- applying an addressing pulse to drive the pixel to a first optical state;
- leaving the pixel undriven for a period of time, thereby permitting the pixel to assume a second optical state different from the first optical state; and
- applying to the pixel a refresh pulse which substantially restores the pixel to the first optical state, the refresh pulse being short relative to the addressing pulse.

In this refresh pulse method, the refresh pulse will typically have an impulse not greater than about 20 per cent of the impulse of the addressing pulse, desirably not greater than about 10 per cent of this impulse, and preferably not greater than about 5 per cent of this impulse. Typically the difference between the first and second optical states does not exceed about 1 unit of L* (where L* has the usual CIE definition:
L*=116(R/R₀)^1/3−16,
where R is the reflectance and R₀is a standard reflectance value); desirably this difference does not exceed about 0.5 unit of L*, and preferably does not exceed about 0.2 unit of L*. A plurality of refresh pulses may be applied to the pixel at regular intervals.

This refresh pulse method does allow refreshing of images on bistable electro-optic displays, thus improving the appearance of the images, but necessarily introduces some DC imbalance. The aforementioned 2005/0270261 also describes various methods for reducing the DC imbalance produced.

The present invention provides a modification of the refresh pulse method described in the aforementioned 2005/0270261 which can give a similar improvement in the appearance of a bistable electro-optic display while, at least in some cases, reducing the DC imbalance introduced.

The present invention also seeks to reduce or eliminate another problem encountered in bistable electro-optic displays which is not discussed in the aforementioned 2005/0270261, namely the so-called “edge ghosting” problem. Electro-optic displays typically have a backplane provided with a plurality of pixel electrodes each of which defines one pixel of the display; conventionally, a single common electrode extending over a large number of pixels, and normally the whole display is provided on the opposed side of the electro-optic medium. The individual pixel electrodes may be driven directly (i.e., a separate conductor may be provided to each pixel electrode) or the pixel electrodes may be driven in an active matrix manner which will be familiar to the those skilled in backplane technology. Since adjacent pixel electrodes will often be at different voltages, they must be separated by inter-pixel gaps of finite width in order to avoid electrical shorting between electrodes. Although at first glance it might appear that the electro-optic medium overlying these gaps would not switch when drive voltages are applied to the pixel electrodes (and indeed, this is often the case with some non-bistable electro-optic media, such as liquid crystals, where a black mask is typically provided to hide these non-switching gaps), in the case of many bistable electro-optic media the medium overlying the gap does switch because of a phenomenon known as “blooming”.

Blooming refers to the tendency for application of a drive voltage to a pixel electrode to cause a change in the optical state of the electro-optic medium over an area larger than the physical size of the pixel electrode. Although excessive blooming should be avoided (for example, in a high resolution active matrix display one does not wish application of a drive voltage to a single pixel to cause switching over an area covering several adjacent pixels, since this would reduce the effective resolution of the display) a controlled amount of blooming is often useful. For example, consider a black-on-white electro-optic display which displays numbers using a conventional seven-segment array of seven directly driven pixel electrodes for each digit. When, for example, a zero is displayed, six segments are turned black. In the absence of blooming, the six inter-pixel gaps will be visible. However, by providing a controlled amount of blooming, for example as described in the aforementioned 2005/0062714, the inter-pixel gaps can be made to turn black, resulting in a more visually pleasing digit. However, blooming can lead to a problem denoted “edge ghosting”.

Consider the common situation of a layer of electro-optic medium disposed between a continuous front electrode and an array of pixel electrodes arranged on a backplane. When a pixel electrode is held at a voltage different from that of the front electrode, an electric field exists in the electro-optic medium. This electric field is primarily directed perpendicular to the layer of electro-optic medium and is of approximately uniform intensity (ignoring for present purposes the distorting effects of space charges and polarization of the electro-optic medium itself). However, near the edge of the pixel, fringe fields form, and these fringe fields extend into the interpixel-regions between the pixel subjected to the electric field and the adjacent pixels. In the fringe field region, the electric field lines are curved, going from the pixel to the top plane or the neighboring electrodes. The electric field in this region is generally weaker than in the central part of the pixel.

