Droplet ejection device

Abstract

The present invention provides a droplet dispensing device that can suppress misjudgments in the detection process for dispensing defects. [Solution] The droplet dispensing device includes a control device that performs the following steps: detecting noise based on a first signal output from a light receiving unit based on light emitted from a light source before detecting whether there is a nozzle dispensing defect; determining the amount of noise to be subtracted based on a first spectrum relating to the first signal, which shows the relationship between signal intensity and frequency, before detecting whether there is a nozzle dispensing defect; obtaining a second spectrum relating to a second signal output from a light receiving unit based on light emitted from a light source when droplets are dispensed by the nozzle; subtracting the determined amount of subtraction from the second spectrum; and detecting whether there is a nozzle dispensing defect based on the second spectrum after the subtraction.

Description

Technical Field

[0001] The present disclosure relates to a droplet ejection device used in a printing device such as an inkjet printer.

Background Art

[0002] Conventionally, a liquid ejection failure detection device including a discharge head having nozzles, a light emitting element, and a light receiving unit has been known (Patent Document 1). The light receiving unit includes a light receiving element, a current-voltage conversion circuit unit, and a high-pass filter unit. The light receiving element outputs a current corresponding to the intensity of the received light. The current-voltage conversion circuit unit converts the current output by the light receiving element into a voltage and outputs it. The high-pass filter unit removes the noise component of the voltage output by the current-voltage conversion circuit unit. For example, the high-pass filter unit removes an offset voltage generated when droplets are not passing through the beam. Also, a liquid ejection failure detection device that detects ejection failure using the average value of a plurality of peak values as a feature amount of the waveform data acquired by the light receiving unit has been known (Patent Document 2).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, since the amount of noise varies depending on the surrounding environment, the conventional method of uniformly removing noise components from the voltage, as described above, results in noise reduction effectiveness varying depending on the surrounding environment. Therefore, there was a risk of misjudgment in the dispensing failure detection process. In addition, when small droplets are separated from a large droplet that has been dispensed, if noise is included in the waveform corresponding to the small droplet, it may be possible to acquire the peak corresponding to the small droplet instead of the peak corresponding to the large droplet that should have been acquired, and thus determine that there is a dispensing failure. Therefore, there was a risk of misjudging dispensing failure.

[0005] Therefore, the present disclosure aims to provide a droplet dispensing device that can suppress misjudgments in the detection process for dispensing defects. [Means for solving the problem]

[0006] The droplet ejection device of the present disclosure comprises: an ejection head having a plurality of nozzles for ejecting droplets onto a printing medium; a light source that emits light toward a flight space through which the droplets ejected from the nozzles fly; a light receiving unit that detects the amount of light received after the light emitted from the light source has passed through the flight space and outputs a signal; and a control device. The control device performs the following processes: detecting noise based on a first signal output from the light receiving unit based on the light emitted from the light source before detecting whether there is a nozzle ejection defect; determining the amount of noise to be subtracted based on a first spectrum relating to the first signal, which shows the relationship between signal intensity and frequency, before detecting whether there is a nozzle ejection defect; obtaining a second spectrum relating to a second signal output from the light receiving unit based on the light emitted from the light source when the droplets are ejected by the nozzles; subtracting the determined amount of subtraction from the second spectrum; and detecting whether there is a nozzle ejection defect based on the second spectrum after the subtraction.

[0007] According to this disclosure, the noise reduction amount is determined before detecting whether or not there is a nozzle discharge defect, so that an appropriate reduction amount according to the surrounding environment can be subtracted from the second spectrum. As a result, the presence or absence of a discharge defect can be detected with high accuracy based on the second spectrum after reduction. Therefore, misjudgments in the discharge defect detection process can be suppressed.

[0008] The droplet ejection device of the present disclosure comprises: an ejection head having a plurality of nozzles for ejecting droplets onto a printing medium; a light source that emits light toward a flight space in which the droplets ejected from the nozzles fly; a light receiving unit that detects the amount of light received after the light emitted from the light source has passed through the flight space when the droplets are ejected by the nozzles and outputs a detection signal; and a control device. The control device performs the following processes: ejecting the droplets from the nozzles; receiving the detection signal corresponding to the ejected droplets from the light receiving unit; obtaining the number of peak waveforms in the detection signal whose maximum amplitude is greater than or equal to a threshold; identifying a target peak waveform, which is one of the target peak waveforms, based on the number of peak waveforms; obtaining the timing of the appearance of the peak in the target peak waveform; and determining whether or not there is a nozzle ejection defect based on a comparison of the voltage of the peak based on the appearance timing with the voltage of a reference peak in a predetermined reference waveform.

[0009] For example, when two peak waveforms are detected, one corresponding to a large droplet and the other to a small droplet, conventionally, the presence or absence of a nozzle discharge defect was determined based on the timing of the peak appearance of the peak waveform with the largest amplitude (i.e., the peak waveform corresponding to the large droplet). Therefore, if the peak waveform corresponding to the small droplet contains noise and its peak is larger than the peak of the peak waveform corresponding to the large droplet, there is a possibility that the presence or absence of a discharge defect will be determined based on the timing of the peak appearance of the peak waveform corresponding to the small droplet, potentially leading to a false positive. In contrast, according to this disclosure, the presence or absence of a nozzle discharge defect is determined based on a comparison between the voltage based on the peak appearance timing, depending on the number of peak waveforms, and the voltage of a predetermined reference peak. In this case, for example, if the number of peak waveforms is a predetermined number, the presence or absence of a discharge defect is determined based on a comparison between the voltage of the peak of the peak waveform corresponding to the large droplet and the voltage of the reference peak. As a result, even if the peak waveform corresponding to the small droplet contains noise as described above, the voltage of the peak that should be acquired can be obtained, allowing for highly accurate detection of the presence or absence of a discharge defect based on the acquired peak voltage, thereby suppressing false positives of discharge defects. [Effects of the Invention]

[0010] According to this disclosure, it is possible to provide a droplet dispensing device that can suppress erroneous judgments in the detection process of dispensing defects. [Brief explanation of the drawing]

[0011] [Figure 1] This is a plan view showing a droplet dispensing device according to one embodiment of the present disclosure. [Figure 2] Figure 1 is a cross-sectional view showing the configuration of the discharge head. [Figure 3] Figure 1 is a block diagram showing the components of a printing apparatus equipped with a droplet ejection device. [Figure 4] This is a block diagram showing the functional configuration of the control device in Figure 3. [Figure 5]This diagram shows how a laser beam is shone onto ink droplets being ejected from the ejection head while they are in flight. [Figure 6] Figure 6A shows an example of the first FFT spectrum, Figure 6B shows an example of the second FFT spectrum, and Figure 6C shows an example of the second FFT spectrum after noise reduction. [Figure 7] Figure 7A shows an example of the first FFT spectrum, Figure 7B shows an example of the second FFT spectrum, and Figure 7C shows an example of the second FFT spectrum after noise reduction. [Figure 8] Figure 8A shows an example of the first FFT spectrum, Figure 8B shows an example of the second FFT spectrum, and Figure 8C shows an example of the second FFT spectrum after noise reduction. [Figure 9] This diagram shows how ink droplets ejected from the ejection head pass through a laser beam. [Figure 10] This table explains the frequency range of signals on which the Fast Fourier Transform (FFT) should be performed. [Figure 11] Figure 11A shows an example of ejecting ink droplets from a nozzle at a position where it does not intersect with the laser beam emitted from the light source, while Figure 11B shows an example of ejecting ink droplets from a nozzle at a position where it intersects with the laser beam emitted from the light source. [Figure 12] This is a block diagram showing the functional configuration of a control device in one embodiment of the present disclosure. [Figure 13] This figure shows the detection waveform when small droplets are separated from a large droplet that has been dispensed. [Figure 14] Figure 14A shows the peak waveform of the detected waveform, and Figure 14B shows the integrated waveform of the detected waveform in Figure 14A. [Figure 15] This diagram illustrates the minimum voltage value in the peak waveform. [Figure 16] This figure shows the derivative waveform of the peak waveform. [Figure 17] This figure shows the integral waveform of the peak waveform. [Figure 18]FIG. 18A is a diagram showing a detection waveform, and FIG. 18B is a diagram showing an integrated waveform of the detection waveform of FIG. 18A. [Figure 19] It is a diagram for explaining a comparison process between the voltage of a peak in an addition average waveform obtained from a detection waveform and the voltage of a reference peak in a reference waveform. [Figure 20] FIGS. 20A to 20C are diagrams for explaining a process of obtaining an addition average waveform from a plurality of detection waveforms. [Figure 21] FIGS. 21A to 21E are diagrams for explaining an example of a process of specifying the appearance timing of a peak.

