DE69123605T2 - SYSTEM FOR WRITING DATA ON AN ELECTROPHORETIC DISPLAY PANEL. - Google Patents

Description

Technical area

The present invention relates to a method of operating an electrophoretic display panel device and, more particularly, to a method of increasing the speed at which information can be written into an electrophoretic display panel.

State of the art

Electrophoretic information displays (EPIDs) are well known today. Several patents issued in the names of Frank J. Disanto and Denis A. Krusos and assigned to the assignees, Copytele, Inc. of Huntington Station, New York, teach various display types and features. For example, U.S. Patent Nos. 4,655,897 and 4,732,830, each entitled ELECTROPHORETIC DISPLAY PANELS AND ASSOCIATED METHODS, describe the basic operation and construction of an electrophoretic display. U.S. Patent No. 4,742,345, entitled ELECTROPHORETIC DISPLAY PANELS AND METHODS THEREFOR, describes a display with improved alignment and contrast. U.S. Patent No. 4,833,464, entitled ELECTROPHORETIC INFORMATION DISPLAY (EPID) APPARATUS EMPLOYING GREY SCALE CAPABILITY, relates to an EPID with the ability to display pixels with varying gray scale intensity. This patent recognizes, among other things, that the duration of applying a voltage gradient to a particular pixel location affects the amount of pigment particles at that location. By controlling the duration of the write pulse, one can therefore obtain gray scale capability - the shorter the pulse, the brighter the line.

The display panels shown in the above-mentioned patents work on the same basic principle, namely that when a suspension of electrically charged Pigment particles in a dielectric liquid are subjected to an applied electrostatic field, the pigment particles migrate through the liquid in response to the electrostatic field. For a given substantially homogeneous suspension of particles having a pigment color different from that of the dielectric liquid, if the applied electrostatic field is localized, this will cause optically observable localized pigment particle migration. The localized pigment particle migration results in either a localized region of concentration or dilution of particles depending on the sign and direction of the electrostatic field and the charge on the pigment particles. The electrophoretic display devices taught in the above U.S. patents are "triode-type" displays having a plurality of independent parallel cathode row conductor elements or "rows" deposited horizontally on a surface of a glass viewing screen. A layer of insulating photoresist material deposited over the cathode elements and photoetched down to the cathode elements to effect a plurality of insulator strips positioned at right angles to the cathode elements forms the substrate for a plurality of independent parallel column or grid conductor elements or "rows" running in the perpendicular direction. A glass cap member forms a liquid-tight seal with the viewing window along the peripheral edge of the cap to contain the liquid suspension and also acts as a substrate for an anode plate deposited on the flat inner surface of the cap. When the cap is in place, the anode surface is in spaced parallel relationship to both the cathode elements and the grid elements. For a given specific particle suspension, the sign of the electrostatic charge that will attract and repel the pigment particles is known. The cathode element voltage, the anode voltage and the grid element voltage can then be determined so that when a certain voltage is applied to the cathode and a different voltage is applied to the grid, the region near their intersection will acquire a net charge sufficient to attract or repel pigment particles in suspension in the dielectric fluid. Since numerous cathode and grid rows are employed, there are numerous discrete intersection points which can be controlled by varying the voltage on the cathode and grid elements to cause localized visible areas of pigment concentration and dilution. Essentially then, the operating voltages on both the cathode and the grid must be capable of assuming at least two states corresponding to a logic one and a logic zero. A logic one for the cathode can correspond to either an attraction or a repulsion of pigment. Typically, the cathode or grid voltage is selected such that only when both are at a logic one at a particular intersection point is there a sufficient electrostatic field at the intersection point relative to the anode to cause a visible bit of information to be written into the display by migration of pigment particles.

The bit can be erased, for example, when the polarity is reversed and a zero-zero logic state occurs at the intersection in combination with an erasure voltage gradient between the anode and cathode. In this way, digitized data can be displayed on the electrophoretic display.

