JP2010014093A - Liquid discharge device with abnormality detecting function and control device therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology easily detecting abnormality of a liquid discharge device. <P>SOLUTION: This invention relates to a control device for controlling the liquid discharge device having a positive displacement pump for supplying approximately constant amount of liquid with respect to each predetermined cycle. The control device can detect an abnormality of at least one of a nozzle and the positive displacement pump by comparing a threshold value in a predetermined range determined in accordance with orifice characteristics of the nozzle and supply flow rate characteristics of the positive displacement pump with a command value of drive power to be supplied to the positive displacement pump. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for discharging and cleaning a liquid at high pressure.

  In recent years, a physical cleaning system has been realized in which a cleaning liquid is pressurized to a high pressure state of 5 MPa to 30 MPa, and the cleaning liquid, which is microdroplets from a microjet nozzle having an orifice diameter of several hundred microns, is jetted (discharged) toward a workpiece at high speed. . In this cleaning system, the inventors of the present application have found that the occurrence of a failure mode such as nozzle clogging of a nozzle having a fine orifice diameter or expansion of the nozzle diameter due to wear is significant compared to other failure modes. A technique for detecting a failure mode is also provided by the present inventor (Patent Document 1).

  However, in this detection method, detection is performed according to the release from the standard value in the normal operating state, and the system specifications (for example, the cleaning area and the kinetic energy distribution of the droplets) vary for each cleaning target. It was difficult. In particular, when the standard value is determined based on the measurement value in the normal operation state, the standard value cannot be set until the test operation is confirmed to be in the normal operation state. For this reason, conventionally, it was necessary to check the normal operation based on whether or not the cleaning operation was normally performed by performing the test operation in the entire production system including the cleaning system, and the inspection load was large. Furthermore, such a problem is not limited to the cleaning device, but is a problem common to coating devices and other liquid discharge devices in general.

JP 2004-225620 A JP 2004-245181 A

  SUMMARY An advantage of some aspects of the invention is that it provides a technique for easily detecting an abnormality of a liquid ejection apparatus.

[Application Example 1]
A control device for controlling the liquid ejection device,
The liquid ejection device includes:
A positive displacement pump for supplying a substantially constant amount of liquid every predetermined cycle;
A nozzle for discharging the supplied liquid;
A piping section for delivering the liquid from the positive displacement pump to the nozzle;
A drive unit for variably driving the positive displacement pump;
With
The controller is
An observation unit for observing a nozzle pressure which is a pressure of a liquid supplied to the nozzle;
A command value determination unit that determines a command value of drive power that the drive unit supplies to the positive displacement pump according to a difference between a preset target pressure value and the observed nozzle pressure;
At least one of the nozzle and the positive displacement pump is compared by comparing a predetermined range threshold determined based on the orifice characteristic of the nozzle and the supply flow rate characteristic of the positive displacement pump and the command value of the driving power. An anomaly detector that detects anomalies
With
The orifice characteristic represents a relationship between the flow rate of the nozzle and the amount of pressure drop,
The supply flow rate characteristic represents a flow rate of a substantially constant amount of liquid supplied for each cycle.

  The control device according to the application example 1 compares the threshold value in a predetermined range determined according to the orifice characteristic of the nozzle and the supply flow rate characteristic of the positive displacement pump with the command value of the driving power, and compares the nozzle with the positive displacement pump. It is possible to detect at least one of the abnormalities. Both the orifice characteristics of the nozzle and the supply flow rate characteristics of the positive displacement pump can be acquired as known information that does not depend on the characteristics of the entire liquid ejecting apparatus. Therefore, before acquiring the characteristic data of the entire liquid ejecting apparatus, The threshold can be determined. Thereby, the reference value of the function confirmation test before the trial operation before the actual operation of the liquid ejection device can be set.

[Application Example 2]
A control device of Application Example 1,
The drive unit includes an AC motor, and an inverter that supplies the AC motor with AC power for causing the AC motor to generate the drive power.
The command value determining unit operates the frequency of the AC power by giving a command value of the driving power to the inverter,
The drive unit operates a cycle of a drive cycle of the positive displacement pump by driving the AC motor according to an operation of the frequency of the AC power,
The abnormality detection unit is a control device that detects the abnormality by comparing a command value of the driving power with a threshold value in the predetermined range.

  The control device of Application Example 2 uses the characteristic that the AC motor as the drive power source of the positive displacement pump has a characteristic that the drive cycle can be directly operated, and is linear with respect to the discharge flow rate of the positive displacement pump. Highly reliable abnormality detection is realized by monitoring the command value of the drive frequency (drive cycle) having the characteristics.

[Application Example 3]
A control device according to application example 1 or 2,
The abnormality detection unit is a control device that determines that an abnormality has occurred in the nozzle when a command value of the driving power deviates below a threshold value in the predetermined range in a preset initial state.

  In the control device of Application Example 3, in the preset initial state, it can be determined that an abnormality has occurred in the nozzle when the command value of the driving power deviates below the predetermined range threshold. Failure detection with high utility value can be realized. The preset initial state includes, for example, a state immediately after assembly of the liquid ejection device.

[Application Example 4]
A control device according to any one of Application Examples 1 to 3,
The abnormality detection unit is a control device that determines that an abnormality has occurred in the positive displacement pump when a command value of the driving power deviates above a predetermined range threshold in a preset initial state.

  In the control device of Application Example 4, in the preset initial state, when the drive power command value deviates above a predetermined range of threshold values, the discharge amount per drive cycle is reduced. It can be determined that an abnormality has occurred.

[Application Example 5]
A control device according to any one of Application Examples 1 to 3,
The abnormality detection unit includes a supply pressure sensor that measures a supply pressure of the cleaning water to the positive displacement pump, and corrects a supply flow rate characteristic of the positive displacement pump according to a measurement value of the sensor.

