JP2004042670A - Method and semiconductor device for driving ink jet head - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of driving an ink jet head which can prevent the influence of suspension capacitance, prevent discharging performance from degrading due to increase in the number of discharging nozzles, and drive numerous nozzles simultaneously. <P>SOLUTION: Capacitors Ca and Ca constituting respective ink chambers are connected between a common switch S0 and individual switches S1, S2 and Sn. The common switch S0 has ON resistance R0, while the individual switches S1, S2 and Sn have an ON resistance R1 and have a wiring resistance Ri on the side of individual electrodes. By reducing the ON resistance R0 of the common switch S0, the number of the nozzles to be simultaneously driven can be increased. <P>COPYRIGHT: (C)2004,JPO

Description

The present invention relates to a method for driving an inkjet head, a method for electrostatically driving a silicon vibrator, a semiconductor device for electrostatically driving a silicon vibrator, and the like.

2. Description of the Related Art Conventionally, an inkjet head that applies electric energy to a diaphragm, which is a capacitive means such as a capacitor, converts the energy into mechanical energy, increases the pressure in an ink chamber, and discharges ink from nozzles includes (1) a piezoelectric element. (PZT) or the like, a capacitive means that is a piezoelectric vibrator is used as a vibrating plate, and the vibrating plate is joined to a pressure plate to bend the pressure plate (for example, Japanese Patent Application Laid-Open No. Hei 6-340075); 2) A method in which electrostatic energy is applied to a capacitive means between a silicon substrate and a glass substrate also serving as an ink chamber, and a thin film portion of the silicon substrate is bent as a diaphragm (for example, Patent Document 2; No. 40535, Patent Document 3, JP-A-7-81088).

JP-A-6-340075 JP-A-7-40535 JP-A-7-81088 JP-A-7-137249

Here, the circuit for driving the piezoelectric vibrator of (1) is as shown in FIG. 5 of Patent Document 1 (JP-A-6-340075), and the silicon vibrator of (2) is operated by electrostatic force. The driving circuit is a circuit as shown in FIG. 22 of Patent Document 3 (Japanese Patent Application Laid-Open No. 7-81088). The piezoelectric vibrator of (1) has its own capacitance of about 10 nF (10 ^-8 farads). Even if the piezoelectric vibrator is opened when ink is not ejected from the nozzle communicating with the ink chamber, the effect of the stray capacitance I do not receive. However, when the silicon oscillator of (2) is driven by electrostatic force, the capacitance between the silicon oscillator and ITO (Indium-Tin-Oxide, tin-added-indium oxide; transparent conductive film) on the glass substrate becomes The space between silicon and glass is air (ε = 1), and the gap is 0.23 μm to vibrate at a low voltage (40 V) and the area is about 25 pF from the area of 0.65 mm ^2. There is a problem of getting it.

Hereinafter, the driving circuit for the silicon vibrator (2) will be described with reference to FIG.
First, in the driving circuit of FIG. 15, the on-resistance of the common transistor Tr0 and the common impedance shared by each oscillator are R0, and the on-resistances of the individual transistors Tr1, Tr2, Tr3, are R1, R2, R3,. And When the number of nozzles is increased to increase the resolution, the on-resistance of the individual transistor Tr1 increases (about 3 kΩ) due to integration. On the common electrode side, the silicon substrate is doped with boron (boron), and the resistance value is low (several tens of ohms). On the other hand, the individual electrode side is densely wired with ITO on the glass substrate. The wiring resistance Ri of ITO is about 3 kΩ for each nozzle. Here, the gap between the silicon substrate also serving as the common electrode and the glass substrate on which the individual electrodes are mounted is as small as about 0.23 μm, and it is necessary to maintain a uniform gap in each ink chamber. Since a high temperature is applied for anodic bonding between the silicon substrate and the silicon substrate, it is necessary that the individual electrodes mounted on the glass substrate be ITO. The wiring resistance of ITO naturally increases (several kΩ).

FIG. 16 shows an equivalent circuit when each nozzle is turned on. In FIG. 16, when the number of simultaneously driven (printing) nozzles is n, the capacity of each vibrator is Ca (Ca, R1, and Ri are common values for each nozzle for a simplified model), and the power supply voltage is Vh. The voltage Va (t) applied to each transducer at the time of ON is
Va (t) = Vh ・ (1-exp (-t / ((n ・ R0 + Ri + R1) ・ Ca))
.... (Equation 1)
It becomes. At this time, the time constant T = (n ・ R0 + Ri + R1) ・ Ca, where R0 = 1 kΩ, Ri = 3 kΩ, R1 = 3 kΩ, and Ca = 25 pF. When n = 1, T (1) = 0.18 μs, n = At 50, T (50) = 1.4 μs.