Embodiments for Carrying Out the Invention

[0012] Hereinafter, a droplet ejection device according to an embodiment of the present disclosure will be described with reference to the drawings. The droplet ejection device described below is merely one embodiment of the present disclosure. Therefore, the present disclosure is not limited to the following embodiments, and additions, deletions, and changes are possible without departing from the spirit of the present disclosure.

[0013] (First Embodiment) FIG. 1 is a perspective view showing a droplet ejection device 100 according to an embodiment of the present disclosure. FIG. 2 is a cross-sectional view showing the configuration of the ejection head 10 in FIG. 1. FIG. 3 is a block diagram showing the components of a printing device 1 including the droplet ejection device 100 in FIG. 1. FIG. 4 is a block diagram showing the functional configuration of the control device 20 in FIG. 3. In FIGS. 1 and 2, directions orthogonal to each other are defined as a first direction Ds, a second direction Df, and a third direction Dz. In the present embodiment, for example, the first direction Ds is the moving direction of a carriage 3 described later, the second direction Df is the conveying direction of a printing medium W described later, and the third direction Dz is the vertical direction. In the following description, Ds is referred to as the moving direction, Df is referred to as the conveying direction, and Dz is referred to as the vertical direction.

[0014] As shown in Figures 1 and 3, the droplet ejection device 100 includes, for example, two ejection heads 10 (10A, 10B), two ultraviolet irradiation devices 40 (40A, 40B), a carriage 3 on which the ejection heads 10 and ultraviolet irradiation devices 40 are mounted, a storage tank 62, a pair of guide rails 63, and a controller unit 19 including a control device 20. In this embodiment, an inkjet head that ejects ultraviolet-curable ink droplets Id (Figure 5 below) is exemplified as the ejection head 10. However, the form of the ejection head 10 is not limited. If the ejection head 10 ejects ordinary ink droplets that are not ultraviolet-curable ink droplets, the ultraviolet irradiation devices 40 are not required. In this embodiment, ink droplets Id correspond to liquid droplets.

[0015] The carriage 3 is supported by a pair of guide rails 63 that extend in the direction of movement Ds, and reciprocates along the guide rails 63 in the direction of movement Ds. As a result, the two discharge heads 10 (10A, 10B) and the two ultraviolet irradiation devices 40 (40A, 40B) reciprocate in the direction of movement Ds. The discharge heads 10 are also connected to the storage tank 62 via tubes 62a.

[0016] In this embodiment, the ejection head 10A ejects ink droplets Id of each color, collectively referred to as color inks: yellow (Y), magenta (M), cyan (C), and black (K). These four ink droplets Id are ejected onto the printing medium W supported by a platen (not shown), thereby printing a color image onto the printing medium W. On the other hand, the ejection head 10B ejects ink droplets Id of white (W) and clear (Cr). When printing a color image on a fabric, for example, as the printing medium W, ink droplets Id of white ink are ejected first as a base ink to reduce the impact on the color and material of the fabric, and then ink droplets Id of color ink are ejected on top of the ink droplets Id of white ink. In addition, ink droplets Id of clear ink are ejected when gloss is to be added or when the printed area is to be protected.

[0017] The storage tanks 62 store ink. Each storage tank 62 is provided for a different type of ink. For example, there are six storage tanks 62, each storing black, yellow, cyan, magenta, white, and clear ink.

[0018] The droplet dispensing device 1a further includes a purge section 50 and a receiving section 54. The receiving section 54 is positioned on one end of the pair of guide rails 63 in the direction of movement Ds so as to overlap with the movement area of ​​the carriage 3. The purge section 50 is positioned on the other end of the pair of guide rails 63 in the direction of movement Ds so as to overlap with the movement area of ​​the carriage 3.

[0019] The purging unit 50 includes a cap 51, a suction pump 52, and a lifting mechanism (not shown) that raises and lowers the cap 51 between a suction position and a standby position. The suction pump 52 is connected to the cap 51. In the standby position, the nozzle surface NM (Figure 2), described later, is separated from the cap 51. On the other hand, in the suction position, the nozzle surface NM is covered by the cap 51, forming a sealed space. When the suction pump 52 is driven while the cap 51 is in the suction position, the sealed space is sucked open, and a purging process is performed in which ink is discharged from the nozzle hole 121a (Figure 2). The receiving unit 54 receives ink droplets Id discharged from the discharge head 10 by the flushing process. The ink discharged by the purging process and the ink discharged by the flushing process are treated as waste liquid.

[0020] As shown in Figure 2, the discharge head 10 has a plurality of nozzles 121 that discharge ink droplets Id using ink from the storage tank 62. The discharge head 10 has a laminate of a flow channel forming body and a volume changing body. An ink flow channel is formed inside the flow channel forming body, and a plurality of nozzle holes 121a open on its lower surface, the nozzle surface NM. The volume changing body is driven to change the volume of the ink flow channel. At this time, the meniscus vibrates in the nozzle holes 121a and ink is discharged.

[0021] The flow channel forming body of the discharge head 10 is a laminate of multiple plates, and the volume changing section includes a diaphragm 155 and an actuator (piezoelectric element) 160. A common electrode 161 is connected to the diaphragm 155.

[0022] Multiple plates are stacked, from bottom to top, including a nozzle plate 146, a spacer plate 147, a first channel plate 148, a second channel plate 149, a third channel plate 150, a fourth channel plate 151, a fifth channel plate 152, a sixth channel plate 153, and a seventh channel plate 154.

[0023] Each plate has holes and grooves of various sizes formed in it. Inside the flow channel forming body formed by stacking the plates, the holes and grooves are combined to form multiple nozzles 121, multiple individual flow channels 164, and a manifold 122 as ink flow channels.

[0024] The nozzle 121 is formed by penetrating the nozzle plate 146 in the stacking direction. On the nozzle surface NM of the nozzle plate 146, multiple nozzle holes 121a, which are the tips of the nozzles 121, are arranged in the transport direction Df to form a nozzle row.

[0025] The manifold 122 supplies ink to the pressure chamber 128 to which discharge pressure is applied. The manifold 122 extends in the transport direction Df and is connected to one end of each of the multiple individual flow channels 164. In other words, the manifold 122 functions as a common flow channel for the ink. The manifold 122 is formed by through holes that penetrate the first flow channel plates 148 to the fourth flow channel plates 151 in the stacking direction, and recesses that are recessed from the lower surface of the fifth flow channel plate 152, overlapping in the stacking direction.