An alternative EPID construction is described in application No. 07/345,825 entitled DUAL ANODE FLAT PANEL DISPLAY APPARATUS, filed on May 1, 1989 for the assignee of the patentees, which relates to an electrophoretic display in which the cathode/grid matrix, as found in triode-type displays, is provided by a plurality of independent, separately addressable "local" anode lines. The local anode lines are deposited on and aligned with the grid lines and isolated from them by intervening lines of photoresist. The local anode lines are in addition to the "remote" anode, which is the layer deposited on the anode end plate or cap as in triode displays. The above-mentioned dual anode construction provides improved performance by eliminating unwanted variations in display brightness between frames, increasing the speed of the display and reducing the anode voltage required during write and hold cycles, as explained in Application No. 07/345,825.

A common goal for EPIDs of both the triode and tetrode types, and for digital displays and computers and digital devices in general, is increased speed of operation. With respect to displays, it is desirable that the display be able to write, erase, and manipulate the displayed image as quickly as possible in response to operator input and computer processing. For example, when a computer having a visual output device for displaying character information, such as a cathode ray tube, is used as a word processor, if the writing and erasing of displayed information is not fast enough, the word processor operator will be slowed in completing his task. Although the computer memory and processor unit can operate at speeds far exceeding the capacity of a human user, if the input and output devices through which the computer communicates with the user are slow, the computer and user must wait for the output devices. Therefore, if a word processor user is browsing a document at high speed, a slow display device may well reduce the page turning speed below that at which the user and/or the computer could possibly operate.

In EPID and other display devices, since there are a plurality of pixels arranged on a coordinate grid or matrix and since the pixels must be independently addressable, display functions are often performed at the pixel level, e.g. each pixel is written sequentially. Sequential functions are inherently time consuming in that the previous function must be completed before the next one can begin. Furthermore, although writing to a single pixel can be done very quickly, there are so many of them that even a small write time is important. A process for independently controlling a single pixel display in which some degree of parallel display processing is performed is described, for example, in U.S. Patent No. 4,742,345, in which display information concerning an entire row of pixels, i.e., on or off, is accumulated in an accumulator or register during a first phase, fed in parallel to a holding memory group in a second phase, and applied in parallel to one of the coordinate grids in a third phase. Applying the display information to one of the coordinate row sets, e.g., the grid rows, which may be oriented in the vertical direction, has been referred to as "loading" the grid with the data. When the information bits (voltages corresponding to logic "1" and logic "0") are applied or "loaded" to, for example, all vertical coordinate lines, a single horizontal line can be written by enabling this line, i.e. by applying a voltage corresponding to logic "1" to this horizontal line. The function of applying an enabling voltage to the line to be written, in this case a horizontal cathode line, has been called "line writing". Of course, this line-by-line writing method also has an upper speed limit.

Regarding EPID, one factor contributing to the Speed at which the display can function is the speed at which the pigment particles can migrate through the electrophoretic fluid under the influence of a certain voltage gradient. Pigment particle migration speed depends, among other things, on particle size and viscosity of the electrophoretic fluid. In addition to particle speed, there is also the factor of spatial distribution in the EPID volume, i.e. since the particles are in suspension, they are relatively evenly distributed within the fluid volume before they are exposed to voltage gradients. There is therefore a range of particle proximity to the "target" element, the target element being the element to which the particles are to be directed to perform a function such as writing or erasing.

These speed and proximity factors in EPID are used in U.S. Patent No. 4,833,464 to control pixel display intensity and grayscale, respectively. In fact, if a voltage gradient of shorter or longer duration is applied, fewer or more particles will accumulate at the "target" electrode and thus affect pixel intensity, i.e., the greater the number of particles, the higher the intensity. It is noted that pixel intensity is perceptible on both sides of the typical EPID, so that a strong accumulation of, for example, light-colored particles on one surface of the EPID will be accompanied by a corresponding lack of light-colored particles on the other surface, which will in all probability appear dark due to the selection of a dark solution or background for the light-colored particles. Thus, writing a character on one end plate of an EPID will result in writing its reverse image on the other plate. Writing a space can be called selective character deletion.

