JPS61227060A - Deflection control type ink jet recorder - Google Patents

Abstract

PURPOSE:To stabilize pulverizing of an ink and contrive stable printing quality, by detecting the flying velocity of an ink and adjusting the flying velocity and the viscosity of the ink. CONSTITUTION:When a electrically charged ink droplet flies and collides against a conductive gutter 9, a pulse is generated. In response to this, a microprocessor reads an upper threshold Pwtu and a lower threshold Pwts of pulse width, and calculates a flight time deviation Cc corresponding to the difference between the actual flying velocity and a target flying velocity. When the deviation Cc is in excess of a predetermined value, a correction absolute value DELTAVpd is read, and updated pulse width signal data Vpd to be subsequently supplied to a pump driver 21 is determined by Vpd=Vpd+ or -DELTAVpd. The correction value DELTAVpd is increased with the increase of the deviation Cc and is decreased with the decrease of the deviation Cc. The deviation is reduced in the manner of a geometric series. When Vpd is not less than Pwtu, a diluent is supplied into a diluent tank 15 through a solenoid-operated ON-OFF valve 14.

Description

Detailed Description of the Invention [Technical Field] The present invention jets vibrated ink from a nozzle, selectively charges the jetted ink at a position where it separates into ink particles using a charging electrode, and transforms the charged ink particles into The present invention relates to an inkjet recording device that deflects the ink using a deflection electrode and collides with a predetermined position on a recording sheet, and particularly relates to control of ink viscosity.

■Prior art In this type of inkjet recording, the distance from the ink jet nozzle to the recording paper is relatively long, and therefore the ink pressure is so high that the ink particles ejected from the nozzle and turned into particles are affected by the charging electric field and the deflection electric field. It is set high so that it can reach the recording paper in a stable flight trajectory. In addition, in order to regularly generate ink particles of a predetermined particle size and make them take a predetermined deflection trajectory accurately, it is necessary to maintain stable ink viscosity, ink pressure, vibration pressure, charge amount, deflection electric field, etc. and must be precisely controlled. Also,
If the charging voltage (pulse) is not applied precisely at the timing when the ejected ink separates into ink particles, the ink particles will not be properly charged.

Therefore, in the past, the ink pressure was adjusted before recording charge control.
A phase search is performed to stabilize the voltage at a constant level and determine the timing of applying the charging voltage pulse relative to the timing of ink particle formation, or vice versa. In this phase search, for example, a charge detection circuit mainly consisting of an amplifier, an integrator, and a comparator is connected to a non-contact or contact type charge detection electrode, and a short charging voltage pulse is applied to the charge electrode. The phase of the charging voltage pulse with respect to ink droplet separation is sequentially shifted at predetermined time intervals. When the charge detection circuit issues a signal indicating "charge", the phase of the charging voltage pulse at that time is determined as the appropriate charging phase.

In deflection control inkjet recording, the ink captured by the gutter is returned to the ink jet head, but the diluent in the ink evaporates before it leaves the head and returns to the ink flow path.
During the continuation of ink jetting, the viscosity of the ink gradually increases. For example, the viscosity changes as shown in FIG. 12a.

Since the amount of deflection is affected by the flying speed of the ink particles, in one embodiment, the flying speed of the ink particles is detected,
The ink pressure is adjusted so that it reaches a predetermined speed. For example, U.S. Patent No. 3,600,9
Specification No. 55 (issued August 1971.

Int, C1, Gold 15/13) and Japanese Unexamined Patent Publication No. 50-105733 disclose techniques for detecting ink speed and adjusting ink pressure.

However, the adjustment range of ink droplet velocity and chemotaxis by adjusting ink pressure is extremely narrow, and with high viscosity, even if the ink pressure is increased, fluctuations in the size of the ink droplets become large and recording density may fluctuate. , the amount of deflection fluctuates, or
The ink particle formation becomes disordered, making it impossible to record. This is used when using a constant pressure pump with a constant discharge pressure, a constant flow pump with a constant discharge volume, or a constant pressure or constant flow rate in order to keep the ink pressure constant or the ink drop size constant. The same applies to any case where the drive of the pump is controlled.

Therefore, it is necessary to adjust the ink viscosity.

Conventionally, it has been difficult to detect ink viscosity and control the viscosity based on the detected value. For example, in JP-A-57-63260 and JP-A-56-136381, the specific gravity of ink is measured, but not only is the specific gravity measuring device complicated, but the viscosity that can be measured with this type of specific gravity measurement is Since the unit is larger than the ink viscosity adjustment unit in inkjet recording, it seems impossible to fully achieve the viscosity control precision required by an inkjet recording apparatus.

■Purpose of the invention The present invention enables viscosity adjustment by minimizing the number of additional elements in the device.
The purpose is to reduce the range of viscosity fluctuation by finely controlling the viscosity.

■Configuration In order to achieve the above object, the present invention detects the flying speed of ink particles ejected from an ink ejecting head and compares this with a set value to determine the current and voltage of the pressure pump. control at least one pressure pump energization parameter, such as the frequency of the pressure pump, and supply diluent to the ink when the pressure pump parameter is outside a predetermined upper limit.

In order to further reduce the variation range of ink viscosity due to such diluent supply control, the present invention further includes:

To make ink pressure loss of ink reaching an ink jet nozzle even larger than before as the viscosity of the ink becomes higher.

For example, when the ink pressure is constant, as shown in Figure 12b, when the viscosity is high, the flying speed of the ink particles is low, and when it is low, the flying speed is high.There is a correlation between the ink viscosity and the flying speed. By detecting the ink viscosity, the ink viscosity can be determined accurately. Therefore, when the viscosity is within a predetermined range, increasing the pump energization parameter will increase the flight speed, and decreasing the viscosity will decrease the speed, and the flight speed will be stabilized at a predetermined value. If the viscosity is extremely high, the flight speed will be low and even pump control will not be able to provide a stable flight speed, but in the present invention, the diluent is supplied when the pump energization parameter increases to a predetermined value before that. The ink viscosity is always controlled by pump control within a range that provides a predetermined flying speed, resulting in stable flying speed control. As will be described later, the ink flying speed can be accurately detected without requiring particularly many or complicated additional elements.

For example, as shown in FIG. 13a, when the water evaporation rate in the ink increases, that is, when the ink viscosity increases,
Pressure loss increases in the ink ejecting head, and the ink pressure (required ink pressure) supplied to the head to provide a constant ink jetting speed increases. On the other hand, the efficiency of the pressure pump (the ink pressure obtained with a given electric power) also decreases as the ink viscosity increases, as shown in FIG. 13b. That is, even in the pressurizing pump, the loss of discharge pressure (ink pressure) due to the increase in ink viscosity increases. As a result, as the ink viscosity increases, the pressure pump energizing power required to keep the ink flying speed constant increases as shown by the solid line in FIG. 13c. Conventional (for example, Japanese Patent Application No. 1123/1989)
No. 13), for example, the water evaporation rate in the ink is 15%.
Set the pressure pump energizing power to the limit at L3.
When Po is required at c, ink dilution liquid is supplied. As shown by the solid line in Figure 13c, the slope of the power required for the pressurizing pump to maintain the ink flying speed constant as the water evaporation rate (ink viscosity) increases is relatively small ( This is due to the conventional method of minimizing pressure loss in the ink flow path).
Conventionally, since control errors are large, a relatively large viscosity limit (water evaporation rate of 15%) is set.