[0026] The nozzle plate 146 is positioned below the spacer plate 147. The spacer plate 147 is made of, for example, stainless steel. The spacer plate 147 has a recess 145 formed by, for example, half-etching, which causes a recess in the thickness direction of the spacer plate 147 from the surface facing the nozzle plate 146, thereby forming a thin-walled portion that forms a damper portion 147a and a damper space 147b. This creates a damper space 147b as a buffer space between the manifold 122 and the nozzle plate 146.

[0027] A supply port 122a is connected to the manifold 122. The supply port 122a is formed, for example, in a cylindrical shape and is provided at one end in the conveying direction Df. The manifold 122 and the supply port 122a are connected by a flow path not shown in the figure.

[0028] Each individual flow path 164 is connected to the manifold 122. The upstream end of each individual flow path 164 is connected to the manifold 122, and the downstream end is connected to the base end of the nozzle 121. Each individual flow path 164 consists of a first communication hole 125, an individual throttling passage which is a supply throttling passage 126, a second communication hole 127, a pressure chamber 128, and a descender 129, and these components are arranged in this order in the direction of ink supply.

[0029] The first communication hole 125 has its lower end connected to the upper end of the manifold 122, extends upward from the manifold 122 in the stacking direction, and penetrates the upper portion of the fifth flow path plate 152 in the stacking direction.

[0030] The upstream end of the supply throttling passage 126 is connected to the upper end of the first communication hole 125. The supply throttling passage 126 is formed, for example, by half-etching and consists of a groove recessed from the lower surface of the sixth flow channel plate 153. The second communication hole 127 has its upstream end connected to the downstream end of the supply throttling passage 126, extends upward from the supply throttling passage 126 in the stacking direction, and is formed by penetrating the sixth flow channel plate 153 in the stacking direction.

[0031] The pressure chamber 128 has its upstream end connected to the downstream end of the second communication hole 127. The pressure chamber 128 is formed by penetrating the seventh flow channel plate 154 in the stacking direction.

[0032] The descender 129 is formed by penetrating the spacer plate 147, the first flow path plate 148, the second flow path plate 149, the third flow path plate 150, the fourth flow path plate 151, the fifth flow path plate 152, and the sixth flow path plate 153 in the stacking direction. The upstream end of the descender 129 is connected to the downstream end of the pressure chamber 128, and the downstream end is connected to the base end of the nozzle 121. The nozzle 121 overlaps the descender 129 in the stacking direction, for example, and is positioned in the center of the descender 129 in the width direction.

[0033] The diaphragm 155 is laminated on the seventh flow path plate 154 and covers the upper end opening of the pressure chamber 128.

[0034] The actuator 160 includes a common electrode 161, a piezoelectric layer 162, and individual electrodes 163, arranged in this order. The common electrode 161 covers the entire surface of the diaphragm 155. The piezoelectric layer 162 covers the entire surface of the common electrode 161. Individual electrodes 163 are provided for each pressure chamber 128 and are arranged on the piezoelectric layer 162. One actuator 160 is composed of one individual electrode 163, the common electrode 161, and the portion of the piezoelectric layer 162 sandwiched between the two electrodes.

[0035] The individual electrodes 163 are electrically connected to the head driver IC 32 (Figure 3). The head driver IC 32 receives a control signal from the control device 20, generates a drive signal (voltage signal), and applies it to the individual electrodes 163. In contrast, the common electrode 161 is always maintained at ground potential. In this configuration, the active portion of the piezoelectric layer 162 expands and contracts in the planar direction together with the common electrode 161 and the individual electrodes 163 in response to the drive signal. Accordingly, the diaphragm 155 deforms in cooperation, changing the volume of the pressure chamber 128 in a direction that increases or decreases it. As a result, the discharge pressure that causes the ink droplet Id to be ejected from the nozzle 121 is applied to the pressure chamber 128.

[0036] In the ejection head 10, ink flows into the manifold 122 via the supply port 122a, then flows from the manifold 122 into the supply throttling passage 126 via the first communication hole 125, and from the supply throttling passage 126 into the pressure chamber 128 via the second communication hole 127. The ink then flows through the descender 129 and into the nozzle 121. At this point, when ejection pressure is applied to the pressure chamber 128 by the actuator 160, an ink droplet Id is ejected from the nozzle hole 121a.

[0037] The printing device 1 is, for example, an inkjet printer capable of printing on a three-dimensional object, the printing medium W. The printing device 1 may also be an inkjet printer capable of printing only on paper. As shown in Figure 3, the printing device 1 includes an operation key 4, a display unit 5, and a reading device 26. The droplet ejection device 100 includes motor driver ICs 30, 31, a head driver IC 32, a transport motor 33, a carriage motor 34, an irradiation device driver IC 35, a purge driver IC 36, a light source driver IC 37, a detection driver IC 38, a light source 65, and a detection element 67. In this embodiment, the detection element 67 corresponds to a light receiving unit.

[0038] The controller unit 19 includes, for example, a control device 20 composed of a CPU, a storage unit (ROM 21, RAM 22, EEPROM 23, HDD 24), and an ASIC 25. The control device 20 is connected to each of the above-mentioned storage units and controls the driver ICs 30-32, 35-38 and the display unit 5.

[0039] The control device 20 performs various functions by executing a predetermined processing program stored in the ROM 21. The control device 20 may be implemented as a single processor in the controller unit 19, or as multiple processors working together. The processing program is read by the reader 26 from a recording medium KB such as a computer-readable magneto-optical disk or USB flash memory and stored in the ROM 21. The RAM 22 stores image data received from an external source and the calculation results of the control device 20. The EEPROM 23 stores various initial setting information entered by the user. The HDD 24 stores various information.

[0040] The ASIC25 is connected to motor driver ICs 30 and 31, head driver IC 32, irradiation device driver IC 35, purge driver IC 36, light source driver IC 37, and detection driver IC 38. When the control device 20 receives a print job from the user, it outputs an image recording command to the ASIC25 based on the processing program. The ASIC25 controls each driver IC 30-32 and 35-38 based on the image recording command. The control device 20 moves the platen in the transport direction Df by driving the transport motor 33 with the motor driver IC 30. As a result, the printing medium W supported by the platen is transported in the transport direction Df. The control device 20 moves the carriage 3 in the movement direction Ds by driving the carriage motor 34 with the motor driver IC 31. As a result, the ejection head 10 moves in the movement direction Ds.

[0041] The control device 20 converts image data acquired from an external device into ejection data for ejecting ink droplets Id onto the printing medium W. Based on the converted ejection data, the control device 20 ejects ink droplets Id from the ejection head 10 using the head driver IC 32. The control device 20 also irradiates ultraviolet light from the light-emitting diode chip of the ultraviolet irradiation device 40 using the irradiation device driver IC 35. The control device 20 drives the purge unit 50 using the purge driver IC 36. The control device 20 controls the laser light emission operation of the light source 65 using the light source driver IC 37 and also receives signals from the detection element 67.

[0042] As shown in Figure 4, the control device 20 includes an FFT unit 20a, a noise subtraction unit 20b, and an inverse FFT unit 20c. The amount of light received from the laser light emitted from the light source 65 is detected by the detection element 67. The detection element 67 outputs a signal corresponding to the detected amount of light received as a first signal to the control device 20.