US 4 804 951 discloses a display in which alternating lines of a frame of display data are written by a first electrode scan and the remaining lines are written by a second electrode scan, so that a second of a pair of adjacent lines can be written sequentially immediately without having to wait for the first electrode scan to "backtrack" before writing the second of the lines. However, it is still necessary to perform the same number of scans to write the entire display frame and there is therefore no significant saving in the total time for writing the display frame.

It is an object of the present invention to provide a method of operating an EPID with any particular pigment particle size, viscosity of the electrophoretic fluid, electrode arrangement and operating voltage levels so that the speed of operation is increased.

Description of the invention

The problems and disadvantages associated with conventional methods of operating electrophoretic displays are overcome by the present invention method, which enables a reduction in the time for writing a frame of display data consisting of a plurality of rows of displayable pixels on an electrophoretic display which requires a minimum amount of time to completely write a line.

According to the invention, there is provided a method for reducing the time required to write a frame of display data consisting of a plurality of rows of displayable pixels on a high-density electrophoretic display which requires a predetermined time period to completely write a line, said method comprising the steps of:

(a) simultaneously writing a set of at least two adjacent lines of said display data in a first period of time that is shorter than said predetermined period of time for completely writing a line;

(b) shifting the elements of said set of rows so that said set contains at least one new row and at least one original row;

(c) describing said shifted set of lines in a second period of time shorter than said predetermined period of time after said step of shifting; and

(d) Repeating steps (b) and (c) until said frame is completely written.

Short description of the drawings

For a better understanding of the present invention, reference is made to the following detailed description of an exemplary embodiment in conjunction with the accompanying drawings. In the drawings:

Figure 1 is a cross-sectional view of a typical triode-type EPID with its essential electrical components;

Figure 2 is a simplified circuit diagram showing an addressable display matrix consisting of horizontal and vertical elements such as a plurality of cathode rows and a plurality of grid rows driven by display drivers such as would be used in known EPID devices such as that shown in Figure 1;

Figure 3 is a simplified diagram of circuits for controlling the x and y display drivers shown in Figure 2;

Figure 4 shows a character that could be displayed on an x-y matrix using the circuit and apparatus of Figures 1-3 controlled and operated in accordance with the method of the present invention;

Figure 5 is a flow chart representation of a method for EPID writing according to the present invention.

Implementation of the invention

Figure 1, taken from US Patent No. 4,732,830, shows an electrophoretic display 10 according to The display 10 has an anode end plate 12 and a cathode end plate 14 sealably secured to either side of an intermediate spacer (not shown) to form a liquid-tight enclosure for enclosing a dielectric pigment particle suspension or electrophoretic liquid. The end plates 12 and 14 are typically flat glass plates upon which conductor elements are deposited to provide the location of electrostatic charge for causing movement of the pigment particles 16 in the electrophoretic liquid. The procedures, materials and dimensions used to form the conductor elements on the end plates and the methods of making and using EPID in general are set forth in U.S. Patent Nos. 4,655,897, 4,732,830 and 4,742,345.

Known EPIDs, such as those shown in Figure 1, include a plurality of independent, electrically conductive cathode rows 18, shown here as horizontal rows, deposited on the cathode end plate 14 using significant deposition and etching processes. Of course, the orientation of the cathode rows 18 is dependent upon the orientation of the screen, which, if rotated 90 degrees, would position the cathode rows vertically. Thus, the cathode rows are arbitrarily defined as horizontal or lying on the x-axis. It is preferred that the cathode elements 18 be made of indium tin oxide (ITO) as described in U.S. Patent No. 4,742,345. Superimposed on the cathode elements 16 are a plurality of independent grid conductor rows 20 perpendicularly (parallel to the y-axis), ie at right angles thereto, which are isolated therefrom by an intermediate photoresist layer 22. The grid elements 20 can be formed by coating the photoresist layer 22 with a metal such as nickel or chromium using sputtering techniques or the like, followed by selective masking and etching to give the overlapping but still isolated structure shown in Figure 1. Each Cathode and grid elements 18, 20 terminate at one end in a contact pad or are otherwise adapted to permit connection to display driver circuits. An anode 26 is formed on an inner surface of the anode end plate 12 by plating with a thin layer of conductor material such as chromium.

