Incremental printing of symbolic information – Ink jet – Controller
Reexamination Certificate
2000-06-15
2001-10-30
Yockey, David F. (Department: 2861)
Incremental printing of symbolic information
Ink jet
Controller
C347S041000
Reexamination Certificate
active
06309041
ABSTRACT:
BACKGROUND
1. Field of the Invention
This invention relates generally to printing desired color images on printing media by construction from individual ink drops deposited in pixel arrays; and more particularly to systems in which a multiple-nozzle ink-discharging pen passes across a printing medium multiple times, each time creating a portion of a respective segment of a desired image, and in which the medium is periodically advanced so that the pen can create portions of successively different segments.
2. Prior Art
U.S. Pat. No. 4,748,453 of Lin et al. introduces one representative basic mechanism for a printing machine that uses liquid inks and pixel-array or -matrix construction of images. Lin also describes a basic problem of so-called “beading” on transparency stock, and reviews prior solutions to that problem.
Lin's system provides another relatively early solution. His system prints in alternation two complementary checkerboard patterns, in which each unit of each checkerboard pattern is an approximately circular ink spot—of diameter approximating a diagonal of an individual cell of the checkerboard pattern.
That system is said to avoid overlap of deposited ink that is still wet. While very useful, Lin's approach provides only modest mitigation of the general drying problem when considered in view of modern demands for even higher throughput (area printed per unit time, for example number of pages per minute) and even higher image quality.
More specifically, while Lin set out to deter beading and banding—by preventing the inking of adjacent pixels with resultant puddling of the deposited ink—actually he did not entirely achieve those goals. Among other reasons, statistical variations between various ink-spot positions allow a certain amount of overlapped-spot printing to continue to occur, along diagonals of the checkerboard pattern.
The two-dimensionally connected character of Lin's checkerboard patterns tends to aggravate this residual puddling. That is to say, his checkerboard-style grids offer a relatively large number of pathways over which ink in diagonally adjacent positions can run together.
One of the present inventors has disclosed—see U.S. Pat. Nos. 4,963,882 and 4,965,593, coowned with the present document—fundamental later-generation techniques for enhancing throughput and quality in analogous machines. The '882 patent describes dual statistical benefits of using different nozzles to lay down plural ink drops in each pixel, and also in each row of pixels.
One of these benefits is in better averaging of ink-spot location and quality while all the nozzles are operating normally; and another is in a tendency to conceal individual-nozzle failure—and thereby permit operation with only slightly diminished apparent print quality despite such failure. These benefits accrue from what that earlier patent calls “double-drop always” inking.
More specifically, for each row some pixels are filled by one nozzle in a first pass, and some by another nozzle in a second pass—and if there are four passes then some other pixels by still another nozzle in a third pass, and by yet another in a fourth pass. Since different nozzles have different imperfections—either directional errors or variations in volume etc.—and since these imperfections tend to be randomly scattered within tolerances about some mean, then contributions to each row by two or more different nozzles tend to cause errors to average out in the composite result.
This technique thereby strongly suppresses an objectionable tendency of virtually all early-generation computerized systems for forming images, sounds etc. that are subject to esthetic evaluation—namely, that irregularities tend to be too highly patterned, or regularized, and accordingly seem artificial. Analogously, when an individual nozzle fails entirely so that it is not providing any inking at all in its pixel row, nevertheless its complementary nozzle is most likely to be still working so that at least some inking does occur in that row.
The '882 patent teaches various ways in which different nozzles can be caused to provide inking, as mentioned above, for a particular pixel or row of pixels. One way is to use plural pens; another suggested way is to use plural nozzle sets aligned along the scan axis.
Still another way is by using overlapping pen swaths—in other words, by advancing the printing medium, between pen scans, by some distance less than the height of the nozzle array (or at any rate the effective height, i. e. the height of that part of the array which is in use) along the direction of printing-medium advance. The teaching of the '882 patent relates to use of such over-lapping pen swaths to permit deposition of what might be called redundant sets of ink drops—that is, more than one drop at each individual pixel location—from different nozzles, to obtain the benefits of double-drop-always inking.
The '593 patent discloses use of nozzle spacings, in orthogonal directions (pen-scan and medium-advance directions respectively), that are two different integral multiples of the pixel-grid spacing; the ratio of the two integral multiples is an irreducible fraction. This technique helps to ensure that no adjacent pixels are printed within a time period less than the “fixing time of the colorant on the printing medium”.
Yet another seminal patent coowned with the present document is U.S. Pat. No. 4,967,203 of Doan et al. , which introduces two major advances in this field. One of these is the use of overlapping pen swaths to provide complementary parts of each image segment.
In this technique, as in that of the '882 patent, overlapping swaths are formed by advancing the medium through a distance which is less than the effective height of the nozzle array on the pen, and accordingly less than the height of a segment of the image which can all be produced in one pass of the pen. In this case, however, within each segment later-arriving nozzles are used not to provide statistical enhancement or nozzle-failure backup for pixel locations nominally inked previously, but rather to fill in pixel locations nominally left uninked previously.
Such a system very greatly improves the apparent quality of printed images, notwithstanding imperfections in manufacturing tolerances of a printing-medium advance mechanism. The system achieves this improvement by reducing the amount of positioning error which is allowed to accrue between steps of the advancing mechanism, and also by hiding the edges of every pen swath within the boundaries of one or more other (earlier and later) swaths.
In addition, both the '882 and the Doan '203 patent techniques offer an important benefit in spreading the inking process, for each image segment, over more than one pen pass. Generalizing the teachings of the Lin patent, the '593 patent and others, it is now recognized as highly beneficial to spread the inking over as many pen passes as possible, so that ink spots deposited in each pass have time to at least become tacky before adjacent and possibly slightly overlapping spots are created.
Such inking distribution as between passes is beneficial in that it promotes maximum ink deposition per unit area—and thereby in turn promotes well-saturated, intense colors and broadest possible color gamut—while maintaining throughput as high as practical. On the other hand it is also recognized that, unfortunately, increasing the number of passes per pen swath tends to decrease throughput.
Inasmuch as competitive commercial pressures for both high throughput and all aspects of image quality continue to drive machine design, it will be understood that every newly discovered way to spread inking between pen passes is a valuable opportunity. It is particularly important, however, to find ways to spread inking as between passes without unduly increasing the overall number of passes per pen swath.
These two applications (redundant and complementary inking) of the overlapping-swath principle both are taught—and have been practiced—in terms of (1)
Hickman Mark Stephen
Meyer William D.
Richtsmeier Brent W.
Hewlett--Packard Company
Potts Jerry R.
Yockey David F.
LandOfFree
Odd number of passes, odd number of advances, and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Odd number of passes, odd number of advances, and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Odd number of passes, odd number of advances, and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2577727