[0043] The FFT unit 20a performs a noise detection process based on a first signal output from the detection element 67 based on the laser light emitted from the light source 65. Specifically, before performing a process to detect whether or not there is a discharge defect in the nozzle 121 (discharge defect detection process), the FFT unit 20a obtains a first FFT spectrum SP1 (Figure 6A below) which is obtained by performing a Fast Fourier Transform on the first signal in the noise detection process and shows the relationship between signal intensity and frequency. In other words, the FFT unit 20a performs a process by Fast Fourier Transform to convert the first signal, which is time-series data showing the relationship between signal intensity and time, into a first FFT spectrum SP1 which is frequency component sequence data showing the relationship between signal intensity and frequency.

[0044] Furthermore, in the ejection failure detection process, the FFT unit 20a performs a Fast Fourier Transform on the second signal output from the detection element 67 based on the laser light Lz emitted from the light source 65 when an ink droplet is ejected from the nozzle 121 (i.e., the laser light Lz after it has been emitted from the light source 65 and passed through the flight space Sh) to detect whether or not there is an ejection failure of the nozzle 121, thereby obtaining a second FFT spectrum SP2 (Figure 6B below). In this embodiment, the first FFT spectrum SP1 corresponds to the first spectrum, and the second FFT spectrum SP2 corresponds to the second spectrum.

[0045] The noise subtraction unit 20b determines the amount of noise to be subtracted from the second FFT spectrum SP2 based on the first FFT spectrum described above. The noise subtraction unit 20b then subtracts the determined amount from the second FFT spectrum SP2 and outputs the resulting second FFT spectrum SP2a (Figure 6C below) to the inverse FFT unit 20c. Details of the processing by the noise subtraction unit 20b will be described later.

[0046] The inverse FFT unit 20c performs an inverse fast Fourier transform on the second FFT spectrum SP2a after the subtraction. That is, the inverse FFT unit 20c performs a process to convert the second FFT spectrum SP2a after the subtraction, which is frequency component sequence data showing the relationship between signal intensity and frequency, into a signal, which is time series data showing the relationship between signal intensity and time, using the inverse fast Fourier transform. The control device 20 then determines whether or not there is a discharge defect based on the waveform after the conversion by the inverse FFT unit 20c.

[0047] Next, Figure 5 shows the process of irradiating a flying ink droplet Id, ejected from the ejection head 10, with laser light Lz.

[0048] As shown in Figure 5, the light source 65 emits laser light Lz in a specific wavelength range. Specifically, the light source 65 irradiates the laser light Lz towards the flight space Sh in which the ink droplet Id ejected from the ejection head 10 flies. In this embodiment, the laser light Lz corresponds to light.

[0049] The light source 65 is positioned on one side of the position of the discharge head 10 with reference to the optical axis DL of the optical axis La of the laser light Lz emitted from the light source 65. Examples of the light source 65 include light-emitting diodes (LEDs) and semiconductor lasers (LDs). The light source 65 is housed in a box-shaped light source housing 65a. The light source housing 65a has a slit 65b on the side facing the direction of emission of the laser light Lz emitted from the light source 65. A lens 65c is positioned inside the light source housing 65a so as to cover the slit 65b from the inside of the light source housing 65a. One or more lenses may be provided in addition to lens 65c. The light source 65 is supported by a frame 71 that extends in the optical axis DL of the laser light Lz.

[0050] On the other hand, the detection element 67 is positioned on the other side of the optical axis direction DL, with reference to the position of the ejection head 10. The detection element 67 generates a current based on the received laser light Lz and outputs the current signal to a current-voltage conversion circuit (not shown). The current-voltage conversion circuit converts the current signal from the detection element 67 into a voltage signal and outputs the voltage signal to an amplification circuit (not shown). The amplification circuit amplifies the voltage signal from the current-voltage conversion circuit and outputs it to the FFT unit 20a described above. The detection element 67 is supported by the frame 71, similar to the light source 65.

[0051] In the above configuration, the laser beam Lz emitted from the light source 65 passes through the lens 65c and is then ejected from the ejection head 10, irradiating the ink droplet Id in flight in the flight space Sh. The detection element 67 detects the amount of light received after the laser beam Lz emitted from the light source 65 has passed through the flight space Sh. The control device 20 uses the first signal output from the detection element 67 to perform ejection failure detection processing. In this ejection failure detection processing, ejection failures such as abnormal velocity of ink droplet Id, abnormal volume of ink droplet Id, and distortion of ink droplet Id are detected. Distortion means that the ink droplet Id flies in a direction different from the normal flight direction.

[0052] Next, the processing by the noise reduction unit 20b will be explained in detail. Figure 6A shows an example of the first FFT spectrum SP1, Figure 6B shows an example of the second FFT spectrum SP2, and Figure 6C shows an example of the second FFT spectrum SP2a after noise reduction. Furthermore, Figure 7A shows another example of the first FFT spectrum SP1, Figure 7B shows another example of the second FFT spectrum SP2, and Figure 7C shows another example of the second FFT spectrum SP2a after noise reduction. In addition, Figure 8A shows yet another example of the first FFT spectrum SP1, Figure 8B shows yet another example of the second FFT spectrum SP2, and Figure 8C shows yet another example of the second FFT spectrum SP2a after noise reduction.

[0053] First, the control device 20 emits laser light Lz from the light source 65 before the ejection failure detection process, that is, without ejecting ink droplets from the nozzle 121. At this time, the FFT unit 20a obtains a first FFT spectrum SP1 as shown in Figure 6A from the first signal output from the detection element 67. Then, the noise subtraction unit 20b determines the first FFT spectrum SP1 obtained by the FFT unit 20a as the subtraction amount Sa, which is the amount of noise to be subtracted from the second FFT spectrum SP2.

[0054] Next, the control device 20 emits laser light Lz from the light source 65 and ejects ink droplets from the nozzle 121. At this time, the FFT unit 20a obtains a second FFT spectrum SP2 as shown in Figure 6B from the second signal output from the detection element 67 based on the laser light Lz emitted from the light source 65 and after passing through the flight space Sh. Then, the noise subtraction unit 20b subtracts the determined subtraction amount Sa from the second FFT spectrum SP2 to obtain a subtracted second FFT spectrum SP2a as shown in Figure 6C.

[0055] Alternatively, the second FFT spectrum SP2a after subtraction may be obtained as follows.

[0056] First, the control device 20 emits laser light Lz from the light source 65 before the ejection failure detection process, that is, without ejecting ink droplets from the nozzle 121. At this time, similar to Figure 6A, the FFT unit 20a obtains a first FFT spectrum SP1 as shown in Figure 7A from the first signal output from the detection element 67. Then, the noise subtraction unit 20b determines a predetermined constant value obtained from the first FFT spectrum SP1 obtained by the FFT unit 20a as the subtraction amount Sa. In this case, the noise subtraction unit 20b can, for example, adopt the maximum value of the signal intensity in the first FFT spectrum SP1 as the subtraction amount Sa.

[0057] Next, the control device 20 emits laser light Lz from the light source 65 and ejects ink droplets from the nozzle 121. At this time, similar to Figure 6B, the FFT unit 20a obtains a second FFT spectrum SP2 as shown in Figure 7B from the second signal output from the detection element 67 based on the laser light Lz emitted from the light source 65 and after passing through the flight space Sh. Then, the noise subtraction unit 20b subtracts the above subtraction amount Sa from the second FFT spectrum SP2 over the entire frequency of the second FFT spectrum SP2 to obtain a subtracted second FFT spectrum SP2a as shown in Figure 7C.

[0058] Alternatively, the second FFT spectrum SP2a after subtraction may be obtained as follows.

