GB2347547A - Addressing scheme for liquid crystal displays - Google Patents

Abstract

A method of addressing pixels of a liquid crystal display comprising pixels arranged in a plurality of columns and rows comprises the steps of: <SL> <LI>a) applying data pulses to the columns of pixels, said data pulses being members of a data set having first and second extreme data types 40, 42; <LI>b) dividing the rows of pixels into a plurality of pairs of rows of pixels and, for each pair in turn, applying normal and inverted strobe pulses 44, 46 respectively to the two rows of each pair; and <LI>c) applying a compensation pulse 56 to each pixel in such a way that, at least for said two extreme data types 40, 42, the switching 48, 60 and non-switching 50, 58 resultants are of substantially the same shape, or at least more similar in shape. </SL>

Description

Addressing Scheme for Liquid Crystal Displays The invention relates to an addressing scheme for liquid crystal displays, and particularly for ferroelectric liquid crystal displays employing a split line addressing system.

Some prior art documents will first be discussed.

It is common to address a matrix of pixels by using an addressing technique in which data pulses are applied to each column of pixels, and each row of pixels is strobed in turn by applying strobe pulses. Each strobe pulse typically comprises a blanking pulse (which blanks the row black or white) followed by a switching pulse which combines with the data pulse to form a resultant pulse which switches the pixel to one of a number of possible greyscale states. It is possible to use a number of different"data sets", each set comprising a number of differently shaped data pulses (or data types), each representing a different greyscale level, with the two"extreme"data types representing black and white.

Japanese Patent Application No 27719/1993 (published 5 December 1993) in the name of Canon Inc describes a dual line addressing technique sometimes known by the acronym"SLAGS", which stands for"split line analogue greyscale". This addressing technique applies strobe pulses to two adjacent rows at a time. Both of these rows therefore see the same data pulses because they are strobed simultaneously. The two rows of pixels can also be thought of as a single row which is split into two. However, the strobe pulses applied to each of the two rows are of opposite polarity.

This addressing technique can be understood by referring to the example shown in Figures 1 and 2. Figures 1 and 2 each show strobe and data pulses applied to a"split line" (ie applied to an adjacent pair of pixel rows)."A"and"B"represent the two extreme data types for the chosen data set,"A"representing white and"B"representing black. The first row (of the pair) uses a strobe pulse 10 comprising a negative blanking pulse 12 and a positive switching pulse 14. The second row (of the pair) uses a strobe pulse 16 which comprises a positive blanking pulse 18 and a negative switching pulse 20.

Figure 1 shows what happens at a given pixel when the strobe pulse 10 is combined with data pulse"A". In this case the strobe pulse blanks the row (and hence the pixel in question) black, and the pixel is then switched white by the combination of switching pulse 14 with data pulse"A", so the pixel ends up white. For the second row, the strobe pulse 16 blanks white, but then the combination of the negative switching pulse 20 and the data pulse"A"fails to switch the pixel, so the pixel remains white. Thus the desired outcome is achieved because both pixels are made white when data type"A"is applied.

Figure 2 shows that the desired outcome is also achieved when data type"B"is applied.

In this case for the first row the strobe pulse 10 blanks black and the pixel then stays black (ie fails to switch) when the switching pulse 14 is combined with data pulse"B".

For the second row, the strobe pulse 16 blanks white, but the pixel is then switched black by the combination of the switching pulse 20 and data pulse"B".

This addressing technique also achieves the desired outcome for each intermediate data type between"A"and"B"representing the different greyscale levels between black and white. However the advantage of such a split line addressing technique is that errors caused by factors such as temperature variation and thickness variation tend to have an opposite (and hence cancelling) effect on each of the two rows of the split line because the opposite polarity systems of the two lines are affected in opposite ways. So, for example, if increasing temperature causes a darkening of greyscales in the first row of the split line, it will cause a lightening of greyscales in the second row, so the overall effect (ie the average of the two) becomes substantially independent of temperature variations.

Such a split line addressing technique requires good overlap between the normal and inverted drive windows. Data and strobe pulses are typically divided into a number of "slots" (ie time interval slots). For example, in the J/A (Joers Alvey) data set each data type (ie data pulse) is divided into four slots. The drive window is, on a graph of strobe voltage amplitude against slot time, the area between the switching curves for the two extreme data types. It is necessary to work within this area in order to ensure that the white data type (at one extreme) switches the pixel white, and the black data type (at the other extreme) does not switch the pixel (which therefore remains black, assuming black blanking).

