GB2392873A - Operating ink jet valves during printing - Google Patents

Abstract

A method of operating an ink jet printer comprising the steps of (a) determining the time elapsed since the or each ink jet was previously activated and (b) increasing the time for which the or each ink jet is subsequently activated, the increase in time being determined in accordance with the elapsed time determined in step (a). The invention further provides A method of operating an ink jet printer comprising the steps of determining a valve activation time; determining an elapsed time variable since the ink jet was previously activated and a first and a second threshold value associated with the elapsed time variable; setting the elapsed time variable to zero; decrementing the valve activation time by the first threshold value and if a valve activation command is not received: increasing the elapsed time variable by a first increment if the elapsed time variable is less than the first threshold value; increasing the elapsed time variable by a second increment if the elapsed time variable is less than the second threshold value; or if a valve activation command is received, incrementing the valve activation time with the elapsed time variable and activating the valve accordingly.

Description

- 1 - DEVICE AND METHOD

The present invention relates to a device that can be used to control the operation of a print head and to a method 5 for controlling a print head under operation.

Ink jet printers are non-contact printers in which dots of ink are ejected from one or more nozzle orifices so as progressively to build up a printed image on a substrate 10 moved relative to the nozzle. One form of ink jet printer comprises a source of ink under pressure, typically a reservoir or bottle of ink which is pressurized to from 0.1 to 2 bar, notably about 1 bar. The pressure is created, for example, by pressurizing the air space above 15 the ink in the bottle or reservoir from which ink is fed to the nozzle orifice(s) in a print head through which it is ejected as a series of droplets onto the surface of the substrate. The flow of ink through the each nozzle orifice is controlled by a solenoid valve. Typically, 20 such a valve comprises an electromagnetic plunger journalled for axial movement within an axially extending electric coil. The distal end of the plunger is located within a valve head chamber through which ink flows from the reservoir to the nozzle orifice. When current is fed 25 through the coil, this generates a magnetic field which

acts on the plunger to move it axially and thus open, or shut, the inlet to nozzle orifice. Typically, the magnetic field acts to retract the plunger against the

bias of a coil spring to create a flow path between the 30 valve head chamber and the nozzle orifice. When the electric current no longer flows in the coil, the magnetic field ceases and the plunger returns under the bias of the

- 2 - spring to seat against sealing ribs, lips or other means located at or around the inlet to a bore leading to the nozzle orifice to close the flow path to the nozzle orifice. For convenience, the term drop on demand printer 5 will be used to denote in general such types of ink jet printer. Conventional ink jet print heads have employed electro-

mechanical control and actuation systems that open the 10 valve for a predetermined period of time so that an ink drop can be ejected. The time for which the valve is held open determines the quantity of ink that is ejected from the valve and hence the size of the drop that will be formed on the substrate that is being printed upon.

15 Adjustment of the valve open time is time consuming and laborious as typically a manual adjustment must be made to each valve within the print head matrix.

A preferred embodiment of the invention and its operation 20 under online software control will now be described by way of illustration only and with respect to the accompanying drawings, in which Figure 1 shows a schematic depiction of a solenoid 25 valve which is suitable for use with the method of the present invention; Figure 2 shows a schematic depiction of a printer apparatus that is operated according to the present invention; 30 Figure 3 shows a graphical depiction of low quality print effects;

- 3 - Figure 4 shows a schematic depiction a method according to the present invention; Figure 5 shows a graphical depiction of a further low quality print effects; 5 Figure 6 shows a schematic depiction a second method according to the present invention; Figure 7 shows a first schematic depiction of a preferred embodiment of printer apparatus that is operated according to the present invention; 10 Figure 8 shows a second schematic depiction of a preferred embodiment of printer apparatus that is operated according to the present invention; and Figure 9 shows a third schematic depiction of a preferred embodiment of printer apparatus that is 15 operated according to the present invention; Figure 1 shows a schematic depiction of a solenoid valve 20 10 which is suitable for use with the method of the present invention. The valve 10 comprises plunger 20, tube 30 and coils 40. The plunger 20 comprises a ferromagnetic material (or any other magnetic material) and is received within the tube 30 so as to be able to 25 move freely along the axis of the tube. The plunger can be impelled, for example towards the open end of the tube, by the application of a current to the coils 40, the current generating a magnetic field within the tube, which

causes a magneto motive force to act upon the plunger.