(4) When applying electrostatic force to the silicon vibrator, there is a problem in that ink is not ejected from the nozzles unless a pulse whose rise and fall are as fast as about 1 to 2 μs is applied. Therefore, when driving the silicon vibrator by applying a drive pulse having a pulse width Pw of about 14 μs, a rise of about 2 μs, and a fall of about 1 μs, the waveform Va (n = 50) of FIG. ), The waveform of Va is distorted, a voltage lower than Vh is applied at the time of rising, and the voltage does not immediately fall to GND at the time of falling. Therefore, when the number of simultaneously driven nozzles increases, the rise and fall times are delayed, and there is a problem that ink cannot be ejected.

Furthermore, when ink is ejected from a large number of nozzles, if the impedance of the individual diaphragms on the silicon substrate side or the wiring impedance on the glass substrate side is different, the applied voltage is different between the nozzles, and the ejection characteristics between the nozzles are different. There is a problem of being different.

In addition, when electrostatic energy is applied to a capacitor between the silicon oscillator and the glass substrate, electric charges are accumulated, and even if the electrostatic energy is released, electric charges remain, and the silicon oscillator is moved to the glass substrate side. There is a problem that a state of being bent is generated (see Patent Document 2: JP-A-7-81088). Therefore, there is a problem that it is necessary to apply electrostatic energy so as to apply an electric charge on the opposite side to prevent permanent bending of the silicon vibrator and reduce stray capacitance.

In addition, a CMOS IC is used for integration, but a pulse voltage of about 40 V with a rise of about 2 μs is repeatedly applied to the diaphragm and a forward / reverse pulse is repeatedly applied. In addition, there is a problem that a whisker or the like is generated in the pulse waveform and the pulse waveform is disturbed, and noise is generated or ejection becomes impossible.

Therefore, an object of the present invention is to solve the above-mentioned problems and to provide the following.
(A) The off-nozzle (non-discharge nozzle) may discharge due to the influence of the on-nozzle (discharge nozzle) due to the influence of the stray capacitance. For this purpose, when the silicon vibrator is an off-nozzle, both ends are short-circuited (short-circuited).
(B) Further, while preventing the influence of the stray capacitance, both forward and reverse pulse potentials are applied to the diaphragm to prevent the accumulation of electric charges on the diaphragm, and the deflection of the silicon diaphragm to the permanent glass substrate. And a simple drive circuit is realized.
(C) Further, when the number of ejection nozzles increases, it is possible to prevent individual potentials applied to the diaphragm from lowering. Therefore, the number of ejection nozzles is detected, and the pulse waveform such as the timing of the drive pulse is changed according to the number of ejection nozzles. Therefore, the timing of the drive pulse is shifted, and the number of nozzles driven simultaneously is reduced. Further, the magnitude of the applied voltage is changed. Furthermore, the rising and falling of the pulse and the pulse width are changed to change the ejection performance.
(D) Also, instead of detecting the number of ejection nozzles, even when the number of ejection nozzles increases, a decrease in individual potential applied to the diaphragm is prevented. For this reason, the value of R0 (the on-resistance of the common transistor Tr1 + the common impedance), which affects the number n in Equation 1, is reduced. Therefore, the size of the transistor inside the CMOS IC constituting the common transistor Tr1 is increased. In addition, the common impedance was reduced, and the wiring resistance on the common impedance side between the respective diaphragms was made equal. Therefore, a metal solid portion was provided on the silicon substrate.
(E) It is another object of the present invention to make the ejection characteristics between the nozzles uniform. Therefore, the wiring resistance on the silicon substrate was made equal. Further, the resistance value on the insulating substrate side was made equal, so that the wiring distance was made equal for each individual electrode, and the on-resistance of the individual switching means was made equal.
(F) A CMOS IC is used to integrate the common switching means and the individual switching means.
(G) Further, when a pulse is generated inside the CMOS IC, the pulse waveform applied to the diaphragm is disturbed by noise, thereby preventing the ejection characteristics from deteriorating. For this reason, a pulse waveform is created from the outside and applied to the CMOS IC without creating a pulse waveform inside the CMOS IC.
(H) The charging / discharging direction of the capacitive means is bidirectional, and the charging time and the discharging time are shortened. For this purpose, a transmission gate is used as a switching means inside the CMOS IC.

