JP2012098733A - Method for driving bistable electro-optic display - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a controller that can provide accurate gray levels in an electro-optic display without the need to flash solid color on the display frequently.SOLUTION: A bistable electro-optic display has a plurality of pixels, and each pixel is capable of displaying at least three gray levels. The display is driven by a method including: a step of storing a look-up table containing data representing impulses necessary to convert an initial gray level to a final gray level; a step of storing data representing at least an initial state of each pixel of the display; and a step of generating an output signal representing an impulse necessary to convert the initial state of at least one pixel of the display to a desired final state thereof, as determined from the look-up table. Furthermore, a method for reducing a remnant voltage of the electro-optic display is provided.

Description

  The present invention relates to a method for driving a bistable electro-optic display and an apparatus using such a method. More specifically, the present invention relates to a driving method and apparatus intended to allow more precise control of the gray state of a pixel of an electro-optic display. The invention also relates to a method that enables long term direct current (DC) balance of drive impulses applied to electrophoretic displays. The present invention specifically contemplates the use of particle-based electrophoretic displays, but is not exclusive. In the electrophoretic display, one or more types of electrically charged particles float in the liquid and are moved through the liquid under the influence of an electric field that changes the appearance of the display.

  In one aspect, the present invention relates to an apparatus that enables an electro-optic medium that is responsive to the polarity of an application that is driven using a circuit that is intended to drive a liquid crystal (the liquid crystal material is not responsive to polarity) display.

  The term “electro-optic” as applied to a material or display is used herein in the conventional sense in imaging technology relating to materials having first and second display states. The first and second display states differ in at least one optical characteristic in which the material changes from the first display state to the second display state by application of the electrical field of the material. The optical properties are usually colors that can be perceived by the human eye, but the optical properties are optical transmission, reflectance, luminescence, or for electromagnetic wavelengths outside the visible range (for displays intended for machine reading). It may be another optical property such as pseudo-color that senses changes in reflectivity.

  The term “gray state” is used herein in the conventional sense of imaging technology to refer to an intermediate state between two extreme optical states of a pixel and the black-white between these two extreme states. The transition is not necessarily implied. For example, some of the patents and published applications describe electrophoretic displays where the extreme states are white and deep blue, so that the intermediate “gray state” is actually light blue. Certainly, as already described, the transition between the two extreme states is not a color change at all.

  “Bistability” and “Bistability” are used herein to refer to the conventional meaning of a technique that refers to a display that includes display elements having first and second display states that differ in at least one optical property. Used in This assumes that any given element is driven by an addressing pulse for a finite period of time and then assumes either its first or second display state, and after the addressing pulse ends, this state will be at least several times It continues for the minimum duration of the addressing pulse required to change the state of the display element (eg at least 4 times). Application No. 10/063, 236, filed Apr. 2, 2002 (see the corresponding International Application Publication No. WO 02/0779869), is a cognate genus of some particle-based electricity that is competent in grayscale. Several other types of electro-optic displays are the same where the electrophoretic display is in the extreme black and white state and the gray state intermediate the electro-optic display. This type of display is correctly called “multi-stable” rather than bistable, and for convenience the term “bistable” is used to cover both bistable and multi-stable displays. Used in the description.

  The term “gamma voltage” is used herein as an internal voltage reference used by a driver to determine the voltage applied to a pixel of a display. It should be understood that the bistable electro-optic medium does not display a one-to-one correspondence type between the applied voltage and the optic state characteristics of the liquid crystal. The use of the term “gamma voltage” herein is not exactly the same as that used in conventional liquid crystal displays. In a conventional liquid crystal display, the gamma voltage determines the inflection point of the voltage level / output voltage curve.

  The term “impulse” is used herein in the conventional sense of voltage integration over time. However, some bistable electro-optic media function as charge converters, and with such media, a further definition of impulse can be used, ie the integration of the overtime current (equal to the total applied charge) . Appropriately defined impulses should be used, and the medium functions as a voltage-time impulse converter or a charge impulse converter.

  Several types of bistable electro-optic displays are known. One type of electro-optic display is a bichromal member type that rotates as described above. For example, U.S. Pat. Nos. 5,808,783; 5,777,782; 5,760,761; 6,054,071; 6,055,091; 6,097,531. No. 6, 128, 124; No. 6, 137, 467 and No. 6, 147, 791 (the type of display is often referred to as a “rotating bike roman ball” display, but “rotating” "Bichromal members" are some of the patents more precisely mentioned above, where the rotating member is not a sphere). Such displays use a large number of small bodies (usually spherical or cylindrical) that have different optical features, two or more different sections, and have internal dipoles. These bodies float in the vesicles filled with liquid in the matrix, and the vesicles are filled with liquid so that the body is free to rotate. The appearance of the display changes to apply an electric field, thus rotating the body to various positions, and the appearance section of the body is visible through the viewing surface.

  Another type of electro-optic medium uses an electrochromic medium, for example, a nanochromic film-like electrochromic medium is at least a portion from a plurality of dye molecules with a reversible color change attached to a metal oxide semiconductor and an electrode. Including electrodes to be formed. For example, O'Regan B.I. Nature 1991, 353, 757 and Wood D. et al. , Information Display, 18 (3), 24 (March 2002). In addition, Bach U. Adv. Mater. 2002, 14 (11), 845. This type of nanochromic membrane has also been described, for example, U.S. Patent No. 6,301,038, International Application Publication No. WO 01/27690 and co-generic application series 60 / 365,368; 60 / 365,369; 60/365, 385 and 60/365, 365 all filed March 18, 2002, application series 60/319, 279; 60/319, 280, and 60/319, 281 (March 31, 2002). Application) and application series 60/319, 438 (filed July 31, 2002).

  Another type of electro-optic display is a particle-based electrophoretic display (which has been the subject of much research and development over the years). A plurality of charged particles move through suspended fluid under the influence of an electric field. Electrophoretic displays have the characteristics of good brightness and contrast, wide-angle field of view, state bistability and low power consumption when compared to liquid crystal displays. Nevertheless, the long-term image quality problems of these displays have prevented their widespread use. For example, the tendency of the particles that make up an electrophoretic display to be stable results in an insufficient and short display life.

  A number of recently encapsulated electrophoretic media patents and applications assigned to or in the name of Massachusetts Institute of Technology (MIT) and E Ink Corporation have recently been published. Such capsule-encapsulated media comprises a number of small capsules, each of which comprises an electrophoretic-mobile particle and a capsule wall around the inner surface, suspended in a liquid suspension medium. Generally, the capsule itself is held in a polymeric binder to form a coherent layer disposed between the two poles. Media encapsulated with this type are described, for example, below. U.S. Pat. Nos. 5,930,026; 5,961,804; 6,017,584; 6,067,185; 6,118,426; 6,120,588; 6, 120, 839; 6, 124, 851; 6, 130, 773; 6, 130, 774; 6, 172, 798; 6, 177, 921; No. 232,950; No. 6,249,721; No. 6,252,564; No. 6,262,706; No. 6,262,833; No. 6,300,932; No. 6,312 304, No. 6,312,971; No. 6,323,989; No. 6,327,072; No. 6,376,828; No. 6,377,387; No. 6,392,785 No. 6, 392, 786; No. 6, 413, 790; No. 6, 42 No. 6,445,374; Nos. 6,445,489 and 6,459,418, and US Patent Publication Nos. 2001/0045934; 2002/0019081; 2002/0021270. No. 2002/0053900; No. 2002/0060321; No. 2002/0063661; No. 2002/0063677; No. 2002/0090980; No. 2002/106847; No. 2002/0113770; No. 2002/0130832; 2002/0131147; 2002/0154382 and published international applications; WO 99/53373; WO 99/59101; WO 99/67678; WO 00/05704; WO 00/20922; WO 00/20922; 3 WO 00/38001; WO 00/36560; WO 00/20922; WO 00/36666; WO 00/67110; WO 00/67327; WO 01/07916; WO 01/08241; No. WO 01/17029; WO 01/17041.

  A number of the above-mentioned patents and applications show that the wall around the separated microcapsules of the encapsulated electrophoretic medium can be replaced by a continuous surface. Therefore, so-called polymer-spreading electrophoretic displays are those in which the electrophoretic medium comprises a plurality of separated water droplets of electrophoretic liquid and a continuous surface of polymeric material, and within such a polymer-spreading electrophoretic display. Liquid separated water droplets can be regarded as capsules or microcapsules, but separated capsule membranes are associated with individual water droplets. See, for example, WO 01/02899, page 10, lines 6-19. See copending genus application serial 09 / 683,903 (filed February 28,2002) and corresponding international application PCT / US02 / 06393. Thus, for the purposes of this application, such polymer-dispersed electrophoretic media are considered an auxiliary type of electrophoretic media encapsulated.

  Encapsulated electrophoretic displays generally refuse to set the failure mode of classification and traditional electrophoretic devices and have the ability to print or coat displays of various widths with a flexible solid base (It is intended to include all forms of printing and coating, including the prescribed coatings such as patch die coating, slot or extrusion coating, using the word “printing”) Unrestricted: Forward or reverse roll coating; Gravure coating; Air knife coating; Spray coating; Meniscus coating; Spin coating; Brush coating; Air knife coating; Silk screen printing process; Air printing process; thermal printing processes; ink jet printing processes and other similar techniques). Therefore, the display can consequently be flexible. Further, since the display media can be printed, the display itself can be made at a low cost.

  A related type of electrophoretic display is the so-called “microcell electrophoretic display”. In microcell electrophoretic displays, charged particles and suspended fluid are not encapsulated in microcapsules, but instead are held in multiple cavities formed in a carrier medium, usually a polymer film. is there. See, for example, International Application Publication No. WO 02/01281 and Published US Application No. 2002-0075556, both assigned to Sipix Imaging, Inc.

  The function of bistable or multi-stable particle-based electrophoretic displays and other similarly functioning electro-optic displays is in striking contrast with the behavior of conventional liquid crystal displays (“LC”). The behavior of the twisted nematic liquid crystal is not bistable or multi-stable, but operates as a voltage converter. Thus, applying a predetermined electric field to a pixel, such as a display, produces a specific gray level at the pixel, regardless of the gray level previously present in the pixel. Furthermore, the LC display is not only in one direction, but a reversal transition from a brighter state to a darker state (from a “dark” but not transmittable state to a transmittable and “bright” state) reduces or eliminates the electric field. The effect comes out. Finally, the gray level of the pixels of the LC display does not react to the electric field polarity, only the magnitude, and in fact, for technical reasons, commercial LC displays usually reverse the polarity that drives the field at periodic intervals. .

In contrast, bistable electrophoretic displays function as impulse converters so that the approximation is independent of the electric field at which the final stage is applied to the pixel and the time for which this field is applied. Furthermore, at least in the case of many particle-based electro-optic displays, it can be seen that: Not necessarily commutative. The impulses required to change a given pixel with equal changes in gray levels (as judged by eye or with standard optics) need not be constant, for example, When considering a display that can be displayed with gray levels of 0 (white), 1, 2, or 3 (black), it is beneficial to be spatially separated. (Reflectivity is linear as the space between levels is measured by the eye and instrument, but other spaces can also be used. For example, the spacing is linear in L ^* or provides a specific gamma Gamma 2.2 fits well in monitors and uses the current display as an alternative to the monitor, but the use of a similar gamma may be desired.) Pixels from level 0 to level 1 (see “0-1 transition” below) It can be seen that the pulse required to change until is not the same as requiring 1-2 or 2-3 transitions. In addition, some systems display a “memory” effect, so that impulses that require a 0-1 transition may cause a somewhat singular pixel to be 0-0-1, 1-0-1 or 3-0- Depends on whether one transition is received. (Here, the symbol “xyz”, where x, y, and z all mean the continuous optical states 0, 1, 2, or 3 that visit successively in the optical state.) These problems Can be reduced or overcome with all pixels displaying one of the crisis states for a substantial period of time, but the resulting solid color of “flash” is unacceptable . For example, reading an ebook can often require a textbook that scrolls the screen if the display flashes to solid black or white and can be diverted or lose place. Moreover, such a display flash can increase the energy consumption of the display and reduce the lifetime of the display. Finally, impulses that require at least a specific transition in some cases are affected by temperature and display operating time by the time a specified pixel lasts in a specific optical state for a predetermined transition. These factors that are desired to ensure accurate gray scale performance are corrected.

