Liquid crystal display apparatus - Patent Review 5815132

Description Submit all comments and votes

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a display apparatus, which uses ferroelectric liquid crystal (FLC), and a method of driving the same, and, more particularly, to a liquid crystal display apparatus which displays image gradation by a matrix drive method and a method of driving the same.

2. Related Background Art

As for the display apparatus, which uses a ferroelectric liquid crystal (FLC), there has been a known device disclosed in Japanese Patent Application Laid-Open No. 61-94023 and constituted in such a manner that ferroelectric liquid crystal is injected into a liquid crystal cell formed by placing two glass plates, each of which has a transparent electrode formed thereon and which have been subjected to an orienting process in such a manner that the two glass plates are placed while having a cell gap of about 1 .mu.m.about.3 .mu.m.

The aforesaid display apparatus which uses ferroelectric liquid crystal has two characteristics. That is, a fact, that the ferroelectric liquid crystal has a spontaneous polarization, causes combining force of an external electric field and the spontaneous polarization to be utilized to be utilized in switching. Another effect can be obtained in that the switching operation can be performed by the polarity of an external electrode because the longer axes of ferroelectric liquid crystal molecules correspond to the directions of the spontaneous polarizations.

The longer axes of the liquid crystal molecule of the ferroelectric liquid crystal are oriented in twisted directions under a bulk condition because the ferroelectric liquid crystal ordinarily uses chiral smectic liquid crystal (SmC* SmH*). However, the aforesaid problem that the longer axes of the lqiuid crystal molecules are undesirably twisted can be overcome by injecting the ferroelectric liquid crystal into the aforesaid cell having the cell gap of 1 .mu.m.about.3 .mu.m. The aforesaid phenomenon has been disclosed in p213 to p234, N. A. CLARK et al., MCLC, 1983, Vol 94 and so forth.

Although the ferroelectric liquid crystal has been mainly utilized as a binary (light and dark) display device having two stable states composed of a light transmissive state and a light shielded state, multi-value images, that is, half tone images can also be displayed. The half tone image display methods are exemplified by a method which realizes a half-tone type light transmissive state by controlling the area ratio in a bi-stable state (the light transmissive state or the light shielded state) in a pixel. Then, the gradation expressing method (hereinafter called an "area modulation method") will now be described.

FIG. 9 is a graph which schematically illustrates the relationship between switching pulse V of the ferroelectric liquid crystal device and transmissive light quantity I of the same, where transmissive light quantity I realized after a single pulse of either polarity is applied to a pixel in an initial state in which it is completely shielded from light (dark state) is plotted as the function of voltage V of the single pulse. If the pulse voltage V is lower than threshold V.sub.th (V<V.sub.th), the transmitted light quantity is not changed, and the transmissive state after the pulse has been applied is, as shown in FIG. 10B, the same as that shown in FIG. 10A. If the pulse voltage V is higher than the threshold, (V.sub.th <V), a portion in the pixel is brought to another stable state, that is, a light transmissive state as shown in FIG. 10C so that the overall light quantity becomes an intermediate quantity. If the pulse voltage is raised to a value higher than saturation value V.sub.sat (V.sub.sat <V), the overall portion of the pixel is brought into a light transmissive state as shown in FIG. 10D, and therefore the light quantity reaches a predetermined value (saturated).

That is, the area gradation method is a method for forming half tone images corresponding to the applied voltage V by performing a control in which the pulse voltage V is caused to meet V.sub.th <V<V.sub.sat.

However, the following problem arises if the aforesaid simple driving method is employed. That is, the fact that the relationship between the voltage and the transmissive light quantity depends upon the thickness of the cell and the temperature will arise a problem in that a different gradation is displayed depending upon the position in the display panel although a pulse voltage of a predetermined level is applied if a cell thickness or the temperature is dispersed in the display panel.