Waveforms are typically designed to achieve correct transitions to desired optical states in central parts of pixels, i.e., far from pixel edges, and thus away from the fringe fields. For example, consider an array of pixels in one extreme optical state (say black), and a situation where one pixel is driven to the opposed extreme optical state (white) by applying to that pixel a drive voltage for a drive period, then the pixel is driven back to black by applying an opposite voltage for the same drive period. The voltages and drive periods are chosen to achieve the appropriate black state in the central part of the pixel after this pair of transitions. However, in the fringe field region, the electric field during the transitions is weaker than in the central part of the driven pixel. Therefore, in this fringe field region, the first voltage pulse does not drive the electro-optic medium from black to white, but instead from black to an intermediate gray. Also in the fringe field region, the second voltage pulse drives the electro-optic medium from this intermediate gray towards black. However, as discussed in several of the related applications mentioned above, the response of many electro-optic media to electric fields is not necessarily linear with the magnitude of the applied field, nor is it necessarily symmetric with respect to the direction of the electric field. Consequently, after the two transitions previously discussed, the inter-pixel region does not necessarily return to exactly its original black state. Generally, the optical state of the fringe field region will be slightly less black than the original state, i.e., it will be a very dark gray rather than a true black, thus leaving the fringe field region lighter in color than the black central part of the pixel. This is referred to as an edge ghost, which is one class of edge artifact. Such edge ghosts, if sufficiently large, are readily detected by the human eye and highly objectionable.

(The foregoing discussion is simplified for purposes of illustration. An area of blooming is not a uniform white or black but is typically a transition zone where, as one moves across the area of blooming, the color of the medium transitions from white through various shades of gray to black. Accordingly, an edge ghost will typically be an area of varying shades of gray rather than a uniform gray area, but can still be visible and objectionable, especially since the human eye is well equipped to detect areas of gray in monochrome images where each pixel is supposed to be pure black or pure white.)

In some cases, asymmetric blooming may contribute to edge ghosting. “Asymmetric blooming” refers to a phenomenon whereby in some electro-optic media (for example, the copper chromite/titania encapsulated electrophoretic media described in the aforementioned U.S. Pat. No. 7,002,728) the blooming is “asymmetric” in the sense that more blooming occurs during a transition from one extreme optical state of a pixel to the other extreme optical state than during a transition in the reverse direction; in the media described in this patent, typically the blooming during a black-to-white transition is greater than that during a white-to-black one.

The present invention provides methods for driving bistable electro-optic displays which can reduce or eliminate edge ghosting.

Another problem associated with blooming is that one pixel may in effect control an area of the display which is intended to be controlled by an adjacent pixel. Ideally, one might desire that blooming be controlled so that the fringe field from one pixel causes a change in optical state extending half-way across gap between it and an adjacent pixel. In practice, however, blooming varies with numerous factors, including particularly the temperature of the electro-optic medium, so that it may not always be possible to control the blooming to the optimum value. Furthermore, as already noted, because of fringe fields varying with distance, an area of blooming does not have a sharp edge between the two extreme optical states, but rather an area of finite width where a transition between the two optical states occurs. Accordingly, if one wishes to ensure that, for example, the white area of a white pixel extends half-way across the inter-pixel gap (so that when the two adjacent pixels are both driven white, the inter-pixel gap is the same white color as the pixels themselves), it may be necessary to tolerate a gray area extending more than half-way across the inter-pixel gap when one pixel is driven white while the other is still black. However, the presence of such a gray area may be problematic in some circumstances. For example, if the display is a color display provided with a color filter array, and the two pixels involved underlie different colored areas of the color filter array, the presence of the gray area when the first pixel is white and the second black may result in “contamination” of the color of the first pixel with some amount of the color of the second, thus adversely affecting the color rendering of the display

SUMMARY OF THE INVENTION

The present invention may be regarded as comprising various modifications of the refresh pulse driving method of the aforementioned 2005/0270261. In the present method, “refresh” pulses are not generated at arbitrary intervals but are keyed to transitions occurring at other pixels. Furthermore, in the present method, the refresh pulses may not be applied to all pixels or all pixels having a given optical state, but may be confined to certain pixels adjacent a pixel undergoing a transition. For convenience, the corrective pulses applied in accordance with the method of the present invention will be referred to as “reinforcing pulses” while the term “refresh pulses” will be used to denote pulses applied in the method of the aforementioned 2005/0270261.