  In one aspect, the present invention seeks to provide a method and controller that can provide accurate gray levels for electro-optic displays without having to frequently flash solid colors on the display.

  Furthermore, as already apparent from the above discussion, the driving requirements of a bistable electro-optic medium are not suitable for use in a bistable electro-optic media-based display (active matrix liquid crystal display). Presents an unmodified driver designed to drive (AMLCD). However, such AMLCD drivers are readily available commercially on an off-the-shelf basis, have a high permissible voltage and a high pin-count package, and are not expensive. Therefore, AMLCD drivers are attractive for bistable electro-optic displays. However, similar driver customs designed for bistable electro-optic media based displays are more expensive and affect substantial design and manufacturing time. Therefore, it is advantageous to modify the cost and development time of modifying an AMLCD driver for use with a bistable electro-optic media based display. The present invention seeks to provide such a method and a modified driver.

  Furthermore, as already mentioned, the present invention relates to a method for driving an electrophoretic display, which enables long-term DC-balancing of drive impulses applied to the display. In order to preserve image stability, maintain symmetrical switching characteristics, and provide the maximum useful service life of the display, the encapsulated and other electrophoretic displays are accurately DC- It is believed that it needs to be driven with a balanced waveform (i.e., the integral of the current for the time of any particular pixel in the display should be kept at zero for the extended operating time of the display). Conventional methods for maintaining accurate DC-balance require accurate regulation power supply, accurate voltage modulation driver for gray scale, and crystal oscillator for timing. These and similar components add significant cost to the display.

  Furthermore, even with such expensive components further, true DC balance cannot be obtained. Experience shows that many electrophoretic media are asymmetric and have current / voltage (I / V curves). Although the invention is not limited by this summary, this asymmetric curve is internal to the medium by an electrochemical voltage source. In this asymmetric curve, the current when the medium deals with the extreme optical state (black) is not the same as the current when the medium deals with the opposite extreme optical state (white). The voltage is carefully controlled to make these two cases exactly the same.

  An extension of the DC imbalance of the electrophoretic medium used in the display is confirmed by measuring the potential of the open circuit electrochemical (referred to below as the “residual voltage” of the medium for convenience). If the pixel's residual voltage is zero, it is fully DC balanced. If the residual voltage is positive, the DC is unbalanced in the positive direction. The present invention uses residual voltage data to maintain the long-term DC balance of the display.

Accordingly, one aspect of the present invention is that, as a conventional display technology having a plurality of pixels, each pixel capable of displaying at least 3 gray levels, the extreme black and white states remain at the gray level. 2 gray levels. Therefore,
Storing a lookup table containing data representing impulses that need to be converted from an initial gray level to a final gray level;
Storing data representing at least an initial state of each pixel of the display;
Receiving an input signal representing a desired final state of at least one pixel of the display;
Generating an output signal representing an impulse that needs to be converted from an initial state of the one pixel to a desired final state determined from the lookup table.

  This method may be referred to as the “look-up table method” of the present invention for convenience in the following.

The present invention also provides a device controller for using such a method. The controller is
Storage means adapted to store both a look-up table containing data representing impulses that need to be converted from an initial gray level to a final gray level and data representing at least the initial state of the display;
Input means for receiving an input signal representing a desired final state of at least one pixel of the display;
A calculation means for determining stored data from an input signal; and an output signal representative of said pixel, look-up table, and impulse before impulse, which is required to change the initial state of said one pixel to a desired final state. And output means for generating a controller.

The present invention also provides a method for driving a bistable electrical optic display having a plurality of pixels, each of which is capable of displaying at least 3 gray levels. The method is
Storing a lookup table containing data representing impulses that need to be converted from an initial gray level to a final gray level;
Storing data representing at least an initial state of each pixel of the display; receiving an input signal representing a desired final state of at least one pixel of the display;
Converting the initial state of the one pixel to a desired final state so as to determine an output signal from the look-up table that substantially represents a period for which a drive voltage is applied to the pixel. Generating an output signal representing the required impulse.

The present invention also provides a device controller using such a method. The controller is
It is tuned to store both a lookup table that stores data that represents impulses that need to be converted from an initial gray level to a final gray level, and data that represents at least the initial state of each pixel in the display. Storage means,
Input means for receiving an input signal representing a desired final state of at least one pixel of the display;
Calculation means for determining stored data representing an initial state of the pixel from an input signal and determining an impulse, a look-up table that requires changing the initial state of one pixel to a desired final state When,
A controller comprising: an output signal representative of the impulse; and an output means for generating an output signal representative of a period for which a substantially constant drive voltage is applied to the pixel.

In another aspect, the present invention provides a device controller for use in the method of the present invention. The controller
Storage means for adjusting to store both data representing impulses that need to be converted from an initial gray state to a final gray state and data representing at least the initial state of each pixel of the display;
Input means for receiving an input signal representing a desired final state of at least one pixel of the display;
Stored data representing an initial state of the pixel from an input signal, a lookup table required to change the initial state of the one pixel to a desired final state, calculation means for determining an impulse;
An output representing a plurality of pulses varying at least one of the impulse output signal voltage and duration and an output signal representing a zero voltage after a predetermined period of time are generated.

In another aspect, the present invention provides a driver circuit having an output line that is tuned to connect to a drive electrode of an electro-optic display. The driver circuit has first input means for receiving a plurality of (n + 1) bit numbers representing voltage and signal electrodes installed at the driver electrodes, and second input means for receiving a clock signal. Upon receiving the clock signal, the driver circuit displays the selected voltage on the output line. In a preferred form of this driver, it can be any one of ²ⁿ isolation voltages between selected R and R + V. Here, R is a predetermined reference voltage (generally, a common front electrode voltage of an active matrix display as will be described in more detail later). V can be a maximum value different from a reference voltage that the driver circuit can assert, or any one of ²ⁿ isolation voltages between R and R-V. These selected voltages can be distributed linearly over the range of R ± V or can be distributed non-linearly. That is, the non-linearity can be distributed to two or more gamma voltages arranged within a specified range. Define a linear regime between each gamma voltage and the adjacent gamma or reference voltage.

  In another aspect, the present invention provides a driver circuit having an output line that is adjusted to be connected to a drive electrode of an electrophoretic display. This driver circuit has first input means for receiving a plurality of 2-bit numbers expressing the voltage and polarity of a signal to be arranged on the drive electrode, and second input means for receiving a clock signal. Upon receiving the clock signal, the driver circuit displays the selected voltage from R + V, R to R−V (where R and V are as defined above) on its output line.

  In another aspect, the present invention provides a method of driving an electro-optic display that displays a residual voltage, specifically an electrophoretic display. This method includes:

(A) applying a first drive pulse to a pixel of the display;
(B) measuring the residual voltage of the pixel after the first drive pulse;
(C) applying a second drive pulse to the pixel following the measurement of the residual voltage and applying to the magnitude of the second drive pulse controlled by the measured residual voltage to reduce the residual voltage of the pixel; It is a method of inclusion.

  This method is referred to for convenience as the “residual voltage” of the present invention as described later.

  As already mentioned above, the look-up table appearance of the present invention provides an electro-optic display method and controller having a plurality of pixels, each pixel capable of displaying at least 3 gray levels. The present invention can of course be applied to electro-optical displays having, for example, a number of gray levels of 4, 8, 16 or more.

  Driving a bistable electro-optic display, as already mentioned, requires a completely different method that is different from the method used to drive a liquid crystal display ("LCD"). In a conventional (non-cholesterol) LCD, applying a specified voltage to a pixel for a sufficient period of time causes the pixel to achieve a specific gray level. Furthermore, LC materials react not only to the magnitude of the electric field, but also to its polarity. In contrast, bistable electro-optic displays display the function as an impulse converter, so there is no one-to-one mapping between applied voltage and the reached gray state. That is, a predetermined gray state that is applied to the pixel and varies with the “initial” gray state of the associated pixel is achieved. Furthermore, since the bistable electro-optic display needs to be driven in both directions (white to black, black to white), it is necessary to specify both the polarity and the magnitude of the required impulse.

  In this regard, it can be seen that it is desirable to define certain terms used herein by conventional means of display technology. Most of the discussion below focuses on one or more pixels of the display that change a single grayscale transition (ie, changing from one gray level to another) from an “initial” state to a “final” state. Obviously, the initial and final states are specified with respect to the single transition considered, so in most cases the pixel will vary from the transition of the previous “initial” state and further transition after the “final” state. Fluctuate. As described below, certain embodiments of the present invention consider not only the initial and final states of the pixel, but also the “previous” state of the pixel that exists before the initial state is reached. If it is necessary to distinguish between a number of previous states, the term “first previous state” is the state of the associated pixel where there is one (non-zero) transition before the initial state. Used as for. The term “second previous state” is used to refer to the state of the associated pixel where there is one (non-zero) transition before the previous state, and so on. The term “non-zero transition” is used to refer to a transition that affects a change in at least one unit of the gray state. The term “zero transition” can be used in reference to “transitions” that do not affect the change in the selected gray scale (other pixels of the display can simultaneously vary non-zero transitions).

  As will be apparent to those skilled in the art of image processing, a simple embodiment of the method of the present invention considers only each pixel and final state. In such a case, the lookup table is two-dimensional. However, as already mentioned above, certain electro-optic media display memory effects and when using such media to generate output signals, not only the initial state of each pixel, but (at least) It is also desirable to consider the same previous pixel state as before. In that case, the lookup table is three-dimensional. Therefore, as a result, the lookup table has a dimension of 4 or more (when only the first and second previous states are considered).

  From a formal mathematical point of view, the present invention provides a display (eg Information about the physical state (temperature and total operating time) can be viewed to include an algorithm that generates a function V (t) that can apply a pixel that affects the transition to the desired final state. From this formal point of view, the controller of the present invention can be viewed as an essentially physical embodiment of this algorithm, which acts as an interface between a display that desires display information and an electro-optic display.

  Since the physical state information has been ignored for some time, the algorithm according to the invention is encoded in the form of a look-up table or transition matrix. This matrix has one dimension for the desired final state. For other state matrices, the (initial and any previous state) matrix uses a calculator. The elements of the matrix contain a V (t) function to be applied to the electro-optic medium.

  The elements of the lookup table or transition matrix can have a variety of forms. In some cases, each element may include the singular. For example, using an electro-optic display, modify a driver circuit capable of outputting many different voltages both above and below a high-precision voltage reference voltage and apply the required voltage to the pixel for a standard predetermined period of time To do. In such a case, each entry in the lookup table may simply have a signed integer format that specifies which voltage is to be applied to a given pixel. In other cases, each element may include a continuous number for different portions of the waveform. For example, the embodiments of the invention described below use single or double prepulse waveforms and specify such waveforms that necessarily require several numbers for different portions of the waveform. In addition, in the following, embodiments of the present invention effect pulse length correction by applying a predetermined voltage to a pixel during a selected one of a plurality of auxiliary scans in a full scan period. Applicable. In such embodiments, the elements of the transition matrix may have a continuous bit format that specifies whether a given voltage is applied during each auxiliary scan period of the associated transition. Finally, as will be discussed in detail later, in the case of a temperature correction display etc., the elements of the lookup table to form a function (in fact, more accurate coefficients for such various terms) are useful. obtain.

Obviously, in some embodiments of the present invention, the lookup table used is very large. Using an extreme example, consider the process of the present invention for a 256 (2 ⁸ ) gray level display (using an algorithm that considers the previous and final two previous states). The mandatory 4D lookup table has 2 ³² entries. The required 4D lookup table has 2 ³² entries. If each entry requires (for example) 64 bits (8 bytes), the total size of the lookup table is about 32 gigabytes. If the step of storing the amount of data on the desktop computer is presented as having no problems, there may be a current problem with the portable device. In practice, however, the size of each large lookup table can be substantially reduced. In many cases, they exist only in a small number of types of waveforms that require a large number of different transitions (eg, each pulse length of a typical waveform that varies between different transitions). As a result, the individual entry length of the lookup table can be reduced by creating each entry. Each entry is (a) a pointer to an entry in the second table that specifies one of the small number of types of waveforms being used, and (b) an aspect that fluctuates the transition to which this general waveform relates. Contains a small number of parameters to specify.