FIG. 11 is a graph which illustrates the aforesaid fact, where the relationship between the pulse voltage V and the transmissive light quantity I is shown similarly to FIG. 9. In FIG. 11, the relationship between the two factors at different temperatures, that is, curve H indicating the relationship held at high temperature and curve L indicating the relationship held at low temperature are shown. In general, a display of a type having a large size frequently encounters a fact that the temperatures are dispersed in the same panel. Therefore, when a half tone image is formed at a certain driving voltage V.sub.ap, a problem arises in that the half tone level is distributed irregularly in a range from I.sub.1 to I.sub.2 in the same panel as shown in FIG. 11 and therefore a uniform gradation image cannot be formed.

In order to overcome the aforesaid problem, a driving method (hereinafter called a "4-pulse method") has been disclosed in Japanese Patent Application No. 2-94384 by the applicant of the present invention (inventor: Okada). As shown in FIGS. 8 and 12, the "4-pulse method" is a method in which a plurality of pulses (pulses A, B, C and D shown in FIG. 12) are applied to all of a plurality of pixels positioned on the same scanning line in one panel and having different thresholds so as to obtain the same quantity of transmissive light as shown in FIG. 8.

However, use of the aforesaid "4-pulse method" will arise the following problem in that optical responses of the pixel with respect to the applied writing pulses (A), (B), (C) and (D) are respectively affected by other pulses previously applied to the aforesaid pixel. during a process in which the reset pulse (A) is applied to the pixel on a selected scanning line and then gradation information writing pulses (B), (C) and (D) are applied as shown in FIGS. 8 and 12. That is, the voltage (threshold), at which the liquid crystal is inverted, is changed when the next pulse is applied. The aforesaid phenomenon will raise a problem at the time of setting the voltage of the pulse (B). Although the error is included by an allowable range (although the accuracy in expressing the gradation deteriorates) if the influence of the other pulse is limited and the degree of the threshold change is also limited, forming of gradation images cannot be performed by the 4-pulse method if the threshold is changed considerably. The reason for this lies in that the aforesaid "4-pulse method" disclosed in Japanese Patent Application No. 3-73127 is a driving method based on a fact that the inversion characteristics of liquid crystal with respect to the voltages of the four pulses applied to the pixel are the same.

Furthermore, domain walls such as i, j and k (the boundary between the oriented region corresponding to the light state and the oriented region corresponding to the dark region) shown in FIG. 8 must be included by the pixel in the case where the other pulses (B), (C) and (D) are applied because bright and dark domains present in the pixel, to which the voltage has been applied, while being mixed with each other (in a state where a half tone image is displayed) although the pulse (A) shown in FIG. 8 can be set to a voltage level sufficiently higher than the threshold because it is a reset pulse. As described above, the positions of the domain walls i, j and k are affected considerably by the voltage pulse applied immediately as well as the writing pulses (B), (C) and (D) in the case where switching is performed with the voltage which extremely approximates the inversion threshold of the liquid crystal. Although the influence of the other pulse applied immediately before the writing pulses are applied does not raise a critical problem in the case where the change of the voltage of the pulses applied immediately is limited, a problem sometimes arises in that the "4-pulse method" drive cannot be performed if the change has been made considerably.

The aforesaid problem taken place in that the displayed gradation image is undesirably affected by the pulse except for the writing pulses also arises by the other pulse immediately after the writing pulse has been applied. In a case where a domain wall is formed by the pulse (C) at the position j shown in FIG. 8, the domain wall can be sometimes translated if the pulse (for example, a voltage pulse due to an information signal at the time of no selection) following the pulse (C) has a certain voltage level. That is, there is a problem in that the displayed gradation image determined by the writing pulses can be easily subjected to a cross talk which takes place due to the influence of the ensuring pulses.

There arises another problem in that writing takes a too long time in addition to the aforesaid problems of the threshold level change and the cross talk. The reason for this lies in that the "4-pulse method" must use four pulses (A), (B), (C) and (D) in comparison to the conventional driving method in which two pulses are used to write one pixel. As a result, the time (the frame time) required to write image information on the entire surface of the panel is lengthened, causing the quality of a displayed kinetic image to deteriorate. If the worst comes to the worst, kinetic images cannot be displayed.

As described above, the "4-pulse method" encounters a problem of the error taken place when a gradation image is formed or another problem of an unsatisfactory display speed.