[0059] First, the control device 20 emits laser light Lz from the light source 65 before the ejection failure detection process, that is, without ejecting ink droplets from the nozzle 121. At this time, similar to Figures 6A and 7A, the FFT unit 20a obtains a first FFT spectrum SP1 as shown in Figure 8A from the first signal output from the detection element 67. Then, the noise subtraction unit 20b determines the subtraction amount Sa for each frequency region by using the reference value obtained from the first FFT spectrum SP1 for each frequency region.

[0060] The frequency ranges may be set to correspond to the peak values ​​in the waveform of the first FFT spectrum SP1. In other words, the number of frequency ranges is the same as the number of peak values. In the example in Figure 8A, four frequency ranges Fr1, Fr2, Fr3, and Fr4 are set as frequency ranges, each corresponding to one of the four peak values ​​in the waveform of the first FFT spectrum SP1. Each of the frequency ranges Fr1, Fr2, Fr3, and Fr4 can be a certain range that includes the frequency corresponding to each peak value. The noise subtraction unit 20b can, for example, adopt the maximum value for each of the frequency ranges Fr1, Fr2, Fr3, and Fr4 in the first FFT spectrum SP1 as the subtraction amount Sa.

[0061] Next, the control device 20 emits laser light Lz from the light source 65 and ejects ink droplets from the nozzle 121. At this time, similar to Figures 6B and 7B, the FFT unit 20a obtains a second FFT spectrum SP2 as shown in Figure 8B from the second signal output from the detection element 67 based on the laser light Lz emitted from the light source 65 and after passing through the flight space Sh. Then, the noise subtraction unit 20b subtracts the respective subtraction amounts Sa determined for each frequency region from each portion of the second FFT spectrum SP2 corresponding to each frequency region, thereby obtaining the subtracted second FFT spectrum SP2a as shown in Figure 8C. Note that a certain value of subtraction amount Sa is also subtracted from the portion of the second FFT spectrum SP2 other than the above frequency regions. Thus, the subtracted second FFT spectrum SP2a is obtained.

[0062] Next, a method for reducing the amount of computation performed by the control device 20 in this embodiment will be described. Figure 9 shows how an ink droplet Id ejected from the ejection head 10 passes through the laser beam Lz.

[0063] In this embodiment, the FFT unit 20a performs a Fast Fourier Transform on a portion of the first and second signals output from the detection element 67 that falls within a predetermined frequency range. The control device 20 obtains the predetermined frequency range based on the diameter D of the laser beam Lz, the velocity v of the ink droplets Id ejected from the nozzle 121, the ejection interval of the ink droplets Id, and the number of ink droplets Id ejected. This will be explained in detail below.

[0064] Figure 9 shows the flight positions of, for example, three ink droplets Id in succession from a single nozzle over time. When, for example, three ink droplets Id are successively ejected from each nozzle in this manner, we focus on the point in time when the signal intensity appears in the waveform of the second signal output from the detection element 67 due to the ink droplets Id passing through the laser beam Lz.

[0065] More specifically, the transit time T1, which is the time from when the first ink droplet Id (i.e., the first ink droplet Id) enters the laser beam Lz until it leaves the laser beam Lz, can be calculated by the diameter D of the laser beam Lz / the velocity v of the ink droplet. Note that the diameter D may be the full width at half maximum of the beam, or the beam diameter at which the illuminance exceeds a certain value.

[0066] Furthermore, the transit time T2, which is the time from when the first ink droplet Id entering the laser beam Lz leaves the laser beam Lz until the last ejected ink droplet (i.e., the third ink droplet Id) leaves the laser beam Lz, can be calculated by the ink droplet ejection interval T3 × (N-1). The above N is the number of ink droplets Id entering the laser beam Lz during the transit time T2, and can be calculated by D / (v × T3).

[0067] Here, the frequency fc of the second signal can be calculated by 1 / Tall. Tall is the sum of the transit times T1 and T2. Therefore, Tall is expressed as D / v + T3 × (N-1). And, as stated above, N = D / (v × T3), so Tall is expressed as D / v + T3 × [D / (v × T3) - 1]. Thus, the control device 20 calculates the frequency fc by 1 / {D / v + T3 × [D / (v × T3) - 1]}.

[0068] Figure 10 is a table illustrating the frequency range of the signal on which the Fast Fourier Transform should be performed by the FFT unit 20a. In Figure 10, as an example, the diameter D of the laser beam Lz is set to 0.5 mm, and the ejection interval of the ink droplet Id is set to 0.01 ms. Figure 10 then shows the frequency fc calculated using the above formula, corresponding to each of the varying ink droplet Id velocities, assuming that the velocity of the ejected ink droplet Id varies by, for example, about 20%. As can be seen from Figure 10, the frequency fc range is 9 kHz to 14 kHz. However, if a margin of, for example, ±2 kHz is provided in the above range of frequency fc, the frequency fc range becomes 7 kHz to 16 kHz. Such a frequency range is determined for each nozzle.

[0069] The FFT unit 20a performs a Fast Fourier Transform on the portion of the signal from the detection element 67 within the above frequency range. The noise subtraction unit 20b subtracts the predetermined subtraction amount Sa from the portion of the second FFT spectrum SP2 corresponding to the above frequency range.

[0070] Next, Figure 11A shows an example of ejecting ink droplets Id from the nozzle 121 at a position where it does not intersect with the laser beam Lz emitted from the light source 65, and Figure 11B shows an example of ejecting ink droplets Id from the nozzle 121 at a position where it intersects with the laser beam Lz emitted from the light source 65. Figures 11A and 11B show a top view of the ejection head 10 as seen from below.

[0071] In the above embodiment, the control device 20 performs noise detection processing before the ejection failure detection processing, that is, when no ink droplet Id is ejected from the nozzle 121. In contrast, in this embodiment, the noise detection processing is performed with ink droplet Id ejected from the nozzle 121. Specifically, in the noise detection processing, the control device 20 ejects ink droplet Id from the nozzle 121 at a position that does not intersect with the laser beam Lz emitted from the light source 65, as shown in Figure 11A, and detects noise based on a first signal output from the detection element 67 based on the laser beam Lz emitted from the light source 65.

[0072] In contrast, in the ejection failure detection process, as shown in Figure 11B, when an ink droplet Id is ejected from the nozzle 121 at a position where it intersects with the laser beam Lz emitted from the light source 65, the presence or absence of an ejection failure in the nozzle 121 is detected based on a second signal output from the detection element 67 based on the said laser beam Lz.

[0073] As explained above, the droplet dispensing device 100 determines the noise subtraction amount Sa before detecting whether or not there is a dispensing defect in the nozzle 121. Therefore, an appropriate subtraction amount Sa according to the surrounding environment can be subtracted from the second FFT spectrum SP2. This allows for highly accurate detection of the presence or absence of a dispensing defect based on the second FFT spectrum SP2a after subtraction. Thus, misjudgments in the dispensing defect detection process can be suppressed.

[0074] Furthermore, in this embodiment, the noise subtraction unit 20b determines the first FFT spectrum SP1 obtained by the FFT unit 20a as the subtraction amount Sa to be subtracted from the second FFT spectrum SP2. This makes it possible to subtract an appropriate amount Sa according to the surrounding environment.

[0075] Furthermore, in this embodiment, the noise subtraction unit 20b may determine the subtraction amount Sa as a predetermined constant value obtained from the first FFT spectrum SP1 obtained by the FFT unit 20a. In this case, since one predetermined value, such as the maximum value of the spectrum, is used as a reference, an appropriate subtraction amount Sa according to the surrounding environment can be easily subtracted.