The driving method of the inkjet head of the present invention includes:
Applying a pulse voltage to a plurality of capacitive means, deforming the capacitive means, ejecting ink droplets from each nozzle, and performing a method of driving an inkjet head for performing recording.
The capacitive means has one end connected to the individual switching means,
When the capacitive means is deformed and ejected, the individual switching means is switched so as to apply the pulse voltage to the capacitive means,
When the capacitive means is in the non-deformed state and in the non-ejection state, the individual switching means is switched so that both ends of the capacitive means have the same potential.

Further, the method of electrostatically driving the silicon vibrator of the present invention,
A pulse voltage is applied to a plurality of silicon vibrators, and in a method of electrostatically driving a silicon vibrator for deforming the silicon vibrator,
The silicon vibrator has one end connected to individual switching means,
When deforming the silicon oscillator, the individual switching means is switched so as to apply the pulse voltage to the silicon oscillator,
When the silicon vibrator is in a non-deformed state, the individual switching means is switched so that both ends of the silicon vibrator have the same potential.

Further, the semiconductor device for electrostatically driving the silicon vibrator of the present invention,
For controlling the driving pulse voltage applied to the silicon vibrator having one end connected to the common terminal and the other end connected to the individual terminal, and controlling the first state or the second state for electrostatic driving of the silicon vibrator. In semiconductor devices,
Common switching means for switching a potential of a common terminal of the silicon oscillator to a driving pulse voltage side or a GND potential side;
A plurality of individual switching means for switching the potential of the individual terminal of the silicon vibrator to a drive pulse voltage side or a GND potential side;
When the silicon vibrator is deformed to the first state, one of the common terminal and the individual terminal is connected to a drive pulse voltage side, the other is connected to a GND potential side, and the silicon vibrator is connected to a second potential. Control means for controlling the common switching means and the individual switching means, so as to connect the common terminal and the individual terminal to the same potential when the state is set,
It is characterized by having.

As described above, the present invention has the following effects.
(A) The influence of the floating capacity can be prevented, and the ink droplet is no longer ejected from the off nozzle.
(B) A simple driving circuit that prevents the accumulation of electric charge on the diaphragm while preventing the influence of the stray capacitance is realized.
(C) Printing became possible even when the number of simultaneous ejection nozzles increased.
(D) The ejection characteristics between the nozzles were uniform, and the printing quality was improved.
(E) A small-sized inkjet head can be provided by using a CMOS IC.
(F) Since the driving pulse is applied from outside the CMOS IC, the influence of noise can be prevented and the print quality can be improved.
(G) Since the on-resistance of the common switching means is reduced, when the capacitor is short-circuited for discharging after the application of the driving pulse, the time constant is reduced and the discharging is performed quickly.
(H) Since a transmission gate is employed, current can flow in both directions, and charging and discharging can be accelerated regardless of the applied pulse voltage and timing.

Hereinafter, embodiments of the present invention will be described.

FIG. 1 is a schematic perspective view of an inkjet recording apparatus 310 equipped with an inkjet head 10 according to the present invention. The inkjet head 10 is mounted on the carriage 302 and ejects ink droplets onto the recording paper 105 wound on the platen 300. The carriage 302 is arranged so as to be movable in parallel to the platen axis.
The inkjet head 10 is supplied with ink from an ink tank 301 via an ink supply tube 306. Of course, the inkjet head 10 itself may have an ink tank, and may be exchangeably mounted on the carriage 302 together with the ink tank.

The ink cap 304 is provided near the home position of the carriage 302. The ink cap 304 is capped on the inkjet head 10 when the inkjet head 10 does not eject for a long time or when ink ejection failure occurs and ink is sucked. To collect the waste ink in the waste ink reservoir 305 via the When collecting the waste ink, an electric or manual pump 303 is used.

FIG. 2 is an exploded perspective view of the inkjet head 10. The inkjet head 10 has a laminated structure in which three substrates 1, 2, and 3 are overlapped.

The intermediate substrate 2 has an edge nozzle 4, a common ink chamber 8, an ink chamber 6, an orifice 7, which communicates with each nozzle by etching a silicon substrate, and a diaphragm 5 also serving as a bottom wall of the ink chamber. A common electrode 17 is provided near the common ink chamber 8. The silicon substrate 2 is doped with boron to form a conductor having a resistivity of about 0.1 Ω · cm.