In contrast, in the present invention, since an ink pressure loss means is actively added, the water evaporation rate in the ink is affected.

The power required to energize the pressurizing pump to maintain a constant ink flying speed has a steep slope, for example, as shown by the two-dot chain line in FIG. 13c. As a result, e.g. the same as before.

When the ink dilution liquid is supplied when the energizing power of the pressurizing pump increases to the predetermined power P1, the evaporation rate is extremely lower than the evaporation rate of 15% at the intersection of the two-dot chain line and Pl in Fig. 13c. Diluent will be supplied. In addition, in the case of the present invention (two-dot chain line in Fig. 13c), the power error (deviation or variation) ΔP of the pressurizing pump with respect to the error or variation ΔI in the evaporation rate, that is, ΔP/ΔI, becomes smaller.
Control error becomes extremely small. Therefore, according to the present invention, the ink viscosity can be finely and precisely controlled within a predetermined range.

As shown by the two-dot chain line in FIG. 13c, in order to keep the ink flying speed constant, there is a means for increasing the gradient of the pressurizing pump energizing power with respect to the water evaporation rate in the ink, that is, a means for increasing the ink pressure loss. The pressure pump itself, a part of the ink pipe from the ink tank to the ink jet nozzle, or the entire ink pipe can be used as the ink pump.
For the pressurizing pump itself, this can be done, for example, by reducing the diameter of the ink suction port and the ink discharge port, and for the ink pipe, for example, by making at least part of the diameter small, or by reducing the ink pressure loss. This is possible by inserting a filter into the pipe.

Incidentally, the ink viscosity also varies depending on the ink temperature t, as shown in FIG. 12c. Therefore, it is preferable to keep the ink temperature or the temperature of the environment in which the inkjet recording apparatus is used to be extremely constant, or to ensure that the flight speed is stable even if the ink temperature fluctuates, and that there is no over-replenishment of the diluent.

Therefore, in a preferred embodiment of the present invention, the ink temperature.

Alternatively, detect the temperature corresponding to the ink temperature, such as room temperature or the temperature of the inkjet recording device, access the pump energization parameter upper limit value corresponding to the detected temperature, and determine the pump energization parameter that should be instructed to the pump driver by pump control. A diluent is supplied when the parameter value to be indicated exceeds the upper limit value by comparing with the upper limit value.

According to this, the contribution of ink viscosity due to ink temperature is compensated, and even if there is a temperature change, over-replenishment or insufficient replenishment of the diluent does not occur.

As the pressure pump activation parameter, at least one of energization pulse width (duty control, applied voltage (voltage level or energization level control), or frequency control (speed control) can be used. The pressure pump is assumed to be energized and controlled by its energization pulse width.According to this, the energization pulse width with respect to the ink viscosity while keeping the pressure constant has a correlation as shown in Fig. 12d when the pulse period is 20 m5ec. When the viscosity is within a predetermined range, a predetermined ink pressure can be obtained by pulse width control, so the pump discharge pressure actually changes according to the pulse width signal to control the ink pressure.Pulse width control is digital. Since the processing can be performed with high precision, it is possible to finely stabilize the ink pressure at the target value, and also to easily perform the comparison processing with the upper limit value according to the ink temperature digitally.

In a preferred embodiment of the present invention, the deviation of the ink droplet flying detection speed from the target value is further determined, the pulse width correction data corresponding to this deviation is read out from the memory, and the This pulse width correction data is added or subtracted from the pulse width signal data currently driving the pump to obtain new pulse width signal data, and the ink pressure pump is driven based on this new pulse width signal data.

According to this, the pulse width correction data is a value that corresponds to the speed of the ink droplet, so the higher the speed, the larger the value, and the smaller the value when the speed is closer to the target value. Because it is.

During the repetition of speed detection → pulse width signal change → speed detection → pulse width signal change..., the change in pulse width is large at the beginning, and as it gets closer to the target, the change in pulse width becomes smaller, resulting in a geometric progression. In this case, the ink pressure (ink speed) converges to the target value in a relatively short time. The convergence error will be small.

In a preferred embodiment of the present invention, a gutter that is placed between the deflection electrode and the recording paper and captures non-printed ink particles is used as a conductor, and a charge detection circuit is connected to this gutter, and the deflection electric field is cut off when adjusting the pressure. The ink droplets are charged, and counting of clock pulses is started from the application of a charging voltage for this charging, and the count value when the charge detection circuit detects charge is obtained as ink droplet velocity detection data. This data is proportional to the inverse of ink drop velocity. That is, when the ink droplet speed is high, the count value is small, and when the ink droplet speed is low, the count value is large.

The pulse width correction value is associated with the error in the ink droplet velocity with respect to the set target value, and is associated with the absolute value regardless of whether the error is positive or negative. Assign a small correction value to a small absolute value, and add or subtract it from the pulse width signal data when detecting the ink droplet velocity (when the count value is larger than the set target value) depending on whether the error is positive or negative. (When the count value is smaller than the set target value), new pulse width signal data is obtained, and the pump is driven based on this data. This process is repeated until the error is within a predetermined range.

According to this embodiment, since the gutter is also used for detecting the speed of ink droplets, there is no need to provide a separate ink droplet detection means and there is no need to make the distance from the nozzle to the gutter particularly long. Incidentally, if the distance is long, the head length of the printer becomes long, which is not only disadvantageous in terms of the device configuration, but also increases the possibility of misalignment of ink flight, which is bad in terms of printing quality.

(2) Embodiment FIG. 1a shows an outline of a mechanical system according to an embodiment of the present invention. The ink in the ink tank 16 in which the ink cartridge 17 is attached is sucked by the pump 1 and pumped to the accumulator 2. The pressurized ink sent to the accumulator 2 has pressure vibrations suppressed by the accumulator 2 and the filter 3, and reaches the ink jet head 5 through the electromagnetic switching valve 4.

In the ink ejection head 5, the electrostrictive vibrator is excited at a constant period and a constant amplitude, and applies pressure vibrations at a constant period and a constant amplitude to the ink in the head S. Ink is ejected from the nozzle of the head 5, and due to this pressure vibration, the ejected ink separates into particles at a position that has advanced a predetermined distance from the nozzle.

This particle formation corresponds to the pressure vibrations applied to the pressurized ink in the head 5, and the ejected ink corresponds to one part of the pressure vibrations.
Particles are formed at a rate of one ink droplet per period. By applying a charging voltage between the charging electrode 6 and the ink in the head 5 at the timing when the ejected ink becomes particles,
The atomized ink particles have an electric charge. Charged ink particles and uncharged ink particles are formed depending on whether or not a charging voltage is applied to the electrode 6 as the ink becomes particles.

