JP2006047691A - Current output type semiconductor circuit and driving method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a current output type semiconductor circuit and a driving method thereof which prevent gradation inversion such as gradation inversion due to deviation in channel size ratio and gradation inversion between adjacent gradations due to variance of transistors by increasing a current value corresponding to the most significant bit by using a raising current source for a driver IC comprises a current source and using different channel sizes of transistors between a low-gradation part and a high-gradation part. <P>SOLUTION: The current output type semiconductor circuit has a function of outputting a current corresponding to a desired gradation to a display panel and a function of outputting a voltage from one terminal is equipped with a voltage generation part comprising a temperature compensating element and at least one resistance element to output the voltage. A precharge voltage generation part has its precharge voltage made variable with temperature when generated by using resistance division and a temperature compensating element such as a thermistor or has its precharge voltage raised with temperature by providing a temperature detecting means on a set side or array module side, inputting its detection result to a controller, and varying a resistor for precharge voltage generation in a source driver from the controller according to variation in temperature, thereby making constant a current flowing to an EL regardless of temperature. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a current output type semiconductor circuit that outputs a current used in a display device that performs gradation display by an amount of current, such as an organic electroluminescent element, and a driving method thereof.

  Since the organic light emitting element is a self light emitting element, a backlight required for a liquid crystal display device is unnecessary, and it is expected as a next generation display device from the advantages such as a wide viewing angle.

  A cross-sectional view of a device structure of a general organic light emitting device is shown in FIG. The organic layer 42 is sandwiched between the cathode 41 and the anode 43. When a DC power supply 44 is connected to this, holes are injected from the anode 43 and electrons are injected from the cathode 41 into the organic layer 42. The injected holes and electrons move in the organic layer 42 to the counter electrode by the electric field formed by the power supply 44. In the middle of movement, electrons and holes recombine in the organic layer 42 to generate excitons. Light emission is observed in the process of exciton energy deactivation. The emission color differs depending on the energy of the exciton, and becomes light having a wavelength of energy corresponding to the value of the energy band gap of the organic layer 42.

  In order to extract light generated in the organic layer to the outside, at least one of the electrodes is made of a transparent material in the visible light region. For the cathode, a material having a low work function is used to facilitate injection of electrons into the organic layer. For example, aluminum, magnesium, calcium and the like. These alloys and materials such as aluminum lithium alloys may be used for durability and further lower work function.

  On the other hand, an anode having a high ionization potential is used because of the ease of hole injection. Further, since the cathode does not have transparency, a transparent material is often used for this electrode. Therefore, generally, ITO (Indium Tin Oxide), gold, indium zinc oxide (IZO), or the like is used.

  In recent years, in an organic light emitting device using a low molecular material, the organic layer 42 may be composed of a plurality of layers in order to increase the light emission efficiency. This makes it possible to share the functions of carrier injection, carrier movement to the light emitting region, and light emission with a desired wavelength in each layer. By using efficient materials for each layer, higher efficiency organic A light emitting element can be formed.

  The organic light emitting device thus formed has a luminance proportional to the current as shown in FIG. 5A and a non-linear relationship with the voltage as shown in FIG. 5B. Therefore, in order to perform gradation control, it is better to perform control based on the current value.

  In the case of the active matrix type, there are two types, a voltage driving method and a current driving method.

  The voltage driving method uses a voltage output type source driver, converts a voltage into a current inside a pixel, and supplies the converted current to an organic light emitting element.

  In this method, since voltage-current conversion is performed by a transistor provided for each pixel, there is a problem in that output current varies depending on variation in characteristics of the transistor, and luminance unevenness occurs.

  The current driving method uses a current output type source driver, has only a function of holding the current value output for one horizontal scanning period inside the pixel, and supplies the same current value as the source driver to the organic light emitting element. .

  An example of the current driving method is shown in FIG. The system in FIG. 6 uses a current copier system for the pixel circuit.

  FIG. 7 shows a circuit during operation of the pixel 67 of FIG.

  When a pixel is selected, as shown in FIG. 7A, a signal is output from the gate driver 35 so that the gate signal line 61a in that row is in a conductive state, and 61b is in a non-conductive state. . The state of the pixel circuit at this time is shown in FIG. At this time, a current flowing through the source signal line 60 that is a current drawn into the source driver 36 flows through a path indicated by a dotted line 71. Therefore, the same current as the current flowing through the source signal line 60 flows through the transistor 62. Then, the potential of the node 72 becomes a potential corresponding to the current-voltage characteristics of the transistor 62.

  Next, in a non-selected state, a circuit as shown in FIG. A current flows from the EL power source line 64 to the organic light emitting element 63 along a dotted line path 73. This current is determined by the potential of the node 72 and the current-voltage characteristics of the transistor 62.

  7A and 7B, the potential of the node 72 does not change. Accordingly, the drain current flowing in the same transistor 62 is the same in FIGS. 7A and 7B. As a result, a current having the same value as the current flowing through the source signal line 60 flows through the organic light emitting element 63. Even if there are variations in the current-voltage characteristics of the transistor 62, the values of the currents 71 and 73 are not affected in principle, and a uniform display can be realized without being affected by variations in the characteristics of the transistors.

  Therefore, in order to obtain a uniform display, it is necessary to use a current driving method. For this purpose, the source driver 36 must be a current output type driver IC.

  An example of the output stage of the current driver IC that outputs a current value corresponding to the gradation is shown in FIG. An analog current output is performed from 104 to the display gradation data 54 by the digital-analog converter 106. The analog-to-digital conversion unit defines a plurality of (at least the number of bits of gradation data 54) gradation display current sources 103, switches 108, and a current value flowing through each gradation display current source 103. The gate line 107 is configured.

  In FIG. 10, an analog current is output to the 3-bit input 105. For example, in the case of data 1, the current source 103 corresponds to one current, and in the case of data 7, the number of current sources 103 corresponding to the bit weight is connected to the current output 104. A current corresponding to the gradation can be output, such as seven currents. A current output type driver can be realized by arranging 106 of this configuration in a number corresponding to the number of outputs of the driver. In order to compensate for the temperature characteristics of the transistor 103, the voltage of the common gate line 107 is determined by the distributing mirror transistor 102. The transistor 102 and the current source group 103 have a current mirror configuration, and the current per gradation is determined according to the value of the reference current 89. With this configuration, the output current varies depending on the gradation, and the current per gradation is determined by the reference current.

Examples of display devices using organic light emitting elements are shown in FIGS. FIG. 21 shows a television, FIG. 22 shows a digital camera or digital video camera, and FIG. 23 shows a portable information terminal. An organic light emitting element is a display panel suitable for these display devices having a high response speed and a high chance of displaying moving images.
JP 2001-147659 A

  In recent years, the number of colors has been increasing in portable information terminals, and display of 65,000 colors or 220,000 colors has become mainstream. When the input signal of the driver IC is an RGB digital interface, 16 bits or 18 bits are required. Therefore, the number of input signal lines is 16 to 18 and only necessary for data transfer. In addition, a signal line is necessary for the operation signal of the shift register and the setting of various registers.

  Therefore, the number of wirings increases, and for example, as shown in FIG. 3, the wiring between the control IC 31 and the source driver IC 36 increases with respect to the display panel 33. For this reason, there is a problem that the cost increases due to an increase in the size of the flexible substrate 32 or the use of a multilayer substrate.

The present invention has been made in view of the above problems, and the first aspect of the present invention is
A current output type semiconductor circuit having a function of outputting a current corresponding to a desired gradation to a display panel and a function of outputting a voltage from the same terminal,
In order to output the voltage, the current output type semiconductor circuit includes a voltage generation unit including a temperature compensation element and at least one resistance element.

The second aspect of the present invention
When a current corresponding to a desired gradation is output to the display panel, a function of outputting a voltage from the same terminal,
A current output type semiconductor circuit having an electronic volume circuit for determining the voltage output value,
A method for driving a current output type semiconductor circuit, wherein the set value of the electronic volume circuit is changed according to temperature.

  By using the invention as described above, in a driver IC composed of current sources using different transistor channel sizes in the low gradation portion and the high gradation portion, gradation inversion due to a difference in channel size ratio, etc., and transistor Gradation inversion between adjacent gradations due to variations was prevented by increasing the current value corresponding to the most significant bit using a current source for raising. In addition, it is possible to select not to connect a current source for raising for each terminal by laser processing or the like, thereby reducing defects due to gradation inversion and increasing the yield.

  Embodiments of the present invention will be described below with reference to the drawings.

  The configuration of the current output type source driver IC 36 in the present invention is shown in FIG. The number of outputs is simply the number of shift registers 21 and latches 22 required per output, the current output stage 23, the precharge voltage application determination unit 56, and the current output / precharge voltage selection unit 25 according to the increase or decrease in the number of outputs. Because it can be realized by increasing or decreasing, it is possible to cope with any number of outputs (however, as the number of outputs increases, the chip size becomes too large and the versatility is lost, so about 600 is the maximum in practical use. is there).

  The video signal of the driver IC 36 of the present invention is input from the control IC 28 through the signal lines 12 and 13. The distribution unit 27 distributes the video signal and various setting signals, and inputs only the video signal to the shift register unit 21. The output is distributed to each output terminal by the shift register unit 21 and the two latch units 22. The distributed video signal is input to the current output stage 23. The current output stage 23 outputs a current value corresponding to the gradation from the video signal and the reference current generated by the reference current generator 26. The precharge determination signal data in the latch unit is input to the precharge voltage application determination unit 56. On the other hand, in the precharge voltage application determination unit 56, a switch for determining whether or not to output the voltage supplied from the precharge power supply 24 to the output 53 is controlled by the precharge determination signal latched by the latch unit 22 and the precharge pulse 52. To generate a signal. Thus, a current output / precharge for selecting whether to supply a current corresponding to the gradation to the outside of the driver IC 36 or to supply a voltage supplied from the precharge power supply 24 in accordance with the output signal of the precharge voltage application determination unit 56. A current or voltage is output to the outside of the driver IC 36 via the voltage selection unit 25.

  The voltage output from the precharge power supply 24 has a voltage value necessary for displaying black on the display panel. This method of applying the precharge voltage is a configuration peculiar to the driver IC 36 for performing gradation display according to the current output in the active matrix display device.

  For example, consider a case where a predetermined current value is written from a source signal line to a certain pixel in an active matrix display device having a pixel configuration shown in FIG. When precharge is not performed, that is, when there is no precharge circuit, a circuit extracted from the circuit related to the current path from the output stage of the source driver IC 36 to the pixel is as shown in FIG.

  A current I corresponding to the gradation flows from the driver IC 36 as a drawn current in the form of a current source 122. This current is taken into the pixel 67 through the source signal line 60. The captured current flows through the driving transistor 62. That is, the current I flows from the EL power supply line 64 to the source driver IC 36 through the drive transistor 62 and the source signal line 60 in the selected pixel 67.

  When the video signal changes and the current value of the current source 122 changes, the current flowing through the drive transistor 62 and the source signal line 60 also changes. At that time, the voltage of the source signal line changes according to the current-voltage characteristics of the drive transistor 62. When the current-voltage characteristics of the drive transistor 62 are as shown in FIG. 12B, for example, if the current value flowing through the current source 122 changes from I2 to I1, the voltage of the source signal line changes from V2 to V1. . This change in voltage is caused by the current from the current source 122.

  A floating capacitance 121 exists in the source signal line 60. In order to change the source signal line voltage from V2 to V1, it is necessary to extract the charge of the stray capacitance. The time ΔT required for the extraction is ΔQ (charge of stray capacitance) = I (current flowing through the source signal line) × ΔT = C (stray capacitance value) × ΔV. If ΔV (signal line amplitude from white display to black display time) is 5 [V], C = 10 pF, and I = 10 nA, ΔT = 50 milliseconds is required. This is shorter than one horizontal scanning period (75 μs) when driving the QCIF + size (number of pixels: 176 × 220) at a frame frequency of 60 Hz. For this reason, black display is performed on the pixel below the white display pixel. If this is the case, the switch transistors 66a and 66b for writing the current to the pixel are closed while the source signal line current is changing, so that the halftone is stored in the pixel, so that the pixel shines at a luminance between white and black. It means to end up.

  Since the value of I becomes smaller as the gradation becomes lower, it becomes difficult to extract the charge of the stray capacitance 121, and the problem that the signal before changing to the predetermined luminance is written inside the pixel becomes more prominent as the low gradation display. Appear in Extremely speaking, when black is displayed, the current of the current source 122 is 0, and it is impossible to extract the charge of the stray capacitance 121 without passing the current.

  Therefore, a voltage source having a lower impedance than that of the current source 122 is prepared and applied to the source signal line 60 as necessary. This voltage source corresponds to the precharge power source 24 of FIG.

  A schematic circuit for one source signal line 60 is shown in FIG. By applying the voltage supplied from the voltage generator 24 to the source signal line 60, the charge of the stray capacitance 121 can be charged and discharged. The voltage supplied from the voltage generator 24 may be able to supply a voltage corresponding to each gradation current in accordance with the characteristics of FIG. Since the conversion unit is required, the circuit scale is increased, and the driving transistor 62 has a characteristic variation for each pixel. Therefore, the corresponding voltage is different for the same gradation current. For this reason, even if a digital / analog converter is provided and a voltage corresponding to the gradation is output, the predetermined current is not written, and the current source 106 needs to correct the predetermined current thereafter. For this reason, in practical terms, it is sufficient in terms of cost (chip area) effectiveness to generate only the voltage corresponding to the black gradation for which the current value is most difficult to be written. It can be said.

  Therefore, the voltage generated from the voltage generator 24 may be one, and it is only necessary to determine whether to output a voltage based on the data and to control the switch 131. That is, a 1-bit signal line (precharge determination signal) for determining whether or not to apply the voltage source 24 is prepared before outputting a current corresponding to a certain video signal.

  FIG. 15 shows the operation timing of the switches 251 and 252 within one horizontal scanning period. At the beginning of the horizontal scanning period, a precharge voltage from the voltage generator 24 is applied to reset the charge of the stray capacitance 121. (Period 151) Since the charge is reset by the voltage, even if this period is short, it may be about 2 μsec at the maximum to achieve the purpose. Next, in the period 152, only the switch 132 is turned on, and a current corresponding to the gradation is supplied to the pixel 67. Note that the operation for writing a predetermined current value during the period 152 is slow, so the period 152 needs to be as long as possible. The period 151 needs to be about 10% of the maximum one horizontal scanning period.

  Since it is necessary to control the output period of the voltage generator 24 as described above, the precharge pulse 52 indicating the precharge application period 151 is input, and the switch 131 is controlled together with the precharge determination signal. For this reason, the application determination part 56 is provided.

  Since the voltage value output from the voltage generation unit 24 is only a voltage corresponding to the current at the time of black gradation (hereinafter referred to as black voltage), for example, white is output over a plurality of horizontal scanning periods in which gradation data 54 is continuous. When the gray scale is displayed, the source signal line repeats the black, white, black, and white states. If precharging is not performed, white states will occur continuously. In other words, precharge causes the change of the signal line to become intense, and depending on the current at the time of white display, white may not be completely generated and there is a possibility that the write current is insufficient.

  Therefore, by using the precharge determination signal, precharge is not performed in a gradation in which a relatively large amount of current flows, and only the gradation that is difficult to change to a predetermined current near the black gradation is assisted by the precharge power supply 24. That's fine. For example, there is a period in which the precharge voltage is applied only at the gradation 0 (black), and it is most effective to prevent the precharge voltage from being applied at other gradation display. By reducing the luminance at the lowest gradation, the contrast is increased and a more beautiful picture can be displayed.

  For example, as shown in FIG. 17A, precharging can be performed only at the gradation 0 by setting the precharge determination signal 55 only when the gradation data 54 is 0.

  If the precharge determination signal 55 is generated when the gradation data 54 is 0 or 1, precharge can be performed when the gradation data is 0 or 1 (FIG. 17B).

  By the way, in a pattern in which the source signal line does not change such that the entire screen is black, if a precharge voltage is applied only at the beginning of one frame, a predetermined gradation flows sufficiently even with only the black current thereafter.

  That is, even when the same black is displayed, the time required for the current to flow to the predetermined current value differs depending on the current value passed through the source signal line in the previous horizontal scanning period. The larger the amount of change, the longer the change takes. For example, although it takes time to display black after white display, in the case of performing black display after black display, the signal line changes only for the variation of the drive transistor 62, so the time required for the change is short.

  Therefore, in synchronization with the gradation data 54, a signal (precharge determination signal 55) for determining whether or not to apply a precharge voltage is introduced for each color, so that an arbitrary gradation or the same gradation can be obtained. It is also possible to introduce a configuration that allows selection with or without precharge.

  A precharge determination signal 55 is added to the gradation data 54. Along with this, the latch unit 22 also needs to latch the precharge determination signal, so that it has a latch unit of the number of video signal bits + 1 bit.

  In FIG. 17C, when the precharge is applied when the gradation is 0 and the gradation in the previous period is not 0 (the precharge is applied when the gradation is 0, but the gradation is 0 when continuous. But precharge is not performed).

  Unlike the previous method, this method has an advantage that it can be selected whether or not to precharge depending on the state of the source signal line before one horizontal scanning period even at the same gradation.

  The precharge determination signal is supplied from the control IC 28. By the command operation of the control IC 28, the pattern of the precharge determination signal 55 can be changed and output as shown in FIGS.

  According to the capacity of the source signal line and the length of one horizontal scanning period, the precharge setting can be flexibly changed from the outside of the source driver IC 36, and there is an advantage that versatility is increased.

  In organic light-emitting elements used as display elements, the element configuration differs for each emission color, and the carrier injection efficiency, carrier mobility, phosphor quantum efficiency, etc. differ, so the emission start current can be different for each emission color. There is sex. Examples are shown from 141 to 143 in FIG. Green emits light when an electric current of I1 or higher, blue of I2 or higher, and red of I3 or higher flows. Then, even if it is assumed that there is no variation in the driving transistor 62 of the pixel 67, the black voltage differs from V1 to V3 for each display color as shown in FIG. Since it takes time to change to the predetermined current as the current becomes lower, the voltage V1 is applied to all the elements when setting the precharge voltage with one power source. In this way, there is no black float that dimly shines at the time of black display. However, when white is next displayed, the red display pixel has a voltage of (V3-V1) as compared with the case where there is no precharge. It is necessary to change more. Therefore, when white display is performed next, there arises a problem that it is difficult to change to white as the voltage change increases.

  Therefore, a precharge power supply 24 is provided separately for each display color. A block diagram is shown in FIG. Here, description will be made assuming that R is red, G is green, and B is output to a blue light-emitting element (in addition to the three primary colors of red, green, and blue, it may be three colors of cyan, yellow, and magenta).

  Three outputs of the voltage generator 24 are provided, the output 161 outputs to the R source signal line, 162 outputs to G, and 163 outputs to B. At this time, the output voltage 161 is set so as to output a voltage substantially equal to the voltage of the source signal line 60 when the driving transistor 62 of the pixel 67 passes the current I3. 162 and 163 may output values substantially equal to the voltage of the source signal line 60 when the currents I1 and I2 flow to the pixel transistor 62, respectively. Thereby, an appropriate voltage value for each display color can be directly applied to the pixel.

  Accordingly, since the source signal line potential to be changed at the time of current output can be reduced, the current can be changed to a predetermined current value in a shorter time, and the writing is less likely to be insufficient.

  FIG. 8 is a diagram showing a reference current generating circuit. The reference current defines a current value per gradation in the configuration of the output stage shown in FIG.

  In FIG. 8, the reference current 89 is determined by the potential of the node 80 and the resistance value of the resistance element 81.

  Further, the potential of the node 80 can be changed by the control data 88 by the voltage adjusting unit 85.

  Depending on the transistor size of the gradation display current source 103 for performing current output, output current variation occurs for each terminal. FIG. 11 shows the relationship between transistor size (channel area) and output current variation. Taking into account the variation of the reference current, the variation between adjacent terminals within the chip and between the chips needs to be within 1%. Therefore, the variation in the output current (current variation in the output stage) in FIG. The transistor size of 103 is preferably 30 square microns or more.

  In order to suppress the variation in the ratio of the reference current to a certain gradation output current between chips, the distribution mirror transistor 102 and the gradation display current source 103 can be designed with the same size and the same layout. desirable. The above area ratio is preferably realized by increasing or decreasing the number of transistors. As a result, even in a display device using a plurality of driver ICs 36 arranged side by side, the variation between the chips in the ratio of the output current to the reference current is reduced, so that a display without block unevenness can be realized.

  In the above method, the resistance element 81 is often formed of a component external to the driver IC 36 in the reference current generating unit for generating the reference current. This is because if the value of the resistance element 81 varies, the reference current 89 varies, and therefore, a current per gradation that differs from chip to chip is output. Therefore, in order to suppress variations as much as possible, chip resistors with small variations are often used.

  However, in order to reduce the number of mounted components and simplify wiring on the array, it is necessary to incorporate a resistance element. The present invention devised a configuration in which the variation of the reference current 89 is reduced even when the resistance element 81 is incorporated.