(4) The borate-based glass substrates 1 and 3 as an insulating substrate are anodically bonded to sandwich the silicon substrate 2 therebetween. The anodic bonding is performed by applying a temperature of 300 to 500 ° C. and a voltage of 500 to 1000 V between the glass substrate 1 and the silicon substrate 2, and the anodic bonding between the glass substrate 3 and the silicon substrate 2 is performed similarly. Ink is supplied to the common ink chamber 8 from above the upper glass substrate 1 via the ink supply tube 306, the connection pipe 16, and the supply port 14.

個別 On the lower glass substrate 2, the individual electrodes 31 are arranged in opposition to the vibration plate 5 together with the terminal portions 33 and the wiring lead portions 32 by ITO. The ITO films 31 and 32 are covered with an insulating film 34. The insulating film 34 may be provided not on the ITO side but on the silicon substrate side. Electric energy is applied from the inkjet head driver circuit 220 to the common electrode 17 and the individual electrodes 31, 31, respectively.

FIG. 3 is a side sectional view of the inkjet head 10. FIG. 4 is a plan view seen from AA in FIG. The vibration plate 5 also serving as the bottom wall of the ink chamber 6 of the silicon substrate 2 forms a capacitor through a gap (air) between the diaphragm 5 and the individual electrode 31 provided on the glass substrate 3. Diaphragm 5 has a thickness of 2 μm. The gap 9 (see FIG. 2) may be obtained by etching the silicon substrate 2 from both sides, or by not etching the silicon substrate 2 from the bottom surface and providing a gap adjusting mechanism between the silicon substrate 2 and the glass substrate 3. May be provided. This gap is 0.225 μm. The ink droplet 13 is ejected from the nozzle 4 and is recorded on the recording paper 105. Note that the nozzle 4 may be provided not on the silicon substrate 2 but on the glass substrate 1 so as to eject ink droplets from the glass substrate 1 by the vibration of the vibration plate 5.

(4) When the driver circuit 220 applies a positive voltage pulse to the individual electrode facing the ink chamber communicating with the nozzle to be discharged, the surface of the individual electrode is positively charged and the surface on the diaphragm 5 side is negatively charged. Therefore, the diaphragm is sucked by the electrostatic force and bends downward. Next, when the positive voltage pulse applied to the individual electrode is turned off, the diaphragm returns to the original position, the internal pressure of the ink chamber rapidly rises, and ink droplets are ejected from the nozzles.

FIG. 5 is a block diagram showing a drive control system of the ink jet head used in the ink jet recording apparatus of the present invention. The printer control circuit 201 is composed of a one-chip microcomputer, and is connected to a RAM 205, a ROM 206, and a CG (character generator) -ROM 207 via internal buses 202, 203, and 204 including an address bus and a data bus. The head drive control circuit 210 is controlled by the printer control circuit 201, outputs a drive selection signal SEL and the like to the head driver circuit 220, and outputs print data DATA to the head driver circuit 220 via the data bus 211. The drive pulse Vp applied to the head driver 220 is formed by the head drive control circuit 210.

The print data DATA is connected to the number-of-nozzles detecting means 212, and the number-of-nozzles detecting means 212 detects the number of simultaneously driven nozzles, and sends out a detection signal to the control means 213.
The control unit 213 controls the head drive control circuit 210 as needed based on the detection signal, and controls the pulse shape and drive timing of the drive pulse Vp applied to the head driver 220.

The head driver 220 is formed of a CMOS-IC, and may be provided integrally on the inkjet head 10 or may be provided on an FPC (flexible substrate) to be unitized with the inkjet head 10.

FIG. 6 is a block diagram of a CMOS-IC for the head driver 220. The head driver CMOS-IC has 64 bits of an individual electrode (SEG) driver 61 and 1 bit of a common electrode (COM) driver 62. When discharging a 64-dot nozzle, one CMOS-IC may be used, and when discharging a 128-dot nozzle, two CMOS-ICs may be used.

(4) A drive voltage pulse Vp output from the head drive control circuit 210 with a voltage value of 31 V, a rise of 2 μs, a width of 14 μs, a fall of 1 μs, and a frequency of 16 kHz (each a Typ value) is input to the Vp terminal. The drive voltage pulse Vp input from the Vp terminal is connected to the SEG driver 61 and the COM driver 62. The recording data DATA is input as 64-bit data as serial data via the bus 211, is subjected to serial-parallel conversion by the serial-parallel conversion circuit 64, and is then input to the SEG driver 61 via the inversion circuit 63. In the inversion circuit 63, the SEG driver outputs XOR (exclusive OR; exclusive OR) of each nozzle data and SEL to 61. The selection signal SEL output from the head drive control circuit 210 is input to the terminal SEL, and is connected to the inversion circuit 63 and the COM driver 62.