  In the control device of Application Example 5, since the supply flow rate characteristic of the positive displacement pump can be corrected according to the supply pressure of the liquid to the positive displacement pump, for example, every drive cycle of the positive displacement pump as the supply pressure decreases. It is possible to improve the abnormality detection accuracy by correcting the supply flow rate characteristics in consideration of the decrease in the discharge flow rate.

  Note that the present invention can be realized not only by a control device but also by various methods such as a liquid ejection device or a control method including the control device, a liquid ejection method, and a computer program.

  According to the present invention, it is possible to easily detect abnormality of the liquid ejection device.

Next, embodiments of the present invention will be described in the following order based on examples.
A. Configuration and operation of the ultra-high pressure cleaning system of the first embodiment:
B. Configuration and operation of the ultra-high pressure cleaning system of the second embodiment:
C. Variation:

A. Configuration and operation of the ultra-high pressure cleaning system of the first embodiment:
FIG. 1 is a hardware block diagram showing a semiconductor polishing apparatus 10 having an ultra-high pressure cleaning system 100 of the first embodiment. The semiconductor polishing apparatus 10 includes an ultra-high pressure cleaning system 100 and a polishing apparatus 200. The polishing apparatus 200 includes a platen 220 in which a polishing pad 221 is mounted on a platen main body 222, and a wafer carrier 210 that transfers a semiconductor wafer S. The wafer carrier 210 has a configuration in which a backing film 212 for holding the semiconductor wafer S is provided on the carrier body 211.

  Polishing of the semiconductor wafer S is performed as follows. The platen 220 is rotated at high speed in the direction of the arrow by a motor (not shown). The wafer carrier 210 presses and polishes the semiconductor wafer S against the polishing pad 221 mounted on the platen 220 rotating at high speed with a predetermined pressure. Polishing waste generated by polishing the semiconductor wafer S is discharged from the surface of the polishing pad 221 with high-pressure cleaning water discharged from the nozzle 160 of the ultra-high pressure cleaning system 100.

  The ultra-high pressure cleaning system 100 includes a pump unit 110 that pressurizes cleaning water with electric power supplied from outside, an accumulator 120 that suppresses pulsation of pressurization by the pump unit 110, a high-pressure filter 130 that filters cleaning water, and a high pressure A valve 140, a pressure sensor 150, a nozzle 160, a high-pressure hose Hp that connects the pump unit 110 to the pressure sensor 150, and a nozzle hose Hn that connects the pressure sensor 150 to the nozzle 160 are provided.

  In the present embodiment, the pump unit 110 pressurizes the cleaning water by feedback control according to the nozzle pressure (pressure measurement value Pm) measured by the pressure sensor 150. The details of the configuration of the pump unit 110 and the control content of the ultra-high pressure cleaning system 100 will be described later.

  The accumulator 120 is equipped to suppress discharge pressure pulsation due to intermittent operation (described later) of the pump unit 110.

  The high-pressure filter 130 plays a role of preventing clogging of the nozzle 160 by filtering contamination contained in the cleaning water. The high pressure filter 130 has a pressure sensor (not shown) that measures the differential pressure between the inlet and the outlet of the high pressure filter 130, and can output the pressure loss in the high pressure filter 130. This pressure loss is used to determine when to replace the filter element (not shown).

  The high pressure valve 140 opens and closes the pipe P <b> 5 connected to the nozzle 160. The nozzle 160 has an orifice shape in which a hole having a diameter of several hundred μm is processed into a slit shape. Each of these nozzles discharges high-speed cleaning water by the orifice effect, and the discharged cleaning water becomes water droplets of several tens of μm and can clean the polishing pad 221. The specific configuration of the nozzle 160 is exemplified in Japanese Patent Application Laid-Open No. 2006-15223 filed by the applicant of the present application, and a detailed description thereof will be omitted.

  FIG. 2 is an explanatory diagram illustrating a configuration of the pump unit 110 according to the first embodiment. The pump unit 110 includes an electric plunger pump 112 and a control unit 111. In this embodiment, the electric plunger pump 112 includes a pump body 112p, an AC motor 112m that drives the pump body 112p, and an inverter 112i that supplies AC power Ap to the AC motor 112m having 6 poles.

  The pump body 112p includes a cylinder 112pc and a plunger 112pp that functions as a piston in the cylinder 112pc. The cylinder 112pc is provided with two check valves Cv1 and Cv3. Plunger 112pp is provided with one check valve Cv2.

  The plunger 112pp has a first piston surface 112ppa1 formed on the top surface and a second piston surface 112ppa2 formed on the opposite side of the first piston surface 112ppa1. In the present embodiment, the second piston surface 112ppa2 has a piston area that is about half that of the first piston surface 112ppa1. Thus, a first cylinder chamber V1 having a cylinder area equal to the area of the first piston surface 112ppa1 is formed on the upper surface of the first piston surface 112ppa1. On the other hand, a second cylinder chamber V2 having a cylinder area equal to the area of the second piston surface 112ppa2 is formed on the lower surface of the second piston surface 112ppa2. In addition, since the mounting example of the pump main body 112p is disclosed by Unexamined-Japanese-Patent No. 2004-245181 by the present applicant, description is abbreviate | omitted.

  The pump body 112p pressurizes the cleaning water by the reciprocating motion of the plunger 112pp driven by the AC motor M. Specifically, while the plunger 112pp is rising, the first piston surface 112ppa1 pressurizes the cleaning water in the cylinder 112pc upward, and the cleaning water passes through the check valve Cv2. The wash water that has passed through the check valve Cv2 is discharged as high-pressure wash water from the check valve Cv3 by an amount corresponding to the difference in cylinder area between the first cylinder chamber V1 and the second cylinder chamber V2.

