Sub-row firing method for single-pass monochrome printing at high speeds

Abstract

A method of printing an image from a printhead module having a plurality of horizontal ink planes M supplied with a same ink. Each ink plane has a nozzle row and the nozzles rows of all ink planes have vertically aligned nozzles. The method includes the steps of: defining contiguous span groups along each nozzle row, each span group containing N nozzles; allocating dot data for each image line of the image to a predetermined number of nozzles P in each span group of each nozzle row; sending the dot data to the printhead module and firing nozzles sequentially from the ink planes to print the image line. Only one nozzle from each span group in a same nozzle row is fired simultaneously, N is an integer multiple of M, and P is N divided by M.

Claims

1. A method of printing an image from a printhead module having a plurality of horizontal ink planes M supplied with a same ink, each ink plane having at least one nozzle row, the nozzles rows of all ink planes having vertically aligned nozzles, the method comprising the steps of: defining contiguous span groups along each nozzle row, each span group containing N nozzles; allocating dot data for each image line of the image to a predetermined number of nozzles P in each span group of each nozzle row; sending the dot data to the printhead module and firing nozzles, based on the dot data, sequentially from each of the M ink planes to print the image line of the image such that all ink planes contribute dots to the printed image line, wherein: only one nozzle from each span group in a same nozzle row is fired simultaneously; N is an integer multiple of M; and P is N divided by M.

2. The method of claim 1, wherein each ink plane comprises a pair of nozzle rows.

3. The method of claim 2, wherein the pair of nozzle rows are offset for printing even and odd dots.

4. The method of claim 1, wherein said method is repeated for printing all image lines of the image.

5. The method of claim 1, wherein span groups of different nozzle rows having different firing nozzles.

6. The method of claim 5, wherein the firing nozzles in the span groups of consecutively fired nozzle rows are horizontally shifted by S nozzles.

7. The method of claim 6, wherein S is 1.

8. The method of claim 1, wherein 1/M.sup.th of the image line is printable by each ink plane.

9. The method of claim 1, wherein the dot data comprises a ‘1’ for an enabled firing nozzle and a ‘0’ for a non-enabled non-firing nozzle.

10. The method of claim 1, wherein all aligned nozzle rows in the M ink planes are fired, based on the dot data, within one row-time, and wherein one row-time is less than or equal to a time period for firing all nozzles in the printhead module divided by the number of nozzle rows.

11. The method of claim 1, wherein one or more steps of said method are repeated to print all image lines of the image.

12. The method of claim 1, wherein the dot data is allocated to a given nozzle row based on a print speed and a position of print media during the sequential firing of nozzles from each of the M ink planes.

13. The method of claim 1, wherein corresponding span groups in different nozzle rows are vertically aligned.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the present invention will now be described by way of example only with reference to the accompanying drawings, in which:

(2) FIG. 1 shows a print chip having a dropped nozzle region;

(3) FIG. 2 is a magnified view of the dropped nozzle region;

(4) FIG. 3 is a schematic view of a printhead having multiple butting print chips;

(5) FIG. 4 shows dot placement errors resulting from dropped nozzle region artefacts in various print modes;

(6) FIG. 5 shows logical distribution of dot data in the print chip;

(7) FIG. 6 shows a physical layout of selected features of the print chip; and

(8) FIGS. 7A and 7B are simulated test prints using printing methods described herein.

DETAILED DESCRIPTION OF THE INVENTION

(9) Referring to FIG. 1, the printing methods described herein employ printhead modules, typically in the form of print chips as described in, for example, U.S. Pat. No. 7,290,852. Accordingly, each print chip comprises horizontal rows of nozzles extending parallel with a longitudinal axis of the print chip. Each nozzle row has a main row portion and a corresponding displaced (“dropped”) row portion, which is vertically offset from its main row portion.

(10) For the sake of convenience, the print chip is defined to have a nominal horizontal axis extending parallel with its length dimension and a nominal vertical axis extending perpendicular to the horizontal axis. As used herein, the terms “horizontal” and “vertical” are not intended to limit the orientation of print chips or nozzles rows in use. Furthermore, the term “dropped” (e.g. “dropped row portion”, “dropped nozzle region” etc) is not intended to limit the orientation of the print chip relative to a media feed direction—a “dropped row portion” merely means that a row portion is displaced, either upstream or downstream relative to a media feed direction, of a corresponding main row portion

(11) Nozzles in the main row portion extend along a majority of the length of the print chip, while nozzles in the dropped row portion are positioned at one end of the print chip. The total number of nozzles in each main row portion and corresponding dropped row portion is the same for all nozzle rows (e.g. 640 nozzles per row). However, the dropped row portions each have different lengths and, as shown in FIGS. 1 and 2, are together arranged in a generally trapezoidal shape in plan view. The multiple dropped row portions having a trapezoidal shape are together referred to as a “dropped nozzle region” of the print chip.