  Lookup table entry variables may be predetermined by an empirical optimization process. In essence, the pixel is set to an initial state, the impulse estimated to be generally equal to what is needed to achieve the desired final state, the final state of the pixel is measured, and the actual final state And the deviation (if any) between the desired final state. The process then repeats the impulse correction until the deviation is less than a predetermined value. The predetermined value can be determined by the performance of the instrument that measures the final state. In the case of a method that considers one or more previous states of a pixel, it is generally convenient to first determine an impulse that requires the initial and all preceding states used when defining the impulse. is there. It is convenient to “good tune” the impulse that allows it to differ from the previous state.

  Preferably, the method provides for impulse correction that allows for variations in display temperature and / or total operating time. Since some electro-optic media “age” and its movement is changed after an expansion operation, a correction of the operation time may be required. Such a modification can be made in one of two ways. First, the look-up table can be extended with additional dimensions for each variable, each variable being considered when calculating the output. Obviously, when dealing with continuous variables such as temperature and manipulation, the number of runs is quantized to keep the look-up table in a practical finite size. To find the waveform and apply it to the pixel, the computing means simply selects the lookup table entry of the table that is closest to the measured temperature. Instead, the calculation means can look up two adjacent look-up tables on either side of the measured continuous variable to provide a more accurate temperature correction, required with variable measured intermediate variables Appropriate inter-interpolation algorithms are applied to obtain correct entries. For example, assume a matrix that includes an entry for a temperature of 10 ° C. If the actual temperature of the display is 25 ° C., the calculator may look up the 20 ° C. and 30 ° C. entries. It should be noted that the change in temperature characteristics of the electro-optic medium is often non-linear, and the set of temperatures at which the lookup table storage entries are not assigned linearly is the fastest at high temperatures. This may be sufficient at a low temperature interval of 20 ° C. between the look-up tables, but may be desired at a high temperature interval of 5 ° C.

  Alternative temperature / operation time correction methods can use look-up table entries in the form of functions of physical variable (s) or possibly use the exact coefficients of the standard terms of such functions There is sex. Considering simply the case of a display using a time correction scheme, the transition for each pixel of different time length in each time correction scheme is processed by applying a constant voltage (of either polarity). Thus, any correction of the environment variable may consist of only a single entry in the look-up table that represents and signs a period of time with a constant current applied and polarity. If it is desired to correct such a display of temperature variables (eg time T when a constant voltage needs to be applied with a specific transition at temperature t), the following equation is given:

T _t = T ₀ + AΔt + B (Δt) ²
Where T ₀ is the time required at a certain standard temperature, and in general, the midpoint intended to manipulate the temperature range of the display and Δt is between t and the temperature at which T ₀ was measured. It is a difference. An entry in the lookup table is composed of a value of T ₀ , A and B of a specific transition of a predetermined entry relationship, and the calculation means can use a coefficient for calculating T _t at the measured temperature. More generally, the calculation means uses a function defined to find the appropriate initial and final state lookup table entries to correlate, and then calculate the true output signal by the entries. The true output signal relates to other variables that should enter the account.

  The correlation temperature to be used for the temperature correction calculation is the correlation temperature of the electro-optic material of the correlation pixel, which can be significantly different from the ambient temperature. Particularly in the case of a display intended for outdoor use, for example, sunlight acting through a protective front sheet can cause the temperature of the electro-optic layer to be substantially higher than the surroundings. In fact, for a large bulletin board outdoor sign, the temperature can vary between different pixels of the same display. For example, if a part of the display falls into the shadow of a nearby building, it will arouse enough sunlight. Thus, it may be desirable to incorporate one or more temperature capsules or other temperature sensors in or near the electro-optic layer. For large displays, it may also be desirable to provide an interpolation method between temperatures by multiple temperature sensors that estimate the temperature of each particular pixel. Finally, for large displays formed from multiple modules that can be individually placed, the method and controller of the present invention can provide different operating times for pixels of different modules.

  The method and controller of the present invention also allows the residence time of the particular pixel being driven (ie, the period since the pixel has continued to undergo a non-zero transition). In some cases, it can be seen that the impulse requires a predetermined transition that varies with the resident time of the optic pixel. Therefore, it is desirable and necessary to vary the impulse applied with a predetermined transition as a function of the resident time of the pixel in the initial optical state. To accomplish this, the lookup table may optionally include an additional dimension indexed by a counter that indicates the resident time of the pixel in the initial optical state. In addition, the controller requires an additional storage area that includes a counter for each pixel of the display. It further requires a display clock that increments for each counter whose value is stored in each pixel at set intervals. The length of this interval must be the integrated frame time of the display. Therefore, it must be less than one frame time. The size of this counter and clock period is determined by the length of time the impulse varies and the required time resolution. For example, the step of storing 4 bits for each pixel may vary in 0.25 second intervals beyond the second period of 4 (4 seconds × 4 counts / second = 16 counts = 4 bits) in the impulse. Enable. The counter is optionally reset when certain events occur, such as a transition of a pixel to a new state. When the counter reaches the maximum value, the counter can be configured to either “roll over” to counter zero or to maintain the maximum value until reset.

  The look-up table method of the present invention is of course modified to take into account any other physical parameters that have a detectable effect on impulses that need to affect any one of the specific transitions of the electro-optic medium. obtain. For example, the method can be modified to incorporate ambient humidity correction if the electro-optic medium is detected to sense humidity.

  For bistable electro-optic media, the look-up table has characteristics that are the same for any zero transition of the initial and final states of the pixel. The entry is zero, or in other words, no voltage is applied to the pixel. As a result, no impulse is applied if the pixels of the display do not change for a predetermined interval. This allows ultra-low power operation, as well as electro-optic media is not overdriven while a static image is displayed. In general, the lookup table should only hold information about non-zero transitions. In other words, for two images I and I + 1, if a given pixel is in the same state at I and I + 1, state I + 1 is not stored in the previous state table, and further information remains until the pixel transitions Not stored.

  As will be apparent to those skilled in the modern electrical arts, the controller of the present invention may have physical form variations and may use convenient processing components. For example, the method of the present invention may be practiced using a multipurpose digital computer (eg, one or more analog converters “DAC”) associated with a suitable device, and digital output from the computer to the appropriate voltage applied to the pixel. Can be converted to Alternatively, the method can be practiced using a specific integrated circuit (ASIC) application. In particular, the controller of the present invention may have the form of a video card that can be inserted into a personal computer, and displays an image generated from the computer on an electro-optic screen instead of or in addition to an existing screen such as an LCD. Make it possible to do. Since the structure of the controller of the present invention is well within the state of the art of image processing technology, its circuitry need not be described in detail herein.

  The preferred physical embodiment of the controller of the present invention is an integrated circuit (IC) timing controller. The IC receives incoming image data and outputs control signals to the collection of data and selected driver ICs. This is to generate the correct voltage at the pixel that produces the desired image. The IC may receive image data via access to a memory buffer that contains the image data (where the IC retrieves the image data) or may receive a signal intended to drive a conventional LCD panel. The IC may also receive a serial signal that includes information that seeks to perform the required impulse calculations. Alternatively, the timing controller can be implemented in software or incorporated as part of the CPU. The timing controller may also have the ability to measure any external parameters that affect the operation of the display, such as temperature.

    The controller can operate as follows. The lookup table (s) is stored in a memory accessible to the controller. For each pixel, all of the required initial, final and (optionally) previous and physical state information in turn are supplied as input. The state information is then used to calculate an index into the lookup table. In the case of quantized temperature or other corrections, the return value from this lookup is one voltage or voltage-time layer. The controller repeats the process for the two blanket temperatures in the lookup table and then interpolates between the values. Algorithmic temperature correction is a lookup return value that is one or more parameters that can be inserted into the equation along the temperature to determine the correct formation of the drive pulse, as described above. This procedure can be similarly achieved for any other system that requires real-time correction of the drive pulse. One or more of these system variables are determined by, for example, programmable register values or memory locations in the EPROM. This memory arrangement sets the display panel during configuration to optimize display operation.

  An important feature of display controllers is that unlike most displays, in the most practical case, several perfect scans of the display are required to complete the image update. A continuous scan requiring one image update should be considered a non-disruptive unit. If the display controller and the image source operate asynchronously, the controller must be used to calculate the applied impulse and ensure that the data remains constant across all scans. This can be achieved in one of two ways. First, incoming image data can be stored in a separate buffer by the display controller (alternatively, if the display controller accesses the display buffer via the dual port memory, the access can be locked out from the CPU). Second, in the first scan, the controller can store the calculated impulse in the impulse buffer. The second option has the advantage that the overhead of scanning the panel collides only once through the transition, and the data that leaves the scan can be output directly from the buffer.

  Optionally, update imaging may be directed in an asynchronous manner. In general, several scans are used to achieve a complete transition between two images, but individual pixels can begin or override a transition that has already begun in the middle frame. In order to achieve this, the controller must keep track of what part of the total transition has been achieved for a given pixel. A request is received and changes the optic state of a pixel that is not currently in transition. The pixel counter can be set to zero and the pixel begins transitioning at the next frame. If the pixel at which a new request is received transitions actively, the controller applies an algorithm to determine how to reach the new state from the current intermediate transition state. For a 1-bit general image follow, a potential algorithm should apply a pulse that simply invalidates the polarity with amplitude and duration equal to the portion of the forward pulse already applied.

  In order to minimize the power required to operate the display and to maximize the image stability of the electro-optic medium, the display controller can stop the display scan, all not applied or applied near zero The voltage of the pixels can be reduced. While having little advantage, the display controller can turn off power to the associated row and column drivers while the display is in such a “hold” state. Therefore, power consumption can be minimized. In this outline, the driver may be able to react when the next pixel transition is required.

FIG. 1 is a schematic description of the device of the present invention, the display driven by the device, and related devices, designed to show the overall architecture of the system. FIG. 2 is a schematic block diagram of the controller unit shown in FIG. 1, showing the output signals generated by this unit. FIG. 3 is a schematic block diagram illustrating aspects of the controller unit shown in FIGS. 1 and 2 (which produces the specific output shown in FIG. 2). FIG. 4 shows two different sets of reference voltages that can be used in the display shown in FIG. FIG. 5 shows two different sets of reference voltages that can be used in the display shown in FIG. FIG. 6 is a schematic illustration of the trade-off (approached with the lookup table method of the present invention) between pulse width correction and voltage correction. FIG. 7 is a block diagram of a custom drive useful with the lookup table of the present invention. FIG. 8 is a flowchart showing a program that can be executed by the controller unit shown in FIGS. 1 and 2. FIG. 9 shows two drive schemes of the present invention. FIG. 10 shows two drive schemes of the present invention. FIG. 11A shows two parts of the third drive scheme of the present invention. FIG. 11B shows two parts of the third drive scheme of the present invention.

  Of the accompanying drawings, FIG. 1 schematically illustrates an apparatus in which the present invention is used (with associated apparatus). The entire apparatus shown in FIG. 1 (generally designated 10) includes an image source shown as a personal computer 12. The personal computer 12 outputs data representing an image through a data line 14. Data line 14 may be any conventional type of line, and may be a single data line or a bus. For example, the data line 14 may comprise a universal serial bus (USB), serial, parallel, IEEE-1394 or other line. The data placed on line 14 can be in the form of a conventional bit mapping image. For example, a bitmap (BMP), a tagged image file format (TIF), a graphic compatible format (GIF), or a Joint Photographic Experts Group (JPEG) file. However, the data placed on line 14 instead may be in the form of a signal intended to drive the video device. For example, many computers provide a video output that drives an external monitor and signal of such output that can be used with the present invention. The apparatus of the present invention, described below to those skilled in the imaging process, should be able to perform substantial file format conversion and / or decoding utilizing input signal separation types. This input signal is used and such conversion and / or decoding is well within the level of the prior art, so that the apparatus of the present invention can process the image data used as the original input by the apparatus. It is described only from the point converted to the format.