  FIG. 9 shows the configuration of the reference current generator when the resistor 81 is built in the driver IC 36, and FIG. 19 shows the relationship of external wiring when two driver ICs 36 are used.

  The resistance element 81 is divided into two parts (11a, 11b).

  By devising the connection between the two resistance elements, the variation in the reference current 89 between different chips can be reduced.

  When the two driver ICs 36 are in contact with each other, the configuration of the current source is the configuration of the two current sources shown in FIG. One of the two required resistance elements 81 is taken in from a different IC 36 by the external wiring 92.

  One resistive element 81 is brought from both adjacent ICs 36. The reference current 89a of the driver IC 36a is determined from the sum of the resistors 81c and 81b, and the reference current 89b of the driver IC 36b is determined from the sum of the resistors 81a and 81d. As the voltages 80a and 80b, as shown in FIG. 8, a voltage obtained by dividing the reference voltage 86 by the resistor 84 is supplied. If the reference voltage 86 is commonly input to the driver IC 36, there is no variation, and further, since the divided voltage is determined by the resistance division ratio of 84, the variation between chips can be reduced, so the variation of the node 80 is small.

  Accordingly, the deviation between the reference currents 89 a and 89 b is caused by the deviation of the resistance element 81. The resistance values of the resistance elements 81a to 81d are Ra, Rb, Rc, and Rd, and the voltage applied to both ends of the resistance is Vd.

  The current 89a is Vd / (Rc + Rb), and the current 89b is Vd / (Ra + Rd).

  In order to create a resistor inside the IC 36, there are a diffused resistor and a polysilicon resistor. In order to create a resistor with less variation, it is better to use a polysilicon resistor, and when including chips and lots, the variation is about 5%. However, when two resistance elements 81 are formed close to each other in the same chip, the variation in resistance value is about 0.1%. Therefore, the variation between resistance elements 81c and 81d (Rc and Rd) and between 11a and 11b (Ra and Rb) shown in FIG. 19 is suppressed to 0.1%. Therefore, the variation between (Rc + Rb) and (Ra + Rd), which causes variation between 89a and 89b, is 0.14%, which is the mean square of 0.1.

  In this way, by taking the resistances that determine the current value from two adjacent chips, it becomes irrelevant to the variation between chips and between lots, and even a polysilicon resistance having a variation of about 5% can be used. Therefore, it is possible to realize a driver IC 36 that does not have built-in resistors and block unevenness.

  As described above, when the constant current source having the configuration shown in FIG. 9 is used, it is possible to reduce the number of mounted components, which is advantageous because the cost can be reduced.

  Now, in a display panel using an organic light emitting element, current flows only to the lit pixel, and no current flows to the non-lit pixel. Therefore, the maximum current flows when the full screen is white and the minimum current flows when the full screen is black.

  The power supply circuit that supplies current to the display panel needs to have a capacity that allows the maximum current to flow. However, there is very little screen display that allows the maximum current to flow. It is wasteful to provide a power supply circuit having a large capacity because of the maximum current that is generated with very few opportunities. In order to reduce power consumption, it is necessary to reduce the maximum current as much as possible.

  Therefore, as a method of reducing the maximum current, when the number of white display pixels is 60% or more, the luminance of all the pixels is reduced by about 2 to 3%. According to this, the maximum current decreases by 2 to 3%, and the power at the peak decreases.

  This method can be realized by changing the value of the reference current 89 generated from the reference current generation unit 26 that determines the current per gradation by about 2 to 3%.

  Therefore, the reference current 89 is changed by changing the value of the control data 88 and changing the voltage at the node 80 according to the display pattern.

  As described above, in order to change the value of the control data in accordance with the display pattern, it is necessary to perform a control of determining the display pattern and changing the control data according to the determination result. Therefore, this determination is normally performed by the control IC 28.

  Therefore, the number of signal lines input from the control IC 28 to the source driver IC 36 is equal to the number of control data lines of the electronic volume in addition to the video signal lines. Therefore, the input / output terminals of both ICs increase. When the electronic volume control is 6 bits and the video signal line is 18 bits (each color is 6 bits), 24 terminals are required.

  Further, when the precharge power supply 24 is built in, a register for setting the output voltage of the precharge power supply 24 exists. Since the precharge voltage is determined by the TFT characteristics of the display panel and the threshold voltage of the organic light emitting element, it is necessary to set a different voltage value for each different panel, and it is necessary to set it from the outside at least once. Providing an external input terminal for one setting is inefficient.

  Reducing the number of input / output signal lines is effective for reducing the chip area and simplifying external wiring.

  Therefore, in the present invention, the number of signal lines is reduced by connecting the data lines and address lines between the control IC and the source driver IC so that the video signal and various setting signals are serially transferred at high speed. The video signal is also serially transferred in the three source colors of red, green and blue.

  FIG. 1 shows a timing chart of data lines and address lines. After the start pulse 16 is input, pixel data for one row is transferred from the data line 12. Thereafter, control data is transferred. For example, the setting value of the electronic volume. In order to determine what data is flowing in the data line 12, the address 13 is transferred in synchronization with the data on the data line 12. In this example, when the data on the address line 13 is 0, red data is displayed, 1 is green data, and 2 is blue data. A value of 4 or more is command data.

  A block diagram of the distribution unit 27 for distributing the serially transferred data is shown in FIG. The distribution unit is composed of a two-stage register for video signals and a one-stage register or latch circuit for other command data.

  The first stage register or latch circuit 182 captures only necessary data, and for the video signal 11, the timing of the three color signals is adjusted so that the carry pulse of the next shift register unit 21 can be lengthened. . Thereby, video data 11 as shown in FIG. 1 is extracted. This data is distributed to each output by the shift register unit 21.

  As a result, the source driver IC 36 having a small number of input signal lines can be realized.

  Up to this point, an example in which the transistor used for the pixel 67 is a p-type transistor has been described, but the same can be realized by using an n-type transistor.

  FIG. 20 shows a circuit for one pixel when a current mirror type pixel configuration is formed by n-type transistors. The direction of current flow is reversed, and the power supply voltage changes accordingly. Therefore, the current flowing through the source signal line 205 needs to flow from the source driver IC 36 toward the pixel 67. The configuration of the output stage is a current mirror configuration of a p-type transistor so as to discharge current outside the driver IC. Similarly, the direction of the reference current needs to be reversed.

  Thus, the transistor used for the pixel can be applied to both p and n.

  FIG. 24 shows a current output stage 23 for outputting an output current in 256 stages for an 8-bit input. For the lower 2-bit signal line, a current source through which the current of “I” flows is prepared according to the weight of the bit, and for the upper 6-bit signal line, “4I” (“I” 4 A current source through which current flows twice) is prepared according to the weight of the bit. As a result, a current of 0 that is the lowest current flows at the gradation 0, and a current of 255I that is the maximum current flows at the gradation 255. The current differs by I per gradation.

  If the current source is composed of transistors, 255 transistors are required when the current source is composed of only the current source of “I”. On the other hand, the configuration of FIG. 24 requires three “I” current source transistors and 63 “4I” current source transistors. The “4I” transistor has a channel width approximately four times that of the “I” transistor. Therefore, in view of only the channel area of the transistor, the same area is required even when only “I” is used or when both “I” and “4I” transistors are used. However, when a transistor is formed, contact portions for gate, source, and drain electrodes are required in addition to the channel region. One of these is required for each transistor. Therefore, in the two methods in which the total transistor channel area is equal, the output stage can be formed with a smaller area in the method of FIG. 24 in which “4I” and “I” are mixed and output because the number of transistors is small.

  FIG. 25 shows an example in which the structure of FIG. 24 is realized by a transistor. For input video signal data D [7: 0], between D [0] and D [1] and between D [2] and D [7], the number of transistors connected to the output for each bit weight. The weighting between the lower 2 bits and the upper 4 bits is determined by the channel width of the transistor. The transistors 251 and 252 are designed so that the channel width of the 252 is about four times. However, since the channel width ratio and output current ratio do not match exactly, the ratio of the channel width of the transistor should be determined between 3.3 and 4.7 times based on simulation and mounting data. Thus, an output stage with higher gradation can be configured.

  Thus, by using transistors having different sizes for the lower bits and the upper bits as current sources, it is possible to form a smaller output stage by reducing the area of the contact portion by reducing the number of transistors.

  The output current is determined by the number of current source transistors connected to each bit, and the output current is changed in such a manner that the amount of current flowing through one transistor is accumulated. In the case of 8-bit output in FIGS. 24 and 25, the gradation and output current characteristics are as shown in FIG. (Note that only the lower 64 gradations are shown due to space limitations) The current shown in the region 262 is output by the upper 6-bit transistor, and the current shown in the region 261 is output by the lower 2-bit transistor. Since the current of 262 changes the current value due to the difference in the number of transistors, the variation in the step size can be reduced to 1% or less. Since most of the output current is the portion 262, even if the current in the portion 261 slightly varies, the linearity of the gradation is not affected. Further, even if the step width of 261 is increased or decreased from a predetermined value, there is only a portion where the step width is different only once every four gradations, and there is no practical problem when the ratio of the output current of 262 and 261 is taken into consideration. In the low gradation region where the current ratio of 262 is small, it is difficult to recognize the luminance difference due to the characteristics of the human eye, and the variation in the step size becomes more inconspicuous, so there is no problem.

  However, the variation needs to be 1% or less in the middle tone (near gradation 128 in the case of 8-bit display) in which the variation between adjacent areas is easy to see. For that purpose, it is sufficient that one transistor used for output is 70 square microns or more from the relationship of FIG. Since only the transistor 252 is used at the gradation 128, only the 252 needs to have an area of 70 square microns or more. In the case of the gradation 127, the transistor 252 outputs a current corresponding to the gradation 124, and the transistor 251 outputs a current corresponding to the gradation 3. Since the current from the transistor 251 is about 2% of the total, even if the current from the transistor 251 varies by about 3%, the total can be kept within 1%. If the channel area of the transistor 252 is 70 square microns, 251 passes a quarter current to 252. Therefore, if the channel length is designed to be equal, the channel width is reduced to a quarter. In this case, the area is 17 square microns. From the relationship of FIG. 11, the variation of the transistor 251 is about 2% (when 3σ is taken). Therefore, even when the gradation is 127 as a whole, the variation between adjacent regions can be kept within 1%.

  Note that when the number of gradations is increased from 128, the number of output transistors 252 is increased. Therefore, the variation is further reduced, and no vertical line is caused by the variation.

  In the current output stage 23 as shown in FIG. 24, a current source corresponding to each bit is prepared, and a current output is obtained by accumulating values of each current source according to input data. Compared to the source, there is an advantage that display is possible even if the output variation of the current source used for the lower bits is large.

  In the low gradation area where only the lower bits are output, it is difficult to observe display unevenness even if the variation is large due to human visual characteristics, and in the halftone region where the variation is most visible, the output from the upper bit side current source Since the majority of the output current occupies, the ratio of the lower bit side current source to the total output current is several percent, and even if the current source on the lower bit side varies by 3%, there is an advantage that it can be realized within 1% as a whole. .

  If the configuration shown in FIG. 25 is used in the high gradation region, the number of transistors used for output increases, and the variation is further reduced.

  From the above, in the method of configuring a current source corresponding to each bit, when current output is performed using transistors of different sizes for the upper N bits and the lower M bits, the conditions are the most severe with respect to variation. What is necessary is just to design so that the variation at the time of (halftone-1 gradation) display is 1% or less.

As for the variation at this time, when the upper N-bit current source output variation is p [%] and the variation is inversely proportional to the square root of the channel area of the transistor as shown in FIG. The variation is 2 ^{(M / 2)} × p [%], and the variation in the (halftone-1 gradation) display of the N + M bit display current output type semiconductor circuit is {(2 ^N −1) × 2 ^M × p + (2 ^M −1) × 2 ^{(M / 2)} × p} / (2 ^{(M + N)} −1).

Summarizing this equation, the equation of variation is represented by (1 + 2 ^{(M / 2−N)} ) × p. Therefore, in a current output type semiconductor circuit having an (N + M) bit output, if the value of M is such that (1 + 2 ^{(M / 2−N)} ) × p is within 1%, the current output type has no display unevenness. A semiconductor circuit can be created. The maximum value of M that can be taken at this time is the maximum value of M, and the minimum value is 1.

  Therefore, in the 8-bit output, it is possible to adopt a driver configuration of N = 7 and M = 1 as shown in FIG. 27 in addition to the driver of N = 6 and M = 2 as shown in FIG.

  In the above description, the driver is described as a monochrome output driver. However, the present invention can also be applied to a multi-color output driver. It is sufficient to prepare the same circuit for the number of display colors. For example, in the case of three-color output of red, green, and blue, three identical circuits may be placed in the same IC and used for red, green, and blue, respectively.

  In the above invention, the transistor has been described as a MOS transistor, but a MIS transistor or a bipolar transistor can be similarly applied.

  The present invention can be applied to any material such as crystalline silicon, low-temperature polysilicon, high-temperature polysilicon, amorphous silicon, and gallium arsenide compound.

  FIG. 28 is a diagram in the case where the upper half (region 281) of the display region is white display and the lower half (region 283) is low gradation display (for example, gradation 1). At this time, the scanning direction is from the top to the bottom of the drawing.

  Since the potential of the source signal line cannot be quickly changed by the stray capacitance 121 as described with reference to FIG. 12 at the boundary between the regions 281 to 283, the current output stage 54 that performs current output based on the gray scale in FIGS. In addition, a line indicated by (a) in the boundary is used by using a method of providing a precharge power supply 24 and quickly changing the source signal line potential to black by the precharge voltage during black display that takes a long time to change. When the precharge voltage 24 is output in (282), as shown in FIG. 28, only the uppermost row in the region 283 where gradation 1 is displayed has a luminance lower than that of gradation 1. There was a problem displayed.

  This indicates that since the current value is small in gradation 1, it takes time even for the voltage change from gradation 0 corresponding to gradation 0 having a small change amount to gradation 1. This phenomenon is particularly prominent in a large panel having a large source signal line capacitance.

  Therefore, as shown in FIG. 29, a precharge pattern control unit 292 is provided instead of the precharge voltage application determination unit 56.

  The precharge pattern control unit 292 changes the output based on the gradation data 54 and the synchronization signal. For example, even when the gradation 0 is input, the precharge power supply 24 is not output to the current output 104 by the frame. It was possible to do something like that.

  For example, it is possible to precharge in 2 out of 3 frames and not to precharge in 1 frame. In the display example of FIG. 28, the display between gradations 0 and 1 is performed in the precharged frame. In the frames that are not performed, the intermediate level between white and black is displayed. In this case, as in the case of the frame rate control, in the row 282, the luminance per frame is obtained by adding the luminance of 2 frames when precharging is performed and the luminance of 1 frame when not performing the precharging and dividing by 3 frames. Will be displayed.

In order to prevent flicker due to a luminance difference between when the precharge is performed and when the precharge is not performed, the black display pixels 302 with the precharge and the black display pixels 303 without the precharge are distributed in the same frame for each frame. The pattern at this time is shown in FIG.

  Further, it may be other than three frames, between two frames, or between arbitrary frames. FIG. 31 shows an example in which the presence / absence of precharge is controlled in two frames. In this case, the luminance of the black display pixel is an average of the luminance when the precharge is performed and the luminance when the precharge is not performed.

  Thereby, even in the same black display pixel, the luminance is different between FIG. 30 and FIG. By utilizing this fact, it is possible to perform display close to a predetermined luminance by changing the ratio of frames to be precharged for each display gradation.

  An example is shown in FIG. In general, the more precharge is performed, the more black it becomes, so the lower the gray level, the greater the precharge insertion ratio. For example, at gradation 0, precharge is performed for all frames, gradation 1 is performed twice out of 3 frames, and gradation 2 is performed once out of 2 frames. In this way, it is possible to obtain a luminance relationship close to the gradation characteristics depending on the number of precharges.

  To further improve the gradation, there is a method of preparing a plurality of precharge power supplies 24 as shown in FIG. The output voltage of 24a is V1, and the output voltage of 24b is V3. (Here, V1> V3) If two types of power supplies are prepared, there are three types of cases where only V1 is applied, when V1 and V3 are applied alternately, and only V3 is applied. It is possible to generate three types of precharge voltages on average in the frame.

  FIG. 34 shows an example in which the application pattern of the precharge voltage is changed according to the gradation.

  When the current output unit is configured by a pull-in current source as illustrated in FIG. 10, the pixel configuration is configured by a p-type transistor as illustrated in FIG. FIG. 12 shows an equivalent circuit when supplying current from the source driver to the pixel circuit. (Only the necessary circuit configuration is shown. Therefore, the circuit configuration of FIG. 6 and FIG. 44 is equivalent to the circuit configuration of FIG. 12). The drain-gate voltage and drain current characteristics of the drive transistor 62 are shown in FIG. Shown in FIG. 35 is a rewrite of this to the gate potential vs. drain current characteristics. The current I1 is set to flow at the gradation 0, and the voltage V2 that is the average of V1 and V3 appears to be applied at the gradation 1, which is equivalent to the current I2. In gradation 2, a current of I3 corresponding to V3 flows. By doing so, it is possible to pass a current value corresponding to the gradation, such as I1 to I3, only by the precharge power supply 24. In addition, since there is a period in which a current corresponding to the gradation flows after the precharge is applied, the current can be changed to a predetermined current even if there is a deviation from a predetermined current value (the precharge voltage is shown in FIG. In order to apply a voltage corresponding to a predetermined current using the relationship of 35, if there is an actual shift, it is due to process variations of the drive transistor 62. This is several nA to a low gradation region. Since the current is about several tens of nA, it is possible to change the current sufficiently.)

  In this way, by preparing multiple voltage sources and combining the method of changing the voltage value to be applied for each frame, a large number of voltage values corresponding to a predetermined current value can be obtained with a small number of voltages. There is an advantage that a display with good gradation can be realized.

  Further, in FIG. 33, assuming that a sufficient current cannot be written in a high current region (= high gradation region), a raising current source 331 is prepared, and the charge of the stray capacitance is increased by a predetermined current + raising current. A method of speeding up charging / discharging can be used in combination.

  By inputting a precharge determination signal instead of gradation data as an input to the precharge pattern control unit 292, the source voltage is changed by changing the voltage applied to each frame only in the row under the white display that is least likely to have a predetermined luminance. It is also possible to output no voltage to the source signal line in the row below the low gradation display that is output to the signal line.

  FIG. 36 shows an example in which three types of precharge voltages are prepared. If the voltage value applied for each frame is not changed, only three types of voltage can be output. However, if different voltages are output for each frame, it is possible to output a voltage value greater than three types as an average value. It becomes.

  For example, if the same or different voltages are output in the even frame and the odd frame, six types of voltage application patterns can be realized as shown in FIG. In this way, by applying different voltages for each frame, there is an advantage that many voltage values can be output with a small power source. In this example, the voltage is varied on an average between two frames, but it is also applicable to three frames or more. As with the gradation display by the frame rate control, if the number of frames is increased, flicker is likely to occur.

  In order to further increase the voltage value, there is a method of increasing the precharge power supply 24, but the switches 131 are required as many as the number of voltages. In particular, a switch requires a large number of power supplies for each source line, so that a large area is required.

  This is because the digital data (grayscale data) is converted to analog values (precharge voltage) in each source line, so a digital-analog converter is required for each source line, so the circuit scale increases as the number of output voltages increases. Becomes larger.

  Therefore, as shown in FIG. 38, only one digital-analog conversion unit 381 is prepared in the semiconductor circuit, converts the serially transferred data into an analog voltage, and then distributes it to each source signal line. For this purpose, the output 382 of the digital-analog conversion unit is input to the distribution unit and hold unit 383, and an analog voltage based on the gradation data is distributed and supplied to each source signal line.

  On the other hand, the method of outputting the current corresponding to the gradation is similar to FIG. 2, in which the gradation data 386 is distributed to each source line by the shift register and latch unit 384, and the gradation is obtained by the current output stage 23 in each source line. A corresponding current is output.

  A current / voltage selector 385 is arranged immediately before output to the source signal line as a part for determining whether to output current or voltage. The precharge determination signal 383 causes the precharge voltage application determination unit 56 to switch the current / voltage selection unit 385 to determine whether to output current, output voltage, or output current after voltage output.

  Thus, if the digital-analog converter 381 has the number of analog output stages corresponding to the number of gradations, it is possible to output a voltage corresponding to the gradations, and a period during which a certain row is selected (in the horizontal scanning period). 1), the source signal line current is first changed to a predetermined value by the voltage, and then the deviation of the current value due to the variation of the transistors of each pixel can be corrected by the current output.

  In order to change the current to a predetermined current value by the current, it often takes a time longer than the horizontal scanning period particularly in the low gradation portion, but the method of changing by the voltage can complete the change in approximately 1 μsec. Since the correction by the current is slight, there is an advantage that it is easy to change the current up to a predetermined current within the horizontal scanning period in the method of flowing the current after applying the voltage.

  In particular, in black display compared to white display, since the current value flowing through the source signal line is small from the equation t = C × V / I, the charge accumulated in the floating capacitance of the source signal line is a charge corresponding to a predetermined gradation. It takes time to charge and discharge to the amount (t is the time required for the change, C is the capacity of the source signal line, V is the source signal line voltage, and I is the current flowing through the source signal line).