(12) The print chip shown in FIGS. 1 and 2 contains five ink planes, which are all supplied with a same color of ink for monochrome printing. Each ink plane contains two nozzle rows (“odd” and “even”) horizontally offset from each other by 1 dot pitch. Since the nozzles within the same nozzle row are spaced apart by 2 dot pitches, then the odd and even nozzle rows in one ink plane can print odd and even dots in one line of print. In the embodiment shown, the odd and even nozzle rows within the same ink plane are vertically offset from each other by 4 dot pitches, while each dropped row portion is offset from its corresponding main row portion by 10 dot pitches (at a nominal 1600 dpi).

(13) While one embodiment is described herein with reference to a Memjet print chip printing at a nominal 1600 (horizontal)×1600 (vertical) dpi, it will of course be appreciated that the present invention is not limited by way of print resolution or print speed.

(14) As best seen in FIG. 2, each dropped row portion is positioned to align horizontally with its corresponding main row portion such that a constant dot pitch is effectively maintained both along the print chip and between neighboring print chips. In this way, the dropped row portions can, in principle, compensate for printing in the join regions between neighboring print chips where nozzles cannot be fabricated due to a lack of available silicon at the edges of the print chips. Nevertheless, due to the problems foreshadowed above, the print chip described U.S. Pat. No. 7,290,852 is not ideally suited for fast printing (e.g. at a nominal 5× print speed) in monochrome for all printing resolutions. For example, as explained above and with reference to FIG. 4, when printing in monochrome at 1200 dpi and 5× print speed, errors of 2.5 DP occur between the main nozzle region and the dropped nozzle region. This error produces noticeable artefacts on the printed page.

(15) Independent Firing of Dropped Nozzle Region

(16) Typically, an inkjet printhead receives its dot data and fires its nozzles row-by-row to eject droplets. A given nozzle device of a row will fire if both a row enable signal and a column enable signal are set to 1 on receipt of a fire signal. In one fire cycle of the print chip, all nozzle rows receive a fire signal within a fire cycle time such that all enabled nozzle devices of the print chip are fired. For a given number of nozzle rows, the fire cycle time is limited by the maximum ejection frequency of each nozzle device—a physical limitation due to the maximum refill rate of each nozzle device.

(17) Within each fire cycle, each nozzle row has an allocated row cycle time, which is the fire cycle time divided by the number of nozzle rows. For dot-on-dot printing (e.g. CMYKK printing), the fire cycle must be completed within one line-time—that is, the time taken for the media to advance by one line or one vertical dot pitch (nominally 1600 dpi for a Memjet® print chip). Memjet® print chip has five ink planes and ten nozzle rows (one pair of nozzle rows, even and odd, per ink plane). Each nozzle row is allocated 1/10.sup.th of the line time to fire its nozzles at a predetermined print speed (nominally 12 inches per second). When printing in monochrome at 5× print speed (nominally 60 inches per second), the media necessarily advances by 5 lines (or 5 vertical dot pitches) during one fire cycle. In other words, only two rows of nozzles are able to print in the time taken for the media to advance by one dot pitch at a nominal 1600 dpi. This leads to droplet placement errors for certain print modes, such 400 dpi and 1200 dpi printing at 5× print speed.

(18) In order to address the problems foreshadowed above when printing in monochrome at 5× printing speed, the print chip according to the present invention is configured to fire nozzles in the dropped nozzle region independently of nozzles in the main nozzle region. Decoupling the firing of nozzle rows in the dropped nozzle region from those in the corresponding main nozzle region enables droplets fired from the dropped nozzle region to align perfectly with droplets fired from the main nozzle region irrespective of the print speed and print resolution.

(19) Hitherto, print chips known in the prior art fired nozzles on a row-by-row basis with all enabled nozzles in the same row firing within an allocated row-time. (In practice, all enabled nozzles in the same row are not fired simultaneously within their allocated row-time due to power constraints. As described in U.S. Pat. No. 7,780,256, the contents of which are incorporated herein by reference, the enabled nozzles are fired in span groups separated by a predetermined ‘span’ and firing is sequenced according to a predetermined ‘shift’ within each span group).