  The data line 14 extends to the controller unit 16 of the present invention and will be described in detail later. The controller 16 generates one set of output signals on the data bus 18 and a second set of signals on the isolated data bus 20. Database 18 connects to two row (or gate) drivers 22, and data bus 20 connects to a plurality of column (or source) drivers 24. (The number of column drivers 24 is greatly reduced in the simplified diagram of FIG. 1.) The row and column drivers control the performance of the bistable electro-optic display 26.

The device shown in FIG. 1 has been selected to show the various units used, and is best suited for the evolutionary “breadboard” unit. In fact, the commercial product controller 16 is typically part of the same physical unit as a conventional notebook PC and personal digital assistant with LCD, such as the display 26, and the image source is also part of this physical unit. obtain. In addition, the present invention is shown in FIG. 1 and described primarily below, with an active matrix display architecture having a single, common transparent electrode (not shown in FIG. 1) on one side of the electro-optic layer. Related to. This common electrode extends beyond all pixels of the display. Generally, this common electrode is between the electro-optic layer and the viewer, forming a viewing surface where the viewer looks at the display. On the opposite side of the electro-optic layer, a matrix of pixel electrodes that arranges the rows and columns is arranged so that the electrodes of each pixel are uniquely determined by the intersection of a single row and a single column. Therefore, the electric field encountered by each pixel in the electro-optic layer is controlled by varying the voltage applied to the associated pixel electrode with respect to the voltage applied to the common front electrode (usually indicated as “V _com ”). Is done. Each pixel electrode is associated with at least one transistor and is typically associated with a thin film transistor. The gate of each row transistor connects to one of the row drivers 22 via a single elongated row electrode. The source electrode of each column transistor is connected to one of the column drivers 24 via a single elongated column electrode. The drain electrode of each transistor is directly connected to the pixel electrode. The assignment of the row and source electrode gates to the column, as well as the source and drain electrode assignments, can be indeterminate and invalid. However, the following description assumes a conventional assignment.

  In operation, the row driver 22 applies a voltage to the gate so that one and only one row transistor is conductive at a given time. At the same time, the column driver 24 applies a predetermined voltage to each column electrode. Therefore, the voltage applied to the column driver is applied to just one row of the pixel electrodes, thus writing (or at least partially writing) a line to the desired image of the electro-optic medium. The row driver is then changed to make the next row conductive transistor. A different set of voltages is applied to the column electrode and the next line of the image is written.

  The present invention is not limited to matrix displays. Once the exact waveform of each pixel of the image is determined by the present invention, an optional switching scheme can be used to apply the waveform to the pixel. For example, the present invention may use a so-called “direct drive” scheme. In that scheme, each pixel is provided with a separate drive line. Basically, the present invention may also use a passive matrix drive scheme of the type used in certain LCDs, although many bistable electro-optic media lack a switching threshold (ie, a small electric field is present). Even when applied for a long period of time, the media changes to the optical state), but such media are not suitable for driving passive matrices. However, since the present invention is found in major applications for active matrix displays, it is mainly described herein with reference to such displays.

  The controller 16 (FIG. 1) has two main functions. First, using the method of the present invention, the controller calculates a two-dimensional matrix of impulses (or waveforms) that must be applied to the pixels of the display that change the initial image to the final image. Second, the controller 16 calculates all timing signals from the matrix of impulses, which are the electrodes of the pixel using a conventional driver designed to use an LCD driving a bistable electro-optic display, There is a need to provide the desired impulse.

  As shown in FIG. 2, the controller unit 16 shown in FIG. 1 has two main sections, ie, the frame buffer 16A allows the controller 16B to be written to the display 26 (FIG. 1) and the controller correctly 16B. Buffer the data representing the final image of the controller labeled. The controller 16B reads data from the buffer 16A pixel by pixel and generates various signals in the databases 18 and 20 as described below.

  The signals shown in FIG. 2 are as follows.

D0: 6-bit voltage value of D5-pixel (obviously, the number of bits of this signal may vary depending on the particular row and column driver utilized).
Pixel polarity for POL-V _com (see below).
A START-start bit is placed in the column driver 24 to allow loading of pixel values.
HSYNC-horizontal sync signal that latches the column driver.
PCLK—A pixel clock that shifts the start bit along the row driver.
VSYNC-Vertical sync signal that loads the start bit into the row driver.
OE—Output enable signal that latches the row driver.

  For these signals, VSYNC and OE supplied to the row driver 22 are substantially the same as the corresponding signals supplied to the row driver of the conventional active matrix LCD. Because the method of scanning a row of the device shown in FIG. 1 is in principle the same as the method of scanning an LCD, of course, the exact timing of these signals depends on the exact electro-optic medium utilized. It can change depending on the situation. Similarly, the START, HSYNC, and PCLK signals supplied to the column driver are substantially the same as the corresponding signals supplied to the column driver of the conventional active matrix LCD, but their exact timing is utilized. It can vary depending on the exact electro-optic medium being played. Accordingly, further explanation of these output signals may not be necessary.

  FIG. 3 shows how the controller 16B shown in FIG. 2 generates the D0: D5 and POL signals in a very schematic manner. As described above, the controller 16B selectively selects the final image 120 (the image that it is desired to write to the display), the initial image 122 previously written to the display, and the display before the initial image. Data representing one or more previous images 123 written is stored. The embodiment of the invention shown in FIG. 3 stores two such previous images 123. (Obviously, the required data storage may be internal to the controller 16B or in an external data storage device.) The controller 16B is responsible for the initial, final, and previous images 120. , 122, and 123 (shown as the first pixel in the first row, as shown by the shadow in FIG. 3) as a pointer to the lookup table 124. Lookup table 124 provides impulse values that must be applied to a particular pixel to change the state of the pixel to the desired gray level of the final image. The output of the result from the lookup table 124 and the output from the frame counter 126 are the voltage v. Supplied to frame array 128 to generate D0: D5 and POL signals.

The controller 16B is designed for use with a TFT LCD driver provided in a pixel inversion circuit, which typically alternates the polarity of adjacent pixels with respect to the top plane. Alternating pixels are shown as even and odd and are connected to the opposite side of the voltage line. In addition, the driver input labeled “Polarity” functions to switch the polarity of even and odd pixels. The driver is provided with four or more gamma voltage levels that can be set to determine the local scope of the voltage level curve. A typical example of a commercially available integrated circuit (IC) having these features is the Samsung KS0652 300/309 channel TFT-LCD source driver. As described above, the driven display utilizes a common electrode on one side of the electro-optic medium, and the voltage applied to the common electrode is referred to as the “upper plane voltage” or “V _com ”.

In one embodiment, as shown in FIG. 4 of the accompanying drawings, the reference voltage of the driver is such that the top plane voltage is one half of the maximum voltage (V _max ) that the driver can supply, ie,
_{V com} = V _max / 2
The gamma voltage is configured to vary linearly above and below the upper plane voltage (FIGS. 4 and 5 are drawn to assume an odd number of gamma voltages, thereby , for example, in FIG. 4, if the gamma voltage VGMA (n / 2 + 1/ 2) is equal to _{V com.} even gamma voltage is present, VGMA (n / 2) and VGMA (n / 2 + 1) _{is, V com} Similarly, in Fig. 5, if there are even gamma voltages, both VGMA (n / 2) and VGMA (n / 2 + 1) are set equal to the ground voltage Vss. )
The pulse length required to achieve all the required transitions is determined by dividing the maximum impulse that needs to create a new image by V _max / 2. This impulse can be converted into many frames by multiplying by the scanning speed of the display. The required number of frames is then doubled to give an equal number of even and odd frames. These even and odd frames correspond to whether the polarity bit is set high or low for the frame. For each pixel in each frame, controller 16B determines whether (1) the pixel is even or odd, (2) whether the polarity bit is high or low for the considered frame (3 An algorithm must be applied which takes as input the desired impulse magnitude is positive or negative, and (4) the magnitude of the desired impulse. The algorithm determines whether pixels can be addressed with the desired polarity during this frame. If addressed, the appropriate drive voltage (impulse / pulse length) is applied to the pixel. If not addressed, the pixel is directed to the top plane voltage (V _max / 2), putting the pixel in a holding state. In the hold state, no electric field is applied to the pixel during this frame.

  For example, consider two adjacent pixels of the display, odd pixel 1 and even pixel 2. Further, when the polarity bit is high, odd pixels can access the positive drive voltage range (ie, above the upper plane voltage) and even pixels can access the negative voltage (ie, , Below the upper plane voltage). If pixel 1 and pixel 2 need to be driven by a positive impulse, the following sequence needs to occur:

  (A) During a positive polarity frame, pixel 1 is driven by a positive voltage and pixel 2 is held at the top plane voltage.

  (B) During a negative polarity frame, pixel 1 is held at the top plane voltage and pixel 2 is driven by a positive voltage.

  Normally, frames with positive and negative polarities are interleaved 1: 1 (ie, alternating with each other), but this is not necessary. For example, all odd frames are grouped together, followed by all even frames. As a result, alternating columns of the display are driven in two separate groups.

  The main advantage of this embodiment is that the common front electrode does not need to be switched during operation. The main drawback is that the maximum drive voltage available for the electro-optic medium is only half of the driver's maximum voltage, and each line can be driven in only 50% of the time. The refresh time of such a display is four times the switching time of an electro-optic medium under the same maximum drive voltage.

In a second embodiment of this form of the invention, the gamma voltage of the driver is configured as shown in FIG. 5, and the common electrode switch switches between V = 0 and V = V _max . By configuring this method of gamma voltage, both odd and even pixels can be driven simultaneously in a single direction, but the common electrode can be switched between opposite drive polarities. Needed. Furthermore, since this configuration is the sun for the top plane voltage, the same voltage is applied to either odd or even pixels as a result of specific inputs to this driver. In this case, the inputs to the algorithm are the desired impulse magnitude and sign, and the polarity of the top plane. This value is output if the current common electrode setting corresponds to the desired impulse sign. If the desired impulse is in the opposite direction, the pixel is set to the top planar electrode so that no electric field is applied to the pixel during this frame.

  As in the previous embodiment, in this embodiment, the required length of drive pulses can be calculated by dividing the maximum impulse by the maximum drive voltage, which is multiplied by the display refresh rate to frame Converted to Also, consider the fact that the number of frames must be doubled and the display can only be driven in one direction relative to the top plane at a certain time.

  The main advantage of this second embodiment is that full voltage drivers can be utilized and all outputs can be driven at once. However, two frames are required to drive in the opposite direction. Thus, the refresh time of such a display is twice the switching time of the electro-optic medium under the same maximum drive voltage. A major drawback is the need to switch the common electrode, which results in unwanted voltage artifacts in the electro-optic medium, the transistor associated with the pixel electrode, or both.

  In one embodiment, the gamma value is typically configured on a linear ramp between the driver's maximum voltage and the top voltage. Depending on the driver design, it may be necessary to indicate one or more gamma voltages at the top surface value to ensure that the driver can actually generate the top surface voltage on the output.

Standards have already been created above to meet the need to apply the method of the present invention to the limitations of conventional drivers designed for use with LCDs. More specifically, LCD conventional column drivers, particularly super twisted nematic (STN) LCDs (which typically handle higher voltages than other types of column drivers), require polarity insensitive LC materials. As such, only one of the two voltages can be applied to the drive line at any given time. In contrast, a minimum of three driver voltage levels are required to drive a polarity sensitive electro-optic display. The three driver voltages required are V− driving a pixel that is negative with respect to the top voltage, V + driving a pixel that is positive with respect to the top voltage, and 0V with respect to the top voltage that maintains the pixel in the same display state. .

  However, the method of the present invention provides that the controller is configured to apply an appropriate sequence of voltages to the input of one or more column drivers to apply the necessary impulses to the pixels of the electro-optic display. If done, this can be done using a conventional LCD driver of this type.

There are two main variations on this approach. In the first variation, all applied impulses must have one of three values:
+ I = − (− I) = Vapp × t _pulse
Here, Vapp is a voltage applied to the upper part of the upper surface voltage, and t _pulse is a pulse length in seconds. This change allows the display to operate in binary (black / white) mode. In the second variation, the applied impulse needs to be an integer multiple of Vapp / freq, where freq is the display refresh frequency, but can vary from + I to -I.