SUMMARY OF THE INVENTION

To this end, an object of the present invention is to provide a liquid crystal display apparatus which uses ferroelectric liquid crystal and which is capable of stably displaying an analog gradation image at high speed.

In order to overcome the aforesaid problems, according to one aspect of the present invention, there is provided a liquid crystal display apparatus comprising: a liquid crystal cell in which ferroelectric liquid crystal is disposed between two electrode substrates disposed to fact each other and an intersection portion between a scanning electrode group and an information electrode group respectively formed on the electrode substrates is made to be a pixel; scanning signal applying means; and information signal applying means, wherein the pixel has a threshold distribution with respect to a gradation information signal at the time of a scanning selection operation, the scanning signal applying means simultaneously applies scanning signals to a plurality of scanning electrodes in synchronization with an operation in which the information signal applying means applies the gradation information signal to an information electrode, and the scanning signals applied simultaneously have different waveforms.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates driving waveforms according to Example 1;

FIG. 2 illustrates the structure of an electrode according to Example 2;

FIG. 3 illustrates the potential gradient realized in Example 2;

FIG. 4 is a block diagram which illustrates a driving circuit according to the present invention;

FIG. 5 is a schematic cross sectional view which illustrates a cell according to the present invention;

FIGS. 6A and 6B illustrate the principle of a gradation expression and correction according to the present invention;

FIG. 7 illustrates the angle of a polarizer of a liquid crystal display device according to the present invention;

FIG. 8 illustrates a conventional gradation driving method;

FIG. 9 illustrates the conventional gradation driving method;

FIGS. 10A to 10D illustrate the conventional gradation driving method;

FIG. 11 illustrates the conventional gradation driving method;

FIG. 12 illustrates waveforms in the conventional gradation driving method;

FIGS. 13A to 13D illustrate the operation of the present invention;

FIGS. 14A and 14B illustrate the operation of the present invention;

FIGS. 15A to 15E illustrate the operation of the present invention;

FIG. 16 illustrates a compensating method according to the present invention;

FIGS. 17A to 17C illustrate the compensating method according to the present invention;

FIG. 18 illustrates the compensating method according to the present invention;

FIG. 19 illustrates the driving waveforms according to Example 3;

FIG. 20 is a graph which illustrates curves indicating the DT-V characteristics of liquid crystal materials according to Examples 1 to 6;

FIG. 21 illustrates a scanning method according to Example 4;

FIG. 22 is a time sequential view which illustrates a driving waveforms according to Example 5;

FIGS. 23A and 23B illustrate the driving waveforms according to Example 5;

FIG. 24 is another time sequential view which illustrates driving waveform according to Example 5;

FIG. 25 illustrates other driving waveforms according to Example 5;

FIGS. 26A and 26B illustrate the compensating method according to the present invention;

FIG. 27 is a time sequential view which illustrates driving waveforms according to Example 6;

FIGS. 28A and 28B illustrate the driving waveforms according to Example 6;

FIGS. 29A and 29B show time sequential views which illustrate driving waveforms according to Example 6;

FIG. 30 illustrates other driving waveforms according to Example 6;

FIGS. 31A to 31C illustrate the compensating method according to the present invention;

FIG. 32 illustrates the other cell structure according to Example 1; and

FIG. 33 is a time sequential view which illustrates other driving waveforms according to Example 5.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

A liquid crystal cell adaptable to the present invention has the thresholds dispersed in one pixel thereof as shown in FIG. 5. Since the thickness of an FLC layer 55 between electrodes is changed in the cell shown in FIG. 5, the switching threshold of the FLC is also dispersed. By raising the voltage to be applied to the aforesaid pixel, switching takes place sequentially from a thinner portion.

The aforesaid phenomenon is shown in FIG. 13A. Symbols T.sub.1, T.sub.2 and T.sub.3 shown in FIG. 13A represent temperatures of portions of the panel which is being observed. The switching threshold voltage of the FLC is in inverse proportion to the temperature as illustrated in FIG. 13A, where the relationships between the applied voltages and the light transmittances at the three temperature levels are designated by three curves.

Although the threshold is changed due to factors in addition to the temperature change, the present invention will be described on the basis of a fact that the threshold is changed mainly due to the temperature change.

