the EPU register. That is, since the WKC value is counted up by each read access and write access to the slave register in each slave, the WKC value is also compared with a predefined expected value, and if there is a difference between the expected number of slaves defined in advance on the master side and the number of slaves during actual communication, it is determined that an abnormality has occurred. If a communication failure is not detected, the process returns to step 1203 and communication continues.

[0091] If the master device detects a communication failure in step 1206, the master device transmits a redundant path switching command to each slave device, and the path switching is executed (step 1207). After the predetermined time (port open time) defined in step 1202 has elapsed since the communication abnormality was detected, it is determined from the total number of slaves in the data frame and the value of WKC whether there is a restored slave that has become able to communicate (step 1208). If there is a restored slave (see slv3 to slv6 in FIG. 11), the EtherCAT communication is terminated if it should be terminated, otherwise the process returns to step 1203 and communication is continued (step 1209).

[0092] If it is determined in step 1205 that the device is not operating in operation mode, it is possible that a break has occurred in one of the loop topologies within the network system, or that a slave device has been shut down or broken, and the communication path has not been restored (see slv1 to 7 immediately after power supply A is shut down in Figure 11(b), or slv1, 2, and 7 after the redundant path has been switched over after a predetermined time (port open time) has passed). Then, it is determined from the total number of slaves and WKC value in the data frame information returned to the master device whether all path abnormalities have been restored (step 1210). If path restoration is not confirmed, the process returns to step 1203 and communication continues.

[0093] When it is confirmed that all the paths have been restored from the abnormality, the redundant path is switched to the operation mode (step 1211).

[0094] As described above, according to the embodiment of the present invention, in a control communication network system including a general-purpose master device and multiple slave devices, adding a slave device (component) does not require adding communication ports connecting the master device and each slave device, and does not require changing the hardware configuration. Furthermore, a system configuration can be realized that can continue EtherCAT communication and communication between other components even when power is lost due to startup or malfunction of some connected components during maintenance of the plasma processing apparatus.

[0095] Furthermore, if EtherCAT communication cannot be continued, the system can switch from the normal communication mode to the redundant path, recovering EtherCAT communication or properly shutting down the system programmatically.In addition, when a slave device failure occurs, it is possible to obtain failure analysis data from the sensor component, which shortens the time it takes to return to normal operation and is expected to improve equipment reliability.

[0096] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications are possible without departing from the spirit of the present invention. For example, it is possible to abstract the slaves of slave set 802 (slv3, 4, 6) and branch connection paths 806 and 904 from the components of the control communication network in Figure 9(c), and to devise a control communication network in which only S-branch paths 905 and 906 are formed as branch connection paths in a ring topology path in which master 200 and slv1, 2, 5, 7, 8 belonging to slave set 801 or 803 are arranged in order.