FIG. 7 shows a main circuit diagram of the COM driver and the n-bit SEG driver, FIG. 8 shows an equivalent circuit diagram, and FIG. 9 shows an equivalent circuit diagram in consideration of the on-resistance of the common transistor and the individual transistor.

8, the COM driver 62 is represented by a three-terminal common switch S0 and the SEG driver 61 is represented by a three-terminal individual switch S1, S2,... Sn. The capacitors between the electrodes 31 are represented by C1, C2,. Assuming that the nozzle number to be ejected is m, the potential on the common switch S0 side and the potential on the individual switch Sm side are different for the capacitor Cm, and a voltage of about 31 V needs to be applied to the capacitor Cm.

If the nozzle number that does not want to be ejected is k, the potential of the common switch S0 and the potential of the individual switch Sk must be the same in the capacitor Ck. By setting the same potential, the capacitor Ck is short-circuited, so that the influence of noise due to stray capacitance can be prevented, and ejection errors due to driving pulses to other ejection nozzles can be prevented.

電 圧 A voltage waveform applied to the capacitor will be described with reference to FIGS. The driving pulse voltage Vp is input to the head driver 220 at a frequency of 16 kHz by repeating Vh and GND. In FIG. 10B, H in the waveforms S0 to S2 indicates that the switch is switched to the drive voltage Vp side, and L indicates that the switch is switched to the GND side. The common switch S0 can repeatedly switch between the drive voltage Vp side and the GND side at half the frequency of the drive pulse voltage Vp by the selection signal SEL output from the head drive control circuit 210.

(4) The inverting circuit 63 and the SEL signal and the DATA signal are used to switch the potentials of the individual switches S1 to Sn in accordance with the timing (see FIG. 6).

Returning to FIG. 10A, at timing 1, the common switch S0 and the individual switch S1 are set to the Vp side so that the nozzle 1 is not ejected. Therefore, the potential VCOM of the common electrode and the individual electrode potential VSEG1 of the nozzle 1 become equal at Vh, and the capacitor 1 is short-circuited, the influence of the stray capacitance is prevented, and no discharge is performed. On the other hand, if the individual switch S2 is set to the GND side to discharge from the nozzle 2, the potential VCOM = Vh is applied to the common electrode side (diaphragm) of the nozzle 2 and the potential VSEG2 of the individual electrode side (individual electrode) of the nozzle 2 is set. Becomes GND, and the potential difference Vh is applied to the capacitor 2, so that the capacitor 2 can be discharged.

At timing 2, the common switch S0 and the individual switch S1 are set to the GND side so that the nozzle 1 is not ejected, the common electrode potential VCOM and the individual electrode potential VSEG1 of the nozzle 1 become equal to GND, and the capacitor 1 is short-circuited. The effect of stray capacitance is prevented and no ejection is performed. On the other hand, when the individual switch S2 is set to the Vp side to discharge from the nozzle 2, the potential of the common electrode side of the nozzle 2 becomes GND, the potential of the individual electrode VSEG2 = Vh, and the potential difference Vh is applied to the capacitor 2. Therefore, ejection can be performed.

よ り When discharging from timings 3 and 4 in FIG. 10A, the potential difference Vh is applied to the capacitor 1, but the potential of VCOM-VSEG1 is inverted between timing 3 and timing 4. In the present invention, an inversion drive circuit is realized by a simple circuit while preventing stray capacitance. Since the charge is neutralized by applying an inversion potential to the capacitors 1 to n, it is possible to prevent the diaphragm 5 from being permanently bent toward the glass substrate 1 due to the accumulation of the charge, and to prevent the diaphragm 5 from being separated from the individual electrode 31. Can be kept constant.

を Referring to FIG. 2, a metal solid portion 17 is provided on the silicon substrate 2. Therefore, the resistance value on the common substrate side from the metal solid portion 17 to each ink chamber can be made equal by doping boron so that the specific resistance of the silicon substrate does not change depending on the location. . Further, by making the distance and width of the individual lead portions 32, 32 provided on the glass substrate 3 equal for each nozzle, the resistance value of ITO can be equalized for each nozzle. Further, by making the etching characteristics of the vibration plates 5 and 5 and the areas of the individual electrodes 31 and 31 equal, the capacitance of the capacitor in each ink chamber can be made equal.