  On the other hand, during the downward movement of the plunger 112pp, the second piston surface 112ppa2 pressurizes the cleaning water inside the second cylinder chamber V2 downward and discharges it as high-pressure cleaning water from the check valve Cv3. Plunger 112pp draws wash water from check valve Cv1 into first cylinder chamber V1 by simultaneously lowering first piston surface 112ppa1.

  Thus, the electric plunger pump 112 can supply a substantially constant amount of liquid for each reciprocation of the plunger 112pp. Such a pump is generally called a positive displacement pump, and can control the flow rate of pressurized washing water by manipulating the frequency of the operation cycle (reciprocating motion) of the plunger 112pp.

  The frequency of the drive cycle of plunger 112pp corresponds to the rotational speed of AC electric motor M. The number of rotations of AC motor M is determined by the frequency fr (Hz) of AC power supplied from inverter 112 i and the number of poles of AC motor M. Specifically, for the sake of simplicity, assuming that there is no slip, the rotational speed Nm (rpm) of the AC motor M is the product of the frequency fr (Hz) and 120 (the number of poles in this embodiment is 6). It is known that it can be calculated by dividing by.

  The control unit 111 stores a set pressure target value Pt (fixed value) and a control program, a storage unit 111m, a pressure target value Pt read from the storage unit 111m, and a measurement nozzle pressure (pressure measurement value Pm). Is calculated by calculating the deviation δ, the frequency command value Cf is determined according to the deviation δ and given to the inverter 112i, and the frequency command value Cf is monitored abnormally. And an abnormality detection unit 111e that executes detection processing.

  FIG. 3 is an explanatory diagram showing a control block diagram of the ultra-high pressure cleaning system 100 of the first embodiment. This control system is configured as general feedback PID control. This control system is enabled by inputting a preset pressure target value Pt to the pump unit 110, and a deviation δ between the measured value Pm of the internal pressure of the nozzle hose Hn and the pressure target value Pt is increased. It is controlled to be smaller. That is, this control system operates so that the measured value Pm of the internal pressure of the nozzle hose Hn approaches the pressure target value Pt.

  This control system is composed of components such as a control unit 111, an electric plunger pump 112, a nozzle 160, and a pressure sensor 150. Each component corresponds to each element in a hardware block diagram or the like (FIGS. 1 and 2). However, for the sake of easy understanding, the accumulator 120, the high pressure filter 130, and the high pressure valve 140 are omitted. The inventor of the present application has confirmed that the high-pressure filter 130 causes only a negligible pressure loss before the operation of the ultra-high pressure cleaning system 100 immediately after assembly.

  The pressure sensor 150 generates a pressure measurement value Pm as a voltage corresponding to the internal pressure at the inlet of the nozzle hose Hn directly connected to the nozzle 160, and the arithmetic processing that the control unit 111 (FIGS. 2 and 3) of the pump unit 110 has. It transmits to the part 111d (FIG. 2, FIG. 3).

  The arithmetic processing unit 111d performs analog-digital conversion on the pressure measurement value Pm as a voltage, and compares (subtracts) the pressure measurement value Pm as digital information with a pressure target value Pt input in advance. Is transmitted to the command value determination unit 111c (FIG. 3).

  The command value determination unit 111 c generates a frequency command value Cf according to the deviation δ and transmits it to the electric plunger pump 112. The frequency command value Cf is generated based on a control program read from the storage unit 111m and loaded into the command value determination unit 111c. The frequency command value Cf can be generated using, for example, an approximate calculation formula possessed by the control program or a map (not shown) representing the relationship between the deviation δ and the frequency command value Cf. In this embodiment, since the frequency command value Cf in the steady state of the ultra-high pressure cleaning system 100 is a problem, the control law included in the control program (FIG. 2) suppresses the steady deviation that is the deviation δ in the steady state. In order to do so, it is desirable to include an integral element.

  Abnormality detection unit 111e (FIG. 2) determines whether frequency command value Cf is within a predetermined range (described later) in accordance with, for example, confirmation that deviation δ is stable at a sufficiently small value. The abnormality detection process is executed based on the above. The monitoring of the deviation δ by the abnormality detection unit 111e is for automating the abnormality detection process and is not an essential component, and is configured to execute the abnormality detection process according to a trigger by an engineer. Also good.

  Electric plunger pump 112 drives plunger 112pp with a value at which the period of reciprocating motion of plunger 112pp substantially corresponds to frequency command value Cf. Since the electric plunger pump 112 is a positive displacement pump as described above, the operation of the cycle of the electric plunger pump 112 is substantially the operation of the discharge flow rate of the electric plunger pump 112. The cleaning water discharged from the electric plunger pump 112 is supplied to the nozzle 160.

  In this embodiment, the nozzle 160 has an orifice shape in which a hole having a diameter of several hundred μm is processed into a slit shape, and generates a pressure loss p corresponding to the supply flow rate, and is open to the atmosphere. The control of the discharge flow rate of the electric plunger pump 112 substantially corresponds to the control of the supply pressure to the nozzle 160. This pressure loss p corresponds to the nozzle pressure p.

  In this way, the present control system controls the pressure supplied to the nozzle 160 by manipulating the operating cycle of the pump, utilizing the characteristics of a positive displacement pump that supplies a substantially constant amount of liquid every predetermined cycle. Will be.

  FIG. 4 is an explanatory diagram showing the system dynamics of the electric plunger pump 112. Electric plunger pump 112 operates the frequency of AC power fr supplied by inverter 112i (in terms of mounting, the effective voltage is also operated according to the frequency), whereby the period of reciprocating motion of plunger 112pp corresponds to frequency command value Cf. The driving of the plunger 112pp is controlled so as to approach the value to be performed. Thus, in the configuration of the first embodiment, the cycle (frequency) of the drive cycle of the electric plunger pump 112 can be directly operated.