If a simple square pulse is applied to a n-V min material the switching curve (on a graph of pulse voltage against duration) has a minimum, hence the term"X-V min". Above the curve, the pulse causes switching of the material, and below the curve the pulse does not cause switching. When such a simple square pulse is combined with a data pulse, the switching curve moves. If the data pulse is an extreme data type"A"the switching curve is moved in one direction, and if the data pulse is the other extreme data type"B" the switching curve is moved in the opposite direction. It is the area between these two switching curves of the two extreme data types which forms the drive window.

In a split line addressing system it is possible to draw separate drive windows for both the normal and inverted strobe pulses. In order for the device to operate properly it is necessary to operate in the area of overlap between these two drive windows thus ensuring that both lines (ie both rows of pixels) will switch properly when the various data types are applied.

Japanese Patent Application No. 9-72198/1997 describes a modification of the above split line system in which the dual line addressing is achieved using an interlacing technique which regains much of the resolution lost when addressing two lines with the same data type. This system also requires good overlap between the normal and inverted drive windows.

British Patent Application No 9710403.8 uses dual line addressing with opposite polarity blanking and strobe to achieve 3 digital levels (both pixels white, one pixel white and one pixel black). This requires good overlap of the normal and inverted drive windows.

British Patent Application No 9801299.0 describes an arrangement in which a low frequency waveform is superimposed onto the row waveforms to increase the RMS seen by the pixels with less increase in power consumption.

European Patent Publication No 0306203A describes an addressing scheme suitable for T-V min materials. The standard data types are [Vd,-Vd] and [-Vd, Vd], which are combined with a strobe pulse [0, Vs]. This is thus a two slot addressing scheme, in which Vd is the amplitude of the data pulse, and Vs the amplitude of the strobe pulse.

This is a digital data set, and there are therefore only two data types. Depending on the data stream the frequency of the data can vary by a factor of two which can cause crosstalk problems. The data pulses are referred to in this specification as J/A (Joers Alvey) data.

European Patent Application No 95307751.8 describes a r-Vmin addressing scheme.

The data is designed to avoid cross-talk problems particularly the data that occurs just before and just after the select period. Each data types has an equivalent RMS, is DC balanced and follows the same polarity behaviour as a function of time. Such data pulses are hereinafter referred to as BPPI data. This is an analogue data set, comprising a number of data types representing different greyscale levels.

British Patent Application No 9718121.8 describes a T-Vmin addressing scheme. The data is designed to avoid cross-talk problems particularly the data that occurs just before and just after the select period. Each data types has an equivalent RMS, is DC balanced and follows the same polarity behaviour as a function of time. This data set is designed to be more symmetric in order to avoid drift of greylevels. Such data pulses are hereinafter referred to as KPPI6 in the text. This is an analogue data set, comprising a number of data types representing different greyscale levels.

The problem to be addressed by the invention will now be discussed.

In a dual line addressing scheme (which uses two strobe pulses of opposite polarity, as discussed above), the combination of a data type which gives a switching resultant when combined with the normal strobe will give a non-switching resultant when combined with the inverted strobe. Similarly, the combination of a data type which gives non-switching when combined with the normal strobe will give switching when combined with the inverted strobe. If J/A data is used (see above), then the two switching resultants are equivalent (ie the same shape), and the two non-switching resultants are also equivalent.

This can be seen by referring to Figure 3, in which the two (digital) J/A data types are labelled"A"and"B". Figure 3 shows the four resultants 26,28,30 and 32 which are formed when the data types"A"and"B"are combined with the normal strobe pulse 22 and the inverted strobe pulse 24. It can be seen that the two switching resultants 26 and 32 are equivalent, as are the two non-switching resultants 28 and 30.

In a dual line addressing scheme it is desirable to have similar drive windows for the normal and inverted strobes, as explained above. This is relatively easy to achieve when using J/A data because the fact that the resultant data shapes are the same means that the normal and inverted drive windows are more likely to overlap. However J/A data is unsuitable for analogue greyscale, and even for some digital, applications due to its extreme pixel pattern dependence (a form of cross talk). Pixel pattern dependence is the dependence of pixel switching on the states of adjacent pixels in the adjacent rows (not the same row). It arises because the data pulses which are applied to a pixel when other rows are being strobed nevertheless have an effect on that pixel.

Methods for producing pixel pattern independent data sets, BPPI and KPPI6, are described in the prior art documents mentioned above. For these data sets however the switching and non-switching data types are generally not the inverse of one another and therefore the switching and non-switching resultants for the normal and inverted strobes will be different. This means that the drive windows are less likely to overlap and some modification to the amplitude and/or width of one of the strobes will be required in order to operate in the two windows simultaneously. In addition even when operating in both drive windows simultaneously it has been found that the hysteresis of analogue greyscale due to switch history can have a different. This means that the drive windows are less likely to overlap and some modification to the amplitude and/or width of one of the strobes will be required in order to operate in the two windows simultaneously. In addition, even when operating in both drive windows simultaneously it has been found that the hysteresis of analogue greyscale due to switch history can have a different width for the different strobes. This means that compensation of error by the dual line technique cannot be achieved for all switch histories.