On the other hand, at the time of white display in which a large amount of current I flows, it is possible to change to a predetermined current within one horizontal scanning period. (For example, when I = 2 μA, V = 5 V, and C = 10 pF, t = 25 μs. When a QVGA panel is operated at a frame frequency of 60 Hz, the horizontal scanning period is about 65 μs and can be sufficiently changed.)
In this case, the dynamic range and resolution of the digital / analog converter 381 can be reduced.

  For example, in a driving semiconductor circuit capable of displaying 256 gradations, if only the current can be sufficiently changed to a predetermined current value in the upper 128 gradations, the voltage may be output for the lower 128 gradations. Therefore, the digital-analog converter 381 has only to have a 7-bit resolution and only needs to output 128 kinds of voltages. When the gradation data 386 is one of the upper 128 gradations, a precharge determination signal 383 is input so that voltage output is not performed. As a result, the current / voltage selector 385 always outputs only current. Since the output signal of the digital-analog converter 381 is not output to the outside of the driving semiconductor circuit, it may have any value. The simplest method is to ignore the upper 1 bit of the input gradation data 386 and output a voltage corresponding to the value of the lower 7 bits.

  When the gradation data 386 is between 0 and 127 gradations, the current / voltage selection unit 385 is controlled by the precharge determination signal 383, and the analog voltage from the digital / analog conversion unit 381 is supplied to the outside of the driving semiconductor circuit. An output period is provided.

  As a result, a circuit with a reduced resolution of the digital-analog converter can be formed. In general, in the case of a current copier using a p-type transistor as shown in FIG. 6 or a current mirror pixel configuration as shown in FIG. Will go down. The voltage change width in the black to halftone range is smaller than the voltage change width in the black to white range. Therefore, when the configuration is such that the voltage is output only when the gradation is from 0 to 127, the dynamic range of the output voltage can be reduced.

  When voltage and current are sequentially output within one horizontal scanning period, the configuration shown in FIG. 45 is adopted. Here, the precharge pulse 451 is a signal that determines a period during which a voltage is output in one horizontal scanning period. An example of the circuit configuration of the precharge application determination unit 56 is shown in FIG. In addition, FIG. 47 shows input signal waveforms when outputting only current, outputting only voltage, or outputting current after voltage output. Here, the precharge determination signal 383 is a 2-bit signal line. This is to determine whether or not to precharge (apply voltage), to perform precharge, to output only the current after the horizontal scanning period, and to output all precharge voltage. Therefore, 2 bits are prepared as the minimum number of bits necessary for distinction. Here, for the sake of explanation, it is assumed that the determination as to whether or not to perform precharge is the most significant bit (383a), and the signal for determining the period during which the voltage is applied is the least significant bit (383b).

  When the input grayscale data is high grayscale data, if the signal line can be changed to a predetermined current value without performing precharge, only the current is output within one horizontal scanning period. The period 471 in FIG. 47 corresponds to this. At this time, if the precharge determination signal 383a is set to the low level, the current / voltage selector 386 always selects the current output from the configuration of FIG. As a result, only current is output.

  On the other hand, in the case of low gradation data and current output, when the source signal line cannot sufficiently change to a predetermined current value, it is necessary to output a precharge voltage. At this time, the precharge determination signal 383a is set to the high level. In the configuration of FIG. 46, the operation of the current voltage selection unit 386 is changed by the precharge pulse 461 and the precharge determination signal 383b.

  In a low gradation part (especially gradation 0 where the current becomes 0) where the state of the source signal line hardly changes to a predetermined current value due to the current, gradation is displayed by the precharge voltage. Therefore, since no current output period is required, a voltage is always output as indicated by the period 472. For this purpose, in the case of the circuit configuration of FIG. 46, the least significant bit of the precharge determination signal is set to a high level so that a voltage is output regardless of the state of the precharge pulse 451.

  On the other hand, if the state of the source signal line can be changed to a state close to a predetermined current value by voltage as shown in the vicinity of the halftone, if the current can be changed to a predetermined current value by current, the voltage output is first output at the beginning of the horizontal scanning period. By doing so, the state of the source signal line is changed to near the predetermined current value. Thereafter, the current is changed to a predetermined current value. The signal that determines the ratio between the voltage application period and the current output period at this time is the precharge pulse 451, and whether the voltage is output or the current is output depending on the state of the precharge pulse 451 by setting the precharge determination signal 383b to a low level. Judgment was made.

  45 and 46, and the waveform input as shown in FIG. 47, gradation display by current and gradation display by voltage can be performed. A source driver IC capable of performing gradation display by current after the change was realized.

  In general, the ease of current change varies depending on the difference in the size of the panel using the source driver IC (the floating capacitance of the source line is different) and the difference in the number of pixels in the scanning direction (the horizontal scanning period is different).

  When the driver IC of this configuration is used, if the precharge pulse 451 is input from the outside of the source driver IC, the precharge determination signal 383 and the gradation data 386 become external signal inputs as shown in FIG. According to the panel, there is an advantage that a gradation range for performing gradation display using both current and voltage can be arbitrarily set. The setting of the gradation range can be controlled by a control IC formed outside as shown in FIG. When the operation of the control IC can be changed by inputting a command, it can be adjusted by inputting the command. In addition to the case where the control IC is configured outside the source driver IC as shown in FIG. 2, the source driver IC and the control IC are integrally formed on the same chip as seen in a part of the liquid crystal source driver. It doesn't matter. At this time, the gradation range may be adjusted by inputting an integrated IC command.

  According to the above invention, in the low gradation portion, since the current flowing through the source signal line is small, the current cannot be changed to a predetermined value within a predetermined time (horizontal scanning period). The problem that the luminance of the image becomes higher than a predetermined value has been solved by inputting a precharge voltage.

  For example, in the case where an image having a white display area of one line in the black display screen as shown in FIG. 39 is not precharged, black display is performed in the line below the white display area 391a shown in 392a of FIG. However, the black display was realized in the 392b row as shown in FIG. 39B by precharging in the 392 row.

  On the other hand, when the number of types of precharge voltage is reduced, the voltage value corresponding to high gradation display is eliminated, and gradation display is performed only by current in the high gradation part, the following problem occurs.

  In a pattern in which one line of black display area (401) is displayed in a white screen as shown in FIG. 40, black display is realized by performing precharge in line 401 as shown in FIG. 40 (B). In this case, the problem that the brightness of white display in the next row (402) after precharge is lower than the brightness of other white display (402b) is caused by the vertical resolution of the panel, the size, and the current value at the time of white display. To do.

  This is because the time t required for the current value to change is expressed by t = C × V / I, and the horizontal scanning period is shortened as the vertical resolution of the panel is increased. Since the range of the time T to be reduced becomes smaller, the current is written to the pixel without changing to a predetermined luminance when t> T, and the capacity of the source signal line increases as the panel size increases. For this reason, if t becomes long or the current value required for white display decreases, t becomes long, and it becomes impossible to change to a predetermined current. Therefore, there is a problem that the predetermined luminance is not obtained even during white display.

  In particular, in the case of FIG. 40, since the luminance is originally the same in the region 402 and the region 403 below the region 402, if the luminance differs between the two regions, luminance unevenness is observed. On the other hand, even if the luminance of the black display row 401 is higher than that of the black display, only one row is displayed. Therefore, there is no unevenness and the influence on the display quality is small.

  On the other hand, in the case of FIG. 39, if the luminance is different between the 392 row and the 393 region in the black display portion, luminance unevenness is observed, but even if the luminance of the white display row 391 is different with and without precharge, Since the display section has only this line, it is not observed as unevenness.

  In these two images, it is better to precharge in the case of the display pattern of FIG. 39, but it is better not to precharge in the display pattern of FIG.

  In other words, in order to prevent luminance unevenness that tends to occur due to precharging on a screen with many white display portions and few black display portions, precharging is not performed on black display pixels, and a predetermined current is applied on a screen with many black display portions. Since the reduction in display quality due to the increase in black luminance due to not becoming black (referred to as a black floating phenomenon) is more conspicuous than the problem that the white luminance does not become the predetermined luminance, it is better to precharge.

  This problem does not occur in a configuration in which precharge voltages corresponding to each gradation are prepared and the change to a predetermined current value is accelerated for all gradations by the configuration as shown in FIGS. However, there is a problem in a configuration in which the number of types of precharge voltages is reduced in order to reduce the circuit scale. The countermeasures when the precharge voltage is low are as follows.

  Whether to precharge or not depends on the lighting rate of the panel, and the number of gradations for precharging is changed.

  The lighting rate of the panel can be calculated by adding all luminance data for one frame. When the lighting rate is high according to the lighting rate value obtained by this method, precharge is not performed, or the type of gradation to be precharged is reduced (for example, only gradation 0), and the lighting rate is low. By pre-charging, the luminance of the low gradation display pixel can be displayed faithfully.

  In this method, the screen for actually precharging and the screen displaying the lighting rate differ by one frame. The calculated screen is the previous frame. When a still image is displayed, since the lighting rate does not change between these two frames, there is no problem in display. On the other hand, when a moving image is displayed, the lighting rate hardly changes during one frame, and adverse effects such as those shown in FIGS. When the rapidly changing frames continue, the display pattern changes every 1/60 seconds, and even if the phenomenon as shown in FIG. Can not.

  Therefore, using the lighting rate data based on the image of the previous frame, changing the gradation for precharging and the proportion of frames for precharging according to the lighting rate is the pattern of both FIG. 39 and FIG. This is effective in preventing display unevenness in In addition, if there is a block that stores data for one frame using a frame memory in the display module, the lighting rate is calculated at the time of storage, and this lighting rate data is given when read, Since the pattern for applying the precharge can be changed using the lighting rate of the frame, it is not necessary to use the data one frame before. When there is no storage means such as a frame memory, data is used one frame before, and the storage means for lighting rate calculation is omitted.

  As an example of changing the precharge pattern in accordance with the lighting rate, when the lighting rate is 10% or less, a precharge voltage is applied at a gradation that is one half from the bottom of all gradations (here, In addition to outputting only the voltage as shown in Fig. 47, the current is output after the voltage is output (including both). When the precharge voltage is applied at the gradation, and the lighting rate exceeds 40% and is 60% or less, the precharge voltage is applied only when the gradation is 0 (black display) and exceeds 60%. Suppose no precharge is performed. Accordingly, it is possible to reduce display unevenness caused by a problem that it is difficult to write a predetermined current even with a small number of precharge voltages.

  By applying a precharge voltage to the black display portion in the display pattern of FIG. 39, the problem that black floating occurs in the row indicated by 392a in FIG. 39A can be solved. However, when a precharge voltage is applied to the entire black display portion 393, the display luminance may vary due to the threshold voltage variation of the pixel drive transistor 62 in the region 393. . For example, when the pixel circuit configuration is as shown in FIG. 6, the pixel in the selected row has an equivalent circuit shown in FIG. When the precharge voltage is output, the same voltage as the precharge voltage is applied to the gate electrode of the driving transistor 62, that is, the node 72. If there is a variation in the gate voltage versus drain current characteristics of the drive transistor 62 depending on the row, the luminance varies between rows to which the same precharge voltage is applied. In order to compensate for the variation, a method of changing the gate potential according to the variation is often taken by passing a current thereafter.

  Since the time required for the current change becomes longer in the low gradation portion, it is desirable to make the current flowing period as long as possible. In the case where the same gradation is displayed over several lines, the change in the source signal line is only for compensating for the variation in the period corresponding to the line displaying the same gradation, so that the amount of change is small. For example, if the state of the source signal line is white, black, or black, the amount of change is large and it takes a long time when it changes from white to black, but since the amount of change is small from black to black, precharge is not performed. Can also be changed.

  Utilizing this fact, the voltage output from the precharge voltage is performed only when the previous row data is referred to and the gradation difference between the previous row data and the data is large. In the previous example, precharge is performed when the color changes from white to black, and precharge is not performed when the color changes from black to black. The time required for correcting the black-to-black variation correction can be increased by not performing precharge, and the correction accuracy can be further improved.

  In general, it is more difficult for the current value to change from white to black than to change from black to white. It is effective to perform precharging when the gradation of the pixel in the previous row is equal to or higher than the halftone when the luminance of the pixel is equal to or lower than the halftone.

  If the previous line is less than halftone, even if the luminance of the pixel is less than halftone, a predetermined gradation can be displayed as much as the amount of change is small.

  On the other hand, when the pixel is more than halftone, the amount of current is large, and it is easy to change to a predetermined current. Therefore, precharge is not necessary regardless of the pixel in the previous row. However, when the resolution is high, the current amount is small even in the halftone, or when the panel size is large or the like is difficult to change, precharge may be performed when the pixel in the previous row is equal to or lower than the halftone. Alternatively, a method of outputting the raising current source 331 of FIG. 33 and increasing the current value to shorten the time required for the change and to make the current easy can be used.

  There is generally a vertical blanking period in which no row is selected during one frame. At this time, the source signal line is disconnected from any pixel by the switching transistor, and there is no current path. When the current output stage of the source driver IC is configured as shown in FIG. 10, only the source signal line is connected to the end of the current output 104 in the vertical blanking period, and the current source 103 for gradation display supplies current. Even if you try to pull in from the source signal line, you cannot pull in because there is no current path.

  For this reason, the gradation display current source 103 attempts to draw a current even if it is impossible to reduce the drain voltage of the transistors constituting the current source 103. The potential of the source signal line also decreases at the same time.

When the vertical blanking period ends and an attempt is made to supply current to the pixels in the first row, the source signal line potential is greatly reduced, and the source signal line potential is also reduced compared to normal white display. (Here, the potential of the source signal line is the lowest when white is displayed and the highest when black is displayed. When the pixel configuration shown in FIG. 6 or FIG. 44 is used) It is difficult to change the potential of the signal line compared to other rows. (Required large change)
When the potential of the source signal line is greatly reduced, the potential is further reduced compared to when white is displayed, and even when white display is performed on the first line, if the change takes time, display is performed with higher brightness than the predetermined brightness. End up. For a row that is scanned immediately after the end of the vertical blanking period, it is desirable to output a precharge voltage regardless of the display gradation.

  Therefore, in the present invention, the vertical sync signal is used, and the precharge determination signal corresponding to the data corresponding to the next row in the vertical blanking period is used to forcibly precharge the luminance of the first row. Solved the problem different from the brightness of other lines.

  Note that as a method of reducing the potential drop of the source signal line as much as possible, black display data is input to the gradation data 54 in the vertical blanking period, and the switch 108 is turned off to reduce the source signal line potential. It may be suppressed. Further, a switch may be provided between the current output 104 and the source signal line, and the switch may be turned off during the vertical blanking period. This switch may also be used as the current / voltage selection unit 385, and the number of switches can be reduced by separating the current output, voltage output, and source signal line by setting the switch state to three values. Is possible.

  FIG. 41 shows a flowchart for performing the precharge method described above.

  When the forced precharge signal is valid from the video signal and the forced precharge signal, a precharge voltage is output regardless of the video signal. The output voltage value may be changed according to the video signal when there are a plurality of voltages. Here, if the forced precharge signal is enabled only when the video signal corresponding to the first row is input, the data of the first row is precharged regardless of the video signal, and the source signal line is used during the vertical blanking period. It is possible to avoid a phenomenon in which the current due to the voltage drop hardly changes to a predetermined value.

  If the forced precharge signal is invalid, the gradation of the input video signal is then determined (412). In a small gradation panel or a low-resolution panel, in a high gradation region where the amount of current is larger than that of a low gradation portion, it is possible to change the current value to a predetermined current value only by a current within a predetermined period (one horizontal scanning period). Therefore, in 412, a determination is made such that precharge is not performed in a gradation in which a predetermined current can be written, and precharge is performed in a gradation in which the current does not become a predetermined current alone.

  Next, if it is below a specific gradation that requires precharging, the process proceeds to 413. (Here, it is preferable that the specific gradation can be set by an external command because the specific gradation differs depending on the display panel.) It is determined whether or not to precharge according to the state of the video signal of the previous line. This corresponds to the fact that when black is displayed after black display and after white display, black can be displayed only with current after black display, but the source signal line cannot change sufficiently to black after white display. Thus, the precharge is applied when the amount of change in the source signal line increases. In step 412, 413 is executed when a video signal having a specific gradation or lower is input. In 413, precharging is performed at a gradation higher than a certain level of data in the previous row (here, specified by 412 and 413). When the gradations can be set to different values), precharge is not performed when the gradation is below a certain gradation.

  Next, as shown in FIGS. 39 and 40, it is determined whether or not to precharge based on the lighting rate of the screen (414). In the case of a display with a high lighting rate as shown in FIG. 40, it is more problematic that white does not have a predetermined gradation than black does not have a predetermined gradation. Do not precharge regardless of the key. Here, the setting of the lighting rate that becomes a boundary of whether or not to perform precharge may be changed by an external command, and a highly versatile semiconductor circuit may be used. If the lighting rate is low, precharge is performed and the process proceeds to 415. Here, if the precharge is intermittently performed between a plurality of frames when the lighting rate is medium, and the precharge is always performed when the lighting rate is low, the number of signal line bits from 414 to 415 is plural. For example, the operation of 415 may be made different for each lighting rate. To prevent flicker that occurs when the screen brightness changes rapidly with respect to the lighting rate when the pre-charge judgment is changed at 414 when a certain frame is reached on a screen where the lighting rate changes gradually. It is effective for.

  Next, at 415, it is determined whether or not to precharge according to the FRC flag. This is for realizing the precharge application pattern shown in FIG. In the case of FIG. 32, the FRC flag outputs a signal indicating that precharge is performed in 2 frames out of 3 frames, and that the data corresponding to the same pixel is not precharged in the remaining 1 frame. Output a signal. As the second FRC flag, a signal is output that precharges in one frame out of two frames and does not precharge in the remaining one frame. By determining which FRC flag is used for the input video signal, it is possible to intermittently precharge between frames. In addition, when precharging is performed in all frames (in the case of gradation 0 in FIG. 32), it is possible to perform precharging every frame by referring to an FRC flag that always outputs a signal to perform FRC. It is.

  In this description, it is determined whether or not to precharge through all the processes from 411 to 415 in order, but all the processes may not necessarily be performed.

  In addition to determining whether or not to perform precharge, there are cases where the precharge voltage varies depending on the gradation as shown in FIG. In the case of using the FRC flag, it can be realized by replacing the precharge with the voltage V2 applied and the non-precharge with the voltage V1 applied. In the case of three or more values, it can be realized by increasing the number of bits of the FRC flag.

  FIG. 42 shows a circuit block for realizing the precharge method according to the present invention. A determination signal as to whether to precharge the video signal 410 as a result of determination by each block is output to 417. Whether to perform precharge on the source driver side is determined by the determination signal 417 output at almost the same timing as the video signal 410. The serial / parallel conversion unit 427 is not necessarily required, and is necessary for matching with the input interface of the source driver 36 when realized in combination with the source driver IC configured by 36 of FIG.

  The video signal 410 is input to the precharge determination unit 1 (421), the precharge determination unit 2 (423), and the FRC insertion means (424).

  The precharge determination unit 1 (421) is a block for performing the determination (413) based on the previous line data in FIG. Based on the video signal 410 and the previous row gradation setting signal 428, if the value of the video signal 410 is greater than the previous row gradation setting signal 428, a signal indicating that precharging is to be output is output to the storage means 422. Is output to the storage means 422. By holding one vertical scanning period value in the storage unit 422, the previous row of data is obtained. Since it is necessary to hold one vertical scanning period value, the storage unit 422 needs the number of bits corresponding to the number of pixels in the horizontal direction. At this time, the data stored in the storage means 422 is stored as either precharged or not.

  Note that the order of the storage unit 422 and the precharge determination unit 1 (421) may be reversed. That is, by holding the video signal 410 in the storage unit 422 for one vertical scanning period, it is possible to determine whether or not to perform the precharge by the precharge determination unit 1 based on this data as the previous data. In this method, since the video signal 410 needs to be held (the number of bits of the video signal 410) × (the number of pixels in the horizontal direction), the number of bits is required. Is desirable. However, if the video signal for one row is stored in the functional block unit other than the precharge, it can be performed using the storage unit 422 in the functional block. May be reversed.

  The video signal 410 is also input to the precharge determination unit 2 (423) at the same time. Similar to the block 421, the precharge determination unit 2 (423) determines whether or not to perform precharge depending on the gradation of the input video signal based on the video signal 410 and the precharge application gradation setting signal 429 (FIG. 41 corresponds to the processing 412).

  With respect to the output of the precharge determination unit 2 (423), the output of the storage means 422, which is the determination result of the previous row of data, is determined after determining whether the previous row data selection unit 400 precharges according to the gradation. It is determined whether or not to use the previous row data when precharging according to the key.

  For this operation, the previous row data selection unit 400 performs a logical product of the 423 output and the 422 output. If white data is input as gradation data, and precharge is applied only at halftone or less with the precharge application gradation setting signal 429, the output of 423 becomes "L" level ("L" level is No precharge). At this time, the output of 400 becomes “L” level regardless of the data in the previous row, which satisfies the condition of FIG.