  FIG. 5 is a flowchart showing a procedure for installing the ultra-high pressure cleaning system 100 in the semiconductor polishing apparatus 10 in the first embodiment. In step S100, the engineer executes a cleaning device assembly process. The cleaning device assembling step is a step of setting the pressure of the cleaning water and the spraying range in accordance with confirmation of the size of the polishing pad 221 to be cleaned and assembling the ultra-high pressure cleaning system 100 based on this setting. .

  Specifically, in the assembly of the ultra-high pressure cleaning system 100 at the site, the kinetic energy distribution of the droplets that collide with the cleaning surface is appropriately set according to the object to be cleaned and the content of the cleaning. Since the setting of the kinetic energy distribution of the droplets can be determined from the viewpoint of the discharge pressure p of the nozzle 160 and the discharge amount (L / min), the discharge pressure p of the nozzle 160 and the number of nozzles 160 are determined. In the present embodiment, it is assumed that the number of nozzles 160 is set to one for ease of explanation.

  In step S200, the engineer executes an operation confirmation process. The operation confirmation step is a step of actually operating the ultrahigh pressure cleaning system 100 to confirm that there is no leakage of cleaning water from the system 100 and that each function is normal. Such operation check of the control system is performed, for example, by confirming that the deviation δ is within a predetermined range by changing the pressure target value Pt (dynamic characteristic check) or maintaining the pressure target value Pt for a certain period of time (static characteristic check). Can do.

  In this way, when the operation confirmation step (S200) is completed, the process proceeds to the preliminary confirmation test (S300).

  FIG. 6 is a flowchart showing the procedure of the preliminary confirmation test (S300) of the ultrahigh pressure cleaning system 100 in the first embodiment. In S310, the engineer executes an inspection reference value setting process. The inspection reference value setting step is a step of setting an inspection reference value for the preliminary confirmation test.

  The inspection reference value may be prepared based on the inspection nozzle pressure for which characteristic data has been acquired in advance. In this way, it is possible to easily realize the detection of a presumed failure mode (for example, nozzle clogging) using the characteristic data of the nozzle 160 and the electric plunger pump 112 at the nozzle pressure for inspection acquired in advance. it can. On the other hand, the inspection reference value may be set as follows according to the actual operation state. In this way, there is an advantage that the operation can be confirmed in accordance with the actual operation state.

  7 and 8 are explanatory diagrams showing calculation formulas for setting the inspection reference value of the ultrahigh pressure cleaning system 100 in the first embodiment. FIG. 7 shows a calculation formula (calculation formula F4) for determining the center value of the threshold value used as the inspection reference value. FIG. 8 shows a calculation formula (calculation formula F6) for determining an allowable range in which a deviation from the set center value is allowed.

The calculation formula of FIG. 7 is a calculation formula set by paying attention to the fact that the theoretical water supply amount q1 of the electric plunger pump 112 matches the discharge amount q2 from the nozzle 160 (calculation formula F3). The theoretical water supply q1 of the electric plunger pump 112 uses the characteristics of the positive displacement pump. The piston area (= (Dp / 2) ² × π) of each plunger (not shown), the stroke length Ls, and the number of plungers Np It can be calculated as the product of the rotational speed Nm of the AC motor M (calculation formula F1).

  On the other hand, the discharge amount q2 from the nozzle 160 can be calculated as the product of the number n of nozzles 160 (one in the present embodiment), the nozzle size z, and the square root of the nozzle pressure p (calculation formula F2). This calculation is possible because the nozzle size z is a value defined as a discharge amount at a reference pressure (for example, 1 MPa).

  In this way, the calculation formula (calculation formula F4) for determining the center value of the threshold used as the inspection reference value is set based on the calculation formulas F1 to F3. As can be seen from the calculation formula F4, the central value of the threshold value is a value that varies according to the number n of nozzles 160 and the nozzle pressure p determined in the cleaning device assembly process.

  The calculation formula (calculation formula F6) in FIG. 8 is a calculation formula for determining an allowable range in which a deviation from the center value is allowed. This permissible width is for determining that there is normal even if there is a deviation within the tolerance permitted in advance by the ultra-high pressure cleaning system 100. Note that the allowable width actually used may include an error due to a measurement error or other factors.

  In this embodiment, this calculation is set as a sum (standard expression F5) of the standard deviation of the nozzle size error Er, which is an error of the nozzle size z, by a statistical method. This calculation formula is set on the assumption that when the nozzle 160 is mass-produced, the probability distribution of the nozzle size error Er generally approximates a normal distribution. Calculation formula F6 is a compilation of calculation formula F5.

  Specifically, in the mass production of the nozzles 160, assuming that the set tolerance is 5% and 95% of all products are acceptable (within a range of ± 2σ), for example, the tolerance accumulation of 25 nozzles 160 is statistical. Above is the nozzle size error Er × 5 (= square root of 25). That is, the sum of the tolerances of the 25 nozzles 160 has a 95% probability (within ± 2σ) that the nozzle size error Er is 5 times or less.

  By such a method, even in a system in which 25 nozzles 160 are connected, the error due to the accumulation of the nozzle size error Er of the 25 nozzles 160 is 25% of the pressure loss of one nozzle 160 ( = 5% × 5). That is, even if 25 nozzles 160 are installed, it can be seen that pressure fluctuation (over 25%) due to clogging of one nozzle 160 can be detected with a probability of 95% or more. In addition, when the actual measurement value of pressure loss is included in the inspection result document of the nozzle 160, the allowable width may be reduced by using it.