As can be seen from FIG. 13A, when the overall body of the pixel is reset to a dark state and voltage of V.sub.i is applied to the pixel at temperature T.sub.1, transmissivity of X% can be obtained. However, if the temperature is raised to T.sub.2 or T.sub.3, the transmissivity is undesirably raised to 100% in the case where the same voltage V.sub.i is applied to the pixel and therefore an image having gradation cannot be displayed correctly. FIG. 13C illustrates a state where a pixel is inverted at each of the aforesaid temperature after writing has been performed. In the aforesaid state, written gradation information can be deleted due to the temperature change, causing a problem to take place in that the way of use of the display device is limited unsatisfactorily.

By displaying information about one pixel over two scanning signal lines S.sub.1 and S.sub.2 as shown in FIG. 13D, a stable gradation display can be realized even if the temperature has been changed. The aforesaid driving method will now be described in detail.

(1) A ferroelectric liquid crystal cell having a pixel in which the threshold is dispersed. The liquid crystal cell may be structured as shown in FIG. 5 in such a manner that the cell thickness in the pixel is continuously changed. Another structure disclosed in Japanese Patent Application No. 62-17186 and filed by the applicant of the present invention may be employed which is arranged in such a manner that the potential is inclined in the pixel, or another structure may be employed in which the capacity is inclined in the pixel. In either of the aforesaid methods, a region (domain) corresponding to the bright state and a region (domain) corresponding to the dark state can be present while being mixed with each other so that a gradation display can be performed by utilizing the area ratio of the domains.

Although the aforesaid method may be used in the case where the light quantity is modulated in a stepped manner (for example, 16 gradations), the light quantity must be changed continuously in order to, in an analog manner, display an image of a type having gradation.

Although the description will be made about the area modulation method, the driving method according to the present invention can be adapted to a device having a pixel, the transmissive light quantity of which can be modulated by voltage or the pulse width or the like. That is, the device must have a threshold distribution which causes the continuous light quantity change to take place. An example of the device is described in Example 7.

(2) Two scanning lines are simultaneously selected. The operation required to select the two scanning lines will now be described with reference to FIGS. 14A and 14B. FIG. 14A is a graph which illustrates the characteristics between the transmissivity and applied voltages realized when pixels on the two scanning signal lines are collected. In FIG. 14A, a portion in which the transmissivity is 0% to 100% is made to be a display region of pixel B on the scanning line 2, while a portion in which the transmissivity is 100% to 200% is made to be a display region of pixel A on the scanning line 1. That is, one pixel is constituted for each scanning signal line. Therefore, a transmissivity of 200%, in which both of the pixels A and B are brought to a complete light transmissive state, is realized when the two scanning signal lines are simultaneously scanned. In this embodiment, the two scanning signal lines are simultaneously selected with respect to one gradation information item in such a manner that a region having an area corresponding to one pixel is allocated to display one gradation information item. This arrangement will now be described with reference to FIG. 14B.

Supplied gradation information is, at temperature T.sub.1, written in a range which corresponds to 0% when the applied voltage is V.sub.0 and is written in a range which corresponds to 100% when the applied voltage is V.sub.100. As can be seen from FIG. 14B, all of the aforesaid ranges (pixel regions) are present on the scanning signal line 2 at temperature T.sub.1 (see a diagonal line portion of FIG. 14B). However, since the threshold voltage of the liquid crystal is lowered when the temperature has been raised from T.sub.1 to T.sub.2, a region larger than the region corresponding to the temperature T.sub.1 is undesirably inverted in the pixel in the case where the same voltage is applied to the pixel.

In order to correct this, a pixel region corresponding to the temperature T.sub.2 is set to spread over the scanning signal line 1 and the scanning signal line 2 (a diagonal line portion of FIG. 14B corresponding to the temperature T.sub.2). The principle to display the pixel region to spread over the two scanning signal lines will be described later.

When the temperature has been further raised to T.sub.3, the applied voltage is changed from V.sub.0 to V.sub.100 so as to set the pixel region to be drawn on only the scanning signal line 1 (a diagonal line portion of FIG. 14B corresponding to the temperature T.sub.3).

