Method and Apparatus for High Resolution Ink Jet Printing

This invention generally relates to methods and apparatus for ink jet printing and plotting but more particularly this invention relates to the field of high resolution ink jet color printing and plotting.

During the last decade the use of electrically controllable ink jets in recording devices such as industrial markers, printers, and color hardcopy output devices for computers, has increased measurably. Two principally different methods have been developed: drop on demand methods and methods using continuous ink jets. Both methods have been used both for the printing of alphanumeric characters and to generate computer generated charts and images, often generated in color. Also ink jets have been applied in the field of facsimile transmission.

As a result of intensive research on methods for ink jet printing, the quality of the print out has been improved appreciably during the last few years. The fact of improved quality of print out is especially true for computer controlled ink jet plotters generating color pictures on paper or transparencies, where the ultimate goal is to have an image quality essentially equal to the image quality of a good photographic color print. However, the effort toward achieving this ultimate goal is limited due to the fact that the drop-on-demand method and most of the continuous jet methods are essentially digital devices, i.e., in each picture element (pixel) of the picture they place a drop of ink or no ink at all.

These plotters usually use only four colors of ink (magenta, yellow, cyan, and black). Consequently only a very limited range of color shades can be printed by different combinations of these colors in each pixel. This limitation has been circumvented by the so called dither techniques described by Jarvis, Judice and Ninke in "Computer Graphics and Image Processing", No. 5, 1976, pp 13-40. In the Jarvis, et al article, the picture is divided into a large number of square matrices, each matrix containing a certain number of pixels. A typical matrix size is 4.times.4 or 8.times.8 pixels, i.e., 4 pixels in each row and 4 pixels in each column. Thus each pixel has an area of 1/16 the area of the matrix. Different shades of color can be obtained by filling different numbers of pixels in each matrix with ink. Hence, using a 4.times.4 matrix 16 different shades of a color and white (no color) can be generated, while an 8.times.8 matrix allows the rendition of 64 shades of color in addition to white. There has been described several ways of generating color shades by the matrix method above, e.g. ordered dither, digital half tone, and special unordered dither algorithms. However, due to the relatively large size of the ink drops, all of these methods result in coarse images, images so coarse that the eye can clearly discern the repetitive pattern of the matrices or at the least, the eye perceives a strong and very apparent graininess in the generated color print.

The image quality of the matrix method can be improved appreciably if the color density in each pixel can be varied continuously. If the pixels which make up the matrix are smaller than 0.1.times.0.1 millimeters, then the unaided human eye can no longer resolve the pixels at a normal viewing distance of 20 centimeters. Therefore, an image made up by 8 to 10 of such continuous density pixels to the millimeter has the same appearance, when viewed by the unaided eye, as a truly continuous tone picture, typified by a high grade photographic color print. The fact, of pixels smaller than 0.1.times.0.1 mm appearing as continuous, is used in conventional high quality color printing using offset or gravure plates, where pixel densities of 6 to 8 pixels per millimeter (160 to 200 per inch) are used and these methods generate color shades by varying the size of each pixel point, i.e., the amount of ink applied to each pixel.

The primary object of the present invention is to essentially apply the method used in offset and gravure printing to ink jet printing resulting in dramatic improvements in image quality and resolution of ink jet color prints.

The central idea of the invention is to control the amount of ink laid down in each pixel by the ink jet. This control of the ink can be attained by using continuous ink jets modulated electrically as described by Hertz in the U.S. Pat. No. 3,916,421. Depending on an electrical signal voltage applied between the ink and a control electrode surrounding the point of drop formation, the drops generated by the continuous jet are either charged or uncharged. When these drops subsequently travel through an electric field, the charged drops are deflected into a catcher while the uncharged drops travel undeflected onto the recording paper.

Now, if the jet diameter is small, e.g., about 10 .mu.m and its velocity is about 40 meters per second, the drops are so small that the mark produced by a single ink drop on the paper cannot be detected by the unaided eye. Therefore, in order to print a pixel in the matrix color method described above, about 30 drops of ink have to be deposited in each pixel. This is achieved by keeping the electrical control signal at 0 volts during the generation of 30 drops, so that their electrical charge is zero which allows them to travel to the paper undeflected. Throughout the remaining portion of this disclosure this electrical signal may be called the print pulse. Obviously, if the number of drops generated per second is constant, the length or time duration t.sub.p of the print pulse will determine the number of drops in the form of a drop train, travelling towards the paper.