[0097] The following are examples of possible embodiments of the present invention, but the present invention is not limited to these. (Aspect 1) A plasma processing and control device including a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, and a control unit that controls the operation of a controlled object (slave) of the plasma processing apparatus, a control communication network in which a master included in the control unit and having a plurality of communication ports (also referred to as "M communication ports") and a plurality of slaves that receive packets transmitted from the master and operate according to the packets are connected via communication paths; The control communication network includes: a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first of the M communication ports to a second of the M communication ports; a branch connection path (also referred to as an "M branch path") formed by one slave (also referred to as an "M branch slave") included in the plurality of slaves directly connected to the third M communication port via a communication path, A plasma processing and control device characterized in that, when the control unit does not detect a communication failure in the control communication network, the M-branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the control unit opens the M-branch path and communication is performed via a redundant path. (Aspect 2) A plasma processing and control device including a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, and a control unit that controls the operation of a controlled object (slave) of the plasma processing apparatus, a control communication network in which a master included in the control unit and having a plurality of communication ports (also referred to as "M communication ports") and a plurality of slaves that receive packets transmitted from the master and operate according to the packets are connected via communication paths; each of the plurality of slaves belongs to any one of a plurality of slave sets including at least one slave; The control communication network includes: a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first of the M communication ports to a second of the M communication ports; In a slave set to which a slave directly connected to any of the M communication ports via a communication path belongs, a branch connection path (also referred to as an "S branch path") is formed between slaves (also referred to as "S branch slaves") in the same set via the communication path, A plasma processing and control device characterized in that, when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path. (Aspect 3) 2. The plasma processing and control apparatus according to claim 1, each of the plurality of slaves belongs to any one of a plurality of slave sets including at least one slave; The control communication network includes: In a slave set to which a slave directly connected to any of the M communication ports via a communication path belongs, a branch connection path (also referred to as an "S branch path") is formed between slaves (also referred to as "S branch slaves") in the same set via the communication path, A plasma processing and control device characterized in that, when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path. (Aspect 4) 4. The plasma processing and control apparatus according to claim 2 or 3, power is supplied to each of the plurality of slaves from any one of a plurality of power sources; 1. A plasma processing control apparatus, wherein the slave set is made up of slaves that are powered by the same power supply. (Aspect 5) 5. The plasma processing and control apparatus according to claim 4, A plasma processing and control device characterized in that, among the slaves located on the ring topology path between any two S-branch slaves that does not include the master, there is a slave that belongs to a slave set that is supplied with power from a power source different from the slave set to which the any S-branch slave belongs. (Aspect 6) The plasma processing and control apparatus according to any one of aspects 1 to 5, The plasma processing and control device is characterized in that the control unit detects the presence or absence of a communication failure in the control communication network using information contained in the packet. (Aspect 7) 7. The plasma processing and control apparatus according to claim 6, The plasma processing and control device is characterized in that communication is performed through the redundant path after a predetermined time has elapsed since the communication failure was detected. (Aspect 8) 1. A method for controlling a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, comprising: A master included in a control unit and having a plurality of communication ports (also referred to as "M communication ports") is connected to a plurality of slaves that receive packets transmitted from the master and operate according to the packets via a communication path; a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first of the M communication ports to a second of the M communication ports; In a control communication network having a branch connection path (also referred to as an "M branch path") formed by one slave (also referred to as an "M branch slave") included in the plurality of slaves directly connected to the third M communication port via a communication path, A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the M-branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the M-branch path is opened and communication is performed via a redundant path. (Aspect 9) 1. A method for controlling a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, comprising: A master included in a control unit and having a plurality of communication ports (also referred to as "M communication ports") is connected to a plurality of slaves that receive packets transmitted from the master and operate according to the packets via a communication path; each of the plurality of slaves belongs to any one of a plurality of slave sets including at least one slave; a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first of the M communication ports to a second of the M communication ports; In a control communication network having a branch connection path (also called an "S branch path") formed between slaves (also called "S branch slaves") in the same slave group via a communication path, the branch connection path is formed between slaves (also called "S branch slaves") in the same slave group, and the slaves are connected directly to any of the M communication ports via a communication path. A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path. (Aspect 10) A method for controlling a plasma processing apparatus according to aspect 8, comprising: each of the plurality of slaves belongs to any one of a plurality of slave sets including at least one slave; the control communication network has a branch connection path (also referred to as an "S branch path") formed between slaves (also referred to as "S branch slaves") in the same slave group to which a slave directly connected to any of the M communication ports via a communication path belongs, and A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path. (Aspect 11) A method for controlling a plasma processing apparatus according to any one of aspects 8 to 10, comprising: The control method for a plasma processing apparatus, wherein the control unit detects the presence or absence of a communication failure in the control communication network using information contained in the packet. (Aspect 12) A method for controlling a plasma processing apparatus according to aspect 11, comprising: A method for controlling a plasma processing apparatus, characterized in that communication is performed through a redundant path after a predetermined time has elapsed since the communication failure was detected. (Aspect 13) A method for controlling a plasma processing apparatus according to aspect 12, comprising: a control method for a plasma processing apparatus, characterized in that, if a communication failure in the control communication network is not detected after the predetermined time has elapsed, communication is switched from the redundant path to communication via the ring topology path; [Explanation of symbols]

[0098] 100... plasma etching apparatus, 101... vacuum vessel, 102... shower plate, 102a...gas introduction hole, 103...dielectric window, 104...processing chamber, 105...waveguide, 106...electric field generating power supply, 107...magnetic field generating coil, 108...electrode substrate, 109...semiconductor wafer, 110...vacuum exhaust port, 111...conductive film (electrode for electrostatic attraction), 112...grounding, 113...susceptor ring, 116...plasma, 120...wafer mounting electrode (first electrode), 120a... placement surface, 120b... upper surface, 120c... recessed portion, 124...high frequency power supply (first high frequency power supply), 125...high frequency filter, 126... DC power supply, 127... High frequency power supply (second high frequency power supply), 128...load matching circuit, 129...matching circuit, 130...load impedance variable box, 131...conductor ring (second electrode), 132... mass flow controller, 140... dielectric film, 150... electric field / magnetic field forming unit, 152... refrigerant flow path, 160... PC, 161... power supply connector, 170...control unit, 171...CPU, 172...MAC, 200...master, 210 to 215...Controlled devices (slaves), 220...Control communication network, 230, 700...Data frame, 300~302...Communication port, 400, 401, 600, 601...Loopback, 610, 611...Forwarding, 701...Ethernet header, 702...Ethernet data area, 703...Padding, 704...Frame Check Sequence (FCS), 711...data area, 712...working counter (WKC), 801-803: slave set, 806, 904-906: branch connection route

Claims

1. A plasma processing and control device including a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, and a control unit that controls the operation of a controlled object (slave) of the plasma processing apparatus, a control communication network in which a master included in the control unit and having a plurality of communication ports (also referred to as "M communication ports") and a plurality of slaves that receive packets transmitted from the master and operate according to the packets are connected via communication paths; The control communication network includes: a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first M communication port to a second M communication port; a branch connection path (also referred to as an "M branch path") formed by one slave (also referred to as an "M branch slave") included in the plurality of slaves directly connected to the third M communication port via a communication path, A plasma processing / control device characterized in that when the control unit does not detect a communication failure in the control communication network, the M branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the M branch path is opened and communication is performed via a redundant path.