  On the other hand, if black data is input and the signal 429 is the same, the output of 423 becomes the “H” level, and the output of 400 changes depending on the output of 422 which is the determination result of the previous row data. If the previous row gradation setting signal 428 is halftone and the previous row data is white, the output of 421 is “H” level and the signal output from the storage means 422 is also “H” level. , 400 also becomes “H”. If the data of the previous line is black, the output of 421 becomes “L” level, and the output of 400 becomes “L” level in the same way.

  That is, the output of the data reference unit 400 one row before is the level indicated by the video signal 410 below the tone indicated by the precharge application tone setting signal 429 and the image signal 410 one row before indicated by the tone setting signal 428 one row before. An “H” level signal indicating that precharge is performed is output only in the case of the key or higher, and precharge is not performed in other cases. As a result, the processes 412 and 413 in FIG. 41 are realized.

  When it is determined whether or not to precharge the input video signal regardless of the previous line data (when the step 413 is eliminated), the output of the storage unit 422 may be always set to the “H” level. For example, the previous row gradation setting signal 428 is set to 0 (black) or the previous row data valid / invalid signal is input to the previous row data reference unit 400, and the logical sum of this signal and the output of 422 is output. May be taken with the output of 423 and ANDed.

  As a result, for example, in the case of full-color black display, it is possible to realize any precharge method in which only the first row is precharged and all pixels are precharged.

The FRC register selection unit 424 is a block for selecting a ratio of frames for performing FRC or precharging according to the gradation of the video signal 410. (Blocks for realizing the table in FIG. 32)
The FRC generation unit 425 includes an FRC register 433. By shifting the FRC register 433 for each clock, horizontal scanning signal, and vertical scanning signal, it is possible to determine whether or not to precharge for each frame.

  The operation of the FRC register 433 is shown in FIG. This FRC register is composed of 3 bits, 1 is 2 and 0 is 1. When 1 is precharged, when 0 is set, there is no precharge. Further, the value of the bit (433c) surrounded by a bold line is output to the FRC register selection unit 424.

  In the initial state, register values are stored in the state of 433a to 433c. This is subjected to a 1-bit shift process for each video signal data. By performing this in order up to the data of the last column, a signal with precharge is output twice in three times, so that a pattern without precharge as shown in the first row of FIG. 49A is formed.

  As the first data in the second row, data obtained by performing shift processing from the register in the state of the first column in the first row is used. The shift process at this time is called a line shift 432. The amount of line shift in this case is 1 on the left. The shift amount may be 1 or 2, but in this case, description will be made with an example of 1. For the sake of convenience, description will be made with the amount of left shift. A 1-bit shift process is performed on the data in the second row in order. Similarly, line shift is performed when the rows change in order of the third row and the fourth row. The line shift values are all the same within one frame.

  In this way, the precharge on / off pattern in one frame shown in FIG. 49A is formed.

  When the frame changes, the value obtained by performing shift processing from the value of the FRC register in the first row and first column of the previous frame is used in the first row and first column. The shift amount at this time is defined as a frame shift 431.

  The frame-shifted register is used as the data in the first row and the first column, and the shift process similar to that in the first frame is performed, whereby the precharge pattern shown in FIG. 49B is formed. Further, when the frame shift 431 is similarly performed in the next frame, the pattern shown in FIG. 49C is obtained. When frame shift processing is further performed in the next frame, the FRC register value in the first row and first column in FIG. 43 is obtained. This scanning is sequentially performed.

  In the three frames shown in FIG. 49, each pixel is precharged twice in three frames. In addition, by making the pattern of pixels that are precharged uniform, it is possible to reduce flicker caused by a luminance difference due to the presence or absence of precharge.

  From this, the FRC register 433 indicates the ratio of frames to be precharged. Generally, when there are M 1s for the N-bit FRC register, it indicates that precharge is performed in M frames out of N frames. .

  FIG. 49 shows an on / off pattern in a monochrome display device. In a color display device, display is generally performed by combining pixels of three primary colors of red, green, and blue as one pixel.

  In general, the video signal 410 is often sent with the three primary colors of red, green, and blue at the same timing, and the processing of FIG. 42 is performed in parallel for each color.

  Although the same FRC register output may be referred to for all colors, it is preferable to change the FRC pattern for each color in order to reduce flicker. Although it is possible to prepare the FRC register 433 for each color, the circuit scale increases. Therefore, the precharge pattern is changed by changing which bit of the FRC register 433 is output for each color. In the example of FIG. 43, if red refers to 433c, for example, green refers to 433b and blue refers to 433c. At this time, how much the reference positions of green and blue differ from red are expressed as G shift and B shift, and G shift is 1 for green which is different from 1 and B shift is 2 for blue which is different from 2 It will be. Therefore, in FIG. 42, the signal line 426 from the FRC register 433 to the FRC register selection unit 424 is composed of bits of the number of display colors.

  The FRC register 1 in FIG. 42 includes a register as shown in FIG. 43, and the FRC register 2 includes a register in which 1 bit is 1 and 1 bit is 0 in 2 bits. If these two registers are used, it becomes possible to precharge twice in three frames and once in two frames.

  Next, the FRC register selection unit 424 will be described. The FRC precharge setting signal 419 is a signal that determines at what ratio precharge is to be performed with respect to the gradation of the video signal 410, and is a signal for setting the relationship as shown in FIG. Depending on the signal 419, for example, it is possible to set such that precharge is applied twice in three frames when the gradation is 10 or less, and precharge is not performed when the gradation is 10 or more.

  Although not shown in FIG. 42, there are cases where precharge is performed in all frames or precharge is not performed in all frames. This can be realized by selecting 1 (when precharging in all frames) or 0 (when not precharging in all frames) instead of selecting one of the outputs of the FRC register 433. is there.

  The lighting rate setting signal 418 and the lighting rate data 420 are input, which are input because the FRC pattern may be changed depending on the lighting rate.

  For example, when the lighting rate is high, precharging is not performed from FIG. 41, so the output of the FRC selection unit is always at “L” level (not precharged). This is to perform the relationship between the gradation of the video signal 410 and the precharge pattern as shown in FIG. 50 when the lighting rate is low, and the relationship as shown in FIG. 32 when the lighting rate is medium. . The lighting rate setting signal 418 is a signal for setting the high, medium and low threshold values of the lighting rate. The FRC precharge setting signal 419 determines the relationship between the precharge pattern and gradation (for example, FIG. 32) for each of the lighting rate high, medium, and low.

  If it is not necessary to change the precharge pattern depending on the lighting rate, the value of the FRC precharge setting signal 419 at each lighting rate may be the same.

  When FRC precharge is not performed (when all frames are precharged or not), the FRC precharge setting signal 419 should not output the value of the FRC register 426 regardless of the video signal 410. .

  FIG. 51 shows a setting example using the FRC precharge setting signal 419 and the lighting rate setting signal 418. Depending on the lighting rate, which of FIGS. 51A, 51B, and 51C is selected is determined by the lighting rate setting signal 418. FIG. For example, FIG. 51A shows a lighting rate of 5% or less. Further, the FRC precharge setting signal 419 determines the line 511 indicating the relationship between the gradation and the ratio of precharged frames in each figure.

  The output of the FRC register selection unit 424 created in this way is input to the FRC insertion means 409. The output of the previous row data reference unit 400 is also input to the FRC insertion unit 409. That is, a signal indicating whether to precharge from the input gradation and the data of the previous row, and a signal that determines the precharge rate from the lighting rate and the input gradation are input. If precharging is performed only when both signals are precharged, portions 412 to 415 in the flowchart of FIG. 41 can be realized.

  Next, the output of the FRC insertion means 409 is input to the forced precharge input means 408, and calculation with the forced precharge signal 416 is performed. As indicated by reference numeral 411 in FIG. 41, when the forced precharge signal is valid, precharge is performed regardless of the gradation. Therefore, in the block 408, when the forced precharge signal 416 is in the valid state (precharge state), the output 417 outputs a signal that precharges regardless of the output of 409.

  If the forced precharge signal 416 is enabled only when the video signal 410 corresponds to the data in the first row, precharge is performed in the first row where the change to the predetermined current value due to the vertical blanking period is delayed. It is feasible to do.

  By simultaneously transferring the value of the output 417 for the video signal 410 to the source driver, a precharge voltage application pattern as shown in FIG. 41 can be applied to the source signal line.

The serial / parallel conversion unit 427 is necessary to match the input interface of the source driver 36 in FIG. 3, and is not necessary when the video signal of each color and the precharge output 417 (for each color) are transferred in parallel. (Output directly to the source driver)
2 shows an example in which the control IC 28 and the source driver 36 are composed of different chips, but an integrated chip composed of the same chip may be used. In this case, the configuration shown in FIGS. 41 and 42 is built in the source driver 36.

  It is preferable that the output voltage value of the precharge power supply 24 can be controlled by an electronic volume or the like. This is because the precharge voltage for causing the predetermined current to flow is determined based on the voltage of the EL power supply line 64. In FIG. 12, when the current I2 is made to flow through the source signal line 60, the potential of the source signal line 60 is (the voltage of the EL power supply line 64) from the relationship between the drain current and the drain-gate voltage of the transistor 62 (FIG. 12B). −V2.

  On the other hand, the EL power supply line 64 is supplied to each pixel through wirings 483 and 484 in the display panel shown in FIG. The maximum current flows to 483 when all pixels display white, and the minimum current flows to 483 when black displays. At this time, due to the wiring resistance of 483, the potential is different at points 485 and 486 during white display. On the other hand, at the time of black display, 485 and 486 have substantially the same potential. That is, the potential of the EL power supply line 64 differs depending on the voltage drop of the wiring 483 when displaying white and displaying black. That is, even when the same current I2 is supplied, the voltage of the source signal line 60 varies depending on the voltage drop amount of the wiring 483. Therefore, if the voltage value of the precharge power supply 24 is not changed by the voltage drop amount 483, there arises a problem that the current of the source signal line changes and as a result, the luminance changes.

  If the voltage of the EL power supply line 64 is different, the voltage applied to the source signal line 60 needs to be different. What is necessary is just to change a voltage using the lighting rate data in 1 frame. When the lighting rate is high, the current flowing through the EL power supply line 483 increases, so that the electronic volume is controlled so that the voltage drop is large and the voltage value of the precharge power supply 24 is lowered. On the other hand, when the lighting rate is low, the voltage drop of the EL power supply line 483 is small, so that the luminance value caused by the wiring resistance of the EL power supply line 483 can be reduced by increasing the voltage value of the precharge power supply 24 using an electronic volume. It can be eliminated.

  In the source driver IC 36, when a block for converting an 8-bit digital signal into an analog current output as shown in FIG. 25 is formed of transistors, output current variation varies depending on the channel area per transistor and the number of transistors used for output. (Inversely proportional to the channel area or the square root of the number of transistors).

  When transistors of the same size are arranged by (display gradation number −1) and gradation is displayed by the number of transistors, the higher the gradation portion, the smaller the variation. That is, when the variation is 1% or less at a certain gradation A, the variation is always 1% or less on the higher gradation side than the gradation A (relationship shown by the dotted line 523 in FIG. 52).

  The allowable variation is 1%, and the dotted line 523 in FIG. The structure of the transistor at this time is shown in FIG. In addition, FIG. 54B shows the transistor size of each bit.

  As the number of gradations increases, the number of transistor groups 531 and 532 used for output increases, so the variation decreases.

  Here, paying attention to gradation 30 or more, the variation is 0.6% or less, with a margin of 40% or more for 1%. That is, at gradation 30 or higher, the variation is within 1% even when the transistor area is halved (in the case of area ½, the variation is increased by 40%, see FIG. 11).

  Therefore, in the present invention, in order to reduce the chip area of the source driver IC, it has been considered to reduce the channel area of only the transistor 534 on the high gradation side.

  Since a transistor with a small size is used when the gradation is 30 or more, the size of the transistor that outputs current for the lower 5 bits of the 8-bit signal is not changed, and is output when the gradation is 32 or more. The channel size of the transistor corresponding to the upper 3 bits of data is reduced so that the drain current value with respect to the gate voltage does not change. The method was realized by shortening the channel width at the same rate as the channel length was shortened. FIG. 54A shows the relationship of the transistor size to each bit at this time. The relationship between the output current variation and the gradation is shown by the solid line 522 in FIG.

  As shown in FIG. 54A, transistors having different sizes are used for the lower 5 bits and the upper 3 bits.

  In FIG. 54 (b), the sizes are different, but this is because the transistor group of 531e is composed of 16 transistors, and the transistor group of 532a is composed of 8 transistors, so that the number of transistors is small. This is because it is necessary to increase the amount of current per one. In this case, 532a must be able to output twice as much current as 531e, and the number of transistors is halved. Therefore, the transistor 534 constituting the transistor group of 532a needs to have a channel length approximately four times as long. There is. This makes the size different.

In contrast, in the present invention, the channel area of the transistor in the high gradation portion is further reduced from that in FIG. 54B (FIG. 54A). In order to maintain the current value for the same gate voltage, the reduction rate of the channel length and channel width was made uniform. The variation increases as the channel area decreases. In this example, since the channel width and the channel length are both (1/2) ^1/2 , the area is 1/2, and the variation is about 1.4 times at gradation 32 or higher as shown in FIG. Such a variation relationship is obtained. The variation increases at gradation 32 or higher, but does not affect the display because it is 1% or less, which is within the allowable range.

  FIG. 2 shows an outline of the source driver IC 36. The arrangement of the transistor groups 531 and 531 in FIG. 53 is included in the current output stage 23. The transistor group occupies 30% of the total area of the driver IC 36. In the configuration of FIG. 54A according to the embodiment of the present invention, the total channel area is reduced by 44% compared to the configuration of FIG. This corresponds to 13% of the entire chip, and the chip size can be reduced by 13%. As a result, the cost of the chip can be reduced by 13%, and the unit price of the driver IC can be reduced.

  Note that the current flowing through one transistor is theoretically proportional to the channel width, but there is actually a shift. Even if the configuration of FIG. 54B is as shown in this table, the current of the 531 transistor group tends to be slightly smaller. An increase in the current value with respect to the gradation does not have a proportional relationship, and when a lamp image is displayed, a step is generated in various places. (In this case, it occurs every 32 gradations) Therefore, the output current is actually simulated so that the increase in current with respect to the gradation becomes uniform. For example, by increasing the channel width of the transistors used in the 531 transistor group from 1.5 to 1.7, the rate of increase is made constant. Since such adjustment varies depending on the specification process, the channel size is described as a theoretical value in the present invention. Similarly in FIG. 54B, the channel size needs to be adjusted. Moreover, it is preferable to adjust so that a transistor may become large in the case of adjustment. When the current in the low gradation portion is small, the channel width of the low gradation portion transistor 533 is increased or the channel length of the transistor 534 in the high gradation portion is increased. When the low gradation portion current is large, the channel width of the high gradation portion transistor 534 is increased or the channel length of the low gradation portion transistor 533 is increased. This is to prevent an increase in variation due to a reduction in the channel area of the transistor due to adjustment. This is because even if the variation is theoretically within 1%, this transistor size adjustment should not exceed 1%.

  Further reduction of the chip size can be realized by reducing the channel size of the transistor groups 532b and 532c. Also in this case, as a method of reducing the channel size, the reduction ratio of the channel length and the channel width is made the same. The lower limit is a value at which the variation in output current does not become 1% or more when current is supplied from all the transistors constituting the transistor group. FIG. 55 shows the relationship of output current variation with respect to gradation when the transistor sizes of 532b and 532c are reduced.

  As described above, the output current of the transistor group corresponding to each bit of the input signal is not changed, and the channel of the transistors constituting the transistor group is within a range where the output current variation is within 1% when only the transistor group is output. By reducing the area, it was possible to realize a source driver IC that was as small as possible without causing luminance unevenness due to a shift in output current.

  In FIG. 55, the output current variation exceeds 1% at gradation 10 or less, but since it is a low luminance region, display unevenness does not occur even when the visibility is low and it is 1% or more. In addition, if the current deviation required for changing one gradation is 10% or less at gradation 10 or less, there is no change from the predetermined gradation to the gradations of the upper and lower gradations. Since it is not visible, it may exceed 1% in such a low gradation part.

  In this example, the transistor size is changed between the lower 5 bits and the upper 3 bits. However, the lower 4 bits and the upper 4 bits may be used. The size of the transistor group can be changed for each arbitrary bit.

As described above, when the channel size of the transistor is changed according to the bit and the current source according to the weight for each bit is formed, the channel size of the transistor and the output current value are not exactly proportional, so the channel size is determined by simulation. Adjust the channel width. At this time, when the simulation and actual measurement do not match, the current increases or decreases compared to the design value, and when the current of the lower bit side transistor becomes larger than the design value, the magnitude relation of the current is reversed at the halftone display as shown in FIG. Inversion occurs. (In this description, I127> I128, but even if I127 = I128, it is a problem because the gradation is reduced by one gradation. Therefore, gradation inversion means that the current at the 128th gradation is the 127th gradation. Less than or equal to the current.)
Gradation inversion is most likely to occur during halftone display.

  The most likely occurrence during halftone display will be described in the case of 8-bit display. As shown in FIG. 58, the difference in current between gradation 127 and gradation 128 is only 0.79% when gradation 128 is the reference. Therefore, gradation inversion occurs when the variation of the current source that outputs these two gradations is 0.79% or more. For example, when the current output of gradation 128 is reduced by 0.9% due to variations (the portion where 591 is reduced), the current magnitude relationship is reversed as shown in FIG. Further, when the current of the lower 2 bits of the transistor becomes larger than a predetermined value, the luminance difference further decreases. For example, when the output of the lower 2 bits is 20% larger, the current output becomes 632 and the luminance difference becomes 0.31%.

  On the other hand, at the time of low gradation display, as shown in FIG. 66, for example, the luminance difference between gradations 3 and 4 is 33%. The output variation is 2.9% (at this time, the variation at the time of halftone display is 0.9% and the variation is dependent on the area), and (brightness difference)> (variation) and gradation inversion does not occur. . Even if the output of the lower 2 bits of the transistor is increased by 20%, the luminance difference is 10%, which is larger than the variation of 2.9%, so that the gradation is not inverted.

  Although the difference in luminance is small at halftones and above, any gradation always has a current output corresponding to gradation 128, so it is only necessary to pay attention to output variations for gradations exceeding 128. Since the current of gradation 128 occupies at least 66% of the entire output current, the output variation for gradation exceeding 128 is 0.34 times or less of the total output current. As a result, even if the luminance difference is small, the variation is small, so that gradation inversion is unlikely to occur. The case of gradation 131 and gradation 132 is shown in FIG.

  The difference between gradations 131 and 132 is 0.75%. Therefore, gradation inversion occurs when the current of 132 varies by 0.75% with respect to the current of 131. Of all the outputs, the output for the gradation of 128 exists for both gradations (672), so that the variation in this output can be ignored. Variation occurs in the portion 671, and the output of gradation 4 has variation of 2.9% from the size ratio of the transistors. However, since the portion 671 is 3% with respect to all outputs, the output variation for all outputs is 0.09%. Since the gradation difference is smaller than 0.75%, the gradation inversion does not occur. Even when the output corresponding to the lower 2 bits is increased by 20%, the luminance difference is 0.30%, which is larger than the variation of 0.09%, so that the gradation is not inverted.

  For 128 gradations or more, output variation is obtained from (output variation of transistors not common between two gradations) × (ratio of non-common output transistors to all outputs), and (ratio of non-common output transistors to all outputs) is Since it is 0.33 at the maximum, it is a factor that the variation becomes smaller than the luminance difference.

  FIG. 65 shows the relationship between the luminance difference and output variation between several gradations. The most severe condition is between the 127th and 128th gradations.

  Since the difference in brightness is large on the low gradation side, it is difficult to invert, and on the high gradation side, the output ratio using the same transistor is large between adjacent neighbors. Hateful.

  The problem is that the number of transistors to be output is different (only the most significant bit is output) and (all other than the most significant bit are output).

  Therefore, in the present invention, in order to eliminate gradation inversion, in addition to the current source 241h corresponding to the most significant bit, a raising current source 572 is connected via the switching unit 571 as shown in FIG. In some cases, the switching unit 571 outputs the current source 572 for raising and the current source 241h together to increase the current at the 128th gradation and prevent gradation inversion. When the gradation is not inverted, the switching unit 571 is connected to the ground potential so that only the output of the current source 241h is output.

  The size of the transistor of the raising current source 572 is designed to have a current output capability of 1/10 or more and 1/2 or less that of the current source 251a of gradation 1. As a result, the current of gradation 128 can be increased by 0.1% to 0.5% by the current source 572 for increasing the current. When raising the height by 0.5% or more, if the raising is not performed for all outputs, the adjacent luminance difference becomes 1% or more, which causes a display defect. For this reason, in the structure of FIG. 57, it is preferable to set it as 0.5% or less. On the other hand, the minimum value of the raising current is determined from the following. Because of the relationship between the adjacent luminance difference of 0.79% and the variation ability of 0.9%, the current of the gradation 128 is smaller by 0.1% at the maximum than the gradation 127. Even at this time, it is necessary to prevent gradation reversal, so it is necessary to increase the height by 0.1%. For this reason, it is necessary to increase the height by at least 0.1%, and the minimum value of the current source is required to be 0.1%.