30 The timing and frequency of the current pulses applied to the coils can be controlled by computer (not shown). The solenoid valve additionally comprises a return mechanism

- 4 - (not shown), such as a spring, that acts to return the plunger to its initial position once the plunger has completed its full range of travel.

5 In practice, a print head will comprise a matrix of such valves that are arranged in a square or rectangular arrangement. Figure 2 shows two exemplary valves 210a, 210b from such a print head matrix 220. Associated with each valve is valve control means 215a, 215b, each of the 10 valve control means being in communication with a central computer system 230. The operation of each valve is controlled by the transmission of control pulses from the central computer system 230 to each of the valve control means 215a, 215b. Rather than transmitting a single pulse 15 to 'fire' the valves for a pre-determined period of time, the central computer system can transmit more complex signals that are interpreted within the respective valve control systems in order to control the behaviour of the valve. For example, the signal may be byte wide and if 20 the value is within a certain range, for example from 25-

255 then this may indicate that the valve be held open for a time that is proportional to the value of the signal, for example for 25-255ps. Certain values of the signal may cause the valve to be held open for a pre-determined 25 period of time, with that time period being calculated or retrieved from memory by the valve control means in accordance with the value of the signal. Certain signal values may also be used to initiate other actions from the valve, for example reporting back a parameter associated 30 with the valve, such as the volume of ink consumed, or as a prompt to the valve control means that new or updated control data is to be transmitted to the valve control

- 5 means that is to be stored in memory within the valve control means. If greater than 256 values are required to provide all of the control signals then the size of the control signal may be increased. If it is desired to 5 transmit individual signals to each valve then the signals may be transmitted using some form of time or frequency division multiplexing. Alternatively additional bits may be added to the control signals so that valves may be addressed individually, or as blocks forming a sub-set of 10 the printer head matrix. Similarly, the time for which each valve is to be held open may be altered by a constant time period, for example 1 or 10 Us, in order to produce spot sizes of a slightly different sizes.

15 It has been observed that when a valve has not been operated for a long period of time (that is, a few seconds or longer) then the drop that is formed when the valve is next operated tends to have less than ideal properties (see Figure 3, and in particular the leading edge of the 20 'W' character). This effect is thought to be due to a part of the previous ink droplet remaining in the nozzle and forming a blockage that must be overcome by the next droplet. This effect can be lessened by operating a timer for each valve that effectively measures the time between 25 successive operations of the valve and then increasing the time for which the valve is held open in accordance with the time value that is measured.

Figure 4 shows a flowchart indicating a method by which 30 the valve open time can be varied in response to the idle time of the valve. Each valve has associated with it a valve_on variable that can be varied in order to change

- 6 the time for which the valve will be held open. In step 300 the idle_time variable is set to zero. If the valve is not operated at step 310 then the idle_time variable is compared with a max_idle_time constant at step 340. If 5 idle_time is less than the maximum value then its value will be incremented and the process will return to step 310. If the idle_time variable is already at its maximum value then the process will return to step 310. If the valve is to be opened at step 310 then the idle_time value 10 will be added to the valve_on variable in step 320 and this increased valve_on value will be used in step 330 to control the time for which the valve is held open. Once the valve has been opened the process returns to step 300 in order to reset the idle timer to zero.

In an alternative process, when the step 310 indicates that the valve is to be opened, rather than adding the idle_time value to the valve _on value, the idle_time value is used to select a scalar value from a lookup table.

20 The valve-on value is multiplied with the selected scalar value in step 320 in order to generate the value that will be used in step 330 to control the time for which the valve is held open. This method is generally preferable to the addition method described above as it enables a 25 proportional increase in the valve-on time. The use of the max_idle_time constant limits the increase in valve opening to that required to provide a suitable drop formation on the substrate. Typically, the valve_on variable would be increased by a maximum of either X ps or 30 Y%.

- 7 - The present inventors have also observed that as the rate at which a valve is operated, a greater volume of ink droplet is formed for a given open time for the valve.