The charged ink particles are deflected by the electric field of the deflection electrode 7 and collide with the recording paper 8, while the uncharged ink particles travel straight and collide with the conductive gutter 9, pass through the filter 11, are sucked by the pump 12, and are sent to the air vent tank. 13 and returns to the ink tank 16.

A diluent tank 1 is connected to the air vent tank 13 via an electromagnetic on-off valve 14.
5 diluents are supplied. Valve 14 is normally de-energized and closed.

When the electromagnetic switching valve 4 is energized, the filter 3 and the head 5 are communicated, and the pipe 4o is cut off from them, and when it is not energized, the head 5 and the pipe 40 are communicated, and the filter 3 is cut off from them. .

FIG. 1b shows an enlarged cross section of the pressure pump 1 shown in FIG. 1a. In this case, an electric coil 4 is connected to a magnetic core 4°.
1 is wound around the core 40, and when the electric coil 4I is energized, the plunger 42 is attracted to the core 40. A diaphragm 48 made of a plate spring material is fixed to the plunger 42. When the electric coil 41 is de-energized, the spring force causes the diaphragm 48 to return downward. A fixed sleeve 47, a leaf spring valve body (leave valve) 46. Ring-shaped rubber valve seat 45. A fixed sleeve 44 and an inlet joint 43 are provided.

Fixed sleeve 47. An enlarged perspective view of the leaf spring valve body 46 and fixed sleeve 44 is shown in FIG. 1C. In this, the inner diameter of the suction pipe (the inner diameter of 47, 44, 43) is set to 4 mm or less, which is slightly smaller than before, and the inner diameter of the rubber valve seat 45 is set to 2 mm or less, which is smaller than before. There is. Since the inner diameter is made small in this way, the resistance of the pipe line is large at the suction port of the pressurizing pump 1, resulting in the characteristics shown by the two-dot chain line in FIG. 13c.

Experiments have shown that as the ink viscosity increases, the resistance increases. Additionally, experiments have shown that when energizing a pressurizing pump at a frequency of 20 Hz or higher, the valve structure is a leaf valve and the valve seat diameter is 2 mm or less, and the pressure is 1°5 kg.
Good viscosity characteristics were exhibited by setting the following valve spring pressures.

The discharge port of pump l is configured similarly to the suction port, but
The inner diameter of the rubber valve seat at the discharge port is larger than the inner diameter of the rubber valve seat 45 at the suction port.

In another embodiment of the present invention, a conventional pressure pump was used and a small diameter tube 50 shown in FIG. 1d was inserted between the filter 3 and the solenoid valve 4, and good results were obtained. Inner diameter 0.1
A thin tube 50 having a length of 20 mm or less and a length of 20 mm or less produces good results as an ink pressure loss means. In another embodiment of the invention.

Good results were obtained by inserting a pore filter 60 shown in FIG. 1e between the conventional filter 3 and the solenoid valve 4 using a conventional pressure pump. The pore diameter of the filter 6o is 3μ or less.

FIG. 2 shows the configuration of an electrical system that energizes each mechanical element shown in FIG. 1 and also performs charge recording control and ink viscosity control.

A sine wave generation/amplification circuit 25 applies an excitation voltage to the electrostrictive vibrator of the ink jet head 5 . A charging signal amplifier circuit 26 applies a pulsed charging voltage to the charging electrode 6, and a deflection voltage generating circuit 27 applies a high voltage at a constant level to the deflection electrode 7.

One end of a core wire of a shielded wire 10 is connected to the conductive gutter 9, and a charge detection circuit 30 is connected to the other end of the core wire.

Pumps 1 and 12 are electrically energized by pump drivers 21 and 12, respectively, and pump drivers 21 and 32, respectively. The electromagnetic switching valve 4 and the electromagnetic on-off valve 14 are electrically energized by valve drivers 24 and 34, respectively.

In this embodiment, in order to charge the ink at a predetermined timing in detecting the flying speed of the ink particles, and to apply a search pulse voltage to the electrode 6 in a phase search in which the center of the charging voltage pulse is aligned with the separation phase of the ink particles, and
In order to apply recording charge voltages with different levels stepwise to the electrodes 6 during printing, a timing pulse for ink pressure search is generated @20. A search charge signal generator 23 and a record charge signal generator 22 are provided, and signals from these generators are selectively applied to a charge signal amplification circuit 26 by a gate circuit 28 . The operation timing of each part of the electric circuit element is determined based on a plurality of types of timing pulses generated by the pulse generator 19.

A printing control device (ink viscosity control means) 29 mainly composed of a microcomputer 31 energizes and controls each part.

In FIG. 3, a pulse generator 19. Gate circuit 28° recording charge signal generator 22. Configurations of the search charge signal generator 23, the timing pulse generator 2o, and the printing control device 29,
Show the connections between them.

The pulse generator 19 is composed of a pulse generator 19a including a crystal oscillator and a frequency division counter, and a frequency division counter 19b, and has a frequency of 3.2 MHz, 800 KHz, 400 KHz,
200KHz, 100KHz, 100/32=3.12
5KHz, 133KHz, 33KHz and 390)
Nine types of 50% duty pulses of 1z are generated.

The 100 KHz pulse is generated by a sine wave generation/amplification circuit 25.
is applied to The circuit 25 generates a sine wave of the fundamental frequency component of the input pulse, amplifies it, and applies it to the electrostrictive vibrator of the head.

As a result, a pressure vibration of 100 KHz is applied to the ink in the head 5, and ink particles are formed at a rate of 100XIO'' particles/sec and fly toward the gutter 9 or the recording paper 7. In other words, at a frequency of lOOKHz Ink particles are generated.

The recording charge signal generator 22 has a counter for frequency division by 172 (
It consists of a T flip-flop (T flip-flop) 22a, a serial-in/parallel-out 8-bit shift register 22b, a data selector 22c, and a charge code generator 22d consisting of a counter, a decoder, and an output gate.

A pulse of 100 KHz is input to the counter 22a, and the counter 22a applies a pulse of 50 KHz to the shift register 22b as input data.The shift clock of the shift register 22b is a pulse of 800 KHz. , eight sets of pulse outputs (CHP) of the shift register 22b are shown.These eight sets of pulses are applied to the data selector 22c.

A 3-bit output control code is further given to the data selector 22c, and this code instructs the output of one set of eight pulses.

Since the shift register 22b is shifted and activated by a pulse of 800 KHz, the eight sets of pulses a = h have the same period and the same duty, with the phase delayed by 0.00125n+sec in this order, and the data selector 22
The data specified by the 3-bit code given to c is selected by the data selector from the AND gates AO to A of the gate circuit 28.
9 is output.