Accordingly, this invention provides a first method of driving a bistable electro-optic display having at least first and second pixels each of which can display first and second extreme optical states (the pixels may or may not have other intermediate optical states), the first and second pixels having adjacent edges separated by an inter-pixel gap. The first method comprises applying to the first pixel a drive pulse effective to cause the first pixel to change its optical state to one of its extreme optical states, and applying to the second pixel, which is in the said one extreme optical state, a reinforcing pulse of the same polarity as the drive pulse, the reinforcing pulse being applied either simultaneously with the drive pulse or within a predetermined period after the end of the drive pulse.

This first method of the present invention (which may hereinafter be referred to as the “reinforcing pulse method”) may be applied to a monochrome display in which each pixel is intended to display only first and second (typically white and black) optical states. In such monochrome displays, when the first pixel is undergoing a transition from its first to its second optical state (say from white to black), while the second pixel is in its second optical state (black) and is to remain in this second optical state, the first pixel is given a pulse of one polarity to effect the desired transition. During this pulse, or shortly thereafter, the second pixel is given a reinforcing pulse of the same polarity as the drive pulse given to the first pixel (i.e., a black-going pulse). Since the second pixel is already black, the reinforcing pulse does not effect a gross change in the black color of the second pixel. However, if the second pixel has been in its black state for some time, so that its color has “drifted” from a true black to a dark gray, the reinforcing pulse serves to drive the second pixel back to a true black, thus avoiding having a dark gray second pixel immediately adjacent a true black first pixel, a situation which is readily apparent to the human eye. The reinforcing pulse also serves to reduce or eliminate edge ghosting in the inter-pixel gap between the first and second pixels.

As discussed in the various applications and patents mentioned in the “Reference to Related Applications” section above, waveforms used to drive pixels of electro-optic displays from one optical state to another may be quite complex and may include drive pulses of both polarities. Thus, the “drive pulse” applied to the first pixel in the reinforcing pulse method of the present invention may in fact be a complex waveform including multiple individual drive pulses some of which may have opposing polarities. For present purposes, the polarity of such a complex waveform may be defined as the polarity of a single drive pulse of constant magnitude which effects the same optical transition of the first pixel as the complex waveform.

The “reinforcing pulse” used in the present method may itself comprise more than a single pulse. In particular, there are two principal variants of the present reinforcing pulse method, which may be termed the single and double reinforcing pulse variants. As its name implies, the single reinforcing pulse method uses only one reinforcing pulse. The double reinforcing pulse method is more complex. Consider the situation discussed above where a display has first and second pixels sharing a common edge, with the first pixel undergoing a transition from white to black, while the second pixel remains in its black state. If a single reinforcing method is being used, there is applied to the second pixel a single pulse of the same polarity as the pulse applied to the first pixel, i.e., a black-going pulse. In the double reinforcing pulse method, there is first applied to the second pixel a pulse of the opposite polarity to that applied to the first pixel (i.e., a white-going pulse), and thereafter there is applied to the second pixel a pulse of the same polarity as that applied to the first pixel (i.e., a black-going pulse). The two successive pulses applied to the second pixel may hereinafter be referred to a the “reverse reinforcing pulse” and the “forward reinforcing pulse” respectively. There may or may not be a pause (a period of zero voltage) between the reverse and forward reinforcing pulses. As discussed in more detail below, it has been found that the double reinforcing pulse method typically more accurately matches the color of the second pixel to that of the first pixel.