According to the invention there is provided a method of addressing pixels of a liquid crystal display comprising pixels arranged in a plurality of columns and rows, the method comprising the steps of : a) applying data pulses to the columns of pixels, said data pulses being members of a data set having first and second extreme data types; b) dividing the rows of pixels into a plurality of pairs of rows of pixels and, for each pair in turn, applying normal and inverted strobe pulses respectively to the two rows of each pair; and c) applying a compensation pulse to each pixel in such a way that, at least for said two extreme data types, the switching and non-switching resultants are of substantially the same shape, or at least more similar in shape.

Because the switching and non-switching resultants are more nearly the same shape, it is easier to achieve overlapping drive windows for the normal and inverted strobe pulses.

Preferably said compensation pulse is a compensation strobe pulse, which is superimposed on the normal strobe pulse, the inverted strobe pulse, or both.

The compensation pulse may be applied only during the select period, being the period of overlap between the data and strobe pulses.

Alternatively, the compensation pulse may be applied continuously.

If the data and strobe pulses are divided into a number of time slots, each having a given duration during which the strobe pulse is substantially constant, the compensation pulse may also be divided into a number of time slots of corresponding durations, during which the compensation pulse is also substantially constant.

In this case, the amplitude of the compensation pulse during each time slot may be chosen to be substantially equal to the inverse of the sum of the two extreme data types during the same time slot.

The liquid crystal device may be a ferroelectric liquid crystal display.

The invention also provides a liquid crystal device comprising pixels arranged in a plurality of columns and rows, the liquid crystal device further comprising: a) data means for applying data pulses to the columns of pixels, said data pulses being members of a data set having first and second extreme data types; b) strobe means for applying normal and inverted strobe pulses to pairs of rows; and c) compensation means for applying a compensation pulse to each pixel in such a way that, at least for said two extreme data types, the switching and non-switching resultants are of substantially the same shape, or at least more similar in shape, wherein said device is arranged to carry out the method described above.

The invention will now be more particularly described with reference to the accompanying drawings, in which: Figures I (a) and (b) show the combination of a first extreme (digital) data type"A" with normal and inverted strobe pulses respectively in a split line addressing technique; Figures 2 (a) and (b) show the combination of a second extreme (digital) data type"B" with the same normal and inverted strobe pulses respectively ; Figure 3 shows the combination of two J/A extreme (digital) data types"A"and"B" with normal and inverted strobe pulses in a split line addressing technique, to form switching and non-switching resultants of the same shape; Figure 4 shows the combination of two KPPI extreme (digital) data types"A"and"B" with normal and inverted strobe pulses in a split line addressing technique, to form switching and non-switching resultants of different shapes; Figure 5 shows the combination of the same two KPPI extreme (digital) data types"A" and"B"with the same normal and inverted strobe pulses, but the inverted strobe pulse is also combined with a compensation strobe pulse to form switching and non-switching resultants of the same shape; Figures 6 (a), (b) and (c) show the drive windows for KPPI data types for the different strobes used in Figure 5. These correspond to the cases of the normal strobe pulse, the inverted strobe pulse, and the inverted strobe pulse plus the compensation strobe pulse respectively; and Figure 7 demonstrates the difference in hysteresis width (for KPPI data) between the normal and inverted strobes of Figure 5.

Figures 1 to 3 have been discussed above.

Figure 4 shows the combination of two extreme KPPI data types"A"and"B" (labelled 40 and 42 respectively) with normal and inverted strobe pulses, labelled 44 and 46 respectively. The resulting pulses 48,50,52 and 54 are shown in Figure 4. Pulses 48 and 54 are switching resultants, and pulses 50 and 52 are non-switching resultants. It can be seen from Figure 4 that, unlike in the case of the J/A data types of Figure 3, the two switching resultants 48 and 54 are not the same shape, and neither are the two nonswitching resultants 50 and 52 the same shape.

Figure 5 shows an embodiment of the invention in which a compensation strobe pulse 56 is applied to the inverted strobe pulse 46. The same reference numerals are used in Figures 4 and 5 for corresponding items. The switching and non-switching resultants 48 and 50 formed by the combination of the data types 40 and 42 (respectively) with the normal strobe pulse 44 are the same as the resultants 48 and 50 shown in Figure 4.