The search charge signal generator 23 is composed of a decoder 23a and a data selector 23b.
Three sets of pulses are applied at 100, 200 and 400 KHz. As a result, the output of the decoder 23a is
There are eight sets a=h shown as SP in FIG. 5, and their pulse widths are 0.00125 msec, and the phases are shifted by the pulse width, that is, 0.00i25IIlsec.

The same control code as the 3-bit control code given to the data selector 22c of the recording charge signal generator 22 is also given to the data selector 23b, and the search charge signal a is generated as follows.
= Output h.

Control code 0000010100111001011
Output of 1011122c abcdefgh Output of 23b defghabc The output pulse of the data selector 23b is applied to the AND gate AIO of the gate circuit 28.

The timing pulse generator 20 includes two J-type flip-flops 20a, 20b, an AND gate 20c, and an inverter 20d.

100/32 KHz to Tamming Pulse Generator 20
pulse is given as a clock pulse, and the microcomputer 31 has a width of 0.36 m5ec (50K
The pulse rHJ of l (width slightly longer than 32 cycles of z) is
One pulse is given each time the ink droplet flying speed is detected.

FIG. 6 shows input and output signals of the tanning pulse generator 20. AND gate 2 of the tamming pulse generator 20
The output of 0c (R27) is given to the gate circuit 28.

FIG. 4 shows a charge detection circuit 30 and a pump driver 21.
The configuration is shown below.

The charge detection circuit 30 includes a resistor 30a that grounds the shield wire 10 to the device, and a field effect transistor (FET) 3.
0 b, anti-phase amplifier 30C, bypass filter 30d
, half-wave rectifier 30e, integrating circuit 30f and comparator 30
It is composed of g.

The resistance value of resistor 30a is between gutter 9 and earth filter 1.
The resistance value is set to approximately LOOKΩ, which is smaller than the resistance value between 1 and 1 when they are connected by ink.

There is a slight stray capacitance between the core wire of the shield wire 10 and the equipment ground, and in this embodiment, the ink particles are negatively charged, and each time a charged ink particle collides with the gutter 9, the core wire of the shield wire 10 is However, due to the time constant of the stray capacitance and the resistor 30a, when the charged ink particles continuously collide with the gutter 9, the potential of the core wire continues to be a negative potential. It has become.

Therefore, if a charge control is performed in which a plurality of consecutive elements are charged and the next plurality of consecutive elements are uncharged, the potential pattern corresponding to this pattern of charges, charges, and uncharges will be applied to the FET 30b.
appears at the gate. This potential pattern is FET30b
The current is converted into a current by an amplifier 30c, inverted and amplified by an amplifier 30c, low-frequency noise is removed by a bypass filter 30d, and then provided to a half-wave rectifier circuit 30e. The charge detection circuit 30 detects P19 when no charged ink particles collide with the gutter 9.
P19=1 (high level), and when charged ink particles are colliding with the gutter 9, an output of P19=0 (low level) is produced.

The pump driver 21 is a microcomputer (hereinafter referred to as C
An AND gate 21a, a D flip-flop 21b, a buffer amplifier 21c, and a transistor 21e receive pump drive signal data P15 from a printing control device 29 mainly consisting of a printer (referred to as PU) 31, and also receive a 390 Hz pulse P27 from a pulse generator 19. , 21f, 21g, etc., and a counter 21 for pump drive pulse width adjustment.
d, an AND gate 21h, a delay circuit 21i, and the like.

A 50 Hz pulse and a pump drive on signal P15 are input from the pulse generating circuit 19 to the AND gate 21a of the pump driver 22. The counter 21d is a counter that can be preset with data, and is a counter that can be preset with data.
Load the pulse width signal data Pd from 9.
Preset with input signal. The load input applies a 50Hz pulse to one input terminal of the Nant gate 21h, and
By inverting the rising edge of the 50Hz pulse with an inverter, delaying it, and inputting it to the Nant gate 21h, the Nant gate 21 at the rising edge of the 50Hz pulse
Generated from h. In other words, the counter 21d is 50H.
Load Pd at the rising point of the z pulse. Then, the counter 21d subtracts 1 from the value indicated by Pd every time a 33 KHz pulse arrives, and when the number of arriving pulses matches the Pd value (when the remainder of the subtraction becomes 0), the carry is sent to the flip-flop. 21b to the clear input terminal CR.

Since the flip-flop 21b is set at the rising edge of the 50 Hz pulse to make the Q output a high level H, the Q output of the flip flop 152 has a frequency of 50 Hz and a pulse width of the high level H equal to Pd times the period of the 33 KHz pulse. This becomes a pulse of minutes and is applied to the output circuit. Output circuit (
218 to 21g) apply a pulse voltage synchronized with the Q output pulse to the pump 1 by switching with this Q output. The ten widths of this pulse voltage correspond to the eight widths of the Q output.

Determined by Pd.

7a and 7b show the characteristics of the pump 1 energized by the output pulse voltage of the pump driver 15. FIG. pump l
By applying the above pulse voltage to the electric coil of
In the pump 1, the armature is pulled by the coil core, the diaphragm is displaced, the volume of the liquid chamber increases, a negative pressure is created, the suction valve opens, and ink is sucked. While the pulse voltage is negative, the diaphragm returns due to the spring force of the diaphragm, and the volume of the liquid chamber becomes smaller, resulting in a discharge stroke. When supplying the same pressure to the head 4, for example 4 Kg/cm,
+When the pulse width is 13 m5ec and 50% or more of the pulse period of 20m5ec, the ink is sucked in the suction stroke at 13m5ec, and the negative pressure in the liquid chamber becomes a small value of about 0.4Kg/c+o2. Further, when the + pulse width is 7 m5ec, which is less than 50% of the pulse period, the negative pressure generated becomes a large value of about 0.8 kg/c2 because the suction stroke time is short. The current peak value at this time is larger than that at 13 m5ec. This large negative pressure generated within the liquid chamber generates bubbles due to the cavitation phenomenon, which causes instability of the discharge pressure.

In particular, since the amount of dissolved oxygen in the ink is large due to the circulation of the recovered ink, it is necessary to keep the value of the generated negative pressure small. Regarding pressure control, the pressure of the ink supplied to the head 5 needs to be controlled within a range of about 4 to 5.5 kg/cm in order to keep the flying speed of one drop constant due to the increase in viscosity due to environmental temperature and moisture evaporation. In order to keep the generated negative pressure low by varying this with the peak value of the pump current using pulse width control, the time width must be within a range of 50% or more of the pulse period, that is, about 10 m5ec = 15 rssec for a 50Hz pulse. The above conditions are satisfied by making this a suction stroke.

Therefore, the value of Pd is set in the range of 11.5 to 13.5 m5ec, and the upper limit of Pd (Pwt,
u) 11゜5~13.5m5ec from the high side
It is divided into 0 stages. The lower limit of Pd can also be changed depending on the ink temperature t, but in this example the lower limit is (P wt, s)
is fixed at 11,5m5ec.

A thermistor 18 is installed inside the accumulator 2 to detect the temperature of the ink inside the accumulator 2.