  In this embodiment, since the pressure loss in the nozzle hose Hn connecting the pressure sensor 150 to the nozzle 160 is not actually measured, this pressure loss is estimated.

  FIG. 9 is data of measured values of pressure loss in the nozzle hose performed by the inventors of the present application. This measured value represents how the pressure loss in the nozzle hose changes when a predetermined pressure is constantly applied to the plurality of nozzles 160. As can be seen from this figure, the pressure loss in the nozzle hose is proportional to the hose length, and the proportionality constant is uniquely determined by the number of nozzles. In the present embodiment, it is possible to estimate the pressure loss generated from the pressure sensor 150 to the nozzle 160 according to the length of the nozzle hose and the number of nozzles with high reliability using such characteristics.

  In this embodiment, estimating the nozzle pressure p of the nozzle 160 by subtracting such pressure loss from the measured value of the pressure sensor 150 corresponds to “observation of nozzle pressure” in the claims.

  Using such calculation formulas and estimated values, the minimum threshold value Th1 (FIG. 6) and the maximum threshold value Th2 for detecting abnormalities such as clogging of the nozzle 160 in the preliminary confirmation test are calculated, and the control of the pump unit 110 is performed. Is input to the unit 111. The minimum threshold Th1 is calculated by subtracting the allowable width (calculated by the calculation formula F6) from the center value (calculated by the calculation formula F4). The maximum threshold Th2 is calculated by adding an allowable width (calculated by the calculation formula F6) to the center value (calculated by the calculation formula F4). The minimum threshold value Th1 and the maximum threshold value Th2 may be configured to be calculated by the control unit 111 according to the number of nozzles 160 and the nozzle pressure p, for example.

  Thus, when the inspection reference value setting step (S310, FIG. 6) is completed, the process proceeds to step (S320).

  In step S320, the engineer sets the pressure target value Pt of the ultra-high pressure cleaning system 100 as a fixed value to set the steady operation state. When the transient state ends and the deviation δ becomes sufficiently small and stable, it is determined that the steady operation state has been reached, and the process proceeds to step (S330).

  In step S330, the engineer starts monitoring the frequency command value Cf. The command value Cf is monitored by, for example, outputting the frequency command value Cf and the measurement value (related data (deviation δ and pressure measurement value Pm) of the frequency command value Cf) to an external measuring machine (not shown) and time-series data. Is done by recording as The engineer analyzes whether or not the ultrahigh pressure cleaning system 100 is functioning normally based on the time series data generated in this way.

  Note that the monitoring of the frequency command value Cf (step S330) may be performed only by automatic detection by the abnormality detection unit 111e, and monitoring by an engineer may be omitted, or both may be executed.

  Further, the abnormality detection unit 111e may be deleted only for monitoring by a technician. In this embodiment, the frequency command value Cf, which is output data output from the control unit 111 to the outside, is used for abnormality detection, so that a sensor for measuring the state quantity of the control target also receives data from the control unit 111. There is an advantage that a circuit for acquisition is not required and can be easily applied.

  In step S340, the engineer can make the following determination based on such analysis. Specifically, for example, when the frequency command value Cf is excessively low, that is, when it is lower than the threshold value Th1, it can be determined that the nozzle 160 may be clogged. On the other hand, for example, when the frequency command value Cf is excessively high, that is, when the frequency command value Cf is higher than the threshold Th2, it can be determined that there is a possibility that an abnormality has occurred in the electric plunger pump 112.

  Such a determination is realized based on the knowledge of the present inventor that the occurrence probability of clogging of the nozzle 160 in the control target (FIG. 3) is remarkably high immediately after the assembly of the ultra-high pressure cleaning system 100. It is. This finding is that the electric plunger pump 112 and the nozzle 160 are individually inspected before assembling at the site, and if a failure occurs after assembling, the foreign matter mixed or generated during the assembling causes the nozzle 160 to be The occurrence probability of the failure mode of clogging is remarkably high. Note that such a state after assembly corresponds to the “preset initial state” in the claims.

  As a result of such determination, when the frequency command value Cf is not maintained between the two threshold values Th1 and Th2, it is determined that there is an abnormality in the ultrahigh pressure cleaning system 100, and the process proceeds to step S100 (FIG. 5). return. On the other hand, when the frequency command value Cf is maintained between the two threshold values Th1 and Th2, it is determined that the ultra-high pressure cleaning system 100 is normal, and the process proceeds to step S400.

  In step S400, the engineer performs a system confirmation test. The system confirmation test is a test for confirming that polishing is normally performed by operating the entire semiconductor polishing apparatus 10 having the ultra-high pressure cleaning system 100. If the polishing has not been performed normally, the entire system including the ultra-high pressure cleaning system 100 is inspected. On the other hand, if the polishing has been performed normally, the process proceeds to step S500.

  In step S500, the engineer executes a standard value setting process. The standard value setting step is a step of setting a frequency command value Cf as a standard value for the ultra-high pressure cleaning system 100. The standard value is a value serving as a reference for determining whether or not the operation is normal in the actual operation of the ultra-high pressure cleaning system 100.

  Specifically, when the frequency command value Cf gradually increases with this standard value as a reference, it can be estimated that the orifice diameter has increased due to wear of the nozzle 160. On the other hand, when the frequency command value Cf suddenly decreases, it can be estimated that at least a part of the nozzle 160 is clogged. Furthermore, when the frequency command value Cf suddenly increases, it can be estimated that an abnormality has occurred in the electric plunger pump 112 as will be described later.

  Thus, when data during normal operation of the ultra-high pressure cleaning system 100 is obtained, precise abnormality detection can be performed by using this data as a standard value. When the standard value is stored in the storage unit 111m, the operation of the semiconductor polishing apparatus 10 is started (step S600).