The above components have been described in previous patents and applications of the present applicants. In addition to these teachings, the benefits and operation of an EPID with a local anode have been recognized and described in the present applicants' application No. 07/345,825. The present invention method could find application in any of these disclosed devices.

Figure 2, also taken from U.S. Patent No. 4,732,830, shows in the simplest schematic form how the cathode 18 and grid 20 rows comprise an addressable x-y matrix which allows the selective display of pixels at the intersections. Each horizontal 18 and vertical 20 row has an associated amplifier/driver 24R or 24C for supplying them with either a logic "1" or a logic "0" so that when both are at "1" at an intersection, that intersection is written to. The horizontal rows have been designated R1...R2200 to show that typically there will be 2200 display rows 18 or rows. As shown by designations C1...C1700, 1700 vertical rows 20 or columns are common.

Figure 3, taken from Patent No. 4,742,345, shows exemplary circuits for supplying input data to the x and y drivers 24R and 24C. As fully described in Patent No. 4,742,345, a large capacity composite series-parallel register 26 can be used as a buffer for accumulating a large number of bits of display data, e.g., 850 bits. After sequentially clocking data into the register 26 and filling it, the data stored in parallel into a latch bank 28 of equal capacity. The data is then gated into the display driver amplifiers 24 by a plurality of AND gates 30. During transfer from the latch bank 28 to the drivers 24, data may be accumulated in the serial register. In Figure 3, the outputs of the AND gates are designated odd-numbered columns 1 through 1699. The data for even-numbered columns would in this case be provided by a twin circuit located on the cathode end plate opposite that for the odd-numbered columns. This arrangement avoids congestion of electrical connections to the grid rows as described in Patent No. 4,742,345. Once the column data is provided to all columns, a row may then be written by sending a "1" along the row or cathode 18 to be written. In combination with any column "1", the row "1" will cause the writing of a pixel at their interface, ie a voltage gradient sufficient at that point to cause an optically observable migration and accumulation of pigment particles 16.

The proportions of the grid 20 and cathode 18 lines of Figures 1 and 2 have been greatly enlarged for illustration purposes. In operational displays, the grid 20 and cathode 18 lines are very thin and elongated. An operational panel would have a large number of intersections, e.g., 2,200 x 1,700, or a total of 3,740,000 separately addressable intersections in a panel approximately 20 x 28 cm. For ease of illustration, only a few cathode 18 lines and grid 20 lines are shown. Additional illustrations of electrophoretic displays, their components and electrical circuits can be seen in U.S. Patents No. 4,742,345 and No. 4,772,820, each assigned to the present inventors.

Figure 4 represents a character, i.e. the letter "T", which can be found on an EPID as described above using the