A temperature detection circuit 38 is connected to this thermistor 18. The configuration of the temperature detection circuit 38 is shown in FIG. The thermistor 18 has low resistance at high temperatures and high negative resistance at low temperatures. The voltage of the thermistor 18 is determined by the operational amplifier 3.
It is applied to the opposite polarity input terminal (-) of 8a. When the temperature is high, the output of the operational amplifier 38a becomes high, and when the temperature is low, the output becomes low.

The output of the operational amplifier 38a is converted into digital data by the A/D converter 38b and is constantly applied to the port 35a.

The printing control device 29, which is an ink viscosity control means, includes a central processing unit (microprocessor) 31° ROM 33.
RAM34. This is a general-purpose computer composed of an I10 board 35, a system controller 36, etc., and controls each element described above.

FIG. 9 shows an overall outline of the control performed by the microprocessor 31, FIGS. 10a and 10b show details of the ink viscosity adjustment control, and FIG. 11 shows the interrupt processing in the ink viscosity adjustment.

First, the overall outline of the control will be explained with reference to FIG. When the power is turned on, the microprocessor 31 initializes the input and output ports and sets each control element to a safe state (
Step 1: The word step will be omitted hereafter). At this time, the deflection voltage generation circuit 27 is turned off, the pump driver 21.32 is also turned off, and the valve driver 24.3 is turned off.
4 is also set to OFF, and AND gate A11 is also set to OFF. In other words, the ink ejection is stopped.

When the initialization is completed, the microproset 31 first outputs and sets a drive instruction signal to the pump driver 21 to put the pump 1 into a driving state.
An energization instruction signal is outputted to 3) to energize the electromagnetic switching valve so that the flow flows through the filter 3 and the spatula 15 (3), and a drive instruction signal is outputted to the pump driver 32 to drive the pump 12.

and turn on the preparation timer (program timer) (
5).

As a result, ink is ejected from the ink ejection head 5, collides with the gutter 9, is sucked by the pump 12, and is ejected into the air vent tank 13, thereby starting an ink chain.

In this embodiment, the pump 1 is of a constant pressure type in which the discharge pressure is constant. Other types of pumps may be used to provide constant pressure control.

There, the microprocessor 31 waits for the preparation timer to time out (8), reads the status of each part while waiting, and outputs status data to the host device or operation board according to the status (6). Each part is ready for printing (7
), 1 phase search when the preparation timer times out (9)
Proceed to. When the phase search is completed, the process proceeds to ink viscosity adjustment (10), and after completing the ink viscosity adjustment, the phase search is executed again (11), and the process proceeds to recording control (12). When the recording is finished,
Proceed to stop processing similar to initialization. Note that while the preparation timer is counting time, the ink pressure from the accumulator 2 onward increases to a pressure corresponding to the operating speed of the pump I, and when the preparation timer times out, it remains stable at a certain pressure.

In the phase detection fi (9, 11), the microprocessor 31 sets the 3-bit code given to the data selectors 22c and 23b to 000, opens the AND gate All (gate-on), and starts timing.

As a result, the data selector 22c selects a of the CHP shown in the fifth rgj, and the data selector 23b selects the S shown in FIG.
d of P is output, but since the printing data is L (not recorded), the output gate of the charge code generator 22d is closed, and the output of the charge code generator 22d (charge level instruction code) is all L. (0000000000), and AND gates AO to A9 are all closed. But the AND gate AIO is that Ic 100/32
= 3. ' A pulse of 125 K Hz is applied so that it is open for 16 periods of 100 K Hz (during the generation of 32 ink particles) and then for the next 16 periods (during the generation of 32 ink particles). ) is closed, and this process is repeated thereafter. As a result, the fifth
Of the SP pulses d shown in the figure, 32 consecutive pulses are given, and then 32 consecutive pulses are given again after a pause period of 32 pulses. is applied to While d of SP is given 32 times in a row,
The D/A:I inverter 28a is given 0010011100 only during the period when d of sp is H, and the D/A converter 28a outputs ootooti to the charged signal amplification circuit 26.
A charge signal of a level corresponding to too is given.

As a result, a charging voltage pulse is applied to the charging electrode 6 in synchronization with SP d, and in a pattern in which 32 consecutive pulses appear, then a pause for 32 consecutive pulses, and then 32 consecutive pulses appear. Ru.

If these charging voltage pulses match the timing of ink particle formation, the ink particles will be charged, but if they do not match, the ink particles will not be charged. When charged, the ink particles fly in a charging pattern in which 32 consecutive charged ink particles are followed by 32 consecutive non-charged ink particles, which are then followed by 32 consecutive charged ink particles.

At this time, the bypass filter 3 of the charge detection circuit 30
At the output end of 0d, the potential is positive while 32 consecutive charged ink particles collide with the gutter 9, and the potential is negative while 32 consecutive non-charged ink particles collide with the gutter 9. 1
A signal with a period of 00/32 = 3.125 KHz appears. This signal is rectified by a half-wave rectifier 30e and integrated by an integrating circuit 30f.

When the integrated voltage exceeds the set value, the output of comparator 30g becomes H.
Inverts from to L.

The microprocessor 31 outputs the output P19 of the comparator 30g.
changes from H to L, then the data selectors 22c, 2
The 3-bit code outputted to 3b is assumed to bring about the proper charging timing that gives the proper charge, and is directly set to be output to the data selectors 22c and 23b, and the AND gate All is turned off to perform the first step IO or 1.
Proceed to step 2.

As mentioned above, set the 3-bit code 000 to the data selectors 22c and 23b, and open the AND gates (
If the output P19 of the comparator 30g does not invert from H to L after the timer is set and the timer times out, the microprocessor 31 will set the data selectors 22c and 23b to 001. Set the 3-bit control code to output, and similarly set the timer. Then, it waits for the output P19 of the comparator 30g to change from H to L. Data selector 22c.

When the 3-bit control code 001 is set in 23b.

The data selector 22c now receives the signal CHP shown in FIG.
The data selector 23b outputs b of .

Output e of signal SP. That is, the data selector 22
Both c and 23b are 1/8 of the last output signal.
Outputs a pulse whose phase is delayed by 00m5ec.

Except for the fact that the signal phase is delayed in this way, the operation of each part is the same as that of the data selector 22c described above.

This is the same as when outputting 000 to 23b.6 Then, when the output P19 of the comparator 30g changes from H to L, the microprocessor 31 selects the data selector 2.
The 3-bit code output to 2c and 23b is assumed to bring about the proper charging timing to give the proper charge, and is outputted as is to the data selectors 22c and 23b, and the AND gate All is turned off, and the ink viscosity adjustment (10 ) or proceed to recording control (11). When the timer times out while the output P19 of the comparator 30g remains at H, the microprocessor 31 sets the output of OIO to the data selectors 22c and 23b.

Similarly, the microprocessor 31 operates the comparator 30g.
As long as the output P19 of the data selector 22c is H, each time the timer times out.