As already noted, in the present method the reinforcing pulse may be applied only to a pixel that shares an edge with the pixel being driven (hereinafter for convenience pixels having this relationship may be called “edge-adjacent pixels” to distinguish them from “corner-adjacent pixels” which only have one corner in common). There are three main variants of the present method, all of which may be applied to the typical type of display comprising a two dimensional array of pixels arranged in rows and columns:

- (a) An “edge-adjacent-only” variant, in which the reinforcing pulse is applied only to the edge-adjacent neighbors of each pixel being driven from one optical state to another; in the common case where the pixels of a display are arranged in a rectangular matrix, most pixels will of course have four edge-adjacent neighbors. It has not been found necessary to apply the reinforcing pulse to pixels corner-adjacent to the driven pixels, although this could of course be done if so desired;
- (b) A “global” variant in which, when one or more pixels are driven to one extreme optical state, the reinforcing pulse is applied to all pixels already in that one extreme optical state; and
- (c) An intermediate “contiguous area” variant in which, when one or more pixels in a contiguous area are driven to one extreme optical state, the reinforcing pulse is applied to all pixels within that contiguous area already in that one extreme optical state, but the reinforcing pulse is not applied to pixels in that one extreme optical state outside the contiguous area.

Variant (c) may require further explanation. Consider the not-uncommon situation of a series of images in which a single object, or small number of objects, are moving against an essentially static background; more specifically, consider a series of monochrome images in which a black plane is moving against a white sky. To effect a transition between successive images, certain pixels around the periphery of the plane must be rewritten. To avoid visually-distracting effects, it may be desirable to apply the reinforcing pulses to black pixels representing the plane, but not to black pixels representing the ground beneath. Similarly, if a display is configured as a series of essentially independent windows, it may be desirable to use a drive scheme in which a “global” updating method is applied to each window independently.

As already mentioned, in the reinforcing pulse method the reinforcing pulse is applied either simultaneously with the drive pulse or within a predetermined period after the end of the drive pulse. It is desirable that the reinforcing pulse appear to a user of the display to be part of the same transition as that of the driven pixel, rather than there appearing to be a first transition for the driven pixel and a second one for the reinforcing pulse. Typically, the reinforcing pulse should be applied either simultaneously with the drive pulse or within a subsequent period equal to the length of the drive pulse, and preferably not exceeding about 400 milliseconds. In most cases, it is conveniently to effect the reinforcing pulse simultaneously with the last part of the drive pulse (i.e., simultaneously with a terminal portion of the drive pulse).

The present invention extends to a bistable electro-optic display, controller or application specific integrated circuit arranged to carry out the reinforcing pulse method of the present invention. Specifically, the invention extends to an electro-optic display comprising a layer of bistable electro-optic medium which can display first and second extreme optical states, first and second pixel electrodes disposed adjacent the layer of bistable electro-optic medium and capable of applying electric fields to the medium, the first and second pixel electrodes having adjacent edges separated by an inter-pixel gap, and a controller for controlling the voltages applied to the first and second pixel electrodes, wherein the controller is arranged to carry out a drive method comprising:

- applying to the first pixel electrode a drive pulse effective to cause the electro-optic medium adjacent the first pixel electrode to change its optical state to one of its extreme optical states, and
- applying to the second pixel electrode, while the electro-optic medium adjacent the second pixel electrode is in the said one extreme optical state, a reinforcing pulse of the same polarity as the drive pulse, the reinforcing pulse being applied either simultaneously with the drive pulse or within a predetermined period after the end of the drive pulse.

The present invention also provides a second drive method, which may hereinafter be referred to as the “inverse reinforcing pulse method” of the present invention. Unlike the first (reinforcing pulse) method described above, which is applied when one of an edge-adjacent pair of pixels is transitioning to one optical state while the other of the pair is remaining in that optical state, the inverse reinforcing pulse method of the present invention is applied where one of an edge-adjacent pair of pixels is transitioning from one optical state while the other of the pair is remaining in that optical state. For example, the inverse reinforcing pulse method may be applied where a first pixel is transitioning from white to black, while an adjacent pixel is staying white. In the inverse reinforcing pulse method, the non-transitioning pixel is given a pulse of the opposite polarity to that applied to the transitioning pixel, i.e., in the foregoing example, the adjacent pixel would be given a white-going pulse.