Accordingly, in FIG. 9, the on-resistance of the common driver + silicon substrate resistance when the common switch S0 is on the drive pulse Vp side is RO, and the on-resistance of the common driver + silicon substrate resistance when the common switch S0 is on the GND side is RO ( Because of the CMOS-IC, the on-resistance on the N-channel side and the on-resistance on the P-channel side can be regarded as substantially equal, and the same applies to the individual driver.) Further, the capacitance between the vibration plate 5 and the individual electrode 31 is set to Ca, and can be made equal in each ink chamber (each nozzle). Furthermore, the resistance of the ITO constituting the individual electrodes 31, 32, 33 is set to Ri, and the resistance can be made equal for each nozzle. The on-resistance of the individual driver when the individual switch S1 is on the drive pulse Vp side is R1, and the on-resistance of the individual driver when the individual switch S1 is on the GND side is R1. Since the integrated circuits are formed by the CMOS IC, the on-resistance R1 of each individual driver can be made substantially equal.

Therefore, the time constant T (see Equation 1) of the pulse waveform applied to each ink chamber becomes equal, and the pulse is applied with the same waveform, so that the ink ejection characteristics between the nozzles are the same, and the printing characteristics are improved. I do.

{Circle around (4)} The drive pulse Vp is not created inside the CMOS IC, but is created outside the head drive control circuit 210 or the like and applied to the CMOS IC. Further, since the capacitor is driven in reverse, the switching frequency of the common switch S0 and the individual switches S1 to Sn in the CMOS IC is only half the frequency of the driving pulse Vp. Further, since the driving pulse Vp is generated externally, it is not necessary to generate a high-voltage (about 31 V) high-speed pulse (about 2 μs for rising and about 1 μs for falling) inside the CMOS IC. Therefore, noise such as a mustache does not occur in the pulse waveform, the pulse waveform is not disturbed, and the capacitor can be driven accurately.

共通 It is more desirable that the common driver 62 in FIG. 7 be a transmission gate as shown in FIG. Further, it is desirable that the individual driver 61 also employs a transmission gate as shown in FIG. In FIG. 14, when SEL is set to H, the transistors 141 and 142 are turned on, and a bidirectional current can flow between Vp and COM regardless of the potential of Vp. When the pulse Vp is applied while SEL is H, the potential of Vp is output to the COM terminal as it is, and the capacitor is charged to the discharge nozzle at the rise of the pulse Vp and discharged at the fall of the pulse Vp. While SEL is L, the transistors 141 and 142 are turned off, the transistor 143 is turned on, and the GND potential is output to the COM terminal. Since the pulse width of the driving pulse Vp is included in the pulse width for switching the SEL to the HL, the discharge capacitor can be charged and discharged in accordance with the potential of the driving pulse Vp formed externally, and the timing can be quickly switched in the CMOS IC. No need to do. Therefore, the charge and discharge time of the capacitor can be shortened, and noise and disturbance of the pulse waveform due to switching of timing inside the CMOS IC do not occur. In order to reduce the on-resistance, the sizes of the three transistors in FIG. 14 are increased.

Next, consider driving printing with multiple nozzles simultaneously. In this case, unless the time constant T = (n ・ R0 + Ri + R1) ・ Ca in Equation 1 is kept small, the applied voltage waveform becomes blunt, the applied voltage becomes small, and ejection becomes impossible. Therefore, in order to reduce the time constant T and increase the rise and fall times, the number n of simultaneous driving nozzles, R0 (on resistance of common driver + silicon substrate resistance), Ri (ITO resistance), R1 (individual driver side) It is necessary to reduce each of the on-resistance and the capacitance Ca. However, the electrostatic energy W is proportional to the capacitance * (square of the voltage), and is also proportional to the pulse width t because a rectangular waveform is used as the voltage waveform.

W∝C · V ² · t (Equation 2)
Therefore, if the capacitance Ca and the applied voltage Va are reduced, the ink droplets will not be ejected, so that the capacitance Ca cannot be reduced. In equation 1, when the number of nozzles n is increased, the term of n · Ca · R0 increases. Therefore, it is necessary to keep R0 small so that the time constant T does not become too large even if the number of nozzles n becomes large. The applicant has
n · Ca · R0 ≦ (Ri + R1) · Ca (Equation 3)
Thus, it has been found that ejection can be performed even when the number of nozzles n increases and the time constant T increases. Here, when the number of nozzles n = 64, the capacitance Ca = 25 pF, the individual electrode side ITO resistance Ri = 3 kΩ, and the individual driver ON resistance R1 = 3 kΩ, if R0 (common driver ON resistance + silicon substrate resistance) = 93Ω. Can be discharged. Since the silicon substrate resistance is about 50Ω, the discharge can be performed if the on-resistance of the common driver in the CMOS IC is set to 43Ω or less. Therefore, the transistor size of the common driver in the CMOS IC is set to be 64 times or more the transistor size of the individual driver. From the viewpoint of reducing the on-resistance of the common driver, the larger the transistor size, the better. However, if the transistor size is too large, it is disadvantageous in terms of integration and manufacturing (yield). Therefore, in this embodiment, the IC chip size is about 4.7 * 4.1 (mm). It has become.