The 3-bit control code given to 23b is updated.

As shown in FIG. 5, the pulse width of signals a to h of the signals sp output by the data selector 23b is 1/800 m5ec.
Since the phase is shifted by the pulse width from each other, while changing the 3-bit control code in the range of 000 to 111,
Each of a - h of P is selectively AND gate AIO
, and outputs one of the 3-bit control codes to the data selectors 22c and 23b, the ink particles become charged, and the comparator 30. output P19 is H
becomes L. The microprocessor 31 then ends the phase search, sets the 3-bit control code output to the data selectors 22c and 23b as is, closes (off) the AND gate All, and starts the next control step (
Proceed to step 10 or 12).

Next, ink viscosity adjustment (10) will be explained with reference to FIGS. 10a, 10b, and 11.

When proceeding to the ink viscosity adjustment, the microprocessor 31 clears the charge detection flag 17) and refers to the output P19 of the charge detection circuit 30 (18). Here, if P19=1 (no charge detected), the charge detection flag is cleared. 30 is in a standby state, the process proceeds to the next step 19, but if P19=O (charge detected), the output P19 of the charge detection circuit 30 becomes 1.
wait until Then, when P19=1, the process advances to the next step 19.

In step 19 - the microprocessor 31;

Enable interrupts on the port that receives the zero-crossing pulse. As a result, when a zero-cross pulse (0 level) arrives, the microprocessor 31 proceeds to interrupt processing (FIG. 11).

Now, when the interrupt enable is set, the microprocessor 31 updates the signal SR applied to the clear input terminal of the counter 37 (FIG. 2) from the clear instruction level O to the count enable level 1 (20), and the timing pulse generator 20 J
-, the set signal P26=1 is output to the flip-flop 20a (21), and the ON signal CE=1 is output to the AND gate 39 (FIG. 2) (22). Then, time counting is started (23). As a result, the counter 37
starts counting up the 133 KHz pulse C27, and when the zero time count value when the charging voltage is applied to the charging electrode 6 reaches 0.361 sec, the microprocessor 31
26 is returned to 0 (26). As a result, the microprocessor 31 maintains the high level H signal P for 0.36 m5ec.
26 (see FIG. 6) is output.

Flip-flop 20a of timing pulse generator 20
is 100/32=3, 125 K applied to the clock pulse input terminal GK after P26 becomes high level H.
Set when the Hz pulse is at low level L. As can be seen by referring to Figure 6, 100/32-3
．． Since the signal P26 is returned to L by the time the 125KHz pulse changes from L to H and becomes L again, 3.12
When the pulse of 5K)lz becomes low again, flip-flop 20a is reset and flip-flop 20b is set.

As a result, the output R27 of the AND gate 20c becomes 3.12 after the signal P26 becomes H, as shown in FIG.
In synchronization with the pulse of 5Kl(z, the high level H becomes high only during that period.32 ink particles are generated during this high level H period.The output R27(H ) is the OR gate R4 in the gate circuit 28.
through the OR gate RO-R3. This causes the D/A converter 28a to receive a code 0 indicating a predetermined charge level while exactly 32 ink particles are generated.
010011100 is applied. Therefore, 3
While two ink particles are being generated, the D/A converter 28a continuously applies a voltage at a predetermined level to the charge signal amplification circuit 26. This charges 32 consecutive ink particles.

After setting P26=0 (26), the microprocessor 31 starts counting the time again (27), and then 1
, wait for the elapse of 60m5ec (2B).

While waiting for 0.36 m5 ec to elapse after setting P26 to 1 (24) and while waiting for 1.60 m5 sec to elapse after resetting P26 to 0 (28),
The microprocessor 31 refers to the presence or absence of the charge detection flag (25.29).

The flight time T of the ink particles traveling straight along the distance from the charge detection electrode 6 to the gutter 9 is approximately 1 m5 ec.

Thirty-two consecutive charged ink particles fly and collide with the conductive gutter 9, and after the first one of the charged ink particles collides, the potential of the core wire of the shield wire 10 decreases in the negative direction. Initially, after all 32 charged ink particles collide, the potential rises and shows an approximately pulse-like change. Corresponding to this potential change, the output of the amplifier 30c changes in the positive direction in an approximately pulse-like manner, and the output of the amplifier 30c when it stands up in the positive direction.

That is, immediately after a series of 32 charged ink droplets collide with the conductive gutter 9, the zero-crossing pulse generator generates a zero-crossing pulse (0 level), and in response, the microprocessor 31 generates a zero-crossing pulse (0 level) as shown in FIG. Proceed to interrupt processing.

In the interrupt processing, first, the gate control signal CE to the AND gate 39 is reset to the off instruction level 0 (52
), this causes the counter 37 to stop counting up. Next, set the charge detection flag (53), read the count value of the counter 27, that is, the detection time (flight time) T, into the register T (54), and clear the counter 37 (reset SR to 0). Return to ink viscosity adjustment in Figure 10a.

When returning to ink viscosity adjustment, the charge detection flag is set, so this is detected in step 25 or 29, and the process proceeds to step 30, where temperature data t is read and set in register t (30).

Next, from among the threshold groups stored in advance at a predetermined address (threshold table) in the ROM 34, an address is specified based on the value of register t, and one pulse width upper limit threshold P is determined.
Read out wt and u, and read out the fixed lower limit threshold P wta (31), and calculate the flight time deviation Cc (32)
, C in Cc=C−Cs is the content T of register T, and C
s is target flight speed data (in units of T) that is input to the printing control device W129 from a keyboard or input device (not shown) and set in the memory.

Referring now to FIG. 10b, microprocessor 31
is the maximum allowable absolute value of data Cc C+a (fixed value)
Compared to 33), if Cc is greater than Cm, the ink jetting is abnormal, such as the ink particles are not charged or the ink is not being jetted, so the ink jetting is stopped and an alarm is set. Do (34). Thereafter, the device power remains in that state until it is turned off and turned on again.

If Cc is less than Cm, it is considered normal, and the deviation Cc is 2.
Check to see if it is within 2 (35). If it is within 2, the deviation Cc of the actual flight speed C from the target flight speed (flight time) Cs is within the allowable range. Since the current pulse width signal data Pd (standard value Pds for the first time) gives an appropriate ink velocity, the viscosity adjustment is completed and the process proceeds to recording (12).

If the deviation Cc exceeds 2, the ink speed reaches the target value Cs.
3.Please refer to the code of Cc.
6), according to the absolute value and sign, convert the absolute value into memory address data, access the memory 28, read out the pulse body signal correction value ΔVpd corresponding to the absolute value, and
Depending on the sign of c, when it is positive, Vpd=Vpd
+ΔVPd, and when negative, Vpd=Vpd−ΔV
The pulse width signal data Vpd to be updated and given to the pump driver 21 in the primary manner is obtained using pd (37.47).