[0076] Furthermore, in this embodiment, the noise subtraction unit 20b may determine the subtraction amount Sa for each frequency area based on a reference value obtained from the first FFT spectrum SP1. In this case, an appropriate subtraction amount Sa can be subtracted according to the surrounding environment.

[0077] Furthermore, in this embodiment, the control device 20 ejects an ink droplet Id from the nozzle 121 at a position where it does not intersect with the laser beam Lz emitted from the light source 65 during the noise detection process, and detects noise based on the first signal output from the detection element 67 based on the laser beam Lz. In this case, noise caused by the ejection operation of the ink droplet Id can be subtracted, thereby enabling more accurate determination of whether or not there is an ejection defect.

[0078] Furthermore, in this embodiment, the noise subtraction unit 20b subtracts a predetermined subtraction amount Sa from the portion of the second FFT spectrum SP2 corresponding to the predetermined frequency range. This reduces the computational load on the control device 20 during the discharge defect detection process.

[0079] Furthermore, in this embodiment, the control device 20 obtains a predetermined range of frequencies based on the diameter D of the laser beam Lz, the velocity v of the ink droplet Id, the ejection interval of the ink droplet Id, and the number of ink droplets ejected. In this case, the frequency of the signal output from the detection element 67 that is included in the predetermined range of frequencies can be obtained with high accuracy.

[0080] In the above embodiment, the noise detection process performed to acquire the first FFT spectrum SP1 may be performed, for example, after the initial installation of the droplet dispensing device 100 as a product, after the start-up of the droplet dispensing device 100, or before each dispensing failure detection process.

[0081] Furthermore, in the above embodiment, the first FFT spectrum SP1 and the second FFT spectrum SP2 were obtained by performing a Fast Fourier Transform on the first and second signals. However, the method for obtaining spectra showing the relationship between signal intensity and frequency is not limited to the Fast Fourier Transform. Other methods such as wavelet transform may also be used.

[0082] Furthermore, in the above embodiment, the nozzle 121 that ejects ink droplet Id at a position where it does not intersect with the laser beam Lz emitted from the light source 65 during the noise detection process, and the nozzle 121 that ejects ink droplet Id at a position where it intersects with the laser beam Lz emitted from the light source 65 during the ejection failure detection process, may be the same nozzle or they may be different nozzles.

[0083] (Second Embodiment) Next, a second embodiment will be described. The configuration of the droplet dispensing device in the second embodiment is basically the same as the configuration of the droplet dispensing device 100 in the first embodiment, so explanations that overlap with the first embodiment will be omitted unless otherwise specified. Figure 12 is a block diagram showing the functional configuration of the control device 20 in the second embodiment. Figure 13 is a diagram showing the detection waveform when small droplets are separated from a large droplet that has been dispensed. Figure 14A shows the peak waveform Wp in the detection waveform Wd, and Figure 14B shows the integrated waveform Wi of the detection waveform Wd in Figure 14A.

[0084] As shown in Figure 12, the control device 20 has a calculation unit 20d. The calculation unit 20d receives a detection signal from the detection element 67 that corresponds to the amount of light received for laser light based on the ink droplet ejected from the nozzle 121, and performs the following calculations based on the detection signal.

[0085] In this case, when small droplets are separated from a large droplet that is ejected, as shown in Figure 13, the detection waveform related to the detection signal will show portions caused by the decrease in laser light intensity due to the large droplet and portions caused by the decrease in laser light intensity due to the small droplet. These portions are those in which the maximum amplitude is greater than or equal to the threshold Th, and will be referred to as peak waveforms below.

[0086] First, the calculation unit 20d obtains the number of peak waveforms Wp whose maximum amplitude is greater than or equal to the threshold Th in the detection waveform Wd related to the detection signal. In this case, as shown in Figure 14A, the calculation unit 20d can obtain two peak waveforms Wp1 and Wp2 as peak waveforms Wp whose maximum amplitude is greater than or equal to the threshold Th. Peak waveform Wp1 is the peak waveform corresponding to a large droplet and has a maximum amplitude greater than the maximum amplitude of peak waveform Wp2, which corresponds to a small droplet.

[0087] Alternatively, as shown in Figure 14B, the calculation unit 20d may acquire an integral waveform Wi relating to the voltage value of the detected waveform Wd in Figure 14A, and acquire the number of portions Pt in the integral waveform Wi whose integral value is greater than or equal to a predetermined slope. In the example of Figure 14B, the calculation unit 20d acquires two portions Pt1 and Pt2, for example, by the least squares method. The peak waveform Wp1 described above corresponds to the first peak waveform of this disclosure, and the peak waveform Wp2 described above corresponds to the second peak waveform of this disclosure.

[0088] Next, the calculation unit 20d identifies a target peak waveform Wpt, which is one of the target peak waveforms Wp, based on the number of peak waveforms Wp. In the example in Figure 14A, the calculation unit 20d identifies peak waveform Wp1, which corresponds to a large droplet, as the target peak waveform Wpt.

[0089] Next, we will describe several methods for obtaining the timing of the appearance of peak Pe in the target peak waveform Wp1, which is the peak waveform Wpt. Figure 15 is a diagram illustrating the minimum voltage value in the peak waveform Wp1. Figure 16 is a diagram showing the differential waveform Wf of the peak waveform Wp1. Figure 17 is a diagram showing the integral waveform Wi of the peak waveform Wp1. Figure 18A is a diagram showing the detected waveform Wd, and Figure 18B is a diagram showing the integral waveform Wi of the detected waveform Wd in Figure 18A.

[0090] When acquiring the timing of the peak appearance in the target peak waveform Wp1, the calculation unit 20d can acquire the timing P1 of the appearance of the maximum amplitude of the voltage value in the peak waveform Wp1 as the timing of the peak appearance Pe, as shown in Figure 15.

[0091] Alternatively, as shown in Figure 16, the calculation unit 20d may acquire a differential waveform Wf with respect to the voltage value of the peak waveform Wp1, and acquire the timing P2 at which the differential value of the voltage value in the peak waveform Wp1 becomes zero, i.e., the differential value in the differential waveform Wf becomes zero, as the timing of the appearance of the peak Pe.

[0092] Alternatively, as shown in Figure 17, the calculation unit 20d may acquire an integral waveform Wi relating to the voltage value of the peak waveform Wp1. In this case, the calculation unit 20d uses the voltage value corresponding to the amount of light received when the ink droplet is not ejected as the reference for integration, and acquires the decreasing region Pd, which is the part of the integral waveform Wi relating to the voltage value of the peak waveform Wp1 in which the integral value decreases, for example by the least squares method. The calculation unit 20d can then acquire the timing of the appearance of the peak Pe on the peak waveform Wp1 corresponding to the midpoint P3 of the integral values ​​before and after the decreasing region Pd (integral values ​​Pm1 and Pm2 in the example of Figure 17).

[0093] Alternatively, as shown in Figure 18B, the calculation unit 20d may acquire an integral waveform Wi relating to the voltage value of the peak waveform Wp1 shown in Figure 18A. The calculation unit 20d then acquires a portion Pt where the approximate straight line La of the integral waveform Wi has a slope greater than or equal to a predetermined slope, for example by the least squares method, and acquires the maximum value P4 and minimum value P5 in that portion Pt. The maximum value P4 is the maximum value in the portion Pt where the difference between the integral value in the integral waveform Wi and the value in the portion Pt is less than a threshold. The minimum value P5 is the minimum value in the portion Pt where the difference between the integral value in the integral waveform Wi and the value in the portion Pt is less than the threshold. The calculation unit 20d can then acquire the timing on the peak waveform Wp1 corresponding to the midpoint P6 between the maximum value P4 and the minimum value P5 in the portion Pt as the timing of the appearance of the peak Pe. Note that the portion Pt corresponds to the slope portion in this disclosure.