Accordingly, the present invention provides a method of driving a bistable electro-optic display having at least first and second pixels each of which can display first and second extreme optical states (the pixels may or may not have other intermediate optical states), the first and second pixels having adjacent edges separated by an inter-pixel gap. The present inverse reinforcing pulse method comprises applying to the first pixel a drive pulse effective to cause the first pixel to change its optical state from one of its extreme optical states, and applying to the second pixel, which is in the said one extreme optical state, an inverse reinforcing pulse of the opposite polarity to the drive pulse, the inverse reinforcing pulse being applied either simultaneously with the drive pulse or within a predetermined period after the end of the drive pulse.

The inverse reinforcing pulse method of the present invention can make use of any of the variants of the main reinforcing pulse method already described; thus, for example, the inverse reinforcing method may be applied on an edge-adjacent, contiguous area or global basis, although typically it will be applied on an edge-adjacent basis. The inverse reinforcing pulse method may also comprise more than one pulse, and in particular may comprise two pulses of opposite polarity as described for the first method of the present invention.

The inverse reinforcing pulse method of the present invention is intended to reduce to eliminate the problems caused by excessive blooming as described above. In the situation already mentioned where the first of a pair of adjacent white pixels is transitioning from white to black while the second pixel is remaining white, there is a tendency for the drive pulse applied to the first pixel to cause the resultant black area to extend more than half-way across the inter-pixel gap, causing the problems noted above. By applying an inverse reinforcing (white-going) pulse to the second pixel, the excessive black blooming is reduced or eliminated.

The present invention extends to a bistable electro-optic display, display controller or application specific integrated circuit (ASIC) arranged to carry out the second method of the invention. Specifically, the invention extends to an electro-optic display comprising a layer of bistable electro-optic medium which can display first and second extreme optical states, first and second pixel electrodes disposed adjacent the layer of bistable electro-optic medium and capable of applying electric fields to the medium, the first and second pixel electrodes having adjacent edges separated by an inter-pixel gap, and a controller for controlling the voltages applied to the first and second pixel electrodes, wherein the controller is arranged to carry out a drive method comprising:

- applying to the first pixel a drive pulse effective to cause the first pixel to change its optical state from one of its extreme optical states, and
- applying to the second pixel, which is in the said one extreme optical state, an inverse reinforcing pulse of the opposite polarity to the drive pulse, the inverse reinforcing pulse being applied either simultaneously with the drive pulse or within a predetermined period after the end of the drive pulse.

The displays of the present invention may make use of any of the types of bistable electro-optic media described above. Thus, for example, the displays of the present invention may comprise a rotating bichromal member or electrochromic medium. Alternatively, the displays may comprise an electrophoretic medium comprising a plurality of electrically charged particles disposed in a fluid and capable of moving through the fluid under the influence of an electric field. The electrically charged particles and the fluid may be confined within a plurality of capsules or microcells. Alternatively, the electrically charged particles and the fluid may be present as a plurality of discrete droplets surrounded by a continuous phase comprising a polymeric material, so that the electrophoretic medium is of the polymer-dispersed type. The fluid may be liquid or gaseous.

The displays of the present invention may be used in any application in which prior art electro-optic displays have been used. Thus, for example, the present displays may be used in electronic book readers, portable computers, tablet computers, cellular telephones, smart cards, signs, watches, shelf labels and flash drives.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 of the accompanying drawings is a graph showing the variation with time of the reflectivities of two pixels during the refresh pulse method of the aforementioned 2005/0270261.

FIG. 2 is a graph similar toFIG. 1 but showing the reflectivities of two pixels during a first, reinforcing pulse, method of the present invention.

FIG. 3 is a graph showing the variation with time of the reflectivities of a “flashing” pixel which is repeatedly cycled between its black and white extreme optical states, and of a second pixel which is subjected to a single reinforcing pulse method of the present invention.

FIG. 4 is a graph similar toFIG. 3 but showing the effect of driving the flashing pixel with longer drive pulses.

FIG. 5 is a graph similar to those ofFIGS. 3 and 4 but showing the effect of using a modified waveform for the flashing pixel.

FIG. 6 is a graph similar to those of FIGS.3 to5 but showing the effect of using a double reinforcing pulse method of the present invention on the second pixel.