This aspect of the invention takes advantage of the fact that, as already described, conventional LCD drivers are designed to reverse polarity at a number of intervals to avoid certain undesirable effects that can be generated in the display. To do. As a result, such drivers are configured to receive from a control voltage that can be either polar or either high or low. When the low control voltage is asserted, the output voltage on any given driver output line takes in one of two, other than the three possible voltages required, for example V1 or V2. In other words, if a high control voltage is asserted, the output voltage on any given line may take one of two different of the three possible voltages required, for example V2 or V3. Thus, only one out of two other than the three required voltages can be addressed at any particular time, and all three voltages can be achieved at different times. The three required voltages usually satisfy the following relationship:
V2 = (V3 + V1) / 2
V1 can be at or near logic ground.

In this method of the invention, the display is scanned 2 × t _pulse × freq times. During these half scans (ie, t _pulse × freq scan), the driver is set to output one of V1 or V2, which are usually equal to -V and _Vcom , respectively. Thus, during these scans, the pixels can be driven negative or maintained in the same display state. During the other half scan, the driver is set to output one of V2 or V3, which are usually equal to _Vcom and + V, respectively. In these scans, the pixels can be driven positive or maintained in the same display state. Table 1 shows below how these options can be combined to produce either directional or sustain driving: driving to approach the dark and driving to the bright and light states. Of course, the negative correlation is a function of the particular electro-optic medium used.

  Table 1. Drive sequence to achieve bi-directional drive pulse maintained by STN driver

There are a number of different ways to place the two positions of the drive scheme (ie, two different types of scans or “frames”). For example, two types of frames can alternate. When this is done at a high refresh rate, the electro-optic medium will appear to brighten and darken simultaneously when actually driven in opposite directions in alternating frames. Alternatively, all of one type of frame occurs before any frame of the second type, which is the result of the appearance of a two-step drive. Other configurations are, of course, possible (eg, two or more opposite types following one type of two or more frames). Furthermore, if there is a pixel that needs to be driven in one of the two directions, its extreme frame will drop, reducing the drive time to 50%.

While only the first variation produces a binary image, the second variation may represent multiple grayscale level images. This is achieved by combining the modulation of the pulse width of the different pixels with the driving scheme described above. In this case, the display is again scanned 2 × t _pulse × freq times, but only the drive voltage is arbitrary during these sufficient scans to ensure that the desired impulse of the particular pixel is achieved. Applied to a specific pixel. For example, for each pixel, the entire applied impulse can be recorded, and if the pixel reaches the desired impulse, the pixel can be maintained at the top voltage during every subsequent scan. For pixels that need to be driven for less than the full scan time, this time of driving pixels (i.e., as opposed to the sustaining portion while the applied voltage simply maintains the display state of the pixels, The portion of time during which the impulse is applied to change the display state of the pixel) can be distributed in various ways within the total time. For example, all drive portions can be set to start at the beginning of the total time, or alternatively, all drive portions can be adjusted to be complete at the end of the total time. Similar to the first variant, if at any time in the second variant, a further impulse of a certain polarity needs to be applied to any pixel, the scan applying that positive pulse can be removed. . For example, if the maximum impulse to be applied in both positive and negative directions is less than the maximum possible impulse, this means that the entire pulse is shortened.

  For the sake of illustration, consider a very simple case. Consider applying the gray scale scheme described above to a display with four gray levels. That is, black (level 0), dark gray (level 1), light gray (level 2), and white (level 3). One possible scheme for such a display is summarized in Table 2 below.

For simplicity of illustration, this drive scheme is assumed to use only 6 frames. In practice, however, more numbers are usually used. These frames alternate between even and odd numbers. White-to-white transitions (ie transitions with increasing gray levels) are driven only in odd frames, while transitions over black (ie transitions with decreasing gray levels) are driven only in even frames. . For any frame when the pixel is not driven, it is held at the same voltage as the common front electrode, as shown by “0” in Table 2. For 0-3 (black-white) transitions, impulses are applied to odd frames, frames 1, 3 and 5 rather than white (ie pixel electrodes tend to increase the gray level of the pixel). Maintained at a voltage with respect to the common front electrode). On the other hand, for the 0-2 (black-light gray) transition, the impulse is applied only from the white to the frames 1 and 3, but the impulse is not applied to the frame 5. Of course, this is optional, for example when no impulse can be applied to frames 1 and 5 than white, no impulse is applied to frame 3. For 0-1 (black-dark gray), an impulse is applied only to frame 1 from white, but no impulse is applied to frames 3 and 5. Furthermore, this is optional, for example, an impulse can be applied to frame 3 from white, but no impulse is applied to frames 1 and 5.

  Transitions from black are handled in a manner that is more similar to transitions than the corresponding white, except that transitions from black are only applied to even frames of the drive scheme. One skilled in the art of electro-optic displays will readily understand how transitions not shown in Table 2 are handled by the following description.

  The set of impulses above may be a stand-alone transition between two images (in a general image flow) or they are part of an impulse train designed to achieve an image transition (in a slideshow waveform). There may be.

  While emphasizing the method of the present invention that allows the use of conventional drivers designed for use with LCDs, the present invention provides accurate control of the gray state of custom drivers and electro-optic displays. It makes available drivers that are intended to be enabled, while achieving quick writing of the display. This quick writing is described next with reference to FIGS.

  As already explained, first, many electro-optic media respond to voltage impulses. This voltage impulse may be expressed as t times V (or more generally as an integral of V over t). Here, V is a voltage applied to one pixel, and t is a time during which the voltage is applied. Thus, the gray state may be obtained by modulating the length of the voltage pulse applied to the display, may be obtained by modulating the applied voltage, or obtained by a combination of the two. Also good.

  In the case of pulse width modulation of active matrix displays, the achievable pulse width resolution is simply inversely proportional to the display refresh rate. In other words, for a 100 Hz refresh rate display, the pulse length can be subdivided into 10 ms intervals. Thus, each pixel of the display is only addressed once per scan when the select line for the row pixel is activated. During that remaining time, the voltage for the pixel can be held in a storage capacitor as described in the aforementioned WO 01/079961. As the electro-optic medium reacts faster, the slope of the reflectivity versus time curve becomes steeper. Therefore, to maintain the same grayscale resolution, the display refresh rate must be increased accordingly. Increasing the refresh rate results in more power consumption and eventually becomes impractical as transistors and drivers are required to change pixel and line capacitance in less time.

  On the other hand, in voltage-modulated displays, the impulse resolution is only determined by the number of voltage steps and is independent of the speed of the electro-optic medium. Efficient resolution can be increased by imposing non-linear spacing on the voltage steps and concentrating them where the voltage / reflectance response of the electro-optic medium is most steep.

The attached FIG. 6 is a schematic representation of the trade-off between pulse width modulation (PWM) and voltage modulation (VM) approaches. The horizontal axis represents the pulse length and the vertical axis represents the voltage. The reflectivity of a particle-based electrophoretic display as a function of these two parameters is represented as a contour plot. Bands and spaces represent 1 L ^* differences in the reflected brightness of the display. L ^* is a normal ICE definition L ^* = 116 (R / R ₀ ) ^1/3 −16
Have
Where R is the reflectivity and R ₀ is the reference reflectivity value (the difference in luminance of 1L ^* is significant for the average target of the dual stimulus experiment). This particular particle-based electrophoretic medium used in the experiment summarized in FIG. 6 had a response time of 200 ms at the maximum voltage (16V) as shown in the figure.

The effect of pulse width modulation can be determined simply by traversing a horizontal plot along the top, while the effect of voltage modulation is simply observed by testing the vertical edges. When a display using this particular medium is driven at a 100 Hz refresh rate in pulse width modulation (PWM) mode, it is not possible to obtain a reflectance within ± 1 L ^* in the intermediate gray region where the contour lines are steepest. It becomes clear from this plot. In voltage modulation (VM) mode, achieving a reflectance within ± 1 L ^* is operating at a frame rate as low as 5 Hz (of course, the voltage holding pixel capability is provided by the capacitor and is high enough) Requires 128 equally spaced voltage levels. Furthermore, these two approaches can be combined to achieve the same accuracy with smaller voltage levels. To further reduce the required number of voltage levels, they can be concentrated in the middle steep part of the curve shown in FIG. 6, but the area outside it is sparse. This can be achieved with a low input gamma voltage. In order to further reduce the required number of voltage levels, they can be concentrated with advantageous values. For example, a very small voltage is not effective in achieving a transition where application of such a small voltage at the allocated address time is insufficient to cause some desired gray state transition. By selecting a voltage distribution that eliminates such a small voltage, the tolerance voltage can be placed more advantageously.

  As described above, since the bistable electro-optic display is sensitive to the polarity of the applied electric field, it is desirable to reverse the polarity of the driving voltage to the continuous frame (image), as is usually done with LCDs. And line switching is unnecessary and is actually counterproductive. For example, LCD drivers with pixel conversion carry voltages that alternate polarity in alternating frames. Therefore, it is only possible to transmit impulses of appropriate polarity in half the frame. This is not a problem in LCDs. In LCDs, the liquid crystal material is less sensitive to polarity, but in the case of bistable electro-optic displays, the liquid crystal doubles the time required to address the electro-optic medium.

  Similarly, since the bistable electro-optic display is an impulse transducer and not a voltage transducer, the display accumulates voltage errors for some time. Thus, the display pixels can be greatly displaced from their desired optical state. This makes it important to use a high voltage accuracy and a driver of ± 3 mV or less.

  A maximum pixel clock rate of 60 MHz is required to allow the driver to address a monochrome XGA (1024 × 768) display panel with a 75 Hz refresh rate. Achieving this clock rate is within the skill of the art.

  As mentioned above, one of the main advantages of particle-based electrophoretic displays and other similar bistable electro-optic displays is their image stability, which allows the display to operate with very low power consumption. Is a result opportunity for. To take full advantage of this opportunity, power to the driver should be disabled when the image is not changing. Therefore, this driver should be designed to power down in a controlled manner without creating any false voltage on the output line. Since entering and leaving such a “sleep” mode is a common occurrence, power-up and power-down sequences should be as fast as possible and have minimal effect on the lifetime of the driver. It is.

In addition, there should be an input pin with all driver output pins at _Vcom . This V _com keeps all pixels in their current optical state without powering down the driver.

  The driver of the present invention is particularly effective for driving medium to high resolution, high information content portable displays (eg, 7 inch (178 mm) diagonal XGA monochrome displays). In order to minimize the many integrated circuits required for such high resolution panels, it is desirable to use a driver with a very large number (eg, 324) of outputs for the package. It is also desirable for the driver to have the option of operating in one or more other modes with fewer driver outputs enabled. A preferred method of attaching an integrated circuit to a display panel is a tape carrier package (TCP), which desirably aligns driver output sizing and spacing to facilitate use of this method.

  This driver is used to drive a small to medium active matrix panel of about 30V. Thus, the driver can drive a capacitive load of about 100 pF.

  A block diagram of a preferred driver of the present invention (usually designed at 200) is given in FIG. 7 of the accompanying drawings. The driver 200 includes a shift register 202, a data register 204, a data latch 206, a digital-analog converter (DAC) 208, and an output × factor 210. This driver is typically used to drive LCDs in that this driver provides for multiple bits associated with each pixel of the display and is above and below the top plate voltage controlled by the associated multiple bits. Different from the driver normally used to generate the output.

  The meaning of the signals for this preferred driver is given in Table 3 below.

The driver 200 operates in the following manner. Initially, a start pulse is provided by setting DIO1 high (for example) to reset the shift register 202 to the start position. As will be readily appreciated by those skilled in the art of display driver technology, various DIOx inputs to the shift register are provided so that the driver can be used by displays having various numbers of columns, of these inputs. Only one is used by any given display, and the other is tied permanently low. Here, the shift register operates in the conventional manner used in LCDs. In each pulse of CLK1, only one and one of the 162 outputs of shift register 202 are high and the others are held low, and the high output is shifted to one location in each pulse of CLK1. The As schematically shown in FIG. 7, each of the 162 outputs of the shift register 202 is connected to two inputs of the data register 204. Here, one input is an odd input, and the other input is an even input.