  In general, in the case of N-bit display, the minimum value of the raising transistor is a halftone current value of (adjacent luminance variation [%] during halftone display) − (halftone adjacent luminance difference [%]) [%] or more. It is designed to output a current of 5% or less.

  The switching unit 571 is connected by connecting the output of 572 to the output of 241h at the terminal whose gradation is inverted according to the inspection result after inspecting the current output of each output of the semiconductor circuit, and directly at the terminal where the gradation is not inverted. Realized by connecting to.

  In this way, gradation inversion as shown in FIG. 56 is eliminated, and gradation luminance characteristics as shown in FIG. 60 are realized. Further, as shown by reference numeral 633 in FIG. 63, even if the current of the lower 2 bits is increased by 20% by using the raised current source, the difference between the gradations 127 and 128 can be expanded from 0.31% to 0.77%. It is possible to prevent gradation inversion due to current output variations.

  When the current value of the current source of the lower bit becomes larger than the specified value, as indicated by 632 in FIG. 63, the current (brightness) difference between the 127th gradation and the 128th gradation becomes small and gradation inversion is likely to occur. . If the variation is 0.9% compared to the luminance difference of 0.31%, gradation inversion may occur at about half of the terminals. In this case, if the switching unit 571 is connected for each terminal, it takes time and productivity is lowered.

  Therefore, as shown in FIG. 61, the raising current source 572 and the current source 241h are connected through the switching means 611, and the raising unit 611 is controlled by the raising signal 612, thereby using the raising signal 612 by external command input. We considered a configuration that can easily increase the current of the 128th gradation.

  The raising signal 612 only needs to be set for each output. In this case, as shown in FIG. 69, a latch 691 for holding the value of the raising signal 612 for each signal line is necessary. A signal can be distributed to each latch by a 1-bit signal input 692 if a shift register used for distributing a video signal is shared. However, since the latch is provided for the signal line, there is a problem that the circuit scale becomes large. When the circuit scale may be increased or when the area of the latch portion 691 occupying the entire area is small by using a fine process, it may be determined whether or not to increase by controlling the raising signal for each output. It should be noted that this signal needs to be raised after normal inspection, and it is necessary to determine the latch data of the latch unit 691 at the time of shipment because it is necessary to always keep the same state as well as to know the unnecessary terminals. Therefore, the command for inputting the raising signal is normally hidden from the user. Further, it is desirable that the latch portion is composed of a nonvolatile memory so that it is not necessary to input a signal every time the power is turned on.

  Therefore, for the purpose of relieving the gradation inversion due to the current value of the current source corresponding to the lower bit being increased, the raised signal line 612 is common to all outputs, and the gradation output is simplified by raising all the outputs. To prevent.

  A block diagram at this time is shown in FIG. A command as to whether or not to raise the driver IC is input by the input signals 12 and 13. This is distributed by the distribution unit 27, and a signal indicating whether or not to raise is applied to the raised signal line 612. Since this raised signal line 612 is branched and connected to each output stage, it is possible to select whether or not to raise all the outputs. Note that this command is usually input as a hidden command since it is input after the IC inspection in order to avoid a defect due to IC gradation reversal. In this case, the raising current source 572 is also formed in the same size as that in FIG.

  In order to make a finer adjustment, the current raising rate can be adjusted more finely by changing the number of current sources 572 that are provided with a plurality of raising current sources 572 and outputting them. FIG. 64 shows a case where two raised current sources are used. If the transistor is designed so that the current output amount of 641 is halved compared to 642, for example, when the current of the lower bit is increased by 20%, only 642 is raised and when it is 10%, only 641 is raised. For example, the current output stage can be realized in accordance with the current capability of the lower-bit transistor, without gradation inversion, and with little gradation skipping.

  Further, there is a method in which the raising by the switching unit 571 by wiring correction in FIG. 57 and the collective raising by the switching means in FIG. 61 are used in combination. The configuration of the output stage at this time is shown in FIG.

  When the current of the lower-bit transistor is increased, gradation inversion is likely to occur at most of the output terminals, so that the current of gradation 128 is raised at all outputs to prevent gradation inversion. Therefore, the switching means 611 is turned on by the raising signal 612.

  On the other hand, in the case of gradation inversion due to the variation of each current source 241 (this may occur when the halftone variation is 0.9% and the luminance difference between adjacent gradations is 0.78%), for each terminal. Make adjustments. Since the number of terminals to be reversed is small, only the relevant terminals are corrected with a laser after inspection. This correction is performed by the switching unit 571, and it is determined by correction whether or not the raised current source 572a is connected. This makes it possible to set a fine gradation current for each terminal.

  Since the way of raising differs depending on the current capacity of the transistor (influences on all terminals) and the case of transistor variations (different on each terminal), the function that can be adjusted for each terminal as shown in FIG. By providing a function that can be adjusted, repair after inspection can be performed in a short time, and cost can be reduced by improving work efficiency. Moreover, there is an advantage that the yield increases and the cost decreases because more ICs pass the repair.

  FIG. 70 shows a schematic configuration of the driver IC when the current output stage is formed by a current copier configuration as indicated by 736 in FIG.

  In the current copier circuit, an input current is passed through the drive transistor 731 via the switches 734 and 735, and the voltage at the node 742 is determined according to the amount of current flowing. In order to hold this voltage, a storage capacitor 732 is provided to hold the voltage by storing charges. After storing the input current, the switches 734 and 735 are turned off to save the input current. When the current is output, the transistor 733 is turned on, so that a current corresponding to the amount of charge stored in the storage capacitor 732 flows to the output 731. Since the input current is stored and output using the drain current-gate voltage characteristics of the same drive transistor 731, there is an advantage that the same current as the input current can be output regardless of variations in transistor characteristics.

  Further, the current copier circuit has a memory function in order to output the input current once stored in the storage capacitor 732. For this reason, after the input data is distributed to the output terminals, the current copier circuit can have a function of a latch unit for aligning the data output timing. Thereby, the video signal transferred serially in the configuration of FIG. 70 can be distributed to each output without using the latch unit.

  Since the analog current can be held in the current copier circuit, the video signal is converted in advance into a gradation current signal 730 that is an analog current corresponding to the gradation by the digital-analog conversion unit 706 and output from the shift register 21. Each output is distributed according to the signal. A current copier circuit is formed in the current holding means 702 for holding the distributed current.

  As described above, the current copier circuit operates to output the current according to the input current after holding the input current once. Therefore, the current cannot be output during the period in which the input current is stored. When outputting, the gradation current signal 730 cannot be captured.

  Since the current output to the display unit has a problem that it takes time to change to a predetermined current in the pixel circuit, it is desirable to continue outputting the current as long as possible within the horizontal scanning period. Therefore, it is preferable that a current is always output from the source driver IC.

  Therefore, in order to continuously output current even in the output stage of the current copier circuit configuration, two current copier circuits are provided at the same output terminal, and when one stores the gradation current signal 730, the other stores the current. The driver IC is configured to output current to the outside.

  A circuit of the output stage is shown in FIG. Two holding circuits 736a and 736b have a current copier configuration. A select signal 738 is a signal for determining which of the two holding circuits is to be output and which is to store the gradation current signal 730. The select signal 738 changes every horizontal scanning period, and by changing the holding circuit 736 every horizontal scanning period, a current output corresponding to the video signal can be made. By changing the state of the current output transistor 733 of the holding circuit 736 according to the select signal 738, the holding circuit used for output can be determined.

  When both the holding circuits 736 do not output, the selection signal 738 and the inverted output 739 of the selection signal are set to low level. Although 738 and 739 do not necessarily need to enter the reverse phase, both signals must not be at a high level. As another method, 738 and 739 are always in opposite phases, a separate enable signal is provided, and the same operation can be performed by inputting the logical product of 738 and 739 to a signal for controlling the switch 733. .

  The gradation current signal 730 can be distributed to each output by the shift register 21 and the current holding means 702. Next, a circuit that generates the gradation current signal 730 will be described. A digital-analog conversion unit 706 is provided to convert a video signal that is a logic signal into a gradation current signal 730 that is an analog signal, and a current corresponding to the video signal is output. A circuit example of the digital-analog converter 706 is shown in FIG.

A current corresponding to each bit of the video signal is input from the outside, and the switch 712 is controlled by the gradation signal 711 corresponding to the current value for the corresponding current (gradation reference current 1 to gradation reference current 8). Thus, the gradation current signal 730 corresponding to the gradation signal 711 is output. When gradation signal 1 (711a) to gradation signal 8 (711h) are sequentially assigned from the least significant bit to the most significant bit, twice the gradation reference current 1 (700c) is the gradation reference current 2 (700d). In general, the current value is set and inputted so that twice the gradation reference current n becomes the gradation reference current (n + 1). (Where n is an integer greater than or equal to 1 and less than the number of bits)
As a result, the sum of the gradation reference current 700 in which the switch 712 is in a conductive state is output as the gradation current signal 730.

  Next, a method of creating the gradation reference current 700 and inputting it to the digital-analog conversion unit 706 will be described.

  As shown in FIG. 78, the gradation reference current 700 is generated by the gradation reference current generation unit 704. A gradation reference current 700 corresponding to the bit of the video signal is output by a current mirror configuration or the like based on a reference current 781 that sets how much current per gradation is to be set. Here, in the case of 8-bit output, there are 8 outputs of the gradation reference current 700. Since it is necessary to accurately output a current such that (current value of gradation reference current n) × 2 = (current value of gradation reference current (n + 1)), the number of mirrored transistors 782 is changed. Thus, it is preferable to change the output current. This method has a drawback that the gradation is high but the circuit area is large. On the other hand, the number of transistors 782 that generate each gradation reference current 700 is one for each forward current, and the gradation reference currents 1 to 8 can be changed by changing the channel width. Since it does not exactly match the channel width, it is necessary to change the channel width according to the process by simulation. For this reason, there is a possibility that the gradation is deteriorated as compared with the method of arranging only the number. Therefore, as shown in FIG. 78, the gradation reference currents are grouped into the low gradation part and the high gradation part, and the current value is changed by changing the channel width between the low gradation part and the high gradation part. The current is changed between the gradation parts and between the high gradation parts by changing the number of transistors.

  In FIG. 78, the low gradation part is the lower 2 bits and the high gradation part is the upper 6 bits, and the transistor surrounded by a dotted line indicated by 783 has a channel of about 1/4 of the transistor surrounded by the dotted line indicated by 784. By forming with a width (-10% or more and less than + 50%, which varies depending on the process), it is possible to realize the gradation reference current generation unit 704 having a small circuit scale while maintaining gradation.

Since there is one circuit for the driver IC, the current may be changed depending on the number of transistors as shown in FIG. (Because the total circuit area is 10% or less)
The reference current 781 can be realized by configuring a constant current source with a resistor, an operational amplifier, etc. as shown in FIG. It is also possible to change the current value of the reference current 781 by 88 control data. The control of the reference current 781 is useful for suppressing power, preventing burn-in, and improving contrast.

  The gradation reference current 700 formed as described above may be input to the digital-analog conversion unit 706. However, when directly connected, when a plurality of source driver ICs 36 are connected, an error of 1% or less in all chips is obtained. Thus, it becomes difficult to supply the gradation reference current 700.

  When the reference current generation unit 703 and the gradation reference current generation unit 704 are provided for each chip, the mean square of the variation in the reference current generation unit 703 in FIG. 81 and the variation in the current mirror in FIG. 78 or FIG. Variation occurs in the gray scale reference current 700, the current value of a certain gray scale may be different depending on the chip, and luminance unevenness occurs in each chip. To reduce the variation due to the mirror ratio deviation of the current mirror, it can be realized by increasing the transistor size of 782 and 801. To reduce the variation to 1% or less, a channel size of 10,000 square microns or more is required. It becomes.

  In order to supply the gradation reference current 700 to each chip with a small size and no variation, the gradation reference current is generated by using one gradation reference current generation 704 from one reference current generation part 703 for one display unit. 700 is generated and distributed to each chip. This concept is illustrated in FIG.

  By supplying the gradation reference current 704 generated by the source driver 36a to all the chips including 36a, a current having no variation is supplied to each chip. Here, it is necessary to prevent the gradation reference current 700 from being supplied to two or more source driver ICs 36 simultaneously. When the current is different from the voltage, it is divided when connected to a plurality of drivers, and the gradation reference current value flowing through one driver IC is different. Therefore, one IC generates the gradation current signal 730 corresponding to the video signal by using the switch 712 included in the digital-analog conversion unit 706 so that the plurality of driver ICs 36 do not capture the gradation reference current 700 at the same time. In other ICs, it was considered that all the switches 712 are in a non-conductive state.

  The gradation current signal 730 is necessary when supplying a current to the current holding means 702 and outputting a signal to be taken in one of the outputs of the shift register 21. In other words, the period from when the start pulse 16 is input and when the pulse is output from the carry output 701 to the cascaded next stage IC 36 is a period in which the gradation current signal 730 is required.

  Therefore, the switch 712 of the digital-analog conversion unit 706 is always in a non-conductive state regardless of the gradation signal 711 except during the period when the shift register 21 is outputting. In order to realize this, a chip enable signal generation unit 707 is provided so that the switch 712 is always in a non-conductive state except during the shift register operation. The chip enable signal generation unit 707 outputs a pulse only until the start pulse 16 is input and the carry output 701 is performed, and permits the video signal to be converted into an analog current. More precisely, it is a period during which the shift register output 719 is output within the same chip. Since the relationship between the start pulse 16 and the shift register output 719 and the carry output 701 and the shift register output 719 may change depending on the relationship between the input data and the start pulse 16 and the configuration 21 of the shift register, the start pulse 16 and the carry output 701 The enable signal 821 is output after adjusting the period. A circuit diagram of the digital-analog converter 706 corresponding to the enable signal is shown in FIG. The chip enable signal 821 is in a high level state after the start pulse 16 is input until the carry output 710 is performed, and the gradation reference current 700 is output to the gradation current signal 730 in accordance with the gradation signal 711. In other periods, the chip enable signal 821 is a low level signal, so that the switch 712 is always non-conductive and no current is supplied.

  FIG. 83 shows a timing chart of the chip enable signal 821, the select signal 738, the gradation current signal 738, and the gradation signal 711 of the driver IC (chip 1) in one horizontal scanning period.

  The select signal 738 changes for each horizontal scanning period by the timing pulse 29, and it is determined which of the two holding circuits 736 for one output stores the gradation current signal 738 and the other outputs the stored current. Decide. In the period 831a, current is output from the holding circuit A (736a), and the gradation current signal 730 is stored in the holding circuit B (736b).

  The gradation current signal 730 is stored one by one in order, and the output to be stored is determined by the shift register output 719. Further, since the wiring for distributing the reference current to the plurality of driver ICs is used, the digital / analog conversion unit 706 is operated by the chip enable signal 821 only during the period during which the shift register is operating in order to prevent the diversion. A current signal 738 flows. The chip enable signal 821 of the chip 1 becomes a high level signal only during the period 832a, which is the period during which the shift register is operating in the chip 1, and the gradation current signal 738 flows. During the period 832b (a shift register other than chip 1 is operating), the chip enable signal 821 is at a low level, and the gradation current signal 738 does not flow. Therefore, since the gradation reference current signal 700 is always input only to one driver IC, it is possible to branch and wire to a plurality of driver ICs as shown in FIG. Compared with distribution using a current mirror, the same current can be supplied accurately because the distribution is divided by time.

  In the method in which a current copier is provided for each output and the gradation current is distributed to each output, the same current as the stored current can be output regardless of the characteristic variation of the drive transistor 731. Therefore, output variation hardly occurs. . However, the output current may vary due to a phenomenon called “penetration”.

When the signal of the gate signal line 741 is set to the high level in the holding circuit of FIG. 73, the gradation current is stored. For example, if a white gradation current is stored, as shown in FIG. 74, the drain current in the driving transistor 731 becomes a white gradation current (here, Iw). At that time, the voltage at the node 742 becomes Vw from the current-voltage characteristics of the driving transistor 731 (FIG. 75). (Period 747)
The period 747 ends and the gate signal line 741 changes to a low level in order to finish storing the current in the holding circuit 736. At this time, the voltage at the node 742 also decreases by VG due to capacitive coupling through the gate capacitance of the transistor 735a when the voltage of the gate signal line 741 is lowered. As a result, the drain current of the driving transistor 731 also decreases from Iw by IG.

  This “piercing” may cause the output current to change depending on the terminal. For example, it is assumed that there is a drive transistor 731 having current-voltage characteristics as indicated by 765 and 766 in FIG. When the voltage of the node 742, that is, the gate voltage of the driving transistor 731 changes to VG due to punching, the drain current becomes Iw1 in the driving transistor of 765, and the drain current becomes Iw2 in the driving transistor of 766, and this current passes through the output signal line 737. It flows to the outside and the output current varies. When the difference between Iw2 and Iw1 is 1% or more with respect to the two average currents, the display quality is affected as luminance unevenness.

  The voltage change amount VG at the node 742 is expressed as VG = Vga × Cgs / (Cgs + Cs), where Cgs is the gate capacitance of the transistor 735, Cs is the capacitance of the storage capacitor 732, and Vga is the amplitude of the gate signal line 741.

  To reduce VG, Cgs or Vga is decreased or Cs is increased. The method of increasing Cs is actually difficult because the chip size increases. Vga basically has an amplitude corresponding to the analog power supply voltage. When this voltage is lowered, the voltage amplitude at the output terminal is lowered, so that the dynamic range of the current that can be outputted is lowered. Further, if the high level voltage of only the gate signal line 741 is lowered, the power supply for the gate signal line 741 is required, so the number of power supplies increases. Since the increase in the number of power supplies leads to an increase in power supply circuits, it is difficult to realize this method.

  Therefore, in the present invention, the gate capacitance Cgs of the transistor 735 is considered to be reduced. When the size of the transistor 735 is simply reduced, the leakage current at the time of turning off increases and the charge held in the storage capacitor 732 moves through the transistor 735, so that the potential of the node 742 changes and a predetermined current cannot flow. A problem occurs.

  It was considered that the transistor 735 is divided into at least two, and the transistor closest to the storage capacitor 732 is made smaller. FIG. 77 shows a circuit of the current holding means 702 when divided into two.

  The transistor 735 is divided into two to have two structures 775 and 772. Compared with the transistor 775, 772 has a smaller channel size. In addition, signal lines connected to the respective gate electrodes are separate, and the transistor 772 is brought into a non-conductive state earlier than 775 by the control of the gate enable signal 771. A timing chart is shown in FIG.

  The advantage of using a plurality of transistors is that the waveforms of the gate signal lines of the two transistors are different, the transistor 772 close to the storage capacitor 732 is first turned off, and then 775 is turned off. “Punch-through” depends on the gate capacitance Cg1 and storage capacitance Cs of the transistor 772 and the gate amplitude Vgate. Since Cgs> Cg1, VG itself can be reduced. Further, in order to hold the charge in the storage capacitor 732, the gate signal line 741 is changed to a low level so that 775 becomes non-conductive after 772 is completely non-conductive. 775 is designed to increase the channel width / channel length value of the transistor in order to reduce the leakage current. There is an advantage that leakage current is reduced by connecting two transistors in series. Further, since the transistor 772 is inserted between the transistor 775 and the storage capacitor 732 in a non-conductive state, there is an advantage that the “piercing” to the node 742 due to the gate signal of 775a does not occur.

  In this way, the transistor connected between the gate and drain electrodes of the drive transistor 731 is divided into a plurality of transistors, and the transistor closest to the storage capacitor 732 is made smaller in channel size and is non-conductive earlier than other transistors. By doing so, it is possible to realize a reduction in the amount of penetration without problems such as charge leakage.

  Furthermore, it is preferable that the value of W / L is small with respect to (channel width) / (channel length) (hereinafter referred to as W / L) of the drive transistor 731.

  FIG. 84 shows current-voltage characteristics. The smaller the value of W / L, the smaller the slope. After storing the grayscale current signal 730, the decrease in the amount of current when the gate voltage of the drive transistor 731 decreases by VG due to “push-through” is a curve 841. Is larger than the curve 842. Therefore, in order to suppress a decrease in drain current due to “penetration”, the W / L of the driving transistor is preferably set to 0.5 or less. In this case, the amount of decrease is 1% or less with respect to the set current (Iw). The lower limit value needs to be 0.002 or more because of the effect of increasing the chip area by extending the minimum creation dimension of the channel width and the channel length.

  As described above, an output stage using a current copier circuit is formed to realize a driver IC with small output variation.

  In a source driver for a large screen panel, there is a problem that the signal line frequency becomes high because the video signal needs to be transferred at high speed, and as a result, electromagnetic noise is emitted. In addition, for televisions and the like, the number of input signal line bits also increases, and there is a problem that the number of signal lines increases.

  Therefore, it was decided to transmit a video signal with a small amplitude signal. FIG. 85 shows connections between the source driver 852, the gate driver 851, the controller 854, and the power supply module 853 at that time. Among them, the small-amplitude signal transmission is performed by a clock 858 having a high signal line frequency, a synchronization signal 857, and a video signal line 856.