Figures 1-3 described using the algorithm flow charted in Figure 5. According to the present invention method it has been observed that the writing time of the EPID can be shortened by simultaneously writing more than one line at a time. That is, in the prior art EPID described above, an entire set of column data for a particular line is applied to the columns, e.g. the grid rows. Then a single line is enabled with a logic order and thus written. The next set of column data is loaded into the grid rows and the next line is enabled or written. This continues sequentially until the entire screen is written. A certain period of time is required for the pigment particles to migrate through the electrophoretic liquid to their "write" position, ie form a cluster large enough to be clearly visible. In conventional operation, therefore, each line had to be held in the logic "1" state for the required write period or write cycle time. According to the present invention, when a row set containing more than one row, e.g. two rows, is simultaneously enabled for a period of time approximately half as long as before, the two rows will both be weakly written with the same display information in half the cycle time. For example, if column data for row 1 is loaded and rows 1 and 2 are written, both row 1 and row 2 will be weakly written with display information of row 1. When new column data, e.g. for row 2, is loaded and the row set is shifted down by one and written, e.g. row sets 2 and 3 are written using data of row 2, the first row, which had been half written, is left untouched. However, assuming that the new column data associated with row 2 is the same as that associated with row 1, the second row will be fully written. Row 3 will also be weakly written with Line 2 data is written. By partially writing subsequent overlapping line sets with abbreviated write cycles, the entire display can therefore be written much faster than if individual lines are fully written sequentially. This line set writing strategy depends on there being a repetition of the pixel pattern from one line to the next. In fact, the probability of this condition occurring is very high. Due to the high line density of the EPIDs in question, the number of lines comprising a single character is large. For example, a 70 line x 25 line matrix of 1750 pixels can be used as the area for expressing a single character. In this case, the pattern of pixels comprising the usual characters is very repetitive. In Figure 4, this principle is illustrated using a matrix of only 22 x 22 lines, that is, the lines centered within the entire 29 x 31 line matrix shown. The top of the "T" starts at (r5, c5) and ends at (r9, c26). The meaning of the X in line 5 is explained below. The stem of the "T" begins at (r10, c13) and ends at (r26, c17). As can be readily seen, the top of the "T" consists of identical rows of pixels and the stem of the "T" consists of 17 identical rows of pixels. The "T" shown in Figure 4 is an example of the application of the present inventive method to writing two sets of lines at half the usual write cycle time (twice the write speed). In particular, the following steps would be performed to display the "I" shown in Figure 4:

Load c1-c29 with data for r1 (0,0,0,0,0,0,0,0...0)

Write r1 and r2 simultaneously (apply "1" to r1 and r2)

Load c1-c29 (the grid rows) with data for r2 (0,0,0,0,...0)

Describe r2, r3

Loading grid with r3 data

Describe r3, r4

Loading grid with r4 data

Describe r4, r5

Note: For the purposes of this example, r5 has been chosen as the first line to have "1s" or written pixels, and it should be the first line of the "top" of the "T". However, since r5 is a transition line, i.e. a transition from non-written to written pixels, it will not be fully written and will instead be only weakly written or half written. The reason for this is that in each write cycle, the information is only "half written" as it is twice the speed of a normal cycle. The next cycle is necessary to fully write the information, but only if the next cycle uses the same data. In the case of a transition line, successive lines have different data. Since there are so many pixel lines in functional displays, the loss of a small number of transition lines and/or pixels does not cause a significant loss of readability. Now returning to the writing process:

Loading grid with r5 data (0,0,0,0,1,1,1,1,1,1,1,1...1,0,0,0)

Describe r5, r6

Load grid with r6 data (same as r5 data)

Write r6, r7 (since r6 was previously "half" written with r5 data in the previous cycle and since the r5 data was the same as the r6 data, r6 is fully written in the following cycle)

Loading r7 data

Describe r7, r8

Loaded r8

Describe r8, r9

Loading r9

Describe r9, r10 (r10 is another partial transition line, ie it is the transition from the top of the "T" to the stem of the "T". Since the r9 data is written on line 10, part of it, that is, the one that should contain unwritten pixels - the Xe - will be weak or half-written.)

Load r10 Describe r10, r11

Continue to line 26, where:

Loaded r26

Describe r26, r27 (forms another transition line)

Loaded r27

Describe r27, r28 etc.