Figure 5 shows a pictorial representation of this effect, 5 with Figure 5a showing a test message printed at a valve rate of 400 Hz and Figure 5b showing the same test message which was printed using a valve rate of 4.6 kHz. The definition of the characters has been significantly reduced and there are a large number of satellite dots 10 that have been formed. It is believed that the inductive effects of the coils in the valves causes an inertial effect that prevents the valve from fully closing. It is further surmised that the inductance of the coil causes there to be a residual current flowing in the coil after 15 the valve should have closed. When the valve is next actuated the current developed in that printing pulse will be larger than is required, causing the valve to be held open for a longer period of time. It follows that the return mechanism will take a longer period of time to 20 close the valve, leading to a larger ink droplet being expelled. One potential solution to this problem is to limit the rate at which the valve is operated, although this may prevent the use of the print head in certain high-speed operations.

A preferred solution is to is to reduce the time for which the valve is held open to reduce the volume of ink that is ejected in each drop. As the rate at which the valve is operated increases the valve open time can be reduced, 30 either by subtracting a fixed amount of time or by reducing the valve open time by a fixed proportion. The values by which the open time can be reduced may be

- 8 calculated by the valve control means or accessed from a look-up table. Different values for reducing the valve open time may be used as the valve operating rate changes, as a greater reduction in valve opening times may be 5 required with greater valve operating rates.

This correction technique may also be utilised in conjunction with the technique described above with reference to Figure 3. Each valve is associated with four 10 variables: valve_open, which controls the time for which the valve is opened, an idle_time variable, which is in turn associated with both a short_term_max value and a long_term_max value. The process will be described with reference to Figure 6, in which valve_open is decremented 15 by short_term_max in step 400 and idle_time is set to zero in step 410. If an instruction is received at step 420 to operate the valve then valve_open is incremented by idle_time at step 430 and this value of valve_open is used in step 440 to operate the valve. The process returns to 20 step 410 to reset idle_time to zero.

If there is no instruction at step 420 to operate the valve then at step 450 idle_time is compared with short_term_max and is incremented at step 460 if it is 25 less than short_term_max. The process then returns to step 420. If idle_time is equal to or greater than short_term max then idle_time will be compared with long_term_max at step 470. If idle_time is less than short_term_max then idle_time will be incremented at step 30 480 and the process returns to step 420. If idle_time is not less than long_term_max then the process returns to step 420. In a preferred embodiment of the process,

- 9 - idle_time is incremented at a greater rate at step 460 than at 480 [what is the reason for this?l The value of the valve_open variable is selected in order 5 to provide a desired size of drop that is to be deposited onto the substrate, absent any additional correction factors. The correction obtained by decrementing valve_open by short_term_max is intended to reduce the effects caused by the valve being held open for too long.

10 The correction obtained by incrementing valve_open by idle_time is intended to reduce the effects that are caused when a valve is not operated for a prolonged period of time.

15 Typical values for these variables are valve_open = 150ps, short_term_max=50ps and long_term_max = 80ps [at what valve rate are these figures for?l although it will be understood that these values may change as the operating rate of the valve changes. Appropriate values may be 20 calculated dependent upon the operating rat of the valve, or may be accessed from look-up tables.

The valve control means must be able to receive, interpret and execute signals that are received from the central 25 computer system. It will be readily understood that the valve control means may be implemented such that each valve has a dedicated control means or alternatively that a number of valves may be controlled by a single control means. In a preferred embodiment, the valve control means comprise a field programmable gate array (FPGA). FPGAs

- 10 comprise memory and logic elements that can be configured by the user to provide a desired functionality.

In the preferred embodiment, the FPGA, and associated 5 devices, is used to control a linear array of 16 valves.

Referring to Figure 7, the valves 610a, 610b,..., 610p are controlled by valve control means that comprise FPGA 616, electrically erasable programmable ROM (EEPROM) 617, RAM 618, programmable ROM (PROM) 619 and input/outputs 622, 10 624, 626. The FPGA 616 is connected to each of the valves 610a, 610b,..., 610p, EEPROM 617, RAM 618 & PROM 619. All three input/outputs 622, 624, 626 interface with the FPGA.