By setting the pixel regions, which form an image having a gradation, on the two scanning signal lines depending upon the temperature while shifting the pixel regions, an image having a gradation can be correctly performed in the temperature range from T.sub.1 to T.sub.3.

(3) The scanning signals to be supplied to the two scanning signal lines, which have been selected simultaneously, are made to be different from each other. In order to compensate the threshold change at the time of the inversion of the liquid crystal due to the temperature change by simultaneously selecting the two scanning signal lines, the scanning signals to be supplied to the two selected scanning signal lines must be made different from each other. The fact will now be described with reference to FIGS. 13A to 13D.

The scanning signals to be supplied to the scanning signal lines 1 and 2 are set in such a manner that the threshold of the pixel B on the scanning signal line 2 and that of the pixel A on the scanning signal line 1 are continuously changed. Referring to FIG. 13B, the transmittance-voltage curve at the temperature T.sub.1 is displayed by the region of the scanning signal line 2 when the transmittance is 100% or less, while the same is displayed by the region on the scanning signal line 1 when the transmittance is 200% or less. As described above, the transmittance-voltage curve must be continuously changed from the pixel B to the pixel A at the same gradient.

Therefore, if the shape of the cell for the pixel A on the scanning signal line 1 and that for the pixel B on the scanning signal line 2 (refer to FIG. 15B) are made to be the same, a display substantially the same as that realized when a continuous threshold characteristics are given to the pixels A and B (the cell shown in FIG. 13B) can be performed.

Then, a method for causing the thresholds of the pixels A and B to be continuously changed by utilizing the change of the thickness of the cell as shown in FIG. 5 will now be described.

In the case where the thickness of the cell in one pixel is changed from d.sub.1 (the thinnest portion) to d.sub.2 (the thickest portion), an image having a gradation can be displayed by making the width of the voltage pulse applied to the pixel B to be .DELTA.T.sub.B and making the width of the voltage pulse applied to the pixel A to be .DELTA.T.sub.A (<.DELTA.T.sub.B) and by making the voltages of the voltage pulses applied to the pixels A and B to be the same. The aforesaid process in which the voltages are made to be the same and the pulse width are made to be different as described above can be performed because the voltage supplied to the pixel is determined by the potential difference between the scanning signal line and the information signal line.

When the aforesaid voltage is raised gradually, the area of the inverted region due to switching is increased from the portion d.sub.1 (the thinnest portion) to the portion d.sub.2 (the thickest portion). The switching operation in the pixel A can be inhibited by setting .DELTA.T.sub.A to be an adequate value which is smaller than .DELTA.T.sub.B.

After the inversion region due to switching has been widened to the portion d.sub.2 (the thickest portion) of the pixel B by further raising the voltage, the aforesaid .DELTA.T.sub.A can be set so as to cause switching to be commenced in the pixel A. As a result of the aforesaid setting, the inversion region is widened to the portion d.sub.2 (the thickest portion) of the pixel A when the voltage is further raised.

As can be understood from the above made description, the continuity of the thresholds enabling the pixel A to start switching when the pixel. B has been switched can be realized by adequately setting .DELTA.T.sub.A and .DELTA.T.sub.B.

A method of determining .DELTA.T.sub.A and .DELTA.T.sub.B enabling the aforesaid continuity of the thresholds to be realized will now be described with reference to FIG. 16.

FIG. 16 is a graph which illustrates the relationship between the voltage pulses to be applied to a pixel of the ferroelectric liquid crystal device structure as shown in FIG. 5 and the voltage, where the axis of ordinate stands for the logarithm of the pulse width and the axis of abscissa stands for the logarithm of the voltage so as to show the conditions which enable the portion having the cell thickness d.sub.1 (the thinnest portion) to be switched.

Referring to FIG. 16, switching of the ferroelectric liquid crystal takes place when a voltage pulse indicated by an arbitrary point positioned to the right of segment PQ (pulse width-voltage curve) at the temperature T.sub.1 is applied to the pixel. However, the voltage pulse indicated by a point positioned to the left of a straight line PQ does not cause switching to take place.