However, the non-switching resultant 58 formed by combining the data type 40, compensation strobe pulse 56 and inverted strobe pulse 46 is modified in shape so that it is the same shape as the non-switching resultant 50. Similarly, the switching resultant 60 formed by the combination of data type 42, compensation strobe pulse 56 and inverted strobe pulse 46 is modified in shape by the compensation strobe pulse 56 so that it is the same shape as the switching resultant 48. The effect of the compensation strobe pulse 56 is therefore to ensure that the two switching resultants 48 and 60 are the same shape, as are the two non-switching resultants 50 and 58.

Each of the data types (40,42) strobe pulses (44,46) and compensation strobe pulse 56 is divided into four equal time slots. In order to ensure switching and non-switching resultants (48,50,58,60) of the same shape it is necessary to ensure that, for each time slot, the voltage Is of the compensation strobe pulse 56 (which is superimposed on the invented strobe pulse, hence the symbol Is) satisfies the equation Is =- (A + B), where A and B represent the voltages of the data types 40 and 42 respectively for the same time slot. This can be seen from the fact that, for each time slot, the following equation must be satisfied: A+N=- [B+Is+I] =- [B +Is-N = > A=-B-Is = > Is=-(A+B) The switching (A) and non-switching (B) data types of the KPPI data set described in British Patent Application No. 9718121.8 are: A =-F2. Vd, 0, F2-. Vd I) B={-#2.Vd,0,#2V.d.0}} and are typically combined with the normal (N) and inverted (I) strobes: N={0,VS,Vs,Vs}} I = {0,-VS,-VS,-Vs}} In order to make the two extreme resultants equivalent (in shape) the following waveform must be superimposed onto the inverted strobe I : Is ={#2.Vd,#2.Vd,-#2.Vd,-#2.Vd}} Figures 6 (a), (b) and (c) show, for the case of KPPI data types, the drive windows 62, 64 and 66 for the normal strobe, inverted strobe and compensated inverted strobe respectively. The drive window 64 for the inverted strobe without compensation is small compared to the drive window 62 for the normal strobe. The effect of the compensation pulse 56 (shown in Figure 5) is to provide a larger drive window 66 which is more similar to the normal strobe drive window 62.

Figure 7 demonstrates the difference in hysteresis width (for KPPI data) between the normal and inverted strobes 44,46. When the compensation waveform 56 is superimposed onto the inverted strobe 46 the hysteresis width decreases and is much more similar to the normal strobe hysteresis.

Claims (8)

1. CLAIMS: 1. A method of addressing pixels of a liquid crystal display comprising pixels arranged in a plurality of columns and rows, the method comprising the steps of : a) applying data pulses to the columns of pixels, said data pulses being members of a data set having first and second extreme data types; b) dividing the rows of pixels into a plurality of pairs of rows of pixels and, for each pair in turn, applying normal and inverted strobe pulses respectively to the two rows of each pair; and c) applying a compensation pulse to each pixel in such a way that, at least for said two extreme data types, the switching and non-switching resultants are of substantially the same shape, or at least more similar in shape.
2. 2 A method as claimed in claim 1, wherein said compensation pulse is a compensation strobe pulse, which is superimposed on the normal strobe pulse, the inverted strobe pulse, or both.
3. 3 A method as claimed in claim 1 or 2, wherein the compensation pulse is applied only during the select period, being the period of overlap between the data and strobe pulses.
4. 4 A method as claimed in claim 1 or 2, wherein the compensation pulse is applied continuously.
5. 5 A method as claimed in any preceding claim, wherein the data and strobe pulses are divided into a number of time slots, each having a given duration during which the strobe pulses are substantially constant, and the compensation pulse is also divided into a number of time slots of corresponding durations during which the compensation pulse is also substantially constant.
6. 6 A method as claimed in claim 5, wherein the amplitude of the compensation pulse for each time slot is chosen to be substantially equal to the inverse of the sum of the two extreme data types for the same time slot.
7. 7 A method as claimed in any preceding claim, wherein the liquid crystal display is a ferroelectric liquid crystal display.
8. 8 A liquid crystal device comprising pixels arranged in a plurality of columns and rows, the liquid crystal device further comprising: a) data means for applying data pulses to the columns of pixels, said data pulses being members of a data set having first and second extreme data types; b) strobe means for applying normal and inverted strobe pulses to pairs of rows; and c) compensation means for applying a compensation pulse to each pixel in such a way that, at least for said two extreme data types, the switching and non-switching resultants are of substantially the same shape, or at least more similar in shape, wherein said device is arranged to carry out the method of any preceding claim.