When the FPGA is powered up, it loads its internal configuration data from PROM 619 and then follows the 15 sequences that have been loaded from the PROM. The EEPROM 617 stores a range of data comprising a look-up table comprising data associated with each of the valves, data specific to the valve control means and FPGA, status information, etc. The FPGA will load this data from the 20 EEPROM and then initialize the RAM 618, by writing zero values into each memory location in RAM. The FPGA will then wait to receive print data or other commands from one of the inputs. Input/output 622 is connected to the computer control system and input/output 624 can be used 25 to connect to a further valve control means (see below with reference to Figure 9). Input 626 provides a series of pulses that are used in co-ordinating the printing process. When the array of valves is printing onto a substrate, the substrate is normally moved underneath the 30 valves. The series of pulses supplied to input 626 may be generated from an encoder applied to a shaft in the

apparatus that is moving the substrate relative to the valves. Figure 8 shows a schematic depiction of a number of 5 registers that are formed with the FPGA when the FPGA configuration data is loaded from PROM 619. The first register 631 is used to write to and read from the EEPROM 617 and is also used when initialization data is read from the EEPROM. Second register 632 receives print data from 10 the computer control system, such as the alphanumeric characters or bitmaps to be printed, or a signal to initiate a printing process. Second register 632 also writes print data to the RAM and is used to initialize the RAM during the start-up phase. The third register 15 receives configuration data from the computer control system such as data controlling the slant that may be applied to the print head. Fourth register 634 receives print data from the RAM and passes it to the fifth register 635, which uses the print data to operate the 20 valves 610.

A desired print image (which may include alphanumerical characters) is entered into the computer control system and this image is then converted into raster data that may 25 be communicated with the valve control means. The valves 610 may be operated for different periods of time so as to provide the appearance of 16-level greyscale images. Thus the print data can be supplied in the form of a raster comprising a 4 bit word for each valve, with the value of 30 the 4-bit word determining the greyscale that is to be generated by the valves. The print data is received by the second register and written into the RAM 618. The RAM

- 12 is logically arranged in 16 rows, with each of the valves corresponding to a row. There are a plurality of columns, each of which corresponds to a time slot. Each raster scan also corresponds to a time slot and the time slot is 5 determined by the frequency at which the shaft encoder supplies pulses to the FPGA.

When print data is received at the FPGA the second register interprets the greyscale data for each valve, 10 obtaining the time that each valve must be opened for in order to generate the desired greyscale from a look- up table held in the first register. In theory, each valve should be held open for the same period of time in order to generate the dame greyscale, but mechanical variations 15 in each valve will lead to each valve having slightly different characteristics. Calibration factors that account for these differences are held in the look-up table. The valve times are then written into the RAM, using as many columns as are necessary to store all of the 20 rasters. A write pointer is set to the first column of the data. Each memory location holds the grey scale value for the associated valve and time slot.

When the next shaft encoder pulse is received the RAM 25 column indicated by the write pointer is read to see which of the 16 valves need to be operated, i.e. which memory locations have non-zero entries. Once the memory locations have been read then all the memory locations in the column are overwritten with zero.

The identity of these valves, along with the time for which the valves are to be held open are then transmitted

- 13 to the fourth register, which may perform further operations on the valve times in order to correct for valve operation at high speed or a long time period between subsequent operations of the valve. The valve 5 times are then passed to the fifth register which calculates the number of shaft encoder pulses that are equivalent to the valve times. The valves are then opened for a period of time equal to that number of shaft encoder pulses. As the valves 610 are electro-mechanical devices, their size provides a limitation to the print resolution that can be obtained. Typically, each valve may be provided at an offset of 4mm from the adjacent valve(s). If a 15 greater resolution (i.e. smaller pixel separation is required) then the matrix may be slanted so that the valves are closer together in one axis. The disadvantage of this is that if no correction is made to the print rasters then the desired image will be printed out 20 slanted.

Such a correction may advantageously be provided using the RAM to provide a slant to the print raster data. Once the greyscale data has been translated into valve open times, 25 rather than writing the valve data into a vertical column, the write data can be offset across a number of columns.

For example, if the desired slant angle is 45 then the valve open time for the first valve should be written into the column indicated by the write pointer, the valve open 30 time for the second valve should be written into the next column along from the column indicated by the write

- 14 pointer, and so on, such that the valve open time is written into the RAM at the desired slant angle.