If, e.g., 30 drops of ink are required to generate maximum color density in a pixel, lighter shades of that color can be produced by laying down a smaller number of drops in that pixel. Actually, if it can be assumed that one could determine exactly the number of drops laid down in the pixel by controlling the length of the electrical print pulse, then in this way 30 shades of a color in addition to white (no color) could be generated in each pixel. By the above method then, the color saturation in each pixel can be varied essentially continuously as in the case of conventional offset or gravure printing.

While the principle of the present invention appears to be and in fact is relatively simple, it is actually very difficult to put into practice. Firstly, the diameter of the drops has to be very small and the drop generation rate must be relatively high and essentially constant. Secondly, because of the high air resistance encountered by the very small drops, not only the individual drops but also groups of drops tend to merge on their way to the paper which gives rise to an appreciable graininess especially in the instances of lighter shades of a color. Finally, drops with undefined charges should be avoided for the same reason, that is, merging will take place as noted above.

It is therefore an object of the present invention to control the number of drops deposited in each pixel. According to the invention this objective can be achieved by realizing at least one or more of the following features or combinations of any or all of the following functions or features:

a. the jet diameter should be small and its velocity high to ensure a high drop generation rate;

b. ultrasonic stimulation of the drop formation process of the jet should be used to insure a high and constant drop generation rate and to decrease the drop diameter and also providing for or causing the generation of all drops having diameters which are essentially equal to each other, thereby decreasing the tendency for the drops to merge because equal size drops will encounter equal air resistance resulting in essentially equal transit time from nozzle to paper;

c. providing for slightly charging the drops, which are meant to reach the paper, with like charges thereby causing a repulsive force between them which further counteracts the merging tendency of the drops or of the groups of drops;

d. obviously the length of the electrical print pulse determining the number of drops that reach the recording medium, generally paper, can vary continuously. This also tends to cause graininess of the image, which can be avoided by the use of suitable electronic circuitry which adjusts the length of the signal such that it equals n/f, where f is the drop formation frequency and n an integer chosen such that the ratio n/f is close to the length of the original signal and additionally, the start of the print pulse can be synchronized with a suitable phase of the ultrasonic stimulation mentioned under b;

e. providing for an air flow directed along the axis of the jet from the nozzle towards the recording paper to decrease or essentially eliminate the air resistance encountered by the drops; and

f. evacuating, at least partially, the space close to the jet also decreases the air resistance and further decreases the graininess of the picture.

Many other advantages, features, and additional objects of the present invention will become manifest to those versed in the art upon making reference to the detailed description and the accompanying sheet of drawings in which a preferred embodiment incorporating the principles of the present invention is shown by way of illustrative example.

FIG. 1 shows the side view of an electrode system that can be used in conjunction with the invention.

FIG. 2 shows the time dependence of the electrical signal used for the controlling of the ink jet.

FIG. 3 is similar to FIG. 2 except that the voltage varies from -20 to +150 volts rather than from +20 to +150 volts.

FIG. 4 shows a block diagram of an electrical circuit used to realize one embodiment of the invention.

FIG. 5 is used to illustrate the transformation of the print pulse.

FIG. 6 shows a block diagram of another embodiment of the invention.

The methods and apparatus of this invention can be accomplished using electrode systems very different from those systems conventionally used to control the continuous jet as used in drum plotters and as described by Hermanrud & Hertz in Journal of Appl. Photogr. Eng. 5 No. 4 (1979). However, for the sake of simplicity an electrode system used with a single jet, and as shown in FIG. 1 and described in U.S. Pat. No. 3,916,421, included herein by reference thereto, will be used when explaining the invention. Here an ink jet 11 issues under high pressure from the nozzle 1 and breaks up into a train of drops 11b at the point of drop formation 11a inside the control electrode 2. This train of normally uncharged drops 11b travels in a line or along an initial axis towards the recording medium or paper 3 which is mounted on or otherwise affixed to a rotating drum 4 of a drum plotter (not shown). On the way from nozzle 1 toward paper 3, the drops 11b pass a transverse electric field generated between the positively charged high voltage electrode 5 and the lower part 2a of the control electrode 2.