[0094] Next, Figure 19 is a diagram illustrating the comparison process between the voltage of peak Pe of peak waveform Wp1 in the averaged waveform Waa obtained from the detected waveform Wd and the voltage of reference peak Pes in the reference waveform Ws. Furthermore, Figures 20A to 20C are diagrams illustrating the process of obtaining the averaged waveform Waa from multiple detected waveforms Wd1, Wd2, and Wd3.

[0095] First, we will explain how to obtain an average waveform Waa from multiple detection waveforms Wd1, Wd2, and Wd3. In Figure 20A, of the three detection waveforms obtained when ink droplets are ejected multiple times (three times in the example of Figure 20A), the timing of the peak appearance of detection waveform Wd1 is denoted as ta, the timing of the peak appearance of detection waveform Wd2 is denoted as tb, and the timing of the peak appearance of detection waveform Wd3 is denoted as tc.

[0096] Next, as shown in Figure 20B, the peaks of detection waveforms Wd2 and Wd3 are superimposed on the peak of detection waveform Wd1, which is used as the reference for superposition. Specifically, the timing of the appearance of the peak of detection waveform Wd1 being superimposed is set to ta' (=Ta), the timing of the appearance of the peak of detection waveform Wd2 being superimposed is set to tb' (=tb-ta), and the timing of the appearance of the peak of detection waveform Wd3 being superimposed is set to tc' (=tc-ta). This makes it possible to time-match the peaks of detection waveforms Wd1, Wd2, and Wd3. Subsequently, by averaging the detection waveforms Wd1, Wd2, and Wd3 in Figure 20B, the averaged waveform Waa shown in Figure 20C can be obtained. The above is the method for obtaining the averaged waveform Waa.

[0097] In this embodiment, the calculation unit 20d obtains the averaged waveform Waa shown in Figure 19 from a plurality of detected waveforms Wd using the same method as described in Figure 20C. In the averaged waveform Waa, the target peak waveform Wpt is the peak waveform Wp1 corresponding to the large droplet.

[0098] Next, the calculation unit 20d superimposes the peak Pe of the peak waveform Wp1 in the averaging waveform Waa onto the reference peak Pes in the reference waveform Ws, which is used as the reference for superposition, using the same method as described in Figures 20A and 20B. As a result, as shown in Figure 19, the reference peak Pes in the reference waveform Ws and the peak Pe of the peak waveform Wp1 in the averaging waveform Waa can be time-matched. The reference waveform Ws, including the above reference peak Pes, is stored in the memory unit in advance.

[0099] Next, the calculation unit 20d calculates the difference Vpes between the voltage of the reference peak Pes in the reference waveform Ws and the maximum voltage in the reference waveform Ws. The calculation unit 20d also calculates the difference Vpe between the voltage of the peak Pe of the peak waveform Wp1 in the averaging waveform Waa and the maximum voltage of the peak waveform Wp1 in the averaging waveform Waa. Then, the calculation unit 20d calculates the voltage difference ΔVd, which is the difference between the difference Vpes and the difference Vpe. Note that the above process of temporally matching the reference peak Pes in the reference waveform Ws and the peak Pe of the averaging waveform Waa is not mandatory.

[0100] The calculation unit 20d then determines that there is no discharge defect if the calculated voltage difference ΔVd is less than the threshold, and determines that there is a discharge defect if the voltage difference ΔVd is equal to or greater than the threshold.

[0101] As explained above, when two peak waveforms are detected—one corresponding to a large droplet and the other to a small droplet—conventionally, the presence or absence of nozzle dispensing defects was determined based on the timing of the peak appearance of the peak waveform with the largest amplitude among the two peak waveforms (i.e., the peak waveform corresponding to the large droplet). Therefore, if the peak of a waveform containing noise in the peak waveform corresponding to a small droplet is larger than the peak of the peak waveform corresponding to the large droplet, there is a possibility that the presence or absence of dispensing defects will be determined based on the timing of the peak appearance of the peak waveform corresponding to the small droplet, potentially leading to a false positive. In contrast, according to the second embodiment, the presence or absence of nozzle dispensing defects is determined based on a comparison between the voltage based on the appearance timing of the peak Pe, which is determined by the number of peak waveforms Wp, and the voltage of a predetermined reference peak Pes. In this case, for example, if the number of peak waveforms is a predetermined number, the presence or absence of dispensing defects is determined based on a comparison between the voltage of the peak of the peak waveform corresponding to the large droplet and the voltage of the reference peak. As a result, even if the peak waveform corresponding to a small droplet contains noise as described above, the voltage of the peak that should be acquired can be obtained, allowing for highly accurate detection of dispensing defects based on the acquired peak voltage, thereby suppressing false positives of dispensing defects.

[0102] Furthermore, in the second embodiment, the presence or absence of a discharge defect can be determined with high accuracy based on the voltage difference ΔVd described above.

[0103] Furthermore, in the second embodiment, the timing of the appearance of peak Pe can be easily obtained by acquiring the timing of the appearance of peak Pe as the timing of the appearance of peak Pe, which is the timing of the appearance of peak Pe at the timing of the maximum amplitude of the voltage value in the peak waveform Wp1, P1.

[0104] Furthermore, in the second embodiment, the timing P2 at which the derivative value of the differential waveform Wf with respect to the voltage value of the peak waveform Wp1 becomes zero may be obtained as the timing of the appearance of the peak Pe. This makes it possible to obtain the timing of the appearance of the peak Pe by simple calculation.

[0105] Furthermore, in the second embodiment, a decrease region Pd, which is the portion of the integral waveform Wi relating to the voltage value in the peak waveform Wp1 where the integral value decreases, may be obtained, and the appearance timing on the peak waveform Wp1 corresponding to the midpoint P3 of the integral values ​​before and after the decrease region Pd may be obtained as the appearance timing of the peak Pe. In this case, the appearance timing of the peak Pe can be obtained with high accuracy because it is less susceptible to the influence of high-frequency noise.

[0106] Furthermore, in the second embodiment, a portion Pt is obtained where the approximate straight line La of the integrated waveform Wi has a slope greater than or equal to a predetermined value, and the timing on the peak waveform Wp1 corresponding to the midpoint P6 between the maximum value P4 and the minimum value P5 in that portion Pt is obtained as the timing of the appearance of the peak Pe. In this case, the influence of high-frequency noise and low-frequency noise is reduced, so the timing of the appearance of the peak Pe can be obtained with higher accuracy.

[0107] Furthermore, the process for identifying the timing of the peak Pe's appearance may be performed as follows. Figures 20A to 20E illustrate modified examples of the process for identifying the timing of the peak Pe's appearance.

[0108] As shown in Figure 20A, the calculation unit 20d may acquire multiple peak waveforms Wp1 for the same nozzle. Figure 20A shows an example in which the calculation unit 20d acquires three peak waveforms Wp1a, Wp1b, and Wp1c. Next, the calculation unit 20d adds the three peak waveforms Wp1a, Wp1b, and Wp1c to obtain a peak waveform Wpa, which is the average value of the three peak waveforms Wp1a, Wp1b, and Wp1c, as shown in Figure 20B.