DETAILED DESCRIPTION

As already indicated, a first aspect of this invention relates to a method of driving a bistable electro-optic display in which a reinforcing pulse, which does not effect a gross change in the optical state of a pixel, is applied to one or more pixels during or shortly after the application to another pixel of a drive pulse which does change the optical state of that pixel. The reinforcing pulse serves to match the color of the pixel receiving the reinforcing pulse to that of the pixel receiving the drive pulse and, if the two pixels are edge-adjacent, reduces edge ghosting between the two pixels.

As compared with the refresh pulse driving method of the aforementioned 2005/0270261, the present method may reduce the number of reinforcing pulses needed, since if the display is not updated for a long period, no reinforcing pulses will be applied. (It is of course possible to combine the present methods with the refresh pulse method by ensuring that, if any pixel does not receive a drive pulse or a reinforcing pulse for a long period, that pixel will receive a refresh pulse at the end of the long period.) Also, the reinforcing pulses need not be applied to every pixel of a display in a given extreme optical state, which further reduces the number of reinforcing pulses needed. Accordingly, the present invention can reduce the amount of DC imbalance in a refresh pulse driving method while still avoiding undesirable visual artifacts.

The DC imbalance introduced by the present reinforcing pulse method may be compensated in a manner similar to that described in the aforementioned 2005/0270261; the display controller may keep track of the DC imbalance of each pixel and adjust the impulse of a drive pulse or waveform used to drive the pixel from one extreme optical state to the other to compensate for accumulated DC imbalance of the pixel. If the drive scheme employed makes use of blanking pulses (i.e., pulses which drive all pixels of the display, or of a particular area thereof, to the same optical state) DC balancing is conveniently effected during application of the blanking pulse.

FIG. 1 of the accompanying drawings is a graph of the optical states of two pixels of a monochrome bistable electro-optic display plotted against time. As shown inFIG. 1, both pixels (designatedPixel1 and Pixel2) are originally in their dark extreme optical state.Pixel2 receives a drive pulse which drives it to its white optical state (represented near the upper end of reflectivity (R) ordinate inFIG. 1). Thereafter, this white state slowly decays, as indicated by the gradual downward slope of thePixel2 curve inFIG. 1. Although not shown inFIG. 1, a pixel driven to its black optical state will also decay in an analogous manner, so that after some time a black-and-white image written on the display becomes a dark gray and light gray image, with reduced contrast ratio. (The extent of the decay is exaggerated inFIG. 1 for ease of illustration.) The human eye is very tolerant of such slow, uniform changes in images, and in practice if no part of the image needs to be changed, an image on a bistable display can be left for a substantial period (often of the order of hours, depending upon the specific electro-optic medium employed) before the reduced contrast ratio becomes objectionable to most observers.

However, if, some time afterPixel2 has been driven to its white state,Pixel1, (assumed for present purposes to be edge-adjacent to Pixel2) is driven to its white state, there is a significant difference between the white, freshly rewrittenPixel1 and the partially-decayed, lightgray Pixel2, as shown by the vertical spacing between the two curves inFIG. 1, which also shows that this difference will persist for some time as the optical states of both pixels gradually decay. The resultant difference in color between adjacent pixels tends to be very obvious to the human eye, depending of course upon the extent of the decay in white state. An exactly analogous situation occurs ifPixel1 andPixel2 are driven to their black optical states at different times.

FIG. 2 shows the same sequence of events asFIG. 1 but with the addition of a reinforcing pulse in accordance with the present invention. As will be seen fromFIG. 2, a reinforcing pulse is applied toPixel2 at essentially the same time asPixel1 is driven to its white state, so that after the reinforcing pulse, bothPixel1 andPixel2 are in the same freshly rewritten white state. Thereafter, the white states of

Pixels

1 and2 decay together, so

Pixels

1 and2 maintain the same color at all later times shown inFIG. 2. As already noted, the reinforcing pulse also helps to reduce or eliminate edge ghosting effects between the edge-

adjacent Pixels

1 and2.