  As described above, in the first embodiment, the characteristics of the positive displacement pump that supplies a substantially constant amount of liquid every predetermined cycle are used, and the pump operation cycle (frequency command value Cf) is within a preliminarily assumed range. Based on whether or not there is an abnormality in the controlled object (FIG. 3), it can be detected.

  Originally, what is confirmed by the functional test of the ultra-high pressure cleaning system 100 is whether or not the kinetic energy distribution of the droplets that collide with the cleaning surface is as set in advance. Therefore, since the confirmation of the discharge pressure and the discharge flow rate at the nozzle 160 is originally required, the flow rate of the cleaning water is measured according to common technical common sense.

However, the inventor of the present application has created an abnormality detection method that does not use a flow measurement sensor. This abnormality detection method was created based on the following failure mode analysis by the present inventors.
(1) Nozzle 160 failure mode: Depletion due to operation of the ultra-high pressure cleaning system 100 and clogging of the nozzle 160 due to contamination.
(2) Failure mode of the pump unit 110: Depletion of the electric plunger pump 112 resulting from the operation of the ultra high pressure cleaning system 100, and defective seats of the check valves Cv1 to Cv3.

  In these failure modes, the possibility of depletion can be eliminated immediately after the assembly of the ultra-high pressure cleaning system 100. Therefore, as the failure mode of the nozzle 160, the clogging of the nozzle 160 due to foreign matter may be taken into consideration. It can be seen that as the failure mode 110, only the seat defects of the check valves Cv1 to Cv3 need to be considered. On the other hand, clogging of the nozzle 160 excessively decreases the frequency command value Cf, and a seat failure of the check valves Cv1 to Cv3 excessively increases the frequency command value Cf. Therefore, these failure modes are the frequency command value Cf. It has also been found by the present inventor that separation can be achieved by monitoring the above.

  The inventor of the present application does not correspond to any of these assumed failure modes, that is, there is almost no possibility of failure as long as the pump operation cycle (frequency command value Cf) is within a previously assumed range. I found out. As a result, if the frequency command value Cf is within a previously assumed range, an appropriate kinetic energy distribution of the droplets is realized with a very high probability by the normal nozzle 160 and the normal plunger pump 112. I found that I can estimate it.

  On the other hand, the flow measurement of such high-pressure liquid generally requires the use of a special sensor (for example, Coriolis flowmeter) that is expensive and difficult to handle. Thus, the present invention realizes abnormality detection with a very simple configuration.

  In addition, when the standard value is determined by the standard value setting step (step S500), the flow rate measurement is substantially unnecessary, so that the above-described special sensor is almost wasted in actual operation. It has also been found to provide a significant effect that the problems inherent to the high pressure cleaning system 100 can also be solved.

B. Configuration and operation of the ultra-high pressure cleaning system of the second embodiment:
FIG. 10 is an explanatory diagram showing the configuration of the pump unit 110a included in the ultrahigh pressure cleaning system of the second embodiment. The pump unit 110a of the second embodiment is different from the pump unit 110 of the second embodiment in that an air plunger pump 112a is provided instead of the electric plunger pump 112. On the other hand, the control unit 111 of the second embodiment is different from the first embodiment only in the control program stored in the storage unit 111m, and has the same hardware.

  The configuration of the air plunger pump 112a includes the air motor 112ma instead of the AC motor M that reciprocally drives the plunger 112pp, and the electric plunger pump 112 of the first embodiment in that it includes an electropneumatic regulator 112r instead of the inverter 112i. Is different. The air motor 112ma is a motor driven by air pressure supplied from the outside.

  FIG. 11 is an explanatory diagram showing a control block diagram of the ultra-high pressure cleaning system of the second embodiment. This control system directly operates the cycle of the drive cycle of the pump body 112p by controlling the nozzle pressure p by operating the air pressure supplied to the air plunger pump 112a according to the air pressure command value Cp. This is different from the electric plunger pump 112 of the first embodiment that controls the nozzle pressure p.

  FIG. 12 is an explanatory diagram showing the system dynamics of the air plunger pump 112a. The air plunger pump 112a varies the driving torque of the air motor 112ma by manipulating the pressure of the compressed air supplied by the electropneumatic regulator 112r. Since this driving torque is used for driving the pump main body 112p, the fluctuation of the driving torque indirectly changes the driving cycle of the pump main body 112p.

  As described above, in the second embodiment, since the drive cycle of the air plunger pump 112a cannot be directly operated, it is desirable to estimate the pressure of the compressed air required for realizing the predetermined drive cycle. Will be. Specifically, for example, the pressure to be supplied to the air motor 112ma can be estimated based on the discharge pressure (effective value) of the pump unit 110a. This was devised by the inventors of the present application focusing on the fact that the water pressure of the washing water is generated by the air pressure of the compressed air and the physical dimensions are the same.

  FIG. 13 is an explanatory diagram showing a time history (discharge pressure characteristics) of the discharge pressure of the pump unit 110a of the second embodiment. As can be seen from this figure, at the time of switching between the ascending process and the descending process (the top dead center or the bottom dead center of the plunger 112pp), the discharge pressure decreases as the plunger 112pp decelerates. In addition, since the plunger 112pp is structurally asymmetric in the pump body 112p (FIGS. 2 and 10) of this embodiment, the pressure is different between the ascending process and the descending process.

  The inventor of the present application has created a method for calculating the effective value of the discharge pressure of the pump unit 110a as the time integration of the actual measurement value of the discharge pressure based on the characteristics of the pump unit 110a. According to this method, it is possible to estimate the pressure to be supplied to the air motor 112ma with respect to the pump unit 110a having any characteristics while compensating for the inherent characteristics. This estimation may be realized by a method of adding an offset value to a value when it is assumed that the discharge pressure of the pump unit 110a is constant based on such characteristics.