When the voltage is gradually raised while fixing the pulse width to .DELTA.T.sub.B on the aforesaid graph, the portion of the pixel B having the cell thickness of d.sub.1 is switched at the voltage V.sub.1 (under condition of point R). With the rise of the voltage, the inversion region due to switching is gradually expanded, and the portion having the cell thickness of d.sub.2 of the pixel B is switched when the voltage has been raised to V.sub.2 (under condition of point S). It is preferable to make the pulse width to be .DELTA.T.sub.A (under condition of point T) to be applied to the pixel A so as to cause the portion of the pixel A having the cell thickness of d.sub.1 to be switched first. When the voltage is raised to V.sub.3 (under condition of point U), the inversion region is expanded to the portion of the pixel A having the cell thickness of d.sub.2.

It should be noted that both of V.sub.2 /V.sub.1 and V.sub.3 /V.sub.2 depend upon the shape of the cell (the distribution of the cell thickness). As a result of the aforesaid characteristics and a fact that the transmittance of the pixel is in proportion to the area of the inversion region, the transmittance-voltage curve of the pixel A and that of the pixel B hold a relationship which are mutually translated in parallel on the graph in which the voltage axis is indicated by the logarithm. That is, the transmittance-voltage curve as shown in FIG. 13B is obtained.

The pulse width-voltage curve shown in FIG. 16 indicates the characteristics of the material of the liquid crystal, the pulse width-voltage curve being translated in parallel depending upon the temperature in a graph in which a straight line P'Q' is shown. Assuming that straight line PQ indicates the characteristics realized at temperature T.sub.1 and straight line P'Q' indicates the characteristics realized at temperature T.sub.2, a relationship T.sub.1 <T.sub.2 is held.

In the case where an image having gradation is displayed, voltage ranged from V.sub.1 to V.sub.2 is, in accordance with gradation information, applied to a panel, the lowest temperature of which is T.sub.1. That is, V.sub.1 is the voltage corresponding to the case where information is written by 0% and V.sub.2 is the voltage corresponding to the case where information is written by 100%.

In the case where V.sub.OP (V.sub.1 <V.sub.OP <V.sub.2) is applied to the scanning signal lines 1 and 2, a required gradation level is written on the scanning signal line 2 by the pulse having the pulse width .DELTA.T.sub.B in the portion of the panel, the temperature of which is T.sub.1. However, overwriting on the scanning signal line 2 takes place because the portion of the panel, the temperature of which is T.sub.2, is switched at low voltage as can be understood from FIG. 16 Another problem takes place in that information is written on the overall portion on the scanning signal line 2. However, a writing method which enables an image having gradation to be displayed in a substantially correct manner in which the writing region is shifted from the scanning signal line 2 to the scanning signal line 1 by writing information on the scanning signal line 1 in response to the pulse having the width .DELTA.T.sub.A to correct the overwritten portion on the scanning signal line 2.

Then, the state in which the pixel is turned on/off in the aforesaid writing operation will now be described with reference to FIGS. 17A.about.17C and 18.

FIG. 17A illustrates an example of the structure of electrodes of a liquid crystal cell which can be operated in the matrix manner, where symbols S.sub.1, S.sub.2, . . . represent scanning signal lines and I.sub.1, I.sub.2, . . . represent information signal lines.

FIG. 17B is an enlarged view which illustrates the pixels A and B.

FIG. 17C illustrates an example of a signal to be written on the pixels A and B.

FIG. 18 illustrates a process of writing on the pixels A and B in an order of �1!.fwdarw.�2!.fwdarw.�3! at the temperatures T.sub.1, T.sub.2 and T.sub.3 (T.sub.1 <T.sub.2 <T.sub.3).

The operation of writing information on the pixel while simultaneously scanning S.sub.1 and S.sub.2 shown in FIG. 17 will now be described.

First, writing information on the pixel at the temperature T.sub.1 will now be described.

�1! The pixel B is deleted by pulse P.sub.1 shown in FIG. 17C (the dark state is realized).