  The transmission format of the video signal line 856 is shown in FIG. A period (data transfer period 865) and a blanking period (866) in which data output to the pixels are transferred are formed within one horizontal scanning period 864. The blanking period does not necessarily exist.

  The data transfer period 865 is divided into the number of panel source signal lines (in the case of a color panel, the number of signal lines / the number of colors (generally three colors)). The divided period is a period 862. Within this period 862, 1-bit precharge flag (862) for determining whether or not voltage application corresponding to each color data (861) of red, green and blue and gray scale is inserted at the beginning of the horizontal period is transferred via the video signal line 856. Is done. The video signal data 861 and the precharge flag 862 can be transferred by an arbitrary method from the case where all bits are transferred simultaneously in parallel to the case where the bits are transferred serially bit by bit due to restrictions on the transfer signal rate and the number of signal lines. It is.

  In a large current driver, the current cannot be changed to a predetermined value within one horizontal scanning period due to an increase in the source signal line stray capacitance due to a large panel size and a shortening of the horizontal scanning period due to an increase in the number of pixels. The problem becomes noticeable. Therefore, it is indispensable to change the state of the source signal line to the vicinity of the predetermined gradation by the voltage once before displaying the predetermined gradation by the current, and then change it to the predetermined current by the current.

  A configuration example of the source driver is shown in FIG. Here, the source driver is the source driver 852 in FIG. Since the video signal is transmitted with a small amplitude signal together with the clock and the synchronization signal, it is input to the differential input receiver 893 for level conversion on the source driver side. The video signal is converted into gradation data 386 of CMOS or TTL level. The gradation data 386 is input to the shift register / latch unit 384 and the precharge voltage conversion unit 884. The gradation data 386 is distributed to each output by the shift register and latch unit 384, and the distributed gradation data is converted into a current amount corresponding to the gradation by the current output stage 23. This makes it possible to output current according to the gradation. On the other hand, the gradation data is input to the precharge voltage converter 884 at the same time. The precharge voltage converter 884 outputs a voltage corresponding to the gradation data as a signal 885 with a circuit configuration as shown in FIG. The output voltage can be changed according to the conversion matrix of the precharge value conversion unit 882 and the value of the resistance element 883.

  An equivalent circuit between the pixel and the source driver during the current writing period is the circuit shown in FIG. At this time, assuming that the current during white display is I3 and the current during black display is I1, the fluctuation range of the precharge voltage output is the range from V3 to V1 from FIG. The values of V3 and V1 vary depending on the channel size of the pixel driving transistor 62. For example, the difference between V3 and V1 increases as the channel width decreases. In the present invention, two resistance elements indicated by 883 in FIG. 88 are externally arranged so that different voltage values can be output depending on the panel (configuration of the pixel transistor), and the resistance value can be arbitrarily set. Voltage output to any panel. Since the current-luminance characteristics of organic light emitting elements are generally different for red, green, and blue, the values of I1 and I3 are different for each color, and as a result, V1 and V3 are also different for each color. Therefore, the precharge voltage converter 884 shown in FIG. 88 is necessary for the source driver for three circuits. The external resistance value is different for each color. In FIG. 85 and FIG. 89, one circuit is described, but there are actually three circuits of red, green, and blue.

  As described above, the voltage output in accordance with the gradation is then distributed to each output by the distribution unit and hold unit 383. As a result, a current corresponding to the gradation and a current corresponding to the gradation were distributed to each output. The current / voltage selection unit 385 selects whether to output current or voltage.

  Which of the current voltages is selected is determined by the precharge voltage application determination unit 56. The precharge voltage application determination unit 56 makes a determination based on the precharge pulse 451 and the precharge enable 895, and applies a voltage only when the precharge pulse 451 is input and the precharge enable 895 outputs a precharge signal. Like that.

  Thereby, as shown in the output 901 of FIG. 90, when the voltage corresponding to the gradation data Dn (n is a natural number) is VDn and the corresponding current is IDn, the precharge determination signal 383 becomes high level and precharges. In this case, IDn is output after VDn is output within one horizontal scanning period. (VDn application period depends on the pulse width of the precharge pulse 451) On the other hand, when it is at low level, VDn is not output and only IDn is output during one horizontal scanning period. By using the precharge determination signal 383 (a rough time chart of current output or voltage output is shown in FIG. 47), in the low gradation part that hardly changes to the current corresponding to the predetermined gradation value, the voltage is After roughly changing the state of the source signal line, the source signal line is changed to a predetermined current value by the current. On the other hand, in the high gradation part and in the second and subsequent lines when the same gradation is continuously displayed in a plurality of lines, the source signal line can be easily changed to a predetermined current value in the high gradation part. In the case of row continuation, the state of the source signal line does not need to change, and therefore it is not necessary to change the voltage to a predetermined gradation value by voltage, so that the precharge is not controlled by the precharge determination signal 383. It becomes possible. (It is better not to apply voltage if there is a risk of luminance unevenness due to variations in the characteristics of the drive transistor 62 of the pixel circuit if the voltage is changed in this state.) The precharge determination signal 383 is thus obtained from the source signal line. There is an advantage that it is possible to decide whether or not to precharge depending on the situation. Therefore, it is necessary to transfer even if the amount of data sent through the video signal line 856 increases by 1 bit for each color.

  The precharge pulse 451 inputs a precharge period to the source driver through the command line 847 so that the pulse width of the precharge pulse 451 can be changed according to the precharge period set value. Thereby, voltage output is performed in a minimum time required for precharging according to the screen size, and the current output period for setting the predetermined brightness is made as long as possible. Make unevenness correction easier. In order to reduce the number of signal lines of the command line 847, as shown in FIG. 87, 1-bit data is sent to the source driver by serial transfer. In addition to the precharge period setting 872, commands required for the source driver are only a reference current setting 871 for changing the reference current value and a driver output enable signal. These signals are not frequently rewritten, and even if they are frequently performed, they may be rewritten once within one horizontal scanning period. In the example of FIG. 87, the total number of bits is 15 bits, and since the clock 871 for the shift register of the source driver may be slower than the time required to change within one horizontal scanning period, signal transmission is possible without the influence of electromagnetic noise. It is. Therefore, the number of signal lines may be one. Also, the data flowing on the command line 847 is also determined by, for example, the reference current setting 871, then the precharge period setting 872, and the output enable signal in the order of the upper bit to the lower bit for 8 bits from the clock next to the timing pulse 849. This eliminates the need for a command discrimination line (address setting). Thus, the source driver can be set with a small number of signal lines. Note that the reference current generating unit 891 to which the reference current setting signal is input has a configuration in which the reference current can be changed by the electronic volume, and the reference current is changed by changing the electronic volume value by the setting signal (see FIG. 8 shows a configuration example).

  When the video signal is composed of even bits for each color (for example, 10 bits for each color, a total of 30 bits), the precharge flag 862 is added to each color by 1 bit, so the total number of bits is always an odd number. (33 bits in the example) When performing low-amplitude signal transmission, the wiring is usually sent by twisted pair wires. When sending a 33-bit signal line, 66 lines are required if the transfer rate is the same as that of the driver. In this case, since the number of wirings is large, the normal transfer speed is transferred at a constant multiple of the driver clock, and the number of wirings is reduced accordingly. For example, when sending at double speed, 34 bits can be transferred by transferring 17 bits at a time. Of these, data is transferred at double speed by putting data in 33 bits. However, one bit of blank data is sent compared to the actual transfer capability of 34 bits. Similarly, when data is transferred at even multiple speed, blank data is always sent for one bit of odd-bit data, and it is understood that the utilization efficiency of the signal line is low. That is, even if the data for one bit is increased, the transfer rate (double clock speed) and the number of signal lines are not affected.

  Therefore, in the present invention, when the data / command flag 911 is added to the red / green / blue video signals and the precharge flag, and the value of the data / command flag 911 is 1, for example, the video signal and the precharge flag are transferred. When 0, it is possible to make various register settings of the source driver. 91A shows data transfer, FIG. 91B shows the configuration of each bit when various registers are set, and FIG. 92 shows the transfer timing of data transfer and various register settings. Various register settings of the source driver are set by the data / command flag 911 using the blanking period after transferring all the video signals of each color and the precharge flag without one horizontal scanning period. Here, as shown in FIG. 91B, the reference current and the period for applying the precharge voltage are set.

  In this way, the command line 847 in FIG. 85 is not necessary, and the number of signal lines can be reduced.

  A block diagram of the source driver is shown in FIG. In order to separate the command data and the video signal from the video signal line 856, a video signal / command separation unit 931 which is a circuit for converting the low amplitude signal to the CMOS level is different from the configuration of FIG.

  As described above, in the source driver IC that needs to transfer the precharge flag in synchronization with the video signal line and perform various register settings, the video signal line and the precharge flag or the video signal line and the precharge flag are set. And various register settings can be transferred at high speed with a low amplitude signal using the same signal line. As a result, the number of wires necessary for the precharge flag and the number of wires for setting various registers can be reduced, and electromagnetic noise during high-speed transfer can be reduced.

  In a small-sized display panel, there is a spatial restriction on the module arrangement, and it is necessary to reduce the number of signal lines drawn out of the panel as much as possible. Since the number of display dots is smaller than that of a large panel, the transfer rate of the video signal line is low. Therefore, as shown in FIGS. 94 and 95, for the gradation display data (red, green, and blue color data, here, R data, G data, and B data) and the gradation display data on the video signal line 856, In addition to multiplexing a precharge flag 862 for determining whether or not to perform precharge, gate driver control data 951 is further transmitted. Signal lines necessary for controlling both the gate driver A (851a) and the gate driver B (851b) are transmitted. The signals to be transmitted are a shift register operation clock, a start pulse, an output enable signal, and a signal for determining a shift direction. Since the output enable signal may change the signal line state in units of several microseconds, the gate driver control data 951 is transmitted not only in the data transfer period 962 but also in the blanking period 963 in FIG. Therefore, as shown in FIG. 95B, in addition to the source driver setting signal, the gate driver control data 951 is transferred. As a result, the signal lines drawn from the panel can be composed of a minimum of two pairs of twist lines and three signal lines in addition to the power supply lines.

  If the number of signal lines is reduced, the transfer rate increases, so that the power consumption of the clock generator attached to the transmission-side controller 854 increases. In general, most of the power for small amplitude transmission is consumed by the clock generator. Therefore, in a device that requires low power consumption, the number of twist lines used for the video signal line 856 is increased, and the power consumption is reduced by lowering the transfer rate. (The power consumed by the signal line is about one-tenth to one-twentieth of the power consumed by the clock generator.) The data string shown in FIG. 95 (a) sent during the period indicated by 964 in FIG. These may be sent serially, or some or all of them may be transferred in parallel according to the number of video signal lines 856.

  In this manner, the data of the video signal line 856 transmitted with a small amplitude is separated by the source driver 852. An internal block of the source driver 852 is shown in FIG. Video signal / command separation for outputting clock 858 and video signal line 856, gradation data 386 synchronized with source driver clock 871 created from clock 858 from start pulse 848, precharge determination signal 383 and gate driver control line 941 It is characterized by having a portion 931. The gate driver control signal is always transmitted corresponding to the video signal and command as shown in FIG. 95, and therefore can be demodulated to a signal synchronized with the source driver clock 871 as shown in FIG. By doing so, it is not necessary to draw out gate signal lines to the outside of the panel, and a display panel with a small number of signal lines can be realized. Further, by outputting in synchronization with the source driver clock 871, there is an advantage that the timing of the source driver and the gate driver can be easily matched. Further, since the control line from the controller 854 to the gate driver 851 becomes unnecessary, the number of output terminals of the controller 854 is reduced, and the controller 851 can be created with a smaller package.

  The configuration in FIG. 98 differs from the configuration in FIG. 93 in the block that generates and outputs the precharge voltage. In FIG. 93, a voltage corresponding to the video signal is generated and distributed to each output using an analog latch. However, in FIG. 98, a plurality of voltage outputs of the precharge voltage generator 981 determined by the voltage setting line 986 are output to each output stage. The precharge voltage selection and application determination unit 982 determines which of a plurality of voltages is output or whether only current is output. As a result, the distribution unit and the hold unit 383 become unnecessary. Compared with a large panel, a small current panel has a long horizontal scanning period and a small floating capacitance of a source signal line, so that a predetermined current value is easy to write. Therefore, in this source driver, the number of generated voltage values is reduced and the circuit scale is reduced on the premise that no voltage is applied in a high gradation portion where writing is possible only with current. In this example, a ternary voltage output is used. The number of voltage values may be changed from 1 to 7 as necessary.

A method for outputting a precharge voltage in accordance with video signal data will be described. A video signal and a precharge flag are transmitted as a pair from the video signal line 856 by the method of FIG. In the case of a color panel, a pair of red, green and blue is transmitted. Since precharge is performed by the same method, a red signal is used here. The R precharge flag 862a and the R data 861a transmitted as a pair are input to the video signal / command separation unit 931. Here, it is converted to a CMOS level, and becomes a precharge determination signal 383 and gradation data 386, respectively. In order to distribute the signals sent one pixel at a time to each output, the signals are input to the shift register and latch unit 384. After the distribution, the gradation data 386 is input to the current output stage 23 via the gradation data line 985, and a current corresponding to the gradation is output from 104. On the other hand, the precharge determination signal 383 is output to the precharge determination line 984. As shown in FIG. 100, the precharge voltage selection and application determination unit 982 controls the decode unit 1001 and the selection unit 1004 by using the precharge determination line 984 and the precharge pulse 451, and outputs the gradation current 104 or the precharge voltage. It is determined whether any one of 983 is output. Here, since one signal is selected from the four inputs, the precharge determination line 984 needs to have a 2-bit width. In general, if the number of bits of the precharge determination line 984 is N (N: natural number), the number of bits is required such that the value of ^2N is equal to or greater than (the number of precharge voltages + 1).

  The precharge pulse 451 is a signal for determining a voltage output period within one horizontal scanning period, as indicated by 473 in FIG. Accordingly, even when any precharge voltage 983 is output by the precharge determination line 984, the voltage is output only during the input period of the precharge pulse 451.

  FIG. 101 shows the relationship between the precharge pulse 451 and the precharge determination line 984 and the output 1005. Thus, by controlling the signal input to the precharge determination line 984 from the controller, it is possible to provide a period for outputting the precharge voltage corresponding to the video signal.

  The precharge voltage is generated by a precharge voltage generator 981. A configuration example of the internal circuit is shown in FIG. Each voltage is generated by resistance division. (An operational amplifier is generally connected to the 983 output) Vp1 is determined by resistance elements 992a and 992b. On the other hand, Vp3 has a configuration in which the voltage can be changed for each color because the required current value differs depending on the emission color. The resistor element 997 and the voltage selection unit 994 are used to select any voltage from Vs1 to Vs4. In the display device having a pixel circuit as shown in FIG. 6, the relationship between the source signal line current (= current flowing through the EL element 63) and the voltage of the source signal line 60 is the current-voltage characteristic of the drive transistor 62 in FIG. Since the values coincide with each other, a current shift per gradation due to a difference in light emission efficiency of the EL element between green and blue appears as a shift of the source signal line voltage. Considering from 0 to 2 gradations that require a precharge voltage, blue requires a lot of current because its luminous efficiency is lower than that of green. It becomes a point. As a result, the voltage value is also different. The voltage selection unit 994 is controlled by the voltage setting line 986. For example, 994c selects Vs4 (995c), and 994b selects Vs1 (995a), so that the precharge voltage value is changed depending on the color as shown in FIG. It is possible. A predetermined voltage can be generated by determining the resistance values of 997 and 998 that match the characteristics of the driving transistor 62. The voltage setting line 986 can be set from the outside, and as shown in FIG. 95 (b), a precharge voltage setting 953 is input during the command period, and the video signal / command separation unit 931 separates the video signal from the voltage setting line. 986 can be taken out. As a result, when performing different voltage settings for each color, it can be realized without newly increasing the number of external signal lines. In FIG. 98, only three precharge voltages 983 are shown, but this shows an example of a single color, and in the case of multi-color, three precharge voltages 983 are required for each color, for a total of nine. Become. The precharge voltage selection and application determination unit 982 has three voltage inputs. This is because the display color is determined for each output, and it is sufficient to input three voltages corresponding to the output color.

  Note that when eight or more voltage values are required, the circuit scale of the decoding unit 1001 and the selection unit 1004 of FIG. 100 becomes large, so the circuit configuration of FIG. 89 is better.

  The configuration shown in FIG. 95, FIG. 98 or FIG. 91 and FIG. 93 may be determined based on the panel size and the number of pixels.

  As a result, a source driver IC capable of outputting current and voltage can be realized with a small number of signal lines.

  The problem with current driver ICs is that the change in the current written to the pixel is slow due to insufficient charge / discharge of the source signal line stray capacitance due to the small output current value, especially in the low gradation part. The time Δt required for the current to change is expressed by Δt = C × ΔV / I (where C is the source line capacitance, ΔV is the source line voltage change amount, and I is the current flowing through the source signal line). Therefore, it can be seen that the lower the gradation, the longer it takes to change. It was also found that the change from black to white took longer when changing from white to black and from black to white.

  For example, if a source signal line current of 10 nA is passed during white display and a source signal line current of 0 nA is displayed during black display, the change in the source signal line current from white to black becomes the waveform shown in FIG. The source signal line current changes to the waveform shown in FIG.

  When one frame is scanned at 60 Hz on a QCIF + (176 × 220 pixel) panel, one horizontal scanning period is approximately 70 μsec. The change in 70 μs from the initial state changes from 94% with respect to the target as shown in FIG. 104 from white to black, but only 5% with respect to the target as shown in FIG. 105 from black to white. It has not changed.

  The reason why the difference between 10 nA and 0 nA is so large is that the change in the value of the source signal line voltage with respect to the source signal line current is a non-linear change. The relationship between the source signal line current and voltage is shown in FIG. The relationship between the current and voltage is determined by the current-voltage characteristic (1063) of the driving transistor 62, and the voltage corresponding to the curve of 1063 becomes the source signal line voltage value according to the current of the source signal line. In the expression Δt = C × ΔV / I of the time required for the current change, the current of the source driver is 0 at the time of change from black to white, and the current of the source driver is 0 at the time of change from white to black. In the initial state, I = 10 nA. Then, when Δt is the same as 70 μs, it can be seen that ΔV is almost equal. When the source potential increases by ΔV from the 10 nA state and when the source potential decreases by ΔV from the 0 nA state, the amount of current change is completely different from the characteristic of the curve 1063. In the direction in which the potential increases, it decreases from 10 nA to 0.6 nA as indicated by 1061, while in the direction in which the potential decreases, it changes only from 0 nA to 0.5 nA. As a result, the current changes as shown in FIGS. 104 and 105 are obtained.

  Here, a change between 10 nA and 0 nA has been described as an example. However, in any combination of gradations, a change from a high gradation to a low gradation is similarly performed from a low gradation to a high gradation. Faster than change.

  Therefore, in the present invention, a method for accelerating the change from a low gradation to a high gradation with a slow change speed has been devised.

  In order to speed up the change, it is necessary to reduce the source signal line capacitance, reduce the voltage change amount, or increase the current. The source signal line capacitance cannot be changed because it is determined by the panel size. In order to reduce the amount of voltage change, the current voltage characteristics of the driving transistor must be changed. Specifically, the channel width of the transistor must be increased or the channel length must be decreased. When the channel width is increased, the transistor size increases, and a small high-definition panel with a small area for one pixel cannot take measures. On the other hand, when the channel length is shortened, the Early effect is more greatly generated. When the drain voltage of the driving transistor 62 is different between writing and EL emission (periods of FIG. 7A and FIG. 7B), the Early effect causes In each case, there arises a problem that the drain current value changes, so that the channel length cannot be shortened. Therefore, it was considered to increase the source signal line current.

  FIG. 108 shows a source driver current output waveform according to the present invention when the current I is written to a certain pixel. It is characterized in that a period during which a current 10 times the predetermined current flows is provided for 10 μsec at the beginning of the horizontal scanning period. For example, as shown in FIG. 107, the current change is changed from the conventional 1072 to 1071 by passing the current 10 times, and the predetermined current can be written in 70 μsec. Thus, by providing a period for increasing the current flowing through the source signal line at the beginning of one horizontal scanning period, the change in the current value is accelerated and a predetermined current can be written.

  When the current is output 10 times the predetermined value, it is necessary to calculate a value 10 times the predetermined current, and it is necessary to provide a function capable of flowing 10 times the current on the source driver side. For this purpose, there is a problem that an arithmetic circuit is required or the current source of the current output stage of the source driver must be increased by 10 times, resulting in an increase in circuit scale. In addition, when the current value per gradation differs depending on the display color, it is necessary to change the magnification for each gradation. Therefore, processing becomes complicated.