Typically the 16-level greyscale can be provided using 5 valve open times between approximately 80s and 250ps. It has been found advantageous to initially open the valve by providing a first voltage for a first period of time and to provide a second voltage, that is lower than the first voltage, for a further period of time in order to hold the 10 valve open. This reduces the possibility that the valve remains open for longer than is required to provide the desired greyscale, leading to decreased printing performance. It has been found particularly advantageous to apply a 36V pulse for approximately 80ps and a second 15 pulse of approximately 5V for the remainder of the time that the valve remains open.

In a further preferred embodiment, the valve control means and valves described above with reference to Figure 8 will 20 be co-located upon a single circuit board 650. A number of circuit boards can then be connected in serial and physically located in a vertical array so that the valves can deposit a two-dimensional matrix on a print substrate.

In such a case (see Figure 9), one of the boards 650a will 25 be connected via serial input/output 622 to the computer control system 230 and to the second board via serial input/output 624. The second board 650b will be connected to the first board via serial input/output 622 and to the third board 650c via serial input/output 624, and so on.

30 The last board in the serial chain can detect its position as its serial input/output 624 will have no connection.

On power up the last board in the serial chain assigns

- 15 itself address 0 and transmits this address to the preceding board, which then assigns itself address 1.

This process continues, with the address value being incremented until each board has an assigned address. The 5 first board 650a will then report its address to the computer control system such that the system is aware of the number of connected boards. The system will prefix any communication with a board with the board's address.

Preferably 16 boards are connected together to provide a 10 16 x 16 printing matrix.

The FPGA used in the preferred embodiment was a Xilinx Spartan II XC2S100 which was preferred as its configuration was determined by the data loaded from the 15 PROM in start up. Such an FPGA may be replaced by a cheaper device in which the FPGA is hardwired, for example by blowing fuses to form logic elements, rather than configurable through software.

20 It will be understood that the present invention is suitable for use with any type of solenoid valve and in any application in which solenoid valves are used. The applicant has found that the invention is of particular advantage when used with high speed solenoid valves that 25 find advantage in drop on demand ink jet printers.

Specifically, the invention is of advantage when used with the high speed solenoid valve described in our copending application GB 0203439.5.

Claims (11)

- 16 CLAIMS

1. A method of operating an ink jet printer, the printer 5 comprising one or more ink jets, the method comprising the steps of (a) determining the time elapsed since the or each ink jet was previously activated and (b) increasing the time for which the or each ink 10 jet is subsequently activated, the increase in time being determined in accordance with the elapsed time determined in step (a).

2. A method according to claim 1, wherein the activation 15 time of the or each ink jet is increased by the elapsed time determined in step (a).

3. A method according to claim 1, wherein the activation time of the or each ink jet is increased by multiplying 20 the activation time by a scalar value, the scalar value being determined in accordance with the elapsed time determined in step (a).

4. A method according to claim 3, wherein the scalar 25 value is determined from a look-up table in accordance with the elapsed time determined in step (a).

5. A method according to any preceding claim, wherein the activation time is not increased above a maximum 3 0 value.

- 17

6. A method of operating an ink jet printer, the printer comprising an ink jet valve, the method comprising the steps of (a) determining a valve activation time; 5 (b) determining an elapsed time variable since the ink jet was previously activated and a first and a second threshold value associated with the elapsed time variable; (c) setting the elapsed time variable to zero; (d) decrementing the valve activation time by the 10 first threshold value and (e) if a valve activation command is not received: (i) increasing the elapsed time variable by a first increment if the elapsed time variable is less than the first threshold value; 15 (ii) increasing the elapsed time variable by a second increment if the elapsed time variable is less than the second threshold value; or (f) if a valve activation command is received, incrementing the valve activation time with the elapsed 20 time variable and activating the valve accordingly.

7. A method of operating an ink jet printer, in which the ink jet printer comprises a plurality of ink jet valves.

8. A method according to claim 7, wherein the first increment is greater than the second increment.

9. A method according to any of claims 6 to 8, wherein 30 one or more of the valve activation time, the first threshold value, the second threshold value the first

- 18 increment or the second increment are determined accordingly to the operating rate of the valve.

10. An ink jet printer for performing a method according 5 to any of the preceding claims.

11. An ink jet printer according to claim 10, wherein the ink jet printer comprises computing means to control the operation of the ink jet printer.