  Thus, in the second embodiment, the relationship between the drive cycle calculated based on the predetermined calculation formulas F4 and F6 (FIGS. 8 and 9), the drive cycle of the pump unit 110a, and the supply pressure to the air motor 112pma. Based on the above, in the configuration in which the threshold value related to the supply pressure is set, it is possible to compensate for the inherent characteristic due to the presence of the dead point of the positive displacement pump and to realize highly accurate abnormality detection.

  As described above, the inventor of the present application can also use a single pump for a pump such as the air plunger pump 112a that cannot directly operate a drive frequency (drive cycle) having linearity with respect to the discharge flow rate of the positive displacement pump. The method of applying the present invention was created by using the characteristic data.

  In order to calculate the time integral of the actual measured value of the discharge pressure representing the characteristics of the pump unit 110a, it is required to measure the time history data of the discharge pressure in each drive cycle. On the other hand, strictly speaking, the characteristics of the pump unit 110a may cause a significant fluctuation depending on the supply pressure or discharge pressure of the cleaning water to the pump unit 110a. Data is required.

  However, if the inventor of the present application obtains an actual measurement value of at least one representative state of the supply pressure or discharge pressure of the cleaning water, it corresponds to the supply pressure or discharge pressure of the cleaning water during actual operation or operation confirmation. For example, it has been found that the time history data of each state can be obtained by correcting by interpolation or extrapolation. Furthermore, the present inventor has also found that the adjustment amount of the pressure command value Cp may be directly corrected by, for example, interpolation or extrapolation.

C. Variation:
As mentioned above, although several embodiment of this invention was described, this invention is not limited to such embodiment at all, and implementation in various aspects is possible within the range which does not deviate from the summary. is there. In particular, elements other than the elements described in the independent claims in the constituent elements in each of the embodiments described above can be omitted as appropriate because they are additional elements. Furthermore, elements described in the independent claims can be appropriately replaced with elements not described in the independent claims within the scope disclosed in the present specification.

  Furthermore, in the above-described embodiments, not all of the above-described advantages and effects lead to the essential constituent elements of the present invention, and the present invention has a degree of freedom in design that can easily realize each of the above-described advantages and effects. As long as it achieves at least one advantage or effect.

C-1. First Modification: In the above-described embodiment, a plunger pump is used, but a vane pump may be used, for example. The pump that can be used in the present invention may be a positive displacement pump that supplies a substantially constant amount of liquid every predetermined cycle. However, even with a positive displacement pump, the liquid supply amount for each predetermined cycle may be significantly affected depending on the supply pressure of the liquid to the positive displacement pump. The abnormality detection accuracy may be improved by correcting the supply flow rate characteristics in consideration of the decrease in the discharge flow rate for each driving cycle of the mold pump.

C-2. Second Modification: In the above-described embodiments and modifications, an AC motor or a pneumatic motor is used. However, for example, a hydraulic motor may be used. The drive unit that can be used in the present invention may be any drive unit that can variably drive a positive displacement pump.

C-3. Third Modification: In the above-described embodiments and modifications, the present invention has been described based on the application examples to the cleaning apparatus, but it can also be applied to, for example, a coating apparatus. The present invention is generally applicable to a liquid ejection apparatus that ejects liquid.

C-4. Fourth Modified Example: In the above-described embodiments and modified examples, there is only one high-pressure valve 140 that opens and closes the supply of cleaning water to the nozzle 160. For example, each of the two or more nozzles 160 has a high pressure. One valve 140 may be provided one by one. In such a configuration, the pressure target value Pt and the frequency command value Cf are set corresponding to the case where only one high-pressure valve 140 is opened or the case where a plurality of high-pressure valves 140 are opened in order to cope with various operation modes. It is preferable to do. Further, each operation mode is preferably configured to be automatically recognized by the control unit 111.

  Further, the function confirmation may be performed by individually opening and closing the high-pressure valve 140 and measuring the frequency command value Cf and the pressure command value Cp for each of the plurality of nozzles 160 and comparing them. According to this method, highly practical abnormality detection can be realized. This is because it is very unlikely that all nozzles 160 will accidentally cause the same degree of clogging.

C-5. Fifth modified example: In the above-described embodiments and modified examples, it is configured in the range of classical control theory in a one-input one-output system. Is possible. Such an extension can be realized, for example, by providing a failure detection observer having a mathematical model representing the system dynamics of an abnormality detection target in the control unit 111. Also in this case, if the steady state is set as the orientation, the abnormality can be easily detected by monitoring the command value.

1 is a hardware block diagram showing a semiconductor polishing apparatus 10 having an ultra-high pressure cleaning system 100 according to a first embodiment. Explanatory drawing which shows the structure of the pump unit 110 of 1st Example. Explanatory drawing which shows the control block diagram of the ultra-high pressure washing system 100 of 1st Example. Explanatory drawing which shows the system dynamics of the electric plunger pump 112. FIG. The flowchart which shows the procedure which installs the ultra-high pressure washing | cleaning system 100 with respect to the semiconductor polishing apparatus 10 in 1st Example. The flowchart which shows the procedure of the preliminary | backup confirmation test (S300) of the ultra-high pressure washing system 100 in 1st Example. Explanatory drawing which shows the calculation formula for setting the test | inspection reference value of the ultra-high pressure washing system 100 in 1st Example. Explanatory drawing which shows the calculation formula for setting the test | inspection reference value of the ultra-high pressure washing system 100 in 1st Example. Data on measured values of pressure loss in the nozzle hose. Explanatory drawing which shows the structure of the pump unit 110a which the ultra-high pressure washing system of 2nd Example has. Explanatory drawing which shows the control block diagram of the ultra-high pressure washing system of 2nd Example. Explanatory drawing which shows the system dynamics of the air plunger pump 112a. Explanatory drawing which shows the time history (discharge pressure characteristic) of the discharge pressure of the pump unit 110a of 2nd Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Semiconductor polishing apparatus 100 ... Super high pressure washing system 110, 110a ... Pump unit 111 ... Control part 111c ... Command value determination part 111d ... Arithmetic processing part 111m ... Memory | storage part 112 ... Electric plunger pump 112a ... Air plunger pump 112ppa1 ... 1st 1 piston surface 112ppa2 ... 2nd piston surface 112pma ... air motor 112i ... inverter 112m ... AC motor 112p ... pump body 112r ... electropneumatic regulator 112pc ... cylinder 112pp ... plunger 120 ... accumulator 130 ... high pressure filter 140 ... high pressure valve 150 ... Pressure sensor 160 ... Nozzle 210 ... Wafer carrier 212 ... Backing film 220 ... Platen 221 ... Polishing pad