  Therefore, in the present invention, since the change from the low gradation to the high gradation is difficult to change, and the gradation 0 changes most slowly even at the low gradation, the gradation 0 is changed to the next gradation. The current value (here, Ip1) is applied at the beginning of one horizontal scanning period and then a predetermined current is applied. The configuration is such that it can be changed to a predetermined current value within one horizontal scanning period. When the predetermined gradation value is larger than Ip1, the current from the gradation 0 to the predetermined gradation is set to 1 over the entire gradation area by causing the predetermined gradation current to flow even during the period of flowing the current of Ip1. It became possible to write within the horizontal scanning period. In this case, a multiplier is not necessary because it is sufficient to provide a period for inserting Ip1 only when the video signal is less than a certain gradation. In the output stage, it is only necessary to provide one current source for outputting Ip1 for each output. The concept is shown in FIG. This can be realized by providing a current source Ip1 (1033) for precharging at the current output 104 in addition to the current source for gradation display. Since this current Ip1 is used only for the purpose of accelerating the speed of changing to a predetermined gradation, there may be variations between adjacent terminals. Therefore, the same current is used as compared with a transistor constituting a current source used for gradation display. It is possible to reduce the total transistor area for output.

  Further, the optimum value of the current Ip1 is determined by the source line capacitance and the current-voltage characteristics of the pixel transistor, and does not depend on the light emission efficiency of the EL element 63. Therefore, it is sufficient that a common current value is entered for each color, and there is no need for individual adjustment for each color, and a small circuit can be configured.

  FIG. 109 shows the configuration of the source driver IC in the case where a function for outputting Ip1 is provided at the beginning of the horizontal scanning period. Here, the current Ip1 output at the beginning of the horizontal scanning period is referred to as a precharge current. A precharge reference current generator 1092 for generating a precharge current, a precharge current output stage 1094 for determining whether to output to the source signal line, and a pulse generator 1097 for setting the period of the precharge current are provided. Is a feature.

  Whether or not to output a precharge current is determined by a precharge determination signal 383. Since the precharge determination signal 383 is transmitted in synchronization with the gradation data 386, whether or not to provide a period for outputting the precharge current for each pixel, and when a plurality of precharge currents are provided, It is possible to set which one to select. It is distributed to each output by the shift register and latch unit 384 together with the gradation data 386 so as to be distributed to each output. The gradation data is input as a gradation data line 985 to the current output stage 23 provided for each output. In the current output stage 23, a current amount corresponding to the reference current value created by the gradation data line 985 and the reference current generator 891 is output to 1093. FIG. 110 shows the configuration of the reference current generation unit 891 and the current output stage 23 in the case of a multi-color driver with an example in which the gradation data line 985 is 3 bits. The signal line potential of 1101 is changed by the reference current setting line 934, and the current value of the constant current circuit including the operational amplifier 1103, the resistor 1102, and the transistor is changed. Thus, it can be seen that the current changes according to the value of the reference current setting line 934. The current of the output 1093 is changed by the gradation data line 985 by changing the number of current source transistors 103 connected to the output depending on the value of the gradation data line 985. In general, since the organic EL element has different light emission efficiency for each emission color, it is necessary to change the current per gradation for each emission color. In the present invention, by configuring the resistor 1102 as an element outside the IC, the resistor 1102 can be easily adjusted, and the current value per gradation is changed by the resistance value so as to achieve white balance. On the other hand, the precharge determination line 984 distributed to each output is input to the precharge current output stage. Further, the precharge current output stage 1094 also receives signals from the precharge reference current generator 1092 and the precharge pulse 1098.

  The pulse width of the precharge pulse 1098 is determined by the pulse generator 1097. The pulse generation unit 1097 outputs a precharge pulse 1098 based on the value of the precharge period setting line 1096 from the timing pulse output using the value and timing pulse of the current precharge period setting line 1096 and a counter circuit with a clock. I have to.

  A precharge reference current generator 1092 that determines the value of the precharge current changes the precharge current by inputting a precharge current setting line 1091.

  These two external setting values (the current precharge period setting line 1096 and the precharge current setting line 1091) are applied to the video signal line 856 using the blanking period of the video signal to reduce the input signal lines of the source driver. A setting signal was sent during the ranking period. Therefore, the current precharge period setting line 1096 and the precharge current setting line 1091 are taken out from the video signal line 856 via the video signal / command separator 931.

FIG. 111 shows a circuit configuration of the precharge current output stage 1094 and the precharge reference current generator 1092. (Two examples of multi-color three-color set)
In the precharge current output stage 1094, one of the precharge current source transistors 1112 to 1114 or the gradation current 1093 is output to the output 104 by the determination signal decoding unit 1111 to which the precharge determination line 984 and the precharge pulse 1098 are input. Whether or not to output a precharge current is selected by making the connection.

  As a result, when the precharge pulse 1098 is at a high level, which of the precharge current sources is output depending on the value of the precharge determination line 984, or whether the gradation current is output without the precharge current Can be decided.

  The pre-charge current may be 1 value, but the required current value varies depending on the panel size, that is, the capacitance value. Therefore, when the IC driver is used for general purposes in any size, the current is adjusted for large size and small size. Thus, it is possible to improve versatility by allowing a plurality of items to be output.

  Although the pulse width of the precharge pulse 1098 depends on the panel size and the length of the horizontal scanning period, it is preferably 5 μsec or more and 50% or less of the horizontal scanning period. When a predetermined gradation cannot be written in this range, this is dealt with by increasing the precharge current. The precharge determination signal 383 is controlled so that the value of the gradation data 386 providing a period for inserting the precharge current is applied when the current output from the current output stage 23 is less than the precharge current based on the gradation data 386. That's fine. The precharge determination signal 383 may be differentially input with a small amplitude in the form shown in FIG. 95 in order to reduce the number of input signal lines and to prevent electromagnetic waves.

  In this way, a desired current can be written by inputting a precharge current even when the data in the next row has a higher gradation than the data in the previous row.

  When changing from a high gradation to a low gradation, the target current value can be written almost as shown in FIG. 104, so it can be left as it is. However, it is better to display black exactly for gradation 0 (black). Further, it is possible to emphasize the advantages of improving contrast and displaying black which is a feature of the self-luminous element.

  For this reason, when changing from a gradation other than 0 to 0 gradation, a voltage for displaying black by a voltage is applied at the beginning of the horizontal scanning period, thereby realizing neat black. When a voltage corresponding to the black current is applied to the source signal line, depending on the applied voltage, a phenomenon in which black floats (slightly emits light) is observed depending on the pixel due to variations in the current-voltage characteristics of the drive transistor 62. In order to prevent this, the applied voltage takes into account variations in the current-voltage characteristics, and the drive transistor 62 through which the current flows most is applied with a voltage (precharge voltage) that does not flow in the drive transistor. The brightness variation due to can be prevented.

  FIG. 112 shows the configuration of a source driver that can apply a precharge current or precharge voltage within a horizontal scanning period. A feature is that a precharge voltage generator 981 and a voltage precharge pulse 451 for specifying a period for performing voltage precharge are input so that the precharge voltage can be supplied.

  In the case of performing precharge with a voltage, the source signal line can be sufficiently precharged when the voltage application period is 0.8 μsec or more and 3 μsec or less. Therefore, since application is performed only for a shorter period than the current precharge, a signal line voltage precharge pulse 451 different from the current precharge pulse 1098 is input. Although the period may be shared with the current precharge, in this case, the period for supplying the current according to the gradation is shortened, so that the variation of the driving transistor due to the current is not sufficiently corrected, and the voltage value of the black display is changed. May cause uneven brightness. For this reason, the voltage application period is shortened as much as possible, and the gradation current output period is lengthened (in each panel, the precharge voltage can be adjusted in accordance with the variation of the drive transistor 62, but in practice In some cases, there is a possibility that the characteristics of the drive transistor 62 are greatly shifted between panels and lots, whereas it can be shared by adjusting the precharge voltage, but it requires an adjustment process and is practical. In order to perform this adjustment function with current, it is preferable that the grayscale current output period is long.In a small panel, since the source line capacitance is relatively small and the horizontal scanning period is long, it is sufficient to be shared. (2) Precharge pulses are shared with priority on chip size.

  Since the two precharge pulses 1098 and 451 have the same start position (beginning of the horizontal scanning period) and differ only in pulse width, they can be generated by a counter generated from the source driver clock 871 and the timing pulse 849. It is. The pulse width is determined by a current precharge period setting line 1096 and a voltage precharge period setting line 933, respectively. Similarly to the configuration of FIG. 109, the signal is transmitted using the blanking period of the video signal line 856 in order to reduce the number of input / output signal lines of the source driver. Since two pulses are output once in one horizontal scanning period, the setting is rewritten at most once even in one horizontal scanning period, so insert a signal to be set in the blanking period in this way. That's fine.

  The precharge voltage value to be applied is generated by the precharge voltage generator 981. If there are a plurality of voltages to be output to the precharge current voltage output stage 112, a configuration similar to that shown in FIG. 99 may be used. However, if there is only one voltage value corresponding to the gradation 0, there are three The voltage may be configured by an electronic volume and an operational amplifier, and the voltage value may be adjusted by the electronic volume. In either configuration, the voltage value is adjusted by the precharge voltage setting line 986. As with the precharge pulse, the setting line is performed during the blanking period of the video signal 856.

  The precharge current voltage output stage 1121 selects which of the precharge voltage, precharge current, and gradation current is output. FIG. 113 shows a circuit configuration of the precharge current voltage output stage 1121. In this example, two precharge current sources 1112 and 1113, a total of three precharge voltage lines 983, and a gradation current 1093 are selected, so the precharge determination line 984 is 2 bits. ing. The decision signal decoding unit 1131 decodes which one of the four is output from the decision line 984 and the precharge pulses 1098 and 451. FIG. 114 shows the relationship between the states of the switching units 1132, 1133, 1134, and 1135 and the input signals. A precharge determination line 984 determines whether precharge is performed or, if so, whether current or voltage is used. Further, when precharging is performed, the precharge is performed only during the period of the current or voltage precharge pulse, and the gradation current is output during the other periods. This realized a source driver IC having a current or voltage precharge function. In FIG. 112 to FIG. 114, the number of voltage precharge voltages is one for each color and the number of current precharge currents is two for each color. However, any number can be realized.

  FIG. 115 shows a flowchart for generating a precharge flag that is a source of the precharge determination line.

  Here, the conditions for precharging are considered. The voltage precharge is performed only when the gradation is 0. Further, when the gradation is 0 before one row, since the signal line does not change during these two horizontal scanning periods, it is not necessary to perform voltage precharge. Next, current precharge, but if it is above a certain gradation, it is possible to write enough with the gradation current no matter what data is in the previous row, so no current precharge is required. It is. In general, current precharge is not necessary for a gray scale that outputs a gray scale current larger than the current value Ip of the current source for current precharge. In the example of FIG. 115, a flowchart for a 3.5-type QVGA panel is shown. In this case, the current precharge is unnecessary because the gradation can be changed to a predetermined gradation at 32 gradations or more. The current precharge is required for the 1st to 31st gradation display rows, and the current precharge is performed when the data of the previous row is larger than the display gradation. If the row data is smaller than the previous row data, or if the gradation is the same, current precharge is not required. When the previous row data is gradation 0, a precharge voltage is often applied, and a voltage higher than a predetermined gradation is applied to prevent luminance variation due to voltage. Therefore, the amount of change in potential of the source signal line is increased, and it becomes difficult to write a predetermined gradation. Therefore, when the previous row data is 0, it is possible to prepare Ip0 whose current precharge current value is larger than Ip and to output this current after gradation 0.

  In order to realize such precharge, as shown in FIG. 115, video signal data is first examined by the flow shown in 1151, and gradation 32 or higher that does not require precharge, gradation 0 that becomes voltage precharge, and the like. Branch to gradation. Since precharge is not necessary at gradation 32 or higher, the precharge flag value is set to 0 by the determination of 1157 (when the determination signal decoding unit 1131 truth table of FIG. 114 is used).

  In the case of gradation 0, the data of the previous row is referred to by the flow of 1152. Since it is not necessary at the time of gradation 0, it is divided into gradation 0 and the others, 1157 is not precharged at gradation 0, the flag is set to 0, and voltage precharge is performed at other than gradation 0. The charge flag is 1.

  In the remaining gradations 1 to 31, if the video signal data of the previous line is larger, 1157 is not precharged and the flag is 0 because precharge is not required. In the case of gradation 0, since the current Ip0 is required as a precharge current, the current precharge (current source 1113) is 1155. Therefore, the flag value is 3. In other cases, since the normal current precharge (current value Ip) is used, the current precharge (current source 1112) is 1156, and the precharge flag outputs 2 (where the current source 1112 is the current source of Ip, The current source 1113 is assumed to be a current source of Ip0).

  Depending on the panel, the value of Ip may increase, and the number of gradations that require precharging may increase accordingly. In preparation for this, the branch instruction 1151 may change the condition of the conditional branch by an external command or the like. Further, when the number of precharge current sources and voltage sources is increased, it is possible to similarly create a flowchart and realize a circuit.

  As shown in FIG. 116, the precharge flag generation unit 1162 that realizes this flowchart receives the video signal 1161 and the output of the line memory 1164 that accumulates data of the previous row as shown in FIG. The small amplitude differential signal conversion unit 1163 is input in synchronization with 1161. Here, in order to reduce the number of signal lines and to prevent electromagnetic noise, the signal is converted into a small amplitude differential signal, and further a source driver control signal is inserted during the blanking period, and a video signal line 856 and a clock 858 are output to the source driver. To do. Note that when the controller and the source driver are composed of one IC, the small-amplitude differential signal conversion unit 1163 is not necessary, and this signal may be input to the shift register and latch unit 384 as it is.

  109 and 112, the gate driver control line 941 is output. This signal is used to reduce the number of controller output signal lines, and the number of output signal lines of the controller is not limited. Is unnecessary.

  It has been found that the necessary amount of current precharge varies depending on the display gradation of the previous line even when the same gradation display is performed. For example, when gradation 16 is displayed, a precharge current corresponding to 64 gradations is required when the gradation before the first row is 0, and a precharge equivalent to 26 gradations when the gradation before the first row is 1. When the current gray level of one row was 2, a precharge current corresponding to 16 gray levels (= not necessary) was obtained. For this reason, when determining the precharge current, it is necessary to refer to the data of the previous row and set the optimum precharge current from the data of the previous row and the value of the row data.

  There is a method of controlling the precharge current by preparing a matrix table or the like for the relationship between the previous row data, the row data and the precharge current value, but the table becomes larger as the number of gradations increases, and the circuit scale at the time of IC design increases. There is a problem that it gets bigger.

  The reason why the precharge current has to be determined by preparing a matrix table is that there is a large difference in change time depending on which state the source signal line is in first. The time required for the current change is represented by (capacity of the source signal line) × (source signal line potential difference between the previous row and the row) / (source signal line current). As shown in FIG. 106, the relationship between the current and voltage of the source signal line follows the characteristics of the driving transistor 62 and is represented by a non-linear curve. The lower the gradation display, the larger the potential difference per gradation. For this reason, even if the gradation difference is the same, the time varies greatly to change to a predetermined current. For example, since the potential difference is 1/2 in the 2nd to 4th gradations compared with the 0th to 2nd gradations, the writing time is 1/2 when the source signal line current is doubled. 4 (When the gradation difference is the same at 2) It is necessary not only to detect the gradation difference but also to determine the precharge from the gradation difference and the display gradation, and at least the data of the previous row and the data of the row Need to see.

  If the gradation difference is proportional to the source potential difference, the source potential difference with respect to the gradation difference 1 is uniquely determined, and the necessary current per gradation difference 1 is determined. Based on this, the amount of current required for an arbitrary gradation difference can be obtained by calculation. Therefore, the necessary current value is determined from the calculation result of the gradation difference. If there is a means capable of storing even the necessary current, the precharge current can be determined.

  However, in the display device of the present invention, the gradation difference and the source potential difference are not proportional to each other, and even if the gradation difference is the same, the source potential difference may be different. With reference to the previous data and the row data, the source signal line potential difference is first calculated therefrom. It is necessary to determine the precharge current based on the source signal line potential difference. It is impossible to calculate the relationship between the previous row of data, the row data, and the source signal line potential difference by calculation, or it is not possible because of the calculation that makes the circuit scale very large. In addition, it is necessary to record the precharge current value in all combinations of gradations so that the necessary current value can be found from the data before one row and the row data.

  In the case of 256 gradations, it is necessary to memorize all 65,000 combinations, and even in this case, it is quite difficult to actually create a circuit. The circuit scale is reduced by not storing the combination of unnecessary gradations, which can be realized with a storage amount of about 10,000).

  Therefore, in the present invention, in order to further reduce the circuit scale of the circuit for determining the precharge current value, a voltage corresponding to gradation 0 is applied by a voltage at the beginning of the horizontal scanning period. Changing the state of the source signal line to gradation 0 by the voltage can be realized in about 1 to 3 μsec. Since the time is changed in a period within 10% of the horizontal scanning period, it is not necessary to greatly sacrifice the time required for writing, and the source signal line can be changed to the state of gradation 0.

  By providing a period (voltage reset period) in which a voltage corresponding to this gradation 0 is applied, the state of the source signal line is always changed from the state of gradation 0, and the state of the previous row needs to be stored. Disappears. Since only the precharge current corresponding to the display gradation is stored (because it is always 0), the storage amount is drastically reduced, and at most about 70 are sufficient.

  After the voltage reset period, a precharge current output period is provided to quickly change to a predetermined current, the current is changed to near the predetermined gradation, and then the current change speed is output by outputting the current corresponding to the predetermined gradation. It can be changed quickly even in a low gradation region where the speed is slow.

  In the method of outputting the precharge current with the optimum value according to the display gradation, a current source corresponding to the optimum precharge current value is required for each type of necessary current value. If a current precharge current source is provided in addition to the grayscale display current source 241, the source driver circuit becomes large and the chip size increases. In addition, since the time required for the current change varies depending on the capacity of the source signal line, there is a possibility that the current value of the current precharge is different in panels of different sizes. Since the precharge current cannot be changed by a driver IC formed with a circuit, for example, a current value selection pattern corresponding to a gradation can be created by creating an extra current value and a larger current value than the number of necessary current sources. However, there is a problem that the circuit scale becomes larger.

  Therefore, in the present invention, a precharge current is applied instead of changing the current value according to the gradation so that optimum current precharge according to a plurality of panel sizes can be performed by an external command operation or the like. The period was changed according to the gradation.

  Specifically, the precharge current is a current corresponding to the current at the maximum gradation display, and if the time for applying this precharge current is changed, the amount of change due to the precharge current is small when the time is short. The current is about the gradation, and when the time is long, the amount of change due to the precharge current increases, so that a high gradation current can be achieved.

  A source driver configuration for realizing this is shown in FIG. FIG. 118 shows a circuit configuration example of a current output unit 1171 that outputs a precharge current and a current corresponding to a gradation.

  In FIG. 118, whether or not the gradation display current source 241 is connected to the output 104 depends on the switching means 1183 controlled by the gradation data line 985. This current source is designed so that the amount of current varies depending on the bit weight of the gradation data line 985. Specifically, as shown in FIG. 25, if a current source is formed by a transistor and the weight of the current is determined by the number, the current can be output accurately.

  The circuit scale of the current source unit is reduced by enabling the precharge current to be output from the same current source. For this purpose, the switching means 1184 for connecting or not to connect the current source 241 to the output 104 is connected in parallel with 1183, and the switching means 1184 is controlled by the current precharge control line 1181, so that the current source is shared. Reduced circuit scale. Thus, the fact that switching means 1183 and 1184 are arranged in parallel for one current source 241 can be realized because the precharge current is the maximum current (white display current). Although the switching means is connected in parallel, the current of the connected current source is output if one of them is in a conductive state. Therefore, these two switches realize an OR circuit, and in the current precharge output period, the current precharge control line 1181 is at a high level, and when not output, the gradation data 985 is set at a low level. In this case, all currents 241 are output by the current precharge control line 241. Therefore, a precharge current can be output regardless of the gradation data 985. Note that the use of the maximum current value has an advantage that the current change is accelerated, the precharge current output period 1203 can be made as small as possible, and the gradation current output period 1204 for accurately performing gradation display can be made long.

  Providing two switching units 1183 and 1184 connected in parallel eliminates the need for a logical operation element, thereby reducing the circuit scale.

  In order to control the precharge current output period by gradation, the high level period of the current precharge control line 1181 may be changed by gradation. Therefore, in the present invention, the pulse selection unit 1175 is provided with a plurality of current precharge pulses, and one of the current precharge pulse groups 1174 is selected according to the value of the precharge determination line 984. The precharge pulse 1174 can change the precharge period by using a signal in which the high level period is changed in advance by command setting.

  The input / output relationship of the pulse selector 1175 is shown in FIG. The state of the current precharge control line 1181 and the voltage precharge control line 1182 changes depending on the value of the precharge determination line 984. When the state of the source signal line does not change, such as when the same gradation is displayed in continuous rows, voltage and current precharge is not necessary. In this example, when the precharge determination line 984 is 0, it corresponds to the gradation. Only current output is performed. Further, since gradation 0 is displayed by voltage precharge at the time of gradation 0, only current precharge is not necessary. Therefore, when precharge determination line 984 is 7, only the current precharge control line is always at a low level. Mode is provided. In the case of other determination values, one of a plurality of current precharge pulses having different pulse widths can be selected.