The above should represent an embodiment of the present inventive method. It is further understood that instead of two line set writing, three, four, or more lines may be written simultaneously with corresponding increases in speed and increases in transition lines which will be of varying intensity depending on the number of repetitions of writing operations to these transition lines. For example, in writing a set of four lines, when a transition occurs from empty to written pixels, there will be three transition lines, the first being the faintest and the last being the darkest. The fourth written line will be completely written. Similarly, in a transition from written to unwritten pixels, there will be three transition lines, the first being the darkest and the last being the faintest. The fourth line will not be written. Of course, when writing a set of four lines, the benefit of increasing the writing speed above the normal speed would be used to achieve a four-fold speed increase.

Figure 5 is a generalized flow diagram of the steps of the present inventive method for operating an EPID in a multi-line write mode. It would be expected that operator selection of the display writing speed would be available so that the operator could select the speed and clarity. This type of selection is currently offered to operators when printing on dot matrix printers, that is, enhanced printing has a higher dot density but takes a longer printing time. Accordingly, the operator first enters the number of lines to be written in each write cycle 32. From this input the write cycle time (write speed) is set 34. The greater the number of lines written simultaneously in each write cycle, the faster the writing speed. Of course, the operator input could be expressed as a selection of the writing speed, with the operator selecting from a range of speeds corresponding to the number of lines written simultaneously. The flow chart shown in Figure 5 concerns the display of a single complete image (frame) on the EPID. This algorithm would be applied over and over again under the control of programming at the next higher level. The operator would not be prompted as to the speed of operation for each frame displayed. Such information would initially be set by request or specification and then changed by interrupt if necessary. After determining the row set size for writing, writing is started from the first row 36. (Of course, it would also be equally conceivable to load rows with data and write columns.) The processor then enters a loop in which data for the current row is loaded into the column rows (here, grid rows) 38. The data is written simultaneously on the current row and the next x-1 rows by enabling the rows with a logical "1" 40, where x is the number of rows in the selected write set. Thus, in the first write cycle in a write mode with a set of 4 rows, row 1 and the next (4-1) or 3 lines, i.e. lines 2, 3 and 4. Note that the "1" state can correspond to a variety of voltages depending on the EPID in question, e.g. whether the EPID is a triode or tetrode. A voltage of 0 volts has been used to enable writing to triodes and in these cases represents a logic "1" or enable state. The line set is written for a write cycle time set (divided) by the size of the line set. This continues until all lines are written 42, 44, whereupon control is returned to the next higher level in the program. Of course, other line writing sequences can be employed using a multiple line writing strategy, for example vertical lines could be written from left to right or right to left, horizontal lines could be written from bottom to top or from the center to the outer perimeter, etc.

Claims (7)

1. A method for reducing the time required to write a frame of display data (36, 38) consisting of a plurality of rows (20) of displayable pixels on a high-density electrophoretic display (10) which requires a predetermined time period to completely write a line, said method comprising the steps of: (a) simultaneously writing (40) a set of at least two adjacent lines (20) with said display data (36, 38) in a first period of time which is shorter than said predetermined period of time for completely writing a line; (b) shifting the elements (44) of said set of rows so that said set contains at least one new row and at least one original row; (c) writing said shifted set of lines (40) in a second period of time shorter than said predetermined period of time after said step of shifting (44); and (d) Repeating steps (b) and (c) until said frame is completely written. 2. The method of claim 1, wherein the length of each of said first and second time periods decreases with increasing numbers of said elements in said row set. 3. The method of claim 2, wherein each of said first and second time periods is approximately said predetermined time period divided by the number of elements in said row set. 4. The method of claim 1, wherein said pixels of said original line are written with a pixel intensity that approaches a predetermined pixel intensity for a pixel that has been written for said predetermined amount of time to completely write a line when said new line contains pixels with a predetermined offset. 5. The method of claim 4, wherein the method further includes the step of selecting the number of rows in the said set (32). 6. The method of claim 5, wherein the method further includes the step of setting the length of said time period (34) to be shorter than said predetermined time period for completely writing a line when a change in the number of lines in said set occurs. 7. The method of claim 6, wherein said predetermined offset is equal to the offset of a reference line located at an edge of said display (10).