In practice, as is the way of the world, matters are not as simple as illustrated in the somewhat idealized graph ofFIG. 2. As mentioned in some of the applications referred to in the “Reference to Related Applications” section above, the extreme optical states of many electro-optic media are not absolutely fixed, and there is a so-called “soft rail” phenomenon such that when one attempts to drive the medium to its extreme optical state the exact optical state achieved depends upon a number of factors including the length of the drive pulse applied, the previous optical states of the pixel and the periods during which the pixel has remained in those optical states. It has been found that repeated application of single reinforcing pulses to a single pixel can pause the pixel to drift towards a more extreme optical state than can be achieved when a pixel is cycled between its two extreme optical states. This is not an unlikely scenario in practice; and would for example, occur in an unchanging pixel adjacent a pixel which is being flashed at regular intervals, for example as part of a timekeeping device.

This situation is illustrated inFIG. 3 of the accompanying drawings, which shows data generated from experimental single pixel displays using encapsulated titania/copper chromite based electrophoretic media substantially as described in Example 4 of the aforementioned U.S. Pat. No. 7,002,728. The broken line inFIG. 3 shows the L* values obtained from a pixel which was repeatedly cycled between its black and white extreme optical states using ±10 V, 240 millisecond drive pulses with 3 seconds between each drive pulse, then finally allowed to rest for 6 seconds (because of the scale of the ordinate inFIG. 3, only the portion of the trace of this pixel adjacent the black extreme optical state is visible, and the apparent broadening of the trace during the rest periods is due to detector noise). It will be seen fromFIG. 3 that (after a minor variation in the first cycle) the behavior of this pixel is consistent in all cycles, with a final L* value of 26.52.

The solid line inFIG. 3 shows the trace obtained from a second pixel which is initially driven to its black extreme optical state using a 10 V 240 millisecond pulse in the same way as the first pixel, and is thereafter subjected to 10 V 40 millisecond reinforcing pulses at 6 second intervals, thus mimicking the pulses applied to a pixel adjacent to the flashing pixel and subjected to a single reinforcing pulse method of the present invention. It will be seen fromFIG. 3 that the reinforcing pulses gradually drive the second pixel to darker optical states, so that after five reinforcing pulses and the final 6 second rest period the pixel has an L* value of 23.67. The difference of almost 3 units between the final L* values of the two pixels is readily perceived by the human eye, and would result in an objectionable ghost, i.e., the pixel which has been flashing would appear as dark gray next to the black unchanging pixel, thus leaving a pale “ghost” of the white flashed image when both pixels were supposed to be in the same final black state.

At first glance, it might appear that this ghosting problem could be overcome simply by increasing the length of the drive pulse applied to the first, flashing pixel, but empirically this has been found not to be the case.FIG. 4 of the accompanying drawings shows traces similar to those ofFIG. 3, except that the flashing pixel was driven with 480 millisecond drive pulses. It will be seen fromFIG. 4 that, although the longer drive pulses do initially drive the flashing pixel to lower L* values than the shorter drive pulses used inFIG. 3, they also cause a considerably greater change in L* values in the 1-2 seconds following the drive pulse (the phenomenon referred to as “self-erasing” or “kickback” in some of the aforementioned E Ink patents and applications) so that the final state of the flashing pixel inFIG. 4 does not differ greatly from that inFIG. 3, and indeed the final L* value of the flashing pixel inFIG. 4 is slightly greater than inFIG. 3, i.e., the flashing pixel actually ends up slightly lighter in color. As inFIG. 3, the difference between the final states of the two pixels inFIG. 4 would readily be perceptible to the human eye and would result in an undesirable ghost image.

Another attempt was made to eliminate the difference between the final states of the two pixels by using for the first, flashing pixel a drive waveform comprising a 240 millisecond drive pulse, a 1 second pause when no voltage was applied, and then a 40 millisecond second drive pulse, both drive pulses being at 10 V; in effect, these experiments applied the 40 millisecond reinforcing pulse to both the flashing and the unchanging pixel. The traces obtained are shown inFIG. 5 of the accompanying drawings. From this Figure it will be seen that the difference between the final states of the two pixels was substantially reduced; the approximately 1 L* unit difference which remained would not be readily apparent to most observers. However, the 1 second pause between the two drive pulses applied to the flashing pixel produced an undesirable “flashy” effect; an observer viewing the two pixels would in effect see an initial ghost image which would disappear one second later.