Now, if a positive control voltage is applied to the control electrode 2 via an amplifier 6 while the ink is grounded via the electrode 7, an electric field is established at the point of drop formation causing each of the drops 11c formed at the point of drop formation 11a to be positively charged. Because of the charge, these drops 11c are deflected into the catcher 8 and cannot reach the recording paper 3. When the control voltage is decreased from such a positive level so as to cause all the drops 11c to fall into catcher 8 a lesser positive voltage level is reached which will permit the drops 11c to reach the paper 3. Thus, it is obvious and apparent from FIG. 1 that the length of time during which the signal voltage or print pulse applied to electrode 2 is zero, or less than the point cut off control voltage, determines the number of drops 11c that reach the recording paper 3, i.e., the drops 11c formed during that period of lesser positive voltage are not charged or are not sufficiently charged, as a result of the charge on the droplet directing electrode 5, so that they get deflected into catcher 8.

In an actual embodiment of the invention an ink jet 11 having a diameter of 10 .mu.m and a velocity of 40 meters per second is used. Such a jet will form into approximately 10.sup.6 drops 11c per second at the point of drop formation 11a. If the drum 4 of the drum plotter rotates with a surface velocity of 3.3 meters per second and the size of a pixel (picture element) is assumed to be 0.1.times.0.1 millimeters, the time required in order to print a pixel is about 30 microseconds. Since 30 drops are formed during the 30 microsecond interval of time, all of these 30 drops will be deposited within a certain pixel if the control voltage supplied to the control electrode 2 was zero or less than the print cut-off voltage during the 30 microseconds. Thus a maximum color density will be generated in the pixel. If the print pulse duration is shorter than 30 microseconds, less than 30 drops will be deposited in the pixel, and thus the actual number of drops and therefore the color density of the pixel will depend on the duration i.e., the length of the print pulse or control voltage. Thus by controlling the length of the print pulses the density or saturation of the color can be controlled in each pixel of the picture generated on the recording paper 3. Since a plurality, typically 3 or 4 or more jets 1 having different colors can be arranged side by side on the drum plotter apparatus as described by Hermanrud and Hertz in Journal of Appl. Photogr. Eng. 5 No. 4 (1979), full color pictures can be generated where the hue and color density can be varied continuously in each pixel.

While the principle of the invention, i.e., varying the color density in each pixel by controlling the number of drops deposited in each pixel, is simple, its realization meets with many difficulties. First, the drop generation rate has to be as large as possible to provide for the generation of a nearly continuous variation of the color density. A high drop generation rate can be attained by using a small nozzle 1, thus producing a high speed jet 11. For letter sized pictures the nozzle size should be about 10 .mu.m and the speed of the jet 11 should be about 40 meters per second. Further, by stimulating the drop formation process by applying ultrasonic vibrations generated by the oscillator 9 and the transducer 10 to the nozzle 1, as exemplified in FIG. 1, the number of drops formed per second can be increased. A further consequence of the ultrasonic vibrations is that the size of each drop 11c usually becomes relatively constant resulting in a decrease in the deleterious effects of air resistance such as drop merger. In the 10 .mu.m jet described above ultrasonic frequencies between 1 and 1.5 MHz (megahertz) will create 10.sup.6 to 1.5.times.10.sup.6 drops per second. Therefore, it is an advantage to stimulate the jet 11 by ultrasonic vibrations.

One of the difficulties encountered in attempting to create exactly the same color density in each pixel is created by the tendency for drops 11c to merge, or groups of drops to merge due to air resistance encountered on the flight from the point of drop formation 11a toward the recording paper 3. Such merging results in slight misplacement of the drops on the recording paper 3 since the paper is traveling at a near constant rate while the drops 11c would arrive at the paper 3 at an irregular rate. This slight misplacement results in a certain amount of graininess, especially in the lighter shades of a color.