A correction code table (memory area) is set in the ROM 33, and pulse width signal correction value data ΔVpd associated with the absolute value of the deviation Cc is stored in this table. It is designed to be accessed. The correction value ΔVpd is made linear with respect to the deviation Cc. Therefore, when the deviation Cc is large, the correction value data Δ is read out from the memory 28 based on it.
Vpd takes a large value, and when the deviation Cc is small, the correction value data ΔVpd read from the correction value code table based on it takes a small value.

Next 4: When calculating with Vpd=Vpd+ΔVpd, compared with V pd tt P vtu, VPd is P
If it is greater than or equal to vtu, the pump energizing pulse width will be outside the predetermined range, so the process proceeds to step 41 and subsequent steps to supply the diluent.

In supplying the diluent, first, the valve open signal 3 is sent to the pulp driver 34.
Set the output to 4S=1 to open the electromagnetic valve 14 (41), set the timer DT (program timer) (42), wait for the time to expire (43), and when the time expires, open the electromagnetic valve 14. Return to closed (34S=0) 44) Turn on timer LT (program timer) 4
5) Wait for the timeout (46).

Note that DT is the time required to supply a predetermined amount of diluent, and LT is the time required for the supplied diluent to be sufficiently stirred in the ink circulation system shown in FIG. When the timer LT times out, since the ink viscosity has changed, the phase search (9: FIG. 9) is executed once again, and then the process returns to the ink viscosity adjustment shown in FIG. 10a.

Calculate Vpd=Vpd+ΔVPd, and calculate Vpd as P vt
Compared with u, when Vpd is less than Pwtu, the pump energizing pulse width is within a predetermined range and there is no need to supply diluent, so the calculated Pv
Set (update) d to output 39), set the Ta timer 40), wait for the time to elapse (53), and when the time elapses, proceed to the phase search (9) in Figure 9, and after that, start the ink again. Return to viscosity adjustment. In addition, after changing the pump energizing pulse width, Ta is calculated as Vpd = Vpd - ΔVpd to the pressure determined by it.
Compare with pd'a:, P wts. Vpd is P
If it is less than vts, the energizing pulse width is within the predetermined range, so the calculated Pvd is output to the pump driver 21 and set (updated) L (39), and the Ta timer is set (40).
, the process waits for the time to elapse (53), and when the time elapses, the process proceeds to phase search (9) in FIG. 9, and after that, returns to ink viscosity adjustment.

Calculate Vpd=Vpd-AVpd and get VpdttPvta
As a result of comparison, when Vpd is less than Pwts, the energizing pulse width is outside the predetermined range, so ink cannot be ejected to bring about the set speed Cs, so the set speed C5t must be updated to a value that is one step lower. 49) to do,
This sets the target flight speed one step higher (
time is set one step shorter). Next, the Ta timer is set (50), the process waits for the time to elapse (51), and when the time elapses, the process proceeds to the phase search (9) in FIG. 9, and when that is completed, the process returns to ink viscosity adjustment.

During the first ink pressure adjustment using the ink pressure W11 of the ink viscosity adjustment described above.

The pump driver 21 has standard data (13
+wsecN equivalent count value), but after that,
New data calculated based on ink speed detection is given as Vpd. If the deviation Cc is large, the correction value ΔVPd is large, so as the above-mentioned ink pressure adjustment is repeated, the deviation Cc is initially large, but later on, the deviation becomes smaller in a geometric progression, and rapidly falls within the extremely small error range. Converge. Therefore, the time from the start to the end of ink pressure adjustment is shortened, and the convergence error Cc when Is is completed can be made extremely small. Therefore, when the pump energization pulse width exceeds a predetermined upper limit determined by the ink temperature during ink pressure adjustment, a diluent is supplied to the ink and the ink viscosity is lowered, so that the ink viscosity can be adjusted to bring about the desired ink velocity. Things deteriorate.

Next, recording control (12) will be explained. In recording control, the microprocessor 31 keeps the AND gate All off and the timing pulse generator 20 reset (R27=L), so that the output of the OR gate R4 of the gate circuit 28 is During control, it is maintained at a low level L, and only the outputs of the AND gates AO to A9 are applied to the D/A converter 28a. Microprocessor 31 then instructs deflection voltage generation circuit 27 to generate a deflection voltage. As a result, a predetermined constant high voltage is applied to the deflection electrode 7.

A reset signal is given to the charge code generator 22d immediately before sending out one character of printing data. When the charge code generator 22d receives the reset signal, it clears the count value and starts counting up from zero. The count pulse is 100 KH output by the pulse generator 19.
This is the pulse of z. The count code is an AND gate AO to A only when the printing data is at a high level H (recording instruction).
9, and this is the output C of the data selector 22c.
While HP (any one of a to h, specified by the 3-bit code given to the data selectors 22c and 23b) is at a high level H, the AND gate A is
It is applied to the D/A:I inverter 28a through O-A9.

The signal CHP output by the data selector 22c is the signal s P (
Among a to h, the one specified by the 3-bit code currently set to be output to the data selectors 22c, 23b) has an H pulse section approximately in the center, and has an H pulse section eight times as long as the H pulse section. Therefore, the recording charging pulse voltage applied to the charging electrode 6 has a relatively wide pulse width that extends from just before the ink separates into particles to immediately after the separation, and the ink particles reliably generate a charging code. @22
d to a level corresponding to the output code. While flying between the deflection electrodes, the charged ink particles are deflected by an amount corresponding to the charge they have and collide with the recording paper 8.

The uncharged ink particles travel straight and collide with the gutter 9, pass through the ground filter 11, and are sucked into the pump 12.

In this embodiment, as explained above, first, a phase search is performed to determine the phase of the charging signal at a timing to reliably charge the ink droplets; and by adjusting the ink viscosity, the ink flying speed is adjusted to the set target value Cs. The pump energization pulse width is controlled, and when this pulse width exceeds an upper limit determined by the ink temperature, a diluent is supplied to the ink, and the flying speed of the ink is set to a desired value; recording charging is then performed.

In this way, the ink viscosity is kept below the set value, the ink flying speed is kept constant at the target value, the ink particles are reliably and stably formed, and the charge is reliably determined, so that the printing quality is extremely high.

Since the pump energizing parameter is changed in accordance with the deviation between the ink droplet flying speed and the target value, the ink flying speed is adjusted in a geometric progression, and the time required to set the target value is short.

In ink viscosity adjustment and phase retrieval, the deflection voltage is cut off, charged ink particles are also allowed to travel straight and are captured by the gutter, and the charges on the ink particles are detected.

In printing, a deflection voltage is applied to a deflection electrode to deflect charged ink particles toward recording paper, and uncharged ink particles are captured by a gutter.

In this way, one conductive gutter is commonly used for three purposes: one-phase search for detecting ink flying speed, and capturing non-printing ink particles during recording and printing, so the above explanation As can be seen from the figure, the mechanical and electrical configurations of the inkjet recording apparatus are both simple. in spite of,
Both the detection of the ink flying speed and the phase search are reliable and stable.