  The display controller (see FIG. 2) receives two 6-bit impulse values D0 (0: 5) and D1 (0: 5) and two 1-bit polarity signals D0POL and D1POL for the input of the data register 204. I will provide a. At the rising edge of each clock pulse CLK1, two 7-bit numbers (D0POL + D0 (0: 5) and D1POL + D1 (0: 5)) are stored in the data register 204 associated with the selected (high) output of the shift register 202. Written to the register. Thus, after 162 clock pulses CLLK1, the 7-bit number of 324 (corresponding to one full line impulse value for one frame) is written into the 324 register in the data register 204.

At the rising edge of each clock pulse CLK2, these 324 7-bit numbers are sent from the data register 204 to the data latch 206. The number thus placed in the data latch 206 is read by the DAC 208 and, in the conventional manner, the corresponding analog value is placed at the output of the DAC 208 and provided via the buffer 210 to the column electrodes of the display. Here, they are applied to pixel electrodes in one column selected in a conventional manner using a column driver (not shown). However, the polarity of each column electrode with respect to V _com is controlled by the polarity bits D0POL or D1POL written to the data latch 206, so these polarities are between adjacent column electrodes in the conventional manner used in LCDs. It will not change.

  FIG. 8 is a flowchart showing a program that can be operated by the control unit shown in FIGS. 1 and 2. This program (generally designed at 300) is intended to use the lookup table method of the present invention (described in detail below). In this look-up table method, all pixels of the display are erased and are therefore readdressed at each time, and the image is written or refreshed.

  This program starts with “power on” in step 302. In this step 302, the controller is typically initialized as a result of user input, for example, the user pressing a power button on a personal digital assistant (PDA). Step 302 may also be triggered, for example, by opening the case of the PDA (opening is detected by either a mechanical sensor or a photodetector) or removing the PDA stylus from the stylus fixture Or may be triggered by a proximity detector that detects when the user's hand approaches the PDA.

  The next step 304 is a “reset” step. In this step, all pixels of the display are driven alternately with pixel states of black and white. It is understood that at least some electro-optic media require such “flushing” of pixels to ensure an accurate gray state during the next writing of the image on the display. It is also understood that in some cases, typically at least 5 flashes (counting a continuous black and white state with one flash) are required. The greater the number of flashes, the more time and energy this step will consume, and the more time that must elapse before the user can view the desired image on the display. Therefore, it is desirable to minimize the number of flashes as much as possible consistent with accurate rendering of the gray state of the subsequently written image. In conclusion, in step 304, all pixels of the display are in the same black or white state.

  The next step 306 is a write or “sending out” step. In this step 306, the controller 16 signals the row driver 22 and the column driver 24 (FIGS. 1 and 2), respectively, in the manner already described, thereby writing the desired image on the display. Since this display is bistable, once an image has been written, there is no need to rewrite the display immediately. Thus, after writing the image, by setting the normal blanking signal (eg, by setting signal BL in FIG. 7 high), the controller causes the column and row drivers to stop writing to the display. Is possible.

  Here, the controller enters the decision loop formed by steps 308, 310 and 312. In step 308, the controller 16 checks whether the computer 12 (FIG. 1) requires a new image display. If so, the controller proceeds and erases the image written to the display in step 306 in erase step 314, thereby essentially returning the display to the state reached at the end of reset step 304. . The controller returns from erase step 314 to step 304 and proceeds to reset as previously described and write a new image.

  If, in step 308, no new image needs to be written to the display, the controller proceeds to step 310. In step 310, it is determined when the image has been retained on the display for a period longer than the predetermined period. As is well known to those skilled in the display arts, an image written on a bistable medium does not last indefinitely, but the image fades gradually (ie loses contrast). Furthermore, some types of electro-optic media, particularly electrophoretic media, often have a trade-off between media writing speed and bistability. That is, a medium that is bistable for hours or days has a substantially longer writing time than a medium that is bistable for only a few seconds or minutes. Thus, it is not necessary to continuously rewrite the electro-optic medium to provide an image with good contrast, as in LCDs, but it is desirable to refresh the image at intervals of a few minutes (for example) possible. Thus, at step 310, the controller determines whether the time elapsed since the image was written at step 306 exceeds a predetermined refresh interval, and if so, the controller proceeds to step 314. Proceed to reset step 304, reset the display as described above, and proceed to rewrite the same image to the display.

(The program shown in FIG. 8 may be modified to utilize both local and global rewriting, as discussed in more detail below. If modified, step 310 may be performed locally. Alternatively, it can be modified to determine whether a global rewrite is required, in this modified program, if step 310 determines that the program has not expired, the action is However, if the predetermined time period has expired, step 310 does not immediately call image erasure and rewriting, but instead step 310 simply globally updates the next image update instead of local. Set a flag (usually in computer terminology) indicating that it should be done The flag is checked the next time the program reaches step 306. If the flag is set, the image is rewritten globally and then the flag is cleared, but the flag is set. If not, the image is simply written locally.)
If it is determined in step 310 that the refresh interval has not been exceeded, the controller proceeds to step 312. In step 312, it is determined whether it is time to shut down the display and / or image source. To conserve energy on the portable device, the controller does not allow a single image to be refreshed indefinitely, but exits the program shown in FIG. 8 after an extended inactivity period. Accordingly, in step 310, the controller determines whether a predetermined “shutdown” period (longer than the refresh period described above) has elapsed since a new image (rather than a refresh of the previous image) has been rewritten to the display. If so, the program ends, as indicated at 314. Step 314 includes turning off the image source. Inevitably, the user still has access to an image that gradually fades on the display after the end of such a program. If the shutdown period has not been exceeded, the controller proceeds from step 312 and returns to step 308.

  Various possible waveforms that implement the lookup table method of the present invention will now be described by way of example only. However, general considerations regarding the waveforms utilized in the present invention are discussed first.

  Bistable display waveforms that exhibit the memory effect described above can be classified into two main classes: corrected and uncorrected. In the corrected waveform, every pulse is adjusted to cause any memory effect on the pixel. For example, a pixel that experiences a series of transitions through grayscale level 1-3-4-2 is slightly different in a 4-2 transition compared to a pixel that experiences transition row 1-2-4-2. Can receive impulses. Such impulse correction can occur by adjusting the pulse length, voltage, or otherwise changing the V (t) profile of the pulse. For uncorrected waveforms, no attempt is made to cause any previous (other than initial) state information. In the uncorrected waveform, all pixels that experience the 4-2 transition receive exactly the same pulse. For an uncorrected waveform to work successfully, one of the two criteria must be consistent. Either the electro-optic material does not show a memory effect in its switching behavior, or each transition must effectively eliminate any memory effect on the pixel.

  In general, uncorrected waveforms are most appropriate for use with systems that can only resolve coarse impulses. Some examples are displays with tri-level or displays that are capable of voltage modulation of only 2-3 bits. Uncorrected waveforms require fine impulse adjustment, which is not possible with these systems. Obviously, preferably a coarse impulse system is limited to uncorrected waveforms, but a system with fine impulse adjustment implements either type of waveform.

  The simplest uncorrected waveform is a 1-bit general image flow (1-bit GIF). With 1-bit GIF, the display transitions smoothly from one pure black and white image to the next. The transition rule of this sequence can be simply described as follows. Impulse I is applied when the pixel switches from white to black. If the pixel switches from black to white, an impulse of opposite polarity -I is applied. If a pixel remains in the same state, no impulse is applied to that pixel. As mentioned above, the mapping of impulse polarity to voltage polarity of the system depends on the response function of the material.

  Another uncorrected waveform that can generate a grayscale image is an n-prepulse slide show (n-PP SS). The uncorrected slide show waveform has three basic sections. First, the pixels are erased to a unified optical state, usually either white or black. The pixel is then driven back and forth between two optical states, usually again white and black. Finally, the pixel is addressed to a new optical state that can be one of several gray scales. The last (or write) pulse is called the addressing pulse, and the other pulses (first (or erase) pulse and inverted (or black) pulse) are generally called pre-pulses. This type of waveform is described below with reference to FIGS.

  The prepulse slide show waveform can be divided into two basic formats: a format with an odd number of prepulses and a format with an even number of prepulses. In the case of an odd pre-pulse, the erase pulse may be equal in impulse and the polarity of the previous write pulse (again, see FIG. 9 and the discussion below) may be reversed. In other words, if a pixel is written from black to gray, the erase pulse has the same polarity as the sum of the previous write pulse and the impulse of the previous write pulse, and the erase pulse transitions completely from black to white Should be equal to the impulse needed to do. In other words, if a pixel is written from black when it is an even prepulse, the pixel must be erased to white.

  After the erase pulse, the waveform contains either zero or an even number of blacking pulses. These blacking pulses are typically equal impulses and opposite polarity pulses, and are configured such that the first pulse is the opposite polarity of the erase pulse. These pulses are usually equal in impulse to a complete black-white pulse, but this is not always the case. It is only necessary that the pulse pairs have equal and opposite impulses, a very diverse set of impulses chained together, ie + I, -I, + 0.1I, -0.1I, + 4I, -4I May exist.

  The last pulse applied is a write pulse. The impulse of this pulse is selected only based on the desired optical state (not based on the current state or any previous state). Usually, though not necessarily, the pulses increase or decrease monotonically with gray state values. Since these waveforms are specially designed for use with coarse impulse systems, the selection of write pulses can map the desired set of gray states to a small amount of possible impulse choices, eg, 9 Mapping four gray states to two possible applied impulses.

  A test of either the even or odd form of the uncorrected n-prepulse slide show waveform reveals that the write pulse always starts in the same direction, ie from black or from white. This is an important feature of this waveform. The principle of the uncorrected waveform is that the pulse length cannot be accurately corrected to ensure that this pixel reaches the same optical state, so the same when approaching from the opposite extreme optical state (black or white) Cannot expect to reach the optical state. Thus, there are two possible polarities of either of these types and can be labeled “from black” and “from white”.

  One major drawback of this type of waveform is that it has a large amplitude optical flash between images. This can be improved by shifting the update sequence by one superframe for half the pixels and interleaving the high resolution pixels as discussed below with reference to FIGS. . Possible patterns include all other columns, all other rows, or checkerboard patterns. Note that this does not mean utilizing the opposite polarity, ie “from black” vs. “from white”. This is because, as a result of this, the gray scales of adjacent pixels do not match. Instead, this is a half pixel (i.e., the first set of pixels corrects the erase pulse and then the second set of pixels when the first set of pixels initiates the first blacking pulse). Can be accomplished by delaying the start of the update by one “superframe” (a group of frames equal to the maximum length of the black-white update). This requires the addition of one superframe for the total update time to allow this synchronization.

First, it can be understood that the ideal method of the present invention is the so-called “general gray scale image flow”. In the general gray scale image flow, the controller configures each writing of the image so that the transition of each pixel transitions directly from the initial gray level to the final gray level. However, in general, the general grayscale image flow suffers from the accumulation of error problems. The impulse applied to any given grayscale transition is necessarily different from what is theoretically necessary. This is because, in fact, such inevitable variations in voltage are output by the driver, producing variations in the thickness of the electro-optic medium or the like. Assume the average error per transition when the difference between the theoretical and actual reflectivity of the display is expressed as ± 0.2 L ^* . After 100 consecutive transitions, the pixel displays an average deviation from each expected state of 2L ^* , and this deviation is apparent to the average observer for certain types of images. It is. To circumvent this problem, by configuring the drive technology utilized in the present invention, a given maximum number before any given pixel passes through some extreme optical state (black or white). It may be desirable to only be able to experience a grayscale transition of These extreme optical states act as “rails” in that the medium cannot become somewhat black or white after a particular impulse is applied to the electro-optic medium. In this way, the next transition away from the extreme optical state can start exactly from the known optical state due to the effect of canceling any previously accumulated errors. Various techniques for minimizing the optical effects of passing pixels through these extreme optical states are discussed below.