The merging of the drops can be partially counteracted by using a voltage slightly different from zero for the print pulses 20 and 22 as shown in FIGS. 2 and 3. In FIG. 2 the voltage applied to the control electrode 2 is switched between +20 and +150 volts and the number of droplets 11c which will impinge on the paper 3 is directly related to the width of the pulses. If the voltage is a positive 150 volts, the drops 11c are strongly charged positively and essentially all of the drops 11c are therefore deflected into the catcher 8. If, however, during the voltage of the print pulses 20 (the negatively directed portion of the control signal) is only 20 volts the drops 11c are only slightly charged and their deflection in the transverse electrical field created by the positive 2000 volts applied to the high voltage electrode, droplet directing electrode 5 is so small that they pass above the catcher blade 8 onto the recording paper 3. As a result of the relatively small positive charge, the drops 11c repel each other, thereby decreasing the tendency toward drop merging. Alternatively, essentially the same results are achieved where the print pulse 22 voltage might be -20 volts and the "off" voltage or the print cut-off control voltage is +150 volts as shown in FIG. 3. It should be clear to those skilled in the art that the voltages given here are only examples of useable voltage levels and the voltages can vary within wide limits. Additionally, the same effect can be achieved by applying a DC offset voltage of e.g. +20 volts to the electrode 7 and use a control signal varying from, for example, zero volts to a positive 150 volts at the control electrode 2. The control signal also might be applied to the electrode 7 as described in U.S. Pat. No. 3,737,914.

It has been pointed out above that irregularities in the drop placement give rise to graininess of the picture. Such graininess is also caused if the number of drops deposited in each pixel varies stochastically in an area of constant color density. This is especially true in the lighter shades of color, which are created by 1 to 5 drops per pixel only. If the length of, or time duration of the print pulse can vary continuously, as compared to incrementally, and/or the pulse 20 or 22 is not synchronized or timed with the drop formation process of the jet 11, then it is obvious that (especially at low drop numbers) variations in the number of drops per pixel for a particular pulse duration may occur, a situation which is aggravated by so-called "historic effects" encountered during the charging of the drops 11c. This problem can be avoided in several ways, two of which are described below.

In an area of constant color density all print pulses 20 or 22 controlling the number of drops deposited in each pixel will necessarily have essentially the same length or duration. As pointed out above, unless there is synchronism between the print pulse and the drop formation, a constant duration print pulse does not prevent certain stochastic variations in the number of drops deposited in each pixel. That is, exactly the same number of droplets 11c will not necessarily be deposited. However, this variation can be avoided if the time duration t.sub.p of the print pulse 20 or 22 can be varied only in discrete steps so that t.sub.p =n/f, where n is an integer number equal to the number of drops to be deposited in the pixel and f the drop formation frequency given in drops per second generated by the combination of nozzle 1 and oscillator 9. That is, the print pulse width does not vary continuously but in incremental steps having a time width increment related to n/f. FIG. 4 illustrates a block diagram of an electrical circuit which in effect controls the incremental width of the pulses. For further clarification FIG. 5 shows a timing diagram of some of the pulses occurring in FIG. 4. Particularly pulses A, B, C and D which would be measured at the indicated location in FIG. 4.

In FIG. 4 the nozzle 1 producing the jet 11 as well as the electrode system 2 and 5 are illustrated. The signal from the oscillator 9 driving the transducer 10 is applied to a monostable multivibrator 12 which produces one short duration pulse for each oscillation of the signal. After an adjustable delay circuit 13, these pulses are applied to a sample-and-hold circuit 14. These pulses are the pulses marked as A and are shown in the timing diagram of FIG. 5.

In most cases the information determining the color density in each pixel is stored in digital form in a memory 15. Each time a pixel is to be printed the corresponding digital value is extracted from this memory and applied to a digital-to-pulse-width or duration converter 17 via another sample-and-hold circuit 16. This transformation of a digital number from the memory 15 to a pulse having a duration t.sub.p proportional to this number is governed by a clock 18 which, in the case of a drum plotter, is derived from the shaft encoder pulses. Thus, for each pixel a print pulse is presented at the output of the digital-to-pulse-width converter 17 as shown in the timing diagram B of FIG. 5.

As shown in FIG. 4 the output pulse B is applied to the sample-and-hold circuit 14. Since the sampling is initiated by the pulses A derived from the transducer driver signal, the width of the output pulse C from the sample and hold circuit 14 is always an integer number of the period of the driver signal. In other words, if f is the frequency of the driver signal, the pulse width (or duration) is equal to n/f, n being an integer and being equal to the number of drops per pixel as indicated in the timing diagram D of FIG. 5. In an area of constant color density n is constant wherefore exactly the same number of drops is deposited in each pixel. This process essentially eliminates or at least materially reduces the amount of graininess in the area if merging of drops due to air resistance is prevented at the same time, i.e., the drops 11c have a small positive or negative charge sufficient to minimize the tendency to merge.