(4) Since the on-resistance R1 of the common switch S0 is reduced, the charging time of the capacitor can be shortened. Furthermore, after applying the drive pulse, when the capacitor is short-circuited for preventing the influence of the stray capacitance and discharging, the discharge can be performed quickly because the time constant T is small. As a result, the charge accumulation time is reduced, permanent deflection is reduced, and the residual charge is reduced, so that the charging time of the next drive pulse voltage is also reduced, and the drive frequency can be increased.

Here, consider a case where R0 is not reduced. FIG. 11 shows a drive pulse waveform used in the present invention. (1) is the rise time of the pulse, the charge curve to the discharge ink chamber which is a capacitive means, (2) is the width at the top of the pulse, (3) is the fall time of the pulse, the discharge curve, ( 4) shows the pulse voltage value, and (5) shows the application timing of each pulse. The more the slope of the charging curve in (1) falls, the lower the amount of current can be, and the voltage drop due to the on-resistance of the driver can be prevented. The discharge curve of (3) has the same effect as the charge curve of (1). By changing the portions forming the pulse waveforms (1) to (5), control according to the number of ejection nozzles becomes possible.

(4) The ejection characteristics can be improved by detecting the number n of nozzles and changing the magnitude of the applied voltage Vh ((4) in FIG. 11) according to the number n of ejection nozzles. Since the applied electrostatic energy of Equation 2 is proportional to the ejection energy, the ejection energy is effective with the square of the voltage. However, according to Equation 1, as n increases, the rise of the pulse waveform is delayed, and the applied voltage decreases. Therefore, the number of simultaneous ejection nozzles n of the print data DATA is detected by the nozzle number detection means 212 (see FIG. 5), and if n is equal to or larger than a predetermined value, a signal is sent to the control means 213. , So that the voltage value Vh of the drive pulse output Vp of the head drive control circuit 210 is increased. In FIG. 12A, when the number of ejection nozzles is 1, a pulse having a voltage value V1 is applied, while when the number of ejection nozzles is 32, a voltage value V2 (> V1) is applied, and the number n of ejection nozzles increases. The amount of voltage applied to each ink chamber at the time was reduced.

{Circle around (2)} By changing the applied pulse width ((2) in FIG. 11), it is also possible to prevent the ejection characteristics from deteriorating due to an increase in the number n of nozzles. As described above, the applied electrostatic energy in Expression 2 is proportional to the ejection energy, and the ejection energy is effective with the pulse width. Therefore, the ejection characteristics can be improved by detecting the number n of nozzles and changing the applied pulse width Pw according to the number n of ejection nozzles. In the present invention, the number of simultaneous ejection nozzles n of the print data DATA is detected by the nozzle number detection means 212, and if n is equal to or more than a predetermined value, a signal is sent to the control means 213. Control is performed to increase the pulse width Pw of the drive pulse output Vp of the control circuit 210. In FIG. 12B, when the number of ejection nozzles is 1, a pulse having a pulse width Pw1 is applied. On the other hand, when the number of ejection nozzles is 32, a pulse having a pulse width Pw2 (> Pw1) is applied. The amount of decrease in the voltage applied to each ink chamber when it becomes larger was compensated by changing the pulse width while keeping the voltage value Vh unchanged.

Of course, according to the number n of nozzles detected by the number-of-nozzles detecting means 212, the voltage value in FIG. 12 (a) is changed analogously one by one, or the pulse width in FIG. 12 (b) is changed analogously one by one. Is desirable.