The gutter 9 is easily soiled by ink droplets that collide with it, but since the core wire of the shielded wire 10 in the charge detection circuit 3o is grounded to the device through the resistor 30a, the ink recovery path is located at a predetermined position (filter 11). Since the equipment is grounded, charge detection is reliable even if the resistance value fluctuates due to ink stains on the gutter.

The flight speed is constant by controlling the energization pulse width of pump 1,
Further, since the ink particle size is controlled to be below a predetermined value by adjusting the viscosity, the particle size of the ink is stabilized and the printing quality is stabilized.

■Effects As described above, according to the present invention, it is possible to accurately detect the ink flight speed (= flight time) without adding special mechanical elements, and this also enables viscosity detection and speed detection).
High precision, accurate flight speed adjustment and viscosity adjustment are performed. Conventional (for example, Japanese Patent Application No. 112313/1989)
The viscosity is adjusted in smaller increments, and the range of viscosity fluctuation becomes smaller. Moreover, the adjustment accuracy becomes higher. Therefore, the particle formation of the ink is further stabilized, and the printing quality is further stabilized.

[Brief explanation of drawings]

FIG. 1a is a side view showing an outline of an ink processing system according to an embodiment of the present invention, and a portion thereof is shown in cross section. FIG. 1b is an enlarged sectional view of the pressure pump 1 shown in FIG. 1a, and FIG. 1c is an exploded perspective view of the inlet portion of the pressure pump 1. Figures 1d and 1e are enlarged sectional views showing ink pressure loss means used in another embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of the electrical system of the embodiment shown in FIG. 1a. FIG. 3 is an electric circuit diagram showing the configuration of a part of the electrical system elements shown in FIG. 2. FIG. 4 is an electric circuit diagram showing the configuration of another part of the electrical system elements shown in FIG. 2. Figure 5 shows the 31i! Recording charge signal generator 22 shown in H
3 is a time chart showing the output signals of the people in FIG. FIG. 6 is a time chart showing the output signal of the timing pulse generation @20 shown in FIG. 7a and 7b are time charts for explaining the operating characteristics of the pump 1. FIG. FIG. 8 is an electric circuit diagram showing the configuration of another part of the electrical system elements shown in FIG. 2. FIG. 9 is a flowchart showing an overview of the control operation of the microprocessor 31 shown in FIG. 10a and 10b are flowcharts showing the ink viscosity adjustment control operation of the microprocessor 31 shown in FIG. 2, and FIG. 11 is a blow chart showing the interrupt processing in the ink viscosity adjustment. Figure 12a is a graph showing the relationship between total ink ejection time and ink viscosity, Figure 12b is a graph showing the relationship between ink viscosity and ink flight speed, and Figure 12c is a graph showing the relationship between ink temperature and ink viscosity. The graph shown in FIG. 12d is a graph showing the relationship between the pump energization pulse width and the ink viscosity when the ink flying speed is constant. FIG. 13a is a graph showing the required ink pressure against the water evaporation rate in the ink in order to keep the ink flying speed constant in a conventional deflection control inkjet recording apparatus. FIG. 13b is a graph showing the efficiency of the pressure pump relative to the evaporation rate of water in the ink when the power of the pressure pump is adjusted to keep the ink flying speed constant in a conventional deflection control inkjet recording apparatus. FIG. 13c shows a deflection control inkjet recording device. This is a graph showing the power required for a pressurizing pump with respect to the water evaporation rate in the ink in order to keep the ink flying speed constant, and the solid line shows the conventional example, and the two-dot chain line shows an example of the present invention.

Claims (10)

[Claims] (1) Ink ejection head equipped with an ink ejection nozzle and a vibrator that applies periodic pressure vibrations to the ink in the ink chamber communicating with the ink ejection nozzle: Pressure pump that supplies pressurized ink to the ink ejection head: Ejected from the nozzle A charging electrode that applies a charging electric field to the charged ink particles: A charging voltage generating means that applies a charging voltage to the charged ink particles; A deflection electrode that applies a deflection electric field to the charged ink particles; A deflection control inkjet recording device comprising: a gutter that captures ink particles; energizing a pressure pump; and at least one of the pressure pump energization parameters, such as the energizing current of the pressure pump, the voltage, and their frequency. a pump driver that controls the ink dilution liquid storage container; a dilution liquid supply means that supplies the ink dilution liquid in the ink dilution liquid storage container to the ink supplied to the ink ejection head; a speed detection means that detects the flying speed of ink particles; The detected speed is compared with the set value, and if the detected speed is smaller than the set value, the pump will be energized with the higher pressure pump energizing parameter, and if the detected speed is greater than the set value, the pump will be energized with the lower pressure pump energizing parameter. an ink viscosity control means that instructs the pump driver to supply the diluent when the pump energization parameter that is instructed to the pump driver is outside a predetermined upper limit; A deflection control inkjet recording apparatus comprising: an ink pressure loss means in an ink flow path for increasing ink pressure loss in the flow path at a high rate of change in accordance with an increase in ink viscosity. (2) Further comprising a temperature sensor that detects the temperature of the ink supplied to the ink jet head or a temperature corresponding thereto; the ink viscosity control means selects an upper limit value corresponding to the detected temperature and instructs the pump driver. Said claim 1 comparing the pump energization parameters with this
Deflection control inkjet recording device as described in 2. (3) The deflection control inkjet recording apparatus according to claim 1 or 2, wherein the pump energization parameter is a pulse width of pulse energization. (4) The gutter is an electrical conductor; the speed detection means detects the charge from a charge detection circuit connected to the gutter and generating a signal corresponding to the collision of charged ink particles with the gutter, and by applying a charge voltage to the charge electrode. The deflection control inkjet recording apparatus according to claim 1 or 2, comprising time counting means for counting the time T until the circuit detects the charge. (5) The pump energization parameter is the pulse width of pulse energization, and the ink viscosity control means reads a parameter correction value from the memory based on the speed detected by the speed detection means, and corrects the pump energization parameter with the correction value. A deflection control inkjet recording apparatus according to claim (1) or (2). (6) The ink pressure loss means has a diameter of the ink suction port of 2 m.
Claim No.
The deflection control inkjet recording device according to item 1) or item (2). (7) The deflection according to claim 1 or 2, wherein the ink pressure loss means is an ink supply pipe with an inner diameter of 4 mm or less between the pressure pump and the ink tank. Control inkjet recording device. (8) The ink pressure loss means is the valve spring pressure 1 of the pressure pump.
．． Claim No. 1, which is a leaf valve weighing 5 kg or less,
The deflection control inkjet recording device according to item 1) or item (2). (9) The ink pressure loss means is a narrow diameter pipe line with an inner diameter smaller than the ink supply pipe between the pressure pump and the ink jet nozzle, as described in claim (1) or (2) above. deflection control inkjet recording device. (10) The ink pressure loss means is a filter that is inserted between the pressure pump and the ink jet nozzle and is provided with one more filter in addition to the filter that captures dust in the ink. The deflection control inkjet recording device according to item 1) or item (2).