[0109] Next, as shown in Figure 20C, the calculation unit 20d calculates the average slope of the falling portion Pt3 and the average slope of the rising portion Pt4 in the peak waveform Wpa. Then, as shown in Figure 20D, the calculation unit 20d calculates the centroid P7 of the average slope of the falling portion Pt3 and the centroid P8 of the average slope of the rising portion Pt4.

[0110] Next, the calculation unit 20d calculates the minimum time P9 in the peak waveform Wpa, as shown in Figure 20E, using the formula P9 = P7 + 0.5 × (P8 - P7). The calculation unit 20d then acquires the calculated minimum time P9 as the timing of the appearance of peak Pe in the peak waveform Wpa. The calculation unit 20d may also calculate the minimum time P9 for each of the peak waveforms Wp1a, Wp1b, and Wp1c. In this case, the calculation unit 20d may use the average of each minimum time P9 as the timing of the appearance of peak Pe corresponding to the nozzle for which the presence or absence of a discharge defect should be determined. [Explanation of symbols]

[0111] 10 Discharge heads 20 Control device 20d Arithmetic unit 65 Light source 67 detection elements 100 Droplet discharge device 121 Nozzles Fr1, Fr2, Fr3, Fr4 frequency zones Id Inkdrop La approximate straight line Lz laser light Pd decrease area Peak waveform corresponding to large Pe droplets Pes reference peak Pt: Part with a slope greater than or equal to a predetermined value P3,P6 midpoint Sa Subtraction Amount Sh flight space SP1 First FFT Spectrum SP2 2nd FFT Spectrum Second FFT spectrum after SP2a subtraction Th threshold Vpe is the difference between the peak voltage and the maximum voltage in the averaged waveform. Vpe is the difference between the peak voltage and the maximum voltage in the reference waveform. W Printing medium Wd detection waveform Wf differential waveform Wi integral waveform Wp, Wp1, Wp2 peak waveforms Wpt Target Peak Waveform Ws reference waveform ΔVd Voltage difference

Claims

1. A dispensing head having multiple nozzles for dispensing droplets onto the printing medium, A light source that emits light toward the flight space in which the droplet ejected from the nozzle is flying, A light receiving unit that detects the amount of light received after the light emitted from the light source has passed through the flight space and outputs a signal, A control device is provided, The control device is Before detecting whether or not there is a nozzle discharge defect, a process is performed to detect noise based on a first signal output from the light receiving unit based on the light emitted from the light source, Before detecting whether or not there is a nozzle discharge defect, a process is performed to determine the amount of noise to be subtracted based on the first spectrum relating to the first signal, which shows the relationship between signal intensity and frequency. A process to obtain a second spectrum relating to a second signal output from the light receiving unit based on the light emitted from the light source when the droplet is ejected by the nozzle, A process of subtracting the determined subtraction amount from the second spectrum, A droplet dispensing device that performs a process to detect whether or not there is a nozzle dispensing defect based on the second spectrum after subtraction.

2. The droplet dispensing apparatus according to claim 1, wherein the control device performs a fast Fourier transform on the first signal to obtain a first FFT spectrum as the first spectrum, and determines the first FFT spectrum as the subtraction amount.

3. The droplet dispensing device according to claim 1, wherein the control device performs a fast Fourier transform on the first signal to obtain a first FFT spectrum as the first spectrum, and determines a predetermined constant value obtained from the first FFT spectrum as the subtraction amount.

4. The control device is The process of performing a Fast Fourier Transform on the first signal to obtain a first FFT spectrum as the first spectrum is performed, and the process of performing a Fast Fourier Transform on the second signal to obtain a second FFT spectrum as the second spectrum is performed. A reference value obtained from the first FFT spectrum for each frequency region is determined as the subtraction amount for each frequency region. The droplet dispensing device according to claim 1, wherein each of the subtraction amounts determined for each frequency region is subtracted from each portion of the second FFT spectrum corresponding to each of the frequency regions.

5. In the process of detecting the noise, the control device, A process of ejecting the liquid droplet from the nozzle at a position where it does not intersect with the light emitted from the light source, A droplet dispensing device according to claim 1, which performs a process of detecting noise based on the first signal output from the light receiving unit based on the light emitted from the light source.

6. The control device is A process to obtain the subtraction amount based on the first FFT spectrum, which is the first spectrum obtained by performing a fast Fourier transform on a portion of the first signal output from the light receiving unit within a predetermined frequency range, The process involves performing a Fast Fourier Transform on the second signal to obtain a second FFT spectrum as the second spectrum, The droplet dispensing device according to claim 1, further comprising the process of subtracting the subtraction amount from the portion of the second FFT spectrum corresponding to the predetermined frequency range.

7. The droplet dispensing device according to claim 6, wherein the control device obtains a frequency within a predetermined range based on the diameter of the light, the velocity of the droplets discharged from the nozzle, the interval between droplet discharges, and the number of droplets discharged.

8. A dispensing head having multiple nozzles for dispensing droplets onto the printing medium, A light source that emits light toward the flight space in which the droplet ejected from the nozzle is flying, A light receiving unit that detects the amount of light received after the light emitted from the light source has passed through the flight space when the droplet is ejected by the nozzle, and outputs a detection signal. A control device is provided, The control device is A process of discharging the liquid droplet from the nozzle, The process of receiving the detection signal corresponding to the ejected droplet from the light receiving unit, A process to obtain the number of peak waveforms in the detection signal whose maximum amplitude is greater than or equal to a threshold, A process to identify a target peak waveform, which is one of the peak waveforms to be targeted, based on the number of peak waveforms, A process to obtain the timing of the appearance of the peak in the target peak waveform, A droplet dispensing device that performs a process to determine whether or not there is a nozzle dispensing defect based on a comparison between the voltage of the peak based on the timing of appearance and the voltage of a reference peak in a predetermined reference waveform.

9. The droplet dispensing device according to claim 8, wherein the control device, when comparing the peak voltage with the reference peak voltage, obtains a voltage difference which is the difference between the peak voltage and the maximum voltage in the target peak waveform, and the difference between the reference peak voltage and the maximum voltage in the reference waveform.

10. The aforementioned peak waveform includes a first peak waveform corresponding to a large droplet and a second peak waveform corresponding to a small droplet. The droplet dispensing device according to claim 8, wherein the control device acquires the timing of the appearance of the maximum amplitude in the first peak waveform as the timing of the appearance of the peak.

11. The aforementioned peak waveform includes a first peak waveform corresponding to a large droplet and a second peak waveform corresponding to a small droplet. The droplet dispensing device according to claim 8, wherein the control device acquires the timing at which the differential value of the voltage value in the first peak waveform becomes zero as the timing of the appearance of the peak.

12. The aforementioned peak waveform includes a first peak waveform corresponding to a large droplet and a second peak waveform corresponding to a small droplet. The droplet dispensing device according to claim 8, wherein the control device uses the voltage value corresponding to the amount of light received when the droplet is not being dispensed as the reference for integration, acquires a decreasing region which is the portion of the integrated waveform with respect to the voltage value in the first peak waveform in which the integrated value decreases, and acquires the appearance timing on the first peak waveform corresponding to the midpoint of the integrated values ​​before and after the decreasing region as the appearance timing of the peak.

13. The aforementioned peak waveform includes a first peak waveform corresponding to a large droplet and a second peak waveform corresponding to a small droplet. The droplet dispensing device according to claim 8, wherein the control device acquires a slope portion which is the portion where the approximate straight line of the integral waveform with respect to the voltage value in the first peak waveform is greater than or equal to a predetermined slope, and acquires the timing on the first peak waveform corresponding to the midpoint in the slope portion as the timing of the appearance of the peak.

Images