The graininess due to stochastic variations in the drop number per pixel can be further reduced by synchronizing the start of, i.e., the leading edge of, the print pulse with a suitable phase of the drop formation mechanism. This synchronization can be achieved, e.g., by a circuit shown in FIG. 6. As was the case in FIG. 4, the memory 15 acts as a data source which forwards the digital value of the density of each pixel via a sample-and-hold circuit 16 to the digital-to-pulse-width converter 17. Again this information flow is governed by a clock 18 derived e.g. from a shaft encoder. However, the start of each digital-to-pulse-width conversion cycle is here synchronized to the driver signal generated by the oscillator 9. Such synchronization is controlled by the short pulses A generated by the monostable multivibrator 12. These monostable or synchronizing pulses are suitably delayed in the adjustable delay circuit 13. If necessary this delay can be automatically adjusted at location 13a of FIG. 6 by suitable electronic means sensing the drop formation process. After this delay the synchronizing pulses are used to start the digital-to-pulse-width conversion cycle. This will insure that the start of the print pulse always coincides with the same phase of the drop formation process. Precise synchronizations of an electric control signal to the drop formation process are difficult, especially with small jets. Thus, the synchronizing method is a bit more difficult to carry out.

By the methods and apparatus described above the number of ink drops and thus the color density in each pixel can be very accurately, if not exactly controlled. However, if the number of drops available per pixel is relatively small i.e., N=30 as was described, then only N+1 discrete steps or variations of color density can be generated. If N is small, e.g. 30, this rather limited number of density steps easily causes contouring in the picture, especially in the lower density range of color.

To avoid this effect the number of density steps has to be increased. Obviously, this can be achieved by increasing the drop number per pixel, which will require jets of very small diameters and high speeds, e.g., 3 .mu.m and 50-100 meters per second. However, such jets are hard to produce reliably wherefore an alternate method to increase the number of density steps will be described below.

Assume that the maximum number of drops that can be laid down in a pixel is N. Then, by changing the duration t.sub.p of the print pulse the number of drops n actually deposited can be varied between 0 and N, n and N being integers. Thus N+1 steps of density can be created in each pixel, i.e., if N=30, then 30 shades of color and white (no color) can be generated.

The number of density steps can be increased to 2N+1 by using the fact that the pixels are very small. Assume there are two adjacent color densities, one generated by n drops per pixel and the other one by n+1 drops per pixel. A new color density can be created between these two density steps by laying down alternatingly n or n+1 drops on successive pixels in the print direction. This can easily be realized by a suitable design of the digital-to-pulse-width converter 17 in FIGS. 4 and 6.

This method of increasing the number of density steps above the maximum drop number N can be extended by including more than 2 pixels in the repetitive pattern in the print direction. As an example, let us examine the case where three adjacent pixels can receive n or n+1 drops. Then two intermediate density steps can be created between the density steps n and n+1 by depositing n, n, n+1, or n, n+1, n+1 drops in the three consecutive pixels respectively. Obviously, this scheme can be extended to 4, 5, or more adjacent pixel positions along the print direction. Also it should be emphasized that this method of increasing the number of density steps is especially important if n=0 or 1, i.e., in the very light shades of a color.

In the description given above the importance of avoiding the merger of drops 11c or groups of drops on their flight towards the recording paper 3 has been pointed out. Since this merging is due to air friction, it is obvious that this effect can be avoided by evacuating, at least partially, the space close to the jet axis or the entire plotting apparatus. Alternatively, an air flow having about the same velocity as the jet velocity could be generated close to the jet axis following the jet 11 from the nozzle 1 to the recording paper 3 in FIG. 1, which would also tend to eliminate or substantially reduce the effect of air resistance.

It is obvious that the invention described above can be applied to different electrode systems controlling one or a plurality of ink jets in one or several colors. Also it can be used with other ink jet control mechanisms, e.g., drop on demand or thermal ink jets (bubble jets). It can be used for single or full color image printing by either drum, flat bed or other plotters on any kind of record receiving surface. Further, its use is not limited to the production of pictures and the detailed description of the invention given above is to be regarded as an example only. Having described this invention, it will be apparent to those skilled in the art that various modifications may be made hereto without departing from the spirit and scope of this invention as defined in the appended claims.