�2! Information is written to the pixel A and B by pulses P.sub.3 and P.sub.2, respectively (a 70% bright state according to this example). However, the pixel A is not changed at temperature T.sub.1 because the voltage of the pulse P.sub.3 is lower than the threshold with respect to the threshold.

�3! A correction signal is supplied to the pixel B by the pulse P.sub.4 (the pulse P.sub.4 has a similar function as that of the pulse (c) used in the 4-pulse method shown in FIG. 12). However, the pixel B is not changed from the previous stage �2! at temperature T.sub.1 (the 70% bright state is maintained).

As described above, an image gradation can be correctly displayed (the 70% bright state) at temperature T.sub.1.

Then, an operation of writing information on the pixel at temperature T.sub.2 will now be described.

In the state where the temperature is T.sub.2, also the pixel A on the scanning signal line S1 is in the state where its thresholds being changed.

�1! The pixel B is deleted (it is brought into the dark state).

�2! Information is written on the pixels A and B by the pulses P.sub.3 and P.sub.2. The pixel B is completely written at temperature T.sub.2 (the pixel B is brought to the complete bright state). Also a portion (a bright portion) is formed in the pixel A, to which information is written, in accordance with the relationship between the pulse and the threshold.

�3! A correction pulse P.sub.4 is applied to the pixel B. A portion of the pixel B on the scanning signal line S.sub.2 is deleted by a degree corresponding to the drop of the threshold due to the temperature change. The deleted portion is used for the next line writing.

Observing the pixels A and B (FIG. 18 �3! at temperature T.sub.2), it can be understood that portions 1, 2 and 3 for indicating gradation information are present on the two scanning signal lines S.sub.1 and s.sub.2.

The portion 1 is a portion which indicates a portion of gradation information corresponding to the scanning signal line (S.sub.1) in front of the scanning signal line S.sub.2.

The portion 2 is a portion which indicates a portion (the 70% bright state similarly to temperature T.sub.1).

The portion 3 is a portion on the scanning signal line ensuing the scanning signal line S.sub.2 in which information is (or has been) written.

Then, an operation of writing information on the pixel the temperature of which is T.sub.3 will now be described.

�1! The pixel B is deleted (is brought to the dark state).

�2! Information is written on the pixels A and B by the pulses P.sub.3 and P.sub.2.

�3! The correction signal pulse P.sub.4 is supplied to the pixel B.

All of gradation information to be written to the pixel B on the scanning signal line S.sub.2 is shifted to the pixel A on the scanning signal line S.sub.1 at temperature T3. Also in this case, the gradation display has, of course, been brought to the 70% bright state.

As a result of the aforesaid principle, an image gradation can be displayed while compensating the threshold change taken place due to the temperature change. Furthermore, the polarity of the pulses of the aforesaid scanning signals can be inverted in such a manner that the adjacent scanning signal lines have opposite polarities.

Then, a method of driving the scanning signal lines for causing the adjacent scanning signals have opposite polarities will now be described.

First, a method of compensating the threshold change will now be described briefly with reference to FIG. 26A and 26B. Assumptions are made here that the transmittance when one pixel is completely bright (white) is 100% and that when the one pixel is completely dark (black) is 0%.

FIG. 26A is a graph in which two pixels A and B are used, and the threshold characteristics with respect to information voltage V are continuously illustrated. As a result, the writing region with information voltage V.sub.i (V.sub.th <V.sub.i <V.sub.sat) is not saturated as shown in FIG. 13B even if the reference threshold characteristics .alpha. has been changed to .beta. or .gamma. due to the temperature change or the like. Hence, the region to which information can be written at V.sub.sat but to which information cannot be written at V.sub.th is translated from the pixel B to the pixel A. That is, possession of a display region corresponding to one information signal over a plurality of pixels having the continued threshold characteristics will compensate the dispersion of the threshold characteristics.

Then, this method will now be described in detail.

(1) A ferroelectric liquid crystal cell having the threshold which is continuously changed in the pixel thereof is prepared. The structure as shown in FIG. 5 may be employed in which the thickness of the cell is continuously changed in the pixel. As an alternative to this, a structure may be employed in which the potential is inclined in the pixel, or another structure may be employed in which the capacity is inclined.