Further, when the number n of the ejection nozzles increases, the method of preventing the drop of the applied voltage (FIG. 11) by shifting the pulses simultaneously applied to the ink chambers (capacitors) and apparently reducing the number of simultaneous drive pulses. (5)) will be described with reference to FIG. It is assumed that the number of simultaneous ejection nozzles is 50 and the nozzle numbers 1 to 50 are simultaneously ejected by the nozzle number detection means 212 (FIG. 5). At this time, the on time of the common switch S0 is lengthened, and the timing of simultaneously turning on the individual switches S1 to S25 and the individual switches S26 to S50 is shifted between the on times of the common switch S0. In order to shift the timing, the detection output of the nozzle number detection means 212 is sent to the control means 213, and the head drive control circuit 210 receives the signal of the control means 213, controls the SEL signal timing, and Output. Accordingly, the number of simultaneous application of the drive pulse to the ink chamber is halved to 25, so that the time constant T does not become too small and simultaneous ejection becomes possible. Various methods of shifting the timing can be adopted, such as using a delay circuit.

In addition to reducing the on-resistance R0 of the common switch, reducing the wiring resistance Ri of the individual electrode and the on-resistance R1 of the individual switch is more effective for simultaneous driving of a large number of nozzles. Further, it is more effective to reduce R0 and change the magnitude or pulse width of the pulse voltage according to the number of ejection nozzles.

FIG. 1 is a schematic perspective view of an inkjet recording apparatus according to the present invention. FIG. 2 is an exploded perspective view of the inkjet head of the present invention. FIG. 2 is a side sectional view of the inkjet head of the present invention. AA arrow line diagram in FIG. FIG. 2 is a block diagram of a head drive circuit of the ink jet recording apparatus of the present invention. FIG. 2 is an internal block diagram of a CMOS IC for a head drive circuit according to the present invention. FIG. 2 is a circuit diagram of a driver section of the present invention. FIG. 2 is a circuit diagram of a driver section of the present invention. FIG. 3 is an equivalent circuit diagram of a driver section of the present invention. 8A is a timing chart of the head drive according to the present invention, FIG. 9A is a voltage timing chart of FIG. 7, and FIG. 8B is a switch timing chart of FIG. FIG. 3 is a timing chart of a head drive pulse according to the present invention. 4A and 4B are diagrams illustrating a head drive pulse according to the present invention, wherein FIG. 4A is a diagram illustrating a change in pulse voltage value, and FIG. FIG. 4 is a timing chart of head driving according to the present invention. FIG. 3 is an equivalent circuit diagram of a driver section of the present invention. FIG. 7 is a circuit diagram of a conventional driver unit. FIG. 5 is a conventional driver equivalent circuit diagram. FIG. 5 is a diagram showing a conventional head drive pulse and applied pulse.

Explanation of reference numerals

2, silicon substrate 3, glass substrate 5, diaphragm 10, inkjet head 17, common electrode (solid metal part)
31 individual electrode 32 individual lead unit 61 SEG driver 62 COM driver 210 head drive control circuit 212 nozzle number detection means 213 control means 220 head driver CMOS IC

Claims (3)

Applying a pulse voltage to a plurality of capacitive means, deforming the capacitive means, ejecting ink droplets from each nozzle, and performing a method of driving an inkjet head for performing recording.
The capacitive means has one end connected to the individual switching means,
When the capacitive means is deformed and ejected, the individual switching means is switched so as to apply the pulse voltage to the capacitive means,
The method of driving an ink-jet head, wherein the individual switching means is switched so that both ends of the capacitive means have the same potential when the capacitive means is in a non-deformed state and in a non-ejection state. A pulse voltage is applied to a plurality of silicon vibrators, and in a method of electrostatically driving a silicon vibrator for deforming the silicon vibrator,
The silicon vibrator has one end connected to individual switching means,
When deforming the silicon oscillator, the individual switching means is switched so as to apply the pulse voltage to the silicon oscillator,
When the silicon vibrator is in a non-deformed state, the individual switching means is switched so that both ends of the silicon vibrator have the same potential. For controlling the driving pulse voltage applied to the silicon vibrator having one end connected to the common terminal and the other end connected to the individual terminal, and controlling the first state or the second state for electrostatic driving of the silicon vibrator. In semiconductor devices,
Common switching means for switching a potential of a common terminal of the silicon oscillator to a driving pulse voltage side or a GND potential side;
A plurality of individual switching means for switching the potential of the individual terminal of the silicon vibrator to a drive pulse voltage side or a GND potential side;
When the silicon vibrator is deformed to the first state, one of the common terminal and the individual terminal is connected to a drive pulse voltage side, the other is connected to a GND potential side, and the silicon vibrator is connected to a second potential. Control means for controlling the common switching means and the individual switching means, so as to connect the common terminal and the individual terminal to the same potential when the state is set,
A semiconductor device for electrostatically driving a silicon vibrator, comprising:

Images