(20) Therefore, independent firing of nozzles from the “same” nozzle row presents challenges both in terms of implementation and chip design. Simplistically, the print chip could be treated as having 20 nozzle rows—10 nozzle rows in the main nozzle region and 10 nozzle rows in the dropped row region. Dot data and fire signals could then be sent to the print chip in 20 separate data pulses in sequence. However, this type of implementation is problematic, because the data pulses would contain non-equal amounts of data. Those data pulses corresponding to the main nozzle region will contain much larger amounts of data than those data pulses corresponding to the dropped nozzle region. And even within the dropped nozzle region and main nozzle region, each nozzle row has a different number of nozzles requiring different amounts of data. However, data transfer should ideally be as smooth as possible with a same data allocation for each data pulse.

(21) Referring to the FIGS. 5 and 6, the print chip 20 according to one embodiment of the present invention is designed to receive dot data and fire signals on a row-by-row basis for each of 10 nozzle rows—that is, each data pulse for each nozzle row contains dot data for the main nozzle region and the dropped nozzle region, such that the data pulses contain an equal amount of data (e.g. 640 bits corresponding to 640 nozzles in each nozzle row). However, second dot data associated with the dropped nozzle region 11 is routed separately from first dot data associated with the main nozzle region 13 by a command unit 22 of the chip. Whereas the command unit 22 sends first dot data associated with the main nozzle region 13 directly to corresponding first data latches 24, second dot data associated with the dropped nozzle region 11 is routed to a dedicated buffer 26. The buffer 26 has a data capacity corresponding to the number of nozzles in the dropped nozzle region 11.

(22) The second dot data stored in the buffer 26 is transferred to second data latches 28 corresponding to the dropped nozzle region 11 only after a predetermined delay retrieved from a dedicated delay register of the command unit 22. The value of the predetermined delay stored in the delay register is configurable based on the print job, and may be set by an upstream print controller (not shown) at the start of each print job based on the print speed and print resolution. In this way, dot data for the same line of print can be transferred to the print chip 20 simultaneously in one data pulse, whilst firing of droplets in the dropped nozzle region 11 is delayed relative to the those in the main nozzle region 13. Since the delay is determined by the print speed and print resolution, unlike the print chip 1 described in U.S. Pat. No. 7,290,852, nozzle rows 3 in the dropped nozzle region 11 may be fired at different times to nozzle rows in the main nozzle region 13, and not necessarily at the same time as any other nozzle row in the main nozzle region.

(23) The order in which nozzles rows 3 are fired is determined based on optimal dot placement and minimum error for a given print resolution and print speed. The row firing order is determined by the print controller communicating with the print chip 20.

(24) Referring to FIG. 6, the print chip 20 has a physical layout and architecture configured for efficient use of available space on the chip. The first and second data latches 24 and 28 corresponding to the main nozzle region 13 and dropped nozzle region 11 are positioned at opposites sides of their respective nozzle arrays. Data and power are received via a row of bond pads 30 positioned along one longitudinal side of the print chip 20 opposite the first data latches 24.

(25) The second data latches 28, which receive second dot data via the buffer 26, are positioned along a trailing row of the dropped nozzle region 11—that is, a longer side of the trapezoidal dropped nozzle region. The first data latches 24, which receive first dot data directly from the command unit 22, are positioned along a leading row of the main nozzle region 13. By positioning the second data latches 28 opposite the first data latches 24, conductive traces can extending from the second data latches across the print chip 20 towards the nozzle devices without fanning outwards from single point. This arrangement therefore avoids a high concentration of current in one region of the chip.

(26) FIGS. 7A and 7B are simulated test prints showing the effect of independent firing of the dropped nozzle region when printing at 400 dpi at a nominal 5× print speed. In FIG. 7A, using the method described in U.S. Pat. No. 7,290,852, the join region between two neighboring print chips is visible as a hump due to imperfect dot placement in the dropped nozzle region. However, as shown in FIG. 7B, with independent firing of the dropped nozzle region, the join region is not visible in the same print mode.