Claims (8)

A control device for controlling the liquid ejection device,
The liquid ejection device includes:
A positive displacement pump for supplying a substantially constant amount of liquid every predetermined cycle;
A nozzle for discharging the supplied liquid;
A piping section for delivering the liquid from the positive displacement pump to the nozzle;
A drive unit for variably driving the positive displacement pump;
With
The controller is
An observation unit for observing a nozzle pressure which is a pressure of a liquid supplied to the nozzle;
A command value determination unit that determines a command value of drive power that the drive unit supplies to the positive displacement pump according to a difference between a preset target pressure value and the observed nozzle pressure;
At least one of the nozzle and the positive displacement pump is compared by comparing a predetermined range threshold determined based on the orifice characteristic of the nozzle and the supply flow rate characteristic of the positive displacement pump and the command value of the driving power. An anomaly detector that detects anomalies
With
The orifice characteristic represents a relationship between the flow rate of the nozzle and the amount of pressure drop,
The supply flow rate characteristic represents a flow rate of a substantially constant amount of liquid supplied for each cycle. The control device according to claim 1,
The drive unit includes an AC motor, and an inverter that supplies the AC motor with AC power for causing the AC motor to generate the drive power.
The command value determining unit operates the frequency of the AC power by giving a command value of the driving power to the inverter,
The drive unit operates a cycle of a drive cycle of the positive displacement pump by driving the AC motor according to an operation of the frequency of the AC power,
The abnormality detection unit is a control device that detects the abnormality by comparing a command value of the driving power with a threshold value in the predetermined range. The control device according to claim 1 or 2,
The abnormality detection unit is a control device that determines that an abnormality has occurred in the nozzle when a command value of the driving power deviates below a threshold value in the predetermined range in a preset initial state. The control device according to any one of claims 1 to 3,
The abnormality detection unit is a control device that determines that an abnormality has occurred in the positive displacement pump when a command value of the driving power deviates above a predetermined range threshold in a preset initial state. The control device according to any one of claims 1 to 4,
The abnormality detection unit includes a supply pressure sensor that measures a supply pressure of the cleaning water to the positive displacement pump, and corrects a supply flow rate characteristic of the positive displacement pump according to a measurement value of the sensor. A liquid ejection device,
A positive displacement pump for supplying a substantially constant amount of liquid every predetermined cycle;
A nozzle for discharging the supplied liquid;
A piping section for delivering the liquid from the positive displacement pump to the nozzle;
A drive unit for variably driving the positive displacement pump;
A control device according to any one of claims 1 to 5;
A liquid ejection apparatus comprising: A control method for controlling a liquid ejection device,
The liquid ejection device includes:
A positive displacement pump for supplying a substantially constant amount of liquid every predetermined cycle;
A nozzle for discharging the supplied liquid;
A piping section for delivering the liquid from the positive displacement pump to the nozzle;
A drive unit for variably driving the positive displacement pump;
With
The control method is:
An observation step of observing a nozzle pressure which is a pressure of a liquid supplied to the nozzle;
A command value determining step of determining a command value of driving power supplied to the positive displacement pump by the drive unit according to a difference between a preset target pressure value and the observed nozzle pressure;
At least one of the nozzle and the positive displacement pump is compared by comparing a predetermined range threshold determined based on the orifice characteristic of the nozzle and the supply flow rate characteristic of the positive displacement pump and the command value of the driving power. An abnormality detection process for detecting abnormalities of
With
The orifice characteristic represents a relationship between the flow rate of the nozzle and the amount of pressure drop,
The control method in which the supply flow rate characteristic represents a flow rate of a substantially constant amount of liquid supplied for each cycle. A computer program for realizing the function of a control device for controlling a liquid ejection device,
The liquid ejection device includes:
A positive displacement pump for supplying a substantially constant amount of liquid every predetermined cycle;
A nozzle for discharging the supplied liquid;
A piping section for delivering the liquid from the positive displacement pump to the nozzle;
A drive unit for variably driving the positive displacement pump;
With
The computer program is
An observation function for observing a nozzle pressure that is a pressure of a liquid supplied to the nozzle;
A command value determining function for determining a command value of driving power supplied to the positive displacement pump by the drive unit according to a difference between a preset target pressure value and the observed nozzle pressure;
At least one of the nozzle and the positive displacement pump is compared by comparing a predetermined range threshold determined based on the orifice characteristic of the nozzle and the supply flow rate characteristic of the positive displacement pump and the command value of the driving power. Anomaly detection function to detect anomalies,
Comprising a program for causing the control device to realize
The orifice characteristic represents a relationship between the flow rate of the nozzle and the amount of pressure drop,
The supply flow characteristic is a computer program representing a flow rate of a substantially constant amount of liquid supplied for each cycle.

Images