Finally,FIG. 6 shows traces similar to those of FIGS.3 to5 but in which the unchanging pixel was driven in accordance with a double reinforcing pulse method of the present invention, there being applied to this unchanging pixel a −10 V, 15 millisecond (white-going) reverse reinforcing pulse, immediately followed by a +10 V, 25 millisecond (black-going) forward reinforcing pulse. It will be seen fromFIG. 6 that this double reinforcing pulse method essentially eliminated the difference between the final optical states of the two pixels; the actual values were 26.61 for the flashing pixel and 26.52 for the unchanging pixel, and the difference of 0.09 L* units is unlikely to be noticed even by expert observers. Although the double reinforcing pulse does produce a slight flash as the nominally unchanging pixel first goes slightly whiter and then slightly darker, this slight flash is much less objectionable than the flash produced inFIG. 5, where the two transitions are one second apart. Furthermore, the slight flash experienced in theFIG. 6 drive scheme can be minimized by careful choice of when the reinforcing pulses are applied relative to the drive pulse applied to the flashing pixel; the eye will be less sensitive to the slight flash of the unchanging pixel is this is timed to coincide with the last part of the much more noticeable white-to-black transition of the flashing pixel.

The double reinforcing pulse method ofFIG. 6 not only produces more accurate matching of the final states of the flashing and unchanging pixels than the single reinforcing pulse method ofFIG. 3 but also reduces the amount of DC imbalance introduced by the reinforcing pulses by a factor of 4; the algebraic sum of the forward and reverse reinforcing pulses in theFIG. 6 method is a 10 V, 10 millisecond pulse, as compared with the 10 V, 40 millisecond pulse used in theFIG. 3 method.

The optimum lengths of the reverse and forward reinforcing pulses in the double reinforcing pulse method of the present invention are dependent upon several factors, including the specific electro-optic medium employed, its temperature and the conductivity of the lamination adhesive which is normally present between the electro-optic medium and one of the electrodes, as in the displays used in the experiments shown in FIGS.3 to6. In general, the higher the conductivity of the lamination adhesive, the shorter can be the reinforcing pulse lengths and the less flash is perceived from these pulses. However, high conductivity lamination adhesives do increase blooming. Consequently, it may be more difficult to apply the methods of the present invention in displays using low conductivity lamination adhesives, such as those typically used in high resolution active matrix displays, without causing a possibly objectionable degree of flashiness in the display. It should also be noted that the reinforcing pulse lengths are not necessarily the same for the black and white states. For example, the 15/25 millisecond combination used inFIG. 6 was found to be optimal for transitions to the black extreme optical state of the medium. However, when similar experiments were conducted for transitions to the white extreme optical state of the same medium at the same temperature, it was found that a 20/25 millisecond combination gave optimal results.

Both the reinforcing pulse method and the inverse reinforcing pulse method of the present invention can be used in the same display at the same time. However, because pixels have multiple edge-adjacent neighbors (for example, in a typical rectangular matrix display most pixels have four edge-adjacent neighbors), the demands of the two methods may conflict and it is necessary to devise rules to enable the two methods to co-exist. For example, a single unchanging black pixel may have one edge-adjacent neighbor which is changing from black to white (suggesting the use of an inverse reinforcing pulse method) and another neighbor which is changing from white to black (suggesting the use of a “regular” reinforcing pulse method) and the optimum reinforcing pulse lengths may not be the same in the two cases. In general, artifacts due to mismatches between the optical states of entire pixels are more objectionable than artifacts due to mismatches within the inter-pixel region. Accordingly, where the demands of the reinforcing pulse method are in conflict with those of the inverse reinforcing pulse method, in general the demands of the former should prevail.

Numerous changes and modifications can be made in the preferred embodiments of the present invention already described without departing from the scope of the invention. Accordingly, the foregoing description is to be construed in an illustrative and not in a limitative sense.