2. A plasma processing and control device including a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, and a control unit that controls the operation of a controlled object (slave) of the plasma processing apparatus, a control communication network in which a master included in the control unit and having a plurality of communication ports (also referred to as "M communication ports") and a plurality of slaves that receive packets transmitted from the master and operate according to the packets are connected via communication paths; each of the plurality of slaves belongs to one of a plurality of slave sets including at least one slave; The control communication network includes: a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first M communication port to a second M communication port; In a slave set to which a slave directly connected to any one of the M communication ports via a communication path belongs, there is a branch connection path (also referred to as an "S branch path") formed between slaves (also referred to as "S branch slaves") in the same set via the communication path, A plasma processing / control device characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path.

3. 2. The plasma processing and control device according to claim 1, each of the plurality of slaves belongs to one of a plurality of slave sets including at least one slave; The control communication network includes: In a slave set to which a slave directly connected to any one of the M communication ports via a communication path belongs, there is a branch connection path (also referred to as an "S branch path") formed between slaves (also referred to as "S branch slaves") in the same set via the communication path, A plasma processing / control device characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path.

4. 4. The plasma processing and control device according to claim 2 or 3, power is supplied to each of the plurality of slaves from any one of a plurality of power sources; 10. The plasma processing and control device, wherein the slave set is made up of slaves that are supplied with power from the same power supply.

5. 5. The plasma processing and control device according to claim 4, A plasma processing and control device characterized in that, between any two S branch slaves, among the slaves located on the ring topology path on the side that does not include the master, there is a slave that belongs to a slave set that is supplied with power from a power source different from the slave set to which the any S branch slave belongs.

6. The plasma processing and control device according to any one of claims 1 to 3, The plasma processing / control device is characterized in that the control unit detects the presence or absence of a communication failure in the control communication network using information contained in the packet.

7. 7. The plasma processing and control device according to claim 6, 2. The plasma processing / control device, wherein communication is performed through a redundant path after a predetermined time has elapsed since the communication failure was detected.

8. 1. A method for controlling a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, comprising: A master included in a control unit and having a plurality of communication ports (also referred to as "M communication ports") is connected to a plurality of slaves that receive packets transmitted from the master and operate in accordance with the packets via a communication path; a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first M communication port to a second M communication port; In a control communication network having a branch connection path (also referred to as an "M branch path") formed by one slave (also referred to as an "M branch slave") included in the plurality of slaves directly connected to the third M communication port via a communication path, A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the M branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the M branch path is opened and communication is performed via a redundant path.

9. 1. A method for controlling a plasma processing apparatus that processes a wafer placed in a processing chamber inside a vacuum vessel using plasma generated in the processing chamber, comprising: A master included in a control unit and having a plurality of communication ports (also referred to as "M communication ports") is connected to a plurality of slaves that receive packets transmitted from the master and operate in accordance with the packets via a communication path; each of the plurality of slaves belongs to one of a plurality of slave sets including at least one slave; a ring topology path formed by the plurality of slaves connected in a daisy chain via a communication path from a first M communication port to a second M communication port; In a control communication network having a branch connection path (also called an "S branch path") formed between slaves (also called "S branch slaves") in the same slave group via a communication path, the branch connection path is formed between slaves (also called "S branch slaves") in the same slave group, the branch connection path being a slave group including a slave directly connected to any of the M communication ports via a communication path, A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path.

10. 9. A method for controlling a plasma processing apparatus according to claim 8, comprising: each of the plurality of slaves belongs to one of a plurality of slave sets including at least one slave; the control communication network has a branch connection path (also referred to as an "S branch path") formed between slaves (also referred to as "S branch slaves") in a slave group to which a slave directly connected to any of the M communication ports via a communication path belongs, and A method for controlling a plasma processing apparatus, characterized in that when the control unit does not detect a communication failure in the control communication network, the S branch path is closed and communication is performed via the ring topology path, and when a communication failure is detected, the S branch path is opened and communication is performed via a redundant path.

11. A method for controlling a plasma processing apparatus according to any one of claims 8 to 10, The control method for a plasma processing apparatus, wherein the control unit detects the presence or absence of a communication failure in the control communication network using information contained in the packet.

12. 12. The method for controlling a plasma processing apparatus according to claim 11, A method for controlling a plasma processing apparatus, characterized in that communication is performed through a redundant path after a predetermined time has elapsed since the communication failure was detected.

13. 13. The method for controlling a plasma processing apparatus according to claim 12, a control method for a plasma processing apparatus, characterized in that, if a communication failure in the control communication network is not detected after the predetermined time has elapsed, communication is switched from the redundant path to communication via the ring topology path;

Images