(27) Sub-Row Firing

(28) Ideally, all nozzles contained in the main row portion of a given nozzle row should be fired simultaneously; and the same is also true of nozzles in the dropped row portion. Simultaneous firing of nozzles would ensure that all droplets corresponding to the same image line land on the passing media simultaneously. However, in practice, and as explained in U.S. Pat. No. 7,780,256, it is impossible to fire all enabled nozzles simultaneously, because print chips have inherent power constraints.

(29) For this reason, nozzles are logically grouped into contiguous span groups, with the number of nozzles in each span group defining a ‘span’. Only one nozzle from each span group can be fired simultaneously and once these nozzles have fired then a subsequent nozzle is selected from each span group for firing. For example, with span of 20, a print chip having 640 nozzles in each nozzle row contains 32 contiguous span groups (each containing 20 nozzles) and every 20.sup.th nozzle can fire simultaneously. Thus, in this example, each nozzle row has 20 firing cycles within its allotted row-time.

(30) The distance of the subsequently fired nozzle from the previously fired nozzle in the same span group is defined as a ‘shift’. Thus, a shift 1 means a neighboring nozzle in each span group is fired. U.S. Pat. No. 7,780,256 describes some criteria for setting the span and shift for optimal ink refilling as well as minimizing fluidic crosstalk aerodynamic interference between ejected droplets.

(31) From the foregoing, it will be apparent that the effects of span and shift inevitably produce print artefacts, because the print media is constantly moving during single-pass printing. With a shift of 1, for example, each line of print is actually printed as a sawtooth. When printing at normal speeds, the effects of span and shift are barely noticeable because, although the media is continuously moving, it is effectively stationary on the timescale of each row-firing cycle. However, when printing at very high speeds, print artefacts arising from span and shift become more noticeable due to the increased movement of the media within one row-firing cycle. For example, when printing at 10× print speed using two aligned monochrome printheads, the media will move by 2 DP within one row-firing cycle. Thus, the ‘height’ of each sawtooth will be 2 DP, which may be unacceptable for some print applications.

(32) In a sub-row firing scheme, nozzles from each of the ink planes share printing of droplets for each image line. Thus, with five ink planes (corresponding to ten even/odd nozzle rows) in a monochrome Memjet® print chip, Rows 0, 2, 4, 6 and 8 can each fire 20% of even droplets, while Rows 1, 3, 5, 7 and 9 can each fire 20% of odd droplets for a given image line. By contrast with conventional row-wise firing, in which all enabled nozzles in the same nozzle row are fired within one row-time, in the sub-row firing scheme, all aligned nozzle rows (e.g. all even nozzle rows or all odd nozzle rows) of the print chip fire their enabled nozzles, based on latched dot data, within one row-time. A row-time is less than or equal to a time period allocated for firing all nozzles in the print chip divided by the number of nozzle rows.

(33) Advantageously, sub-row firing facilitates mapping of data for a given line of print to whichever nozzle row is best placed for horizontal alignment of a printed line of dots. So instead of an error of 2 DP over one row-firing cycle in the example above, the error can be reduced to less than 1 DP with suitable mapping of dot data over the 5 usable nozzle rows in each row-firing cycle. Effectively, sub-row firing enables the height of the sawtooth artefact described above to be reduced by a factor of 5.

(34) In order to enable sub-row firing, the number of nozzles N in each span must be an integer multiple of the number of ink planes M. For example, with five ink planes in a Memjet print chip, the span should be 5, 10, 15, 20 etc. Consequently, a predetermined number of nozzles P in each individual span that are used for firing is N divided by M.

(35) In one preferred sub-row firing scheme, the span is 5 and the shift is 1, with different ink planes sequentially printing from shifted nozzles along each span (e.g. Row 0 prints with 0.sup.th nozzle from each span within first 20% of one row-time, Row 2 prints with 1.sup.st nozzle from each span with second 20% of one row-time, Row 4 prints with 2.sup.nd nozzle from each span within third 20% of one row-time, Row 6 prints with 3.sup.rd nozzle from each span with fourth 20% of one row-time, and Row 8 prints with 4.sup.th nozzle from each span within final 20% of one row-time). Advantageously, sub-row firing when combined with appropriate mapping of line data to each nozzle row reduces the effects of span and shift artefacts at very high print speeds. A further advantage is that a shift value of 1 will not generate any problems associated with fluidic crosstalk or ink refilling, since shifted nozzles are not in the same nozzle row.

(36) It will, of course, be appreciated that the present invention has been described by way of example only and that modifications of detail may be made within the scope of the invention, which is defined in the accompanying claims.