  First, referring now to a simple 2-bit grayscale system having black (level 0), dark gray (level 1), light gray (level 2), and white (level 3) optical states, A useful simple drive technology is described. Transitions are implemented using pulse width modulation techniques and transition look-up tables as presented in Table 4 below.

n is a number depending on the particular display, and -n indicates a pulse having the opposite polarity but the same length as pulse n. Further, at the end of the reset pulse 304 of FIG. 8, it is assumed that all pixels of the display are black (level 0). As will be explained below, all transitions occur via an inverted black state, but only transitions to or from this gray state are performed. Thus, the required look-up table size is significantly reduced, and obviously the factor by which the look-up table size is reduced increases with the number of gray levels of the display.

  FIG. 9 shows one pixel transition associated with the drive technology of FIG. At the beginning of the reset step 304, the pixel is in some arbitrary gray state. During the reset step 304, the pixels are driven to each other in three black states and two inverted white states, ending in that black state. The pixel is then written at 306 with the appropriate gray level to the first image and assumed to be level 1. The pixel stays at this level for as long as the same image is displayed. The length of this display period is greatly reduced in FIG. 9 for ease of explanation. At some point, the new image needs to be written, at which point the pixel returns to black (level 0) in the erase step 308, and then in the second reset step, indicated at 304 ', A reset pulse is received alternately in white and black. This returns the pixel to the black state at the end of this reset step 304 '. Finally, in a second writing step, indicated by 306 ', the pixels are written to the second image with the appropriate gray level and assumed to be level 2.

  Many variations of the drive technology shown in FIG. 9 are of course possible. One useful variation is shown in FIG. Steps 304, 306, and 308 shown in FIG. 10 are the same as the steps shown in FIG. However, in step 304 ′, five reset pulses are utilized (obviously, different odd pulses can be utilized), so that at the end of 304 ′, the pixel is in the white state (level 3) and the second In the writing step 306 ′, pixel writing is performed from this white state instead of the black state of FIG. Successive images are then written alternately from the black and white state of the pixel.

  In a further variation of the drive technology shown in FIGS. 9 and 10, the erase step 308 is performed to drive pixel white (level 3) rather than black. An odd number of reset pulses are then applied so that the pixel finishes the reset step with the white state and the second image is written from this white state. Similar to the drive technique shown in FIG. 10, this technique writes successive images alternately from the black and white states of the pixels.

  It will be appreciated that in all the above techniques, the number and duration of reset pulses can vary depending on the characteristics of the electro-optic medium utilized. Similarly, voltage modulation, rather than pulse width modulation, can be used to change the impulse applied to the pixel.

  During the drive technology reset step described above, the black and white flash that appears on the display is, of course, visible to the user and can be unpleasant for many users. To reduce the visual effect of such a reset step, it is convenient to divide the display pixels into two (or more) groups and apply different types of reset pulses to the different groups. More particularly, if any given pixel needs to utilize a reset pulse that drives black and white to alternate, the pixel is divided into at least two groups, It is convenient to configure the drive technology so that another group is driven to black at the same time as it is driven to black. If the spatial distribution of the two groups is carefully selected and the pixel is small enough, the user will experience a reset step as the gray spacing on the display (possibly with some flicker) The spacing is usually less annoying than a series of black and white flashes.

  For example, as one form of such a “reset two groups” step, odd columns of pixels can be assigned to one “odd” group, and even columns of pixels are in a second “even” group. Can be assigned. While odd pixels can utilize drive technology as shown in FIG. 9, even pixels can utilize a variation of this drive technology, during which the pixel is white instead of black. Driven to state. Both groups of pixels then receive an even number of reset pulses during the reset step 304 ′, so that the two groups of reset pulses are substantially 180 ° out of phase and the display is grayed out throughout this reset step. is there. Finally, during the writing of the second image in step 306 ', odd pixels are driven from black to their final state, while even pixels are driven from white to their final state. . To ensure that each pixel is reset for a long time in a similar manner (and thus the reset method does not introduce any artifacts on the display), the controller switches drive technology between successive images. , Whereby a series of new images are advantageously written on the display and each pixel is written alternately from the black and white state to the final state.

  Obviously, a similar technique can be utilized in that the odd rows of pixels form a first group and the even rows of pixels form a second group. In a further similar drive technology, the first group includes pixels in odd columns and odd rows and even columns and even rows, and the second group includes pixels in odd columns and even rows and even columns and odd rows. , Whereby the two groups are arranged in a checkerboard fashion.

  Instead of or in addition to dividing the pixels into two groups and configuring one group of reset pulses to be 180 ° out of phase with the other groups, the pixel is the number and frequency of the pulses. Can be divided into groups that utilize different reset steps. For example, one group may utilize the 6 pulse reset sequence shown in FIG. 9 and the second group may utilize a similar sequence having 12 pulses of double frequency. In a more detailed technique, the pixels can be divided into four groups, the first and second groups using a technique that is 180 ° out of phase but six pulses, The group of 4 utilizes a technique in which the phases are 180 ° out of phase with each other but there are 12 pulses.

  Another technique for reducing the unpleasant effect of the reset step will now be described with reference to FIGS. 11A and 11B. In this technique, the pixels are also divided into two groups, with the first (even) group following the drive technique shown in FIG. 11A and the second (even) group following the drive technique shown in FIG. 11B. Also in this technique, all gray levels in between black and white are a first group of dark gray levels adjacent to the black level and a second group of light gray levels adjacent to the white level. This division is the same for both groups of pixels. Preferably, but not necessarily, there are the same number of gray levels in these two groups. If there are an odd number of gray levels, the middle level can be arbitrarily assigned to either level. For ease of explanation, FIGS. 11A and 11B show this drive technology applied to an 8-level grayscale display, the levels being shown from 0 (black) to 7 (white), gray level 1 2, and 3 are dark gray levels, and gray levels 4, 5, and 6 are light gray levels.

  In the drive technology of FIGS. 11A and 11B, the transition from gray to gray is handled according to the following rules.

  (A) In the first group of even pixels, at the transition to the dark gray level, the last pulse applied is always a white-going pulse (i.e., the pixel from black state to white state). In the transition to the light gray level, the last pulse applied is always a black-going pulse.

  (B) In the second group of odd pixels, the last pulse applied at the transition to dark gray level is always a pulse from black, and at the transition to light gray level, the last pulse applied The pulse is always a pulse rather than white.

  (C) In all cases, the pulse from black can only follow the pulse from white after the white state has been achieved, and the pulse from white can only follow the pulse from black after the black state has been achieved. possible.

  (D) Even pixels cannot be driven from dark gray level to black level by pulses than single black, and odd pixels are driven from light gray to white using pulses rather than single white. I don't get it.

(Obviously, in both cases, the white state can ultimately only be achieved with pulses than white, and the black state can only be achieved with pulses than black eventually.)
Application of these rules allows each gray to gray transition to be performed using a maximum of three consecutive pulses. For example, FIG. 11A shows an even pixel experiencing a transition from black (level 0) to gray level 1. This is accomplished with a pulse from a single white (of course, shown as having a positive slope in FIG. 11A), as shown at 1102. The pixel is then driven to gray level 3. Since gray level 3 is a dark gray level, according to rule (a), a pulse must be achieved over white, so a level 1 / level 3 transition can be handled by pulse 1104 over a single white. This white pulse 1104 has an impulse different from that of the pulse 1102.

  Here, the pixel is driven to gray level 6. Since this is a light gray level, the pulse must be achieved over black according to rule (a). Therefore, the application of rules (a) and (c) requires that this level 3 / level 6 transition be performed by two pulse rows, ie pulse 1106 from the first white. This first white pulse 1106 drives the pixel white (level 7), followed by the second black pulse 1108, and this second black pulse 1108 from level 7 to the desired level. Drive the pixel to 6.

  The pixel is then driven to gray level 4. Since this is a light gray level, the level 6 / level 4 transition is implemented by a pulse 1110 rather than a single black, with exactly the same discussion used for the level 1 / level 3 transition previously discussed. The The next transition is to level 3. Since this is a dark gray level, the level 4 / level 3 transition is more than the two pulse rows, namely the first black, by the same discussion as used for the level 3 / level 6 transition previously discussed. Handled by pulse 1112. This first black pulse 1112 drives pixel black (level 0), followed by a second white pulse 1114, and this second white pulse 1114 from level 0 to the desired level 3 Drive pixels to.

  The final transition shown in FIG. 11A is from level 3 to level 1. Since level 1 is a dark gray level, a pulse must be achieved over white according to rule (a). Thus, applying rules (a) and (c), the level 3 / level 1 transition is three pulses including pulse 1116 from the first white, pulse 1118 from the second black, and pulse 1120 from the third white. Must be handled by line. From this first white pulse 1116 drives pixel white (level 7), from this second black pulse 1118 drives pixel black (level 1), and from this third white pulse 1120 Drive the pixel from black to the desired level 1 state.

  FIG. 11B shows an odd number of pixels achieving a gray state 0-1-3-3-6-4-3-1 sequence equivalent to the even number of pixels of FIG. 11A. However, the pulse sequence used is expected to be very different. Rule (b) requires level 1, dark gray levels to be approximated by pulses rather than black. Thus, the 0-1 transition is achieved by pulse 1122 from the first white driving the pixel white (level 7) and later by pulse 1124 from black driving the pixel from level 7 to the desired level 1. Is done. 1-3 transitions require three pulse sequences. It is pulse 1126 from the first black that drives pulse 1140 from black (level 0) and white of the pixel, pulse 1128 from the second white that drives pixel white (level 7), and the desired level from level 7 3 is a pulse 1130 from the third black driving pixel. Next is a transition to level 6, which is a light gray level. It is approximated by pulses rather than white by rule (b), the level 3 / level 6 transition is pulse 1132 than black driving the pixel black (level 0), and white driving the pixel to the desired level 6 This is achieved by two pulse sequences including pulse 1134. The level 6 / level 4 transition has three pulse sequences: pulse 1136 from white driving the pixel white (level 7), pulse 1138 from black driving the pixel black (level 0), and the desired level. 4 achieved by a pulse sequence that drives the pixel. The level 4 / level 3 transition is achieved by two pulse sequences including pulse 1142 from white driving the pixel white (level 7), and later and pulse 1144 from black driving the pixel to the desired level 3. . Finally, the level 3 / level 1 transition is achieved by pulse 1146 from a single black.

  According to FIGS. 11A and 11B, this drive scheme allows each pixel to move from black to white without changing the direction of the pixel (obviously, the pixel is at any intermediate gray level for short or long periods). It can be seen that the pixels are guaranteed to follow a “serrated” pattern. Then, it moves from white to black without changing the direction. Thus, rules (c) and (d) above can be replaced by a single rule (e) as follows:

  (E) Once a pixel is driven from one extreme optical state (ie, white or black) to the opposite extreme optical state by a pulse of a certain polarity, the pixel is Do not receive pulses of opposite polarity until the condition is reached.

  This drive scheme thus ensures that a pixel will undergo at most many transitions equal to (N-1) / 2 transitions. Here N is the number of gray levels before being driven to some extreme optical state. If severe distortion of the grayscale image is apparent to the viewer, this prevents minor errors in individual transitions that remain permanently at that point. Moreover, this drive scheme is designed so that even and odd pixels always approximate a given intermediate gray level from the opposite direction. That is, the final pulse of the sequence will be white in some cases and black in other cases. This “opposite direction” property minimizes flashing of the region when a substantial area of the display containing substantially equal numbers of even and odd pixels is written to a single gray level.

  To implement the sawtooth drive scheme of FIGS. 11A and 11B to implement the sawtooth drive scheme of FIGS. 11A and 11B in order to be similar to the pixels described above that relate to other drive schemes that divide the pixel into two separate groups, the placement of pixels in even and odd groups Careful consideration should be given to. This arrangement allows any substantial configuration of the display to include a substantially equal number of odd and even pixels and that the maximum size of a similar group of consecutive pixels can be easily discerned by the average observer. It is preferably ensured that it is small enough so that it cannot. As already discussed, placing two groups of pixels in a checkerboard pattern meets these requirements. Probabilistic screening techniques can also be used to place two groups of pixels.

  However, in a sawtooth drive scheme, the use of a checkerboard pattern tends to increase the energy consumption of the display. In any given row of such a pattern, adjacent pixels belong to opposing groups, and in a contiguous area of substantial size when all pixels undergo similar gray level transitions, adjacent pixels are Tend to require impulses of opposite polarity at any given time. Applying an impulse of opposite polarity to successive pixels in any column requires that the column (source) electrodes of the display be discharged and recharged, as each new line is written. It is known to those skilled in the art that discharging and recharging the column electrodes is a major factor in the energy consumption of the display. Thus, the check board arrangement tends to increase the energy consumption of the display.

  A reasonable compromise between energy consumption and the desire to avoid large contiguous areas of pixels of the same group is that each group of pixels assigned to a rectangle (although several pixels along its column are spread over , All pixels in the same column). With such an arrangement, when rewriting an area with a similar gray level, only discharging and recharging the electrodes of the column is necessary when shifting from one rectangle to the next. As desired, the rectangles are 1 × 4 pixels and are arranged so that the rectangles in adjacent columns do not end in the same row, ie, the rectangles in adjacent columns have different “topologies”. Rectangular assignments in a row to phase can be brought about by either a random or periodic manner.

  One advantage of the sawtooth drive scheme shown in FIGS. 11A and 11B is that as part of the overall display update, any region of the image that is monochromatic is a single pulse (either black to white or white to black). Is simply updated. The maximum time it takes to rewrite such a monochromatic area is only half of the maximum time it takes to rewrite the area required for the transition from gray to gray. It can be used to benefit from quick updates of image features such as input, drop-down menus, etc. The controller may check whether the image update requires any gray-to-gray transition; otherwise, areas of the image that need to be rewritten can be re-created using a quick monochromatic update mode. Can be written. Thus, the user may have fast updates of display input characters and other user interaction features that are seamlessly overwritten by slower updates of typical grayscale images.

  As discussed in the aforementioned co-pending application Nos. 09 / 561,424 and 09 / 520,743 of electro-optic media, particularly particle-based electrophoretic media, over time, certain pixels It is hoped that the drive scheme used to drive such mediation will be in direct current (DC) equilibrium, in the sense that the algebraic sum of the currents passing will be zero or as close to zero as possible. In rare cases, the drive scheme of the present invention is designed with this criterion in mind. More specifically, the look-up table used in the present invention is designed so that any sequence of transition start and end in one extreme optical state (black or white) of a pixel will be in DC equilibrium. From what has been described above, it is first apparent that such a DC equilibrium state cannot be achieved until the impulse, so the current through the pixel required for any particular gray-to-grey transition is substantially Constant. However, this is only true to the first approximation and has been found empirically, but at least in the case of particle-based electrophoretic media, 50 ms pulses spaced 5 times apart are applied to the pixel ( The (approximate) effect is not equal to the application of a single 250 millisecond pulse of the same voltage. Thus, there is some degree of freedom in the current through the pixel to achieve a given transition, and this degree of freedom can be used to help achieve DC equilibrium. For example, the lookup table utilized in the present invention may store multiple impulses for a given transition along with all the values of current provided by each of these impulses, For each pixel, a register may be maintained that is arranged to store an algebraic sum of impulses applied to the pixel from a certain time (eg, the pixel ended in black). When a particular pixel is driven from a white or gray state to a black state, the controller examines the register associated with the pixel and DC balances the entire sequence of transitions from the previous black state to the next black state One of the multiple stored impulses of the white / gray to black transition that needs to determine the current required for the state and precisely reduce the associated resistor to zero or at least as small as possible (If the associated resistor holds this remaining value and applies it to the current applied during the slower transition). It is clear that the repeated application of this process can achieve a long-term accurate DC balance for each pixel.

  The serrated drive scheme shown in FIGS. 11A and 11B can pass between successive passes of any given pixel where only a limited number of transitions go through the black state, and in fact, on average, a pixel is its Note that this drive scheme can be successfully applied to the use of such DC balancing techniques as it ensures that it passes through the black state of half of the transition.

  The undesired effects of the reset step can be further caused by local updates rather than global updates, i.e. rewritten by rewriting only those parts of the display that change between successive images. The part to be selected is selected on a “local area” or pixel basis for each pixel. For example, for diagrams showing movement of parts in mechanical devices and examples in diagrams used to reconstruct accidents, it is unusual to find a series of images where a relatively small object moves to a larger background Not. In order to use local updates, the controller needs to compare the final image with the first image to determine that the region is different between the two images and therefore needs to be rewritten. is there. The controller can identify one or more local regions that are typically rectangular regions with sides arranged in a pixel grid containing the pixels that need to be updated, or need to be updated. A pixel here can simply be identified. Next, any drive scheme already described is applied to update only local pixels or individual pixels that are identified as requiring rewriting. Such a local update scheme can substantially reduce the energy consumption of the display.

  The drive scheme described above can be varied in a number of ways depending on the characteristics of the particular electro-optic display used. For example, in some cases it may be possible to eliminate many of the reset steps in the drive scheme described above. For example, if the electro-optic medium used is bistable for a long period and the impulse required for a particular transition does not change much in the period of the period when the pixel was in the initial gray state, the look-up table Can be configured to cause a direct transition from gray state to gray state without arbitrarily switching to the black or white state, and after a substantial period of time, the stepwise transition from nominal gray level to pixel Reset the display that comes only when “drift” causes a noticeable error in the presented image. Thus, for example, if a user uses the display of the present invention as an electronic book reader, it may be possible to display multiple screens of information before a display reset was required; Along with the waveform and the driver, about 1000 screens of information could be displayed before a reset was needed, and therefore, during an actual reset, it is not necessary during a typical reading session of an ebook reader .

  It will be readily apparent in the art that a single device of the present invention can be effectively provided using different drive schemes used in different circumstances. For example, in the drive scheme shown in FIGS. 9 and 10, the reset pulse consumes a substantial portion of the total display energy consumption, so the controller resets the display at a number of intervals, thus the first drive scheme (and thus gray). As well as a second scheme that resets the display only at longer intervals (thus reducing energy consumption but accepting larger grayscale errors). Switching between the two schemes is effected either by external parameters or manually; for example, if the display is used in a laptop computer, the first drive scheme is when the computer is primarily powered up Can be used. The second drive scheme can be utilized while the computer starts up with an intermediate battery power source.

  From the above description, the present invention is expected to provide a driver for controlling the operation of an electro-optic display. Electro-optic displays are well applied to the properties of bistable particle-based electrophoretic displays and similar displays.

  From the above description, the present invention is for controlling the operation of an electro-optic display that allows for precise control of gray scale without the need to undesirably flush the entire display to one of its extremes at several intervals. It is anticipated to provide a method and controller. The present invention further allows for precise control of the display and operation during that time, regardless of the motif, while reducing the power consumption of the display. These advantages can be provided inexpensively because the controller can be constructed from commercially available components.

In the residual voltage method of the present invention, the measurement of the residual voltage is preferably provided by a high impedance voltage measuring device (eg, a metal oxide semiconductor (MOS) comparator). If the display is one with small pixels, for example a 100 dot per inch (DPI) matrix display, where each pixel is an area of 10 ⁻⁴ square inches or 6 × 10 ⁻² mm ² , the comparator is A single pixel needs to have an ultra-low input current, such as a resistance on the order of 10 ¹² Ω. However, a suitable comparator is readily available commercially; for example, if only about 20 pA is used for the input current, the Texas Instruments INA111 chip is appropriate. (In general, this integrated circuit is an instrumentation amplifier, but if its output is passed through a Schmitt trigger, it serves as a comparator). For displays with large signal pixels (eg large direct drive displays used for signing (defined below)), the need for comparators is urgent when individual pixels can have an area of several square centimeters Almost any commercial FET input comparator can be used, such as, for example, the LF311 comparator by National Semiconductor Corporation.

  For cost and other reasons, mass-produced electronic displays usually have drivers in the form of application specific integrated circuits (ASICs), and in this type of display, a comparator is typically provided as part of the ASIC. It will be readily apparent to those skilled in the art of electronic displays. This approach requires the provision of feedback circuitry within the ASIC, but has the advantage of producing a simpler and smaller power supply in the area and the oscillator section of the ASIC. This approach also produces a simpler and smaller ASIC driver section in the area when a three-stage general image flow drive is required. Thus, this approach typically reduces the cost of the ASIC.

  Conveniently, the drive pulses are applied using a driver that can apply a drive voltage that electrically shorts or flows the pixel. In each addressing cycle that results in a DC balance correction, when using such a driver, the pixels are addressed, electrically shorted, and then flowed. (The term “addressing cycle” is used herein for the sake of convenience to those skilled in the electro-optic display art, and is the total cycle required to change from the first to the second image on the display. As indicated above, due to the relatively slow switching speed of electrophoretic displays (generally 10 to 100 milliseconds), a single addressing cycle can include multiple scans of the entire display. .) After a short delay time, the residual voltage across the pixel is measured using a comparator to determine if its sign is positive or negative. If the residual voltage is positive, the controller may slightly extend the duration of the addressing pulse going negative in the next addressing pulse (or increase the voltage of the addressing pulse slightly). However, if the residual voltage is negative, the controller may slightly extend the duration of the positive voltage pulse in the next addressing cycle (or may increase the voltage of the addressing pulse slightly).

  Accordingly, the residual voltage method of the present invention places the electro-optic medium in a bang-bang feedback loop and adjusts the length of the addressing pulse so that the residual voltage goes to zero. As the residual voltage approaches zero, the media exhibits ideal properties and improved lifetime. In particular, the use of the present invention may allow improved control of gray scale. As described earlier, it was observed that the grayscale level obtained in the electro-optic display is not only a function of the starting grayscale level and applied impulse, but also a function of the previous state of the display. One of the reasons for this “history” effect on grayscale levels is that the residual voltage affects the electric field experienced by the electro-optic medium; the actual electric field that affects the behavior of the medium causes the electrode and residual voltage to Is the sum of the voltages actually applied via Thus, controlling the residual voltage according to the present invention ensures that the electric field experienced by the electro-optic medium exactly matches the electric field applied through the electrodes, allowing for improved control of gray scale.

  The residual voltage method of the present invention is particularly useful for so-called “direct drive” type displays, which are divided into a series of pixels provided to the individual electrodes, which further display the voltage applied to each individual electrode. Including switching means configured to be individually controlled. Such direct drive displays are useful for text displays or other limited sets of characters (eg, numerical numbers), and are described in particular in the above-mentioned International Application Publication No. 00/05704. However, the residual voltage method of the present invention can also be utilized in other types of displays (eg, an active matrix display using an array of at least one transistor integrated with each pixel of the display). Activating the gate line of a thin film transistor (TFT) used in such an active matrix display connects the pixel electrode to the source electrode. Since the residual voltage is small compared to the gate voltage (the absolute value of the residual voltage usually does not exceed about 0.5V), the gate drive voltage still turns on the TFT. Therefore, since the source line is electrically floated and can be connected to the MOS comparator, the residual voltage of each pixel of the active matrix display can be read out.

  Although the residual voltage on a pixel of an electrophoretic display is closely related to the range in which the current flow through that pixel has become DC-equilibrium, a residual voltage of zero necessarily results in a complete DC-equilibrium state. Note that it means. But from a practical point of view this is a little different. This is because it is clear that it is the residual voltage itself, not the DC-equilibrium history, that is responsible for the negative effects shown herein.

Since the purpose of the residual voltage method of the present invention is to reduce residual voltage and DC impedance, it can be applied with sufficient frequency to prevent long-term DC impedance build-up in a particular pixel. If provided, it will be readily appreciated by those skilled in the art of display that this method need not be applied to every addressing cycle of the display. For example, during a refresh or blanking pulse, all pixels are driven to the same display state, usually one of the extreme display states (more generally, all pixels are initially in one extreme display state). If the display is one that requires the use of a “refresh” or “blanking” pulse in the interval (as driven to the next extreme display state), then the method of the present invention Can be performed only during a refresh or blanking pulse.

  Although the residual voltage method of the present invention has been described primarily for application in sealed electrophoretic displays, the method is not sealed electrophoretic displays, and other types of displays that display residual voltages (eg, electrochromic displays). ) Can also be used.

  From the above description, the residual voltage method of the present invention provides increased display lifetime while reducing the cost of equipment required to ensure DC balance of the display pixels, window and long-term It can be seen that there is provided a method for driving electrophoretic displays and other electro-optic displays that provides manipulation of the optical performance of the display.

Claims (1)

The invention described herein.