  As a result, as shown in FIG. 120, signals output from the precharge determination line 984, the voltage precharge pulse 451, and the current precharge pulse 1174 to the output 104 are determined. When the relationship shown in FIG. 119 is followed, the output is precharged during the first horizontal scanning period, and then has a precharge current output period 1203 corresponding to the current precharge pulse of 1174d. Period 1204 is entered. In the next one horizontal scanning period, only the gradation current output period 1204 exists. By doing so, it is possible to change the period during which the current precharge is performed by the precharge determination line 984, and the high-level period of each current precharge pulse 1174 is designed to be changed by an external input. For example, an optimal current precharge can be performed according to the panel size and horizontal scanning period, and a source driver corresponding to an arbitrary panel size and the number of pixels can be realized.

  In the present invention, a current precharge pulse group 1174 and a voltage precharge pulse 451 are generated by a pulse generator 1122 as shown in FIG. A current precharge period setting line 1096 and a voltage precharge period setting 933 are input to the pulse generator 1122 from the outside via the video signal / command separator 931, so that an arbitrary command has an arbitrary pulse width. A precharge pulse can be realized.

  Further, in a display device using an organic light emitting element, there is a problem that the current value per gradation is different for each color because the light emission efficiency is different for each display color, thereby changing the precharge current value. In the most efficient display color, since the white display current value is small, there is a possibility that the current cannot be sufficiently changed to a predetermined gradation. Therefore, in the present invention, the current precharge pulse group 1174 is prepared for each color as 1174g, 1174h, and 1174i, thereby solving the above-mentioned problem by adjusting the current application period. Specifically, in the most efficient color, the width of the precharge pulse is increased as a whole because the current is small.

  The fact that the predetermined current can be obtained by changing the length of the precharge pulse 1174 in accordance with the gradation will be described using the state of current change in FIG. 124 (in this case, the driver output is 8 bits, 256 gradations). The description will be made on the assumption that the output can be performed, and the number of gradations can be explained in the same manner even with a driver having an arbitrary number of bits if considered according to the number of bits actually used.

  Assuming that the period of the current precharge pulse is 1174a, for example, the current rapidly changes in the precharge current output period 1242, and then changes slowly because a predetermined current is output. As shown in FIG. 124 (b) The current change indicated by.

  On the other hand, when a current precharge is output for a longer time, for example, when a precharge current is output for a period of 1174c, it quickly changes for a period of 1243, and then slowly changes to a gradation 30 with a predetermined current (curve diagram 124 ( c)).

  Further, when the current precharge pulse is always applied, the change is as shown in FIG.

  It can be seen that the current can be changed most quickly by performing current precharge on the current change curve of FIG. 124 (d) until it approaches the predetermined gradation value and then outputting the predetermined gradation current. The higher the gray level, the longer the precharge current output period, and the shorter the gray level becomes, the shorter the gray level becomes, the less the precharge current value itself can be changed to a predetermined gray level only in the application period.

  FIG. 123 shows the relationship between the necessary precharge current period and gradation in the 3.5-type QVGA panel. As the gray level becomes higher, the precharge current period becomes longer. In addition, it is known that the precharge current period is unnecessary for 36 gradations or more. Therefore, by associating the necessary current period and the current precharge pulse as shown in FIG. 123 and specifying the high level period of each current precharge pulse as the period shown in FIG. 123 by an external command, one precharge is performed. With the current source, it is now possible to display the predetermined gradation properly on the next line for all gradation changes by an external command operation.

  The correspondence between the gradation and the current precharge pulse is replaced with the correspondence between the precharge determination line 984 and the current precharge pulse. A control IC or the like generates and supplies a precharge determination signal corresponding to the gradation data so that a desired precharge pulse is selected for the display gradation, and the gradation can be associated with the current precharge pulse. .

  This is advantageous in that the current precharge pulse for the gradation can be changed by the control of the control IC when the correspondence between the gradation and the current precharge pulse is changed.

  When the current value per gradation is large, a predetermined gradation can be displayed without current precharge even at a lower gradation. For example, when the current is twice as large as that in the case of FIG. 123, theoretically, writing can be performed without current precharge for 18 gradations or more. In this case, it is possible to cope with the problem by changing the processing in the control IC that controls the relationship between the gradation and the precharge determination line 984 and rewriting the relationship.

  For this reason, a precharge determination line is prepared separately from the gradation signal in this way, and a current precharge pulse is selected by this precharge determination line, so that the same source driver can be used even when the light emission efficiency of the organic light emitting device changes. It became possible to display using.

  In the method of selecting one of the precharge pulses 1174 having a plurality of pulse widths according to the value of the precharge determination line 984, the pulse widths of the plurality of precharge pulses 1174 can all be controlled by commands from the outside. In order to do so, a signal defining a large number of pulse widths is required. If all these signals are directly input from the outside of the driver IC 36, many input pins are required, which is not practical. Accordingly, in the present invention, the precharge pulse width is increased without increasing the number of external signal lines by serially transferring all set values via the video signal line 856 within the blanking period using the blanking period of the video signal. Can be set.

  FIG. 121 shows a signal input method for inputting a command using the video signal line 856. While the video signal is transmitted, each display color data 861 (here, red, green, and blue are assumed as shown in FIG. 121A. Note that the data of any color is not limited to these three colors, depending on the display device. For example, three colors of cyan, yellow, and magenta) and a precharge flag 862 that is a signal for determining whether or not to precharge each data 861 are input correspondingly. A data / command flag 950 for determining that it is a video signal is also transmitted. For example, if it is 1 for data and 0 for command, it is possible to identify whether the transmitted signal is a video signal or a command by referring to this bit.

  Next, a command is transmitted during the blanking period. The data / command flag 950 is set to 0 so that it can be identified as a command. This is not necessary if all commands can be set in one transfer. However, in the present invention, since the number of commands is large, some bits are used as addresses, and which data depends on the address value. Determine if it corresponds to a command. In the example of FIG. 121, at address A1211, it is determined whether the signal is a current precharge setting signal or other signal. In FIG. 121 (b), necessary signals are set in addition to the setting of the current precharge period, and a reference current setting signal 912 that defines a precharge voltage value, a voltage precharge period, and current per gradation is transmitted. is doing. In FIG. 121 (c), since it is necessary to set six current precharge output periods for each color, each address B1212 is further provided, and the pulse width of which current precharge pulse depends on the value of the address B1212. Decide whether to set

  Since the pulse width of the current precharge pulse is about 0.4 μsec from FIG. 123, the step width is 0.2 μsec or 0.4 μsec, and the variable range is arbitrary as long as it is about 6.4 μsec. Adjustment to the panel is possible. It suffices if 32 or 16 levels can be set. Since 1174a to 1174f do not need to have the same pulse width, they should be set to different values. Furthermore, 1174a has the minimum pulse width and 1174f has the maximum pulse width, and the roles of the pulses are shared. In this case, for example, the adjustment range of 1174a is 0.2 μs to 6.6 μs (32 step adjustment), and the range of 1174f is 2.0 μs to 8.4 μs (32 step adjustment). The pulse width can be set from 2 μs to a maximum of 8.4 μs. In this way, by setting the variable range of the pulse width of each pulse slightly shifted for each pulse, the variable range can be reduced, the setting signal line width is reduced, and the circuit scale is reduced. can do.

  As described above, by enabling various values to be set by the external input command, the source driver IC 36 that can quickly output a current corresponding to the gradation of the display device at an arbitrary panel size and resolution is realized.

  The current output unit 1171 according to the present invention includes a plurality of switching units connected in parallel to one current source 241 as shown in FIG. 118, and each bit and current of the gradation data line 985 as shown in FIG. This can also be realized by a method in which the logical sum of the precharge control lines 1181 is used to control the switching unit 1221 connected to the current source 241. In the process in which the switching units 1183 and 1184 can be formed small, the circuit scale in FIG. 118 is small. However, if it cannot be reduced, there is a case in which adding a logical sum circuit that can be created according to the logic signal rule may be small.

  Whichever of these two circuits is taken may be selected in consideration of the process rule.

  In this example, the same pulse is input as the voltage precharge pulse 451 regardless of the display color. This is because the state change speed depends on the driving capability of the output operational amplifier to change the state of the source signal line by the voltage. Since there is no influence of different signals for each display color, such as current per gradation, one voltage precharge pulse 451 is used to reduce the circuit scale. If the circuit scale is not a problem, three pulses may be provided so that each color can be individually specified.

  In the source driver IC 36 having the configuration of the output stage of FIG. 118 or 122, an output having the precharge current output period 1243 can be performed with the relationship between the gradation and the precharge pulse as shown in FIG. On the other hand, if the precharge current output period 1243 is determined according to the relationship of FIG. 123, for example, even when the same gradation in which the source signal line does not change is continuously output, precharge is performed.

  As shown in FIG. 125, after the signal line changes to the black display state in the precharge voltage application period 1251 at the beginning of the horizontal scanning period, the state of the source signal line changes to a value close to a predetermined current value in the precharge current output period 1252. In the final gradation current output period 1253, the current value changes to a predetermined current value, and the source signal line current once becomes black at the beginning of the horizontal scanning period. The possibility that the state of the signal line is changed to cause insufficient writing has been increased.

  Therefore, in the present invention, as shown in FIG. 126, when the same gradation current output is continuously output, the precharge current output period 1252 is not provided in the subsequent row, and only the gradation current output period 1253 is provided. It was made difficult to generate an insufficient writing state by reducing the state change of the source signal line.

  In the case of the display pattern shown in FIG. 127 (the pattern in which the areas 1272 and 1274 have the same luminance and the area 1273 has a lower luminance than the areas 1272 and 1274), the first row that becomes the area 1273, the area 1274, In the first row, current precharge is performed. The output current waveform of the source signal line corresponding to the column 1271 is as shown in FIG. Since the output current does not change in the period corresponding to the region 1272, only the gradation current output period is included in the horizontal scanning period 1281.

  In the first horizontal scanning period 1281d after moving to the region 1273, since the source signal line current changes, a precharge voltage application period 1251d and a precharge current output period 1252d are provided for the purpose of quickly changing the current. The current corresponding to the region 1273 can be output in a shorter period than when the precharge current is not output (1282). Similarly, even when the display of the region 1273 is continuous, the change of the source signal line current is minimized by performing only the grayscale current output without providing the period for outputting the precharge current and the precharge voltage.

  Further, when the source signal line performs output corresponding to the display of the region 1274, the voltage and current precharge are performed only in the first horizontal scanning period 1281g. Note that the precharge current output period 1252g is longer than 1252d. This corresponds to the fact that the higher the gradation, that is, the longer the current, the longer the precharge current output period, from the relationship between the gradation and the current precharge output period in FIG. If the region 1274 has gradation 0, the gradation current output period 1253g follows the precharge voltage application period 1251g, and the precharge current output period 1251g disappears. (Because the precharge current output period 1251 exists depending on the gradation, it does not necessarily exist.) By performing this precharge, the output current value is changed only by the gradation current output without the conventional precharge. In comparison with the case (1283), the current of the source signal line could be changed to the predetermined current value in a short time.

  In order to perform voltage precharge and current precharge or voltage precharge only when the state of the source signal line changes in this way, in addition to the relationship with the gray level in FIG. As a result of the comparison, it is necessary to perform the precharge according to the relationship of FIG. 123 only when the video signal is changed.

  FIG. 129 shows a flow for determining whether to perform precharge. The current gradation value is detected from the video signal 1291. (1292) If the gradation is 0, only voltage precharge is performed as in FIG. 123, and then a current corresponding to the gradation is output (1293).

  At the gradation 36 or higher, the current changes up to a predetermined gradation without performing precharge, so only current output corresponding to the gradation is performed (1296).

  In gradations 1 to 35, the processing changes depending on the gradation of the previous line (1294), and only current output corresponding to the gradation is performed in the same gradation as the current gradation (1296). This is done to reduce the waveform change as shown in FIG. 126 when the same gradation is displayed continuously.

  On the other hand, in the processing of 1294, when the gray level one row before and the current gray level change, after the precharge voltage is output, the current pre-charge for the period corresponding to the gray level and the current output corresponding to the gray level for the remaining period are performed ( 1295). This corresponds to the operation in the horizontal scanning periods of 1281d and 1281g in FIG.

  If the signal of the precharge determination line 984 is generated so that the relationship between the gradation and the precharge current output period of FIG. 123 is obtained when the determination result of FIG. The driver IC can output as shown in FIG. In the state of 1296, the relationship of FIG. 123 is not used, and the value of the precharge determination line 984 may be determined so that the gradation current is always output.

  As a result, the change of the source signal line is minimized, and the current can be changed rapidly at the change point, so that the boundary of the region can be displayed properly even in the display as shown in FIG.

  In the gradation 0 display, a precharge voltage is applied to the gate electrode of the drive transistor 62 in the pixel circuit through the source signal line so that a current corresponding to black display (current of 1.3 nA or less) flows. However, in this case, since the voltage is converted into a current in the drive transistor 62, the drain current with respect to the input voltage changes with a change in temperature. For example, as shown in FIG. 130, when the drive transistor 62 is made of low-temperature polysilicon, the current is higher when the temperature is higher (FIG. 130 (a)) than when the temperature is lower (FIG. 130 (b)). Flows well. Therefore, there is a problem that the current during black display increases and black floating occurs (in the case of the circuit configuration as shown in FIG. 6, the drain current of the drive transistor 62 is a current flowing through the EL element. As the current flowing through the EL element is increased, the EL element is lighted slightly and black floating occurs.

  For example, when the temperature is low (a) and the precharge voltage is adjusted to VBk2, the drain current of the transistor 62 flows through IBk. This current is below a level (1.3 nA) where black float is not recognized. When the temperature rises in this state and the characteristics of the transistor 62 change to the curve shown in FIG. 130 (b), the current ID flows, and the current increases to a level where black floating can be seen. In order to eliminate black floating even in a high temperature state, it is necessary to increase the gate voltage to VBk1.

  When the channel size of the pixel transistor is designed with a width of 25 microns and a length of 15 microns, if (a) is −20 ° C. and (b) is + 50 ° C., the voltage of VBk2 is (64 voltage values) − The voltages of 1 [V] and VBk1 are (64 voltage values) −3 [V]. The voltage between the source and drain of the pixel transistor 62 is 1V and 3V, respectively.

  If the necessary source-drain voltage differs depending on the temperature, the precharge voltage applied to the transistor 62 may be changed depending on the temperature. When generating a reference voltage by resistance division when generating the precharge voltage, as shown in FIG. 131, if a temperature compensation element 1311 such as a thermistor is attached in parallel to one of the resistance elements 1312, the temperature is increased. As a result, the voltage at the dividing point 1314 changes. In the case of the thermistor, the resistance value decreases as the temperature rises. Therefore, the temperature compensation element 1311 is connected in parallel to the resistance element 1312a connected to the power supply side 64 among the two resistance elements 1312. If the value of each resistance element, the resistance value of the thermistor, and the temperature coefficient are adjusted, as shown in FIG. 132, a setting can be made such that the precharge voltage increases as the temperature increases.

  A specific circuit configuration is shown in FIG. Description will be made with the source driver 36 and a pixel circuit for one pixel. The circuit of the source driver 36 is described only with respect to an analog output unit that performs voltage precharge. The entire circuit configuration is as shown in FIG. 117, for example. When voltage precharge is performed, the voltage generated by the precharge voltage generator 1313 is output to the current output line 104 by the voltage precharge control line 1182.

  The output voltage is applied to the node 72 in the pixel circuit 67 selected by the gate signal line 61 through the source signal line 60.

  When the pixel selection period ends, the switches 66a and 66b are turned off and the switch 66c is turned on, and a current flows through the EL element 63 based on the relationship between the gate voltage and the drain current of the transistor 62. Since the relationship between the gate voltage and the drain current at this time is as shown in FIG. 130, when the precharge voltage outputs a constant value regardless of the temperature, the node 72 (= the gate voltage of the transistor 62) is also constant. From the relationship of FIG. 130, the current flowing through the EL element 63 changes.

  Therefore, in the present invention, in the precharge voltage generator 1313, the voltage before being buffered by the operational amplifier is not generated by the electronic volume 1341, but via the external connection terminal, using the resistance element 1312 and the temperature compensation element 1311. As a result, the precharge voltage, that is, the voltage at the node 74 is changed according to the temperature, and the current flowing through the EL element 63 is made constant regardless of the temperature.

  133 shows the relationship between the drain current of the transistor 62 (= current flowing through the EL element 63) and temperature when the precharge voltage is constant.

  A solid line 1332 in FIG. 133 shows a change of the current value with respect to the temperature when the precharge voltage is changed. In the case of 1332, it can be seen that the drain current of the transistor 62 is constant regardless of the temperature. By selecting the resistance element 1312 and the temperature compensation element 1311 so that the current value is 1.3 nA or less, it is possible to realize a display with no black floating.

  In the configuration of FIG. 134, a temperature compensation element is used to compensate for a change in current based on temperature characteristics. However, when there is an electronic volume 1341, the value of the electronic volume 1341 may be changed depending on the temperature.

  Since the electronic volume 1341 is generally controlled by the controller 1351, the electronic volume control command may be changed on the controller side according to the temperature. For this purpose, the controller 1351 receives a signal from the temperature detection means 1350.

  In this figure, the electronic volume control signal 1353 is used to set the electronic volume, and the controller 1351 controls the source driver 36. In the source driver as shown in FIG. 117, the voltage value of the precharge voltage generator 981 is set. Is received from the video signal line 856 via the video signal / command separator 931. As described above, since there is a method of separating signals after serial transfer from the controller to the source driver using another signal line, the electronic volume control signal 1353 is not necessarily required. It is only necessary that the controllable signal line be connected between the source driver and the controller, alone or in common with other signals for electronic volume control.

  When the voltage value is controlled by the electronic volume 1341, since the input is a digital signal, the voltage value cannot be increased in a proportional relationship with the temperature. As shown by the solid line in FIG. The output voltage (ie, precharge voltage) of the electronic volume changes.

  Even in this case, in order to make the current flowing through the EL element 63 equal to or less than 1.3 nA in all temperature ranges, the value of the electronic volume is set so as not to fall below the voltage value of the broken line 1362 changed by the temperature compensation element. What is necessary is just to make it change an electronic volume output voltage with respect to temperature like the changed solid line 1361. FIG.

  Thus, the drain current of the transistor 62 flows with respect to the temperature as indicated by 1371 in FIG. As a result, the current flowing through the EL element 63 can be reduced to 1.3 nA or less regardless of the temperature, and a display with no black floating can be realized even at a high temperature as compared with the conventional 1331 in which the precharge voltage is not changed. It was.

  In the present invention, the organic light-emitting element has been described as the display element. However, any display element such as an inorganic electroluminescent element or a light-emitting diode that has a proportional relationship between current and luminance can be used.

  The current output type semiconductor circuit according to the present invention is useful as a current output type semiconductor circuit that reduces the number of input signal lines and outputs current used in a display device that performs gradation display according to the amount of current, such as an organic electroluminescent element. .

The figure which showed the input signal waveform of the current output type semiconductor circuit in this invention Block diagram of the driver IC when it is possible to select from the outside whether or not to precharge each video signal for one dot A diagram showing a display panel using a plurality of source driver ICs Diagram showing structure of organic light emitting device Diagram showing current-voltage-luminance characteristics of organic light-emitting elements The figure which showed the circuit of the active matrix type display device using the pixel circuit of the current copier configuration Diagram showing the operation of the current copier circuit Diagram showing an example of a constant current source circuit Diagram showing the arrangement of each element of the reference current source The figure which showed the circuit for outputting the electric current to each output of the current output type driver The figure which showed the relationship between the transistor size of the gradation display current source 103 of FIG. 10, and output current dispersion | variation. The figure which showed the equivalent circuit when the source signal line current flows to the pixel in the pixel circuit of the current copier configuration The figure which showed the relationship between the current output in one output terminal, the precharge voltage application part, and a changeover switch. Diagram showing the difference in luminance vs. current characteristics depending on the emission color The figure which showed the relationship between the period which performs the precharge voltage within 1 horizontal scanning period, and the period which outputs the electric current based on gradation data A diagram showing a circuit block that can output different precharge voltages for each display color Diagram showing the relationship between gradation data and precharge determination signal Diagram showing a circuit that distributes input serial current to each signal The figure which showed the connection relation of several driver IC in embodiment which distributes a reference current to each output The figure which showed the pixel circuit using the current copier at the time of using an n-type transistor The figure which showed the case where it applied to a television as a display apparatus using embodiment of this invention The figure which showed the case where it applied to a digital camera as a display apparatus using embodiment of this invention The figure which showed the case where it applied to a portable information terminal as a display apparatus using embodiment of this invention The figure which showed the concept of the current output part of the semiconductor circuit using embodiment of this invention FIG. 24 is a diagram showing a case where the current source is configured by a transistor in the configuration of FIG. The figure which showed the relationship of the gradation of an input signal with respect to the output current by the current output part shown in FIG. 24 or FIG. The lower 1 bit of 8-bit data is output in a certain size transistor configuration, and the remaining upper 7 bits are provided with a transistor with a larger drain current than the lower 1 bit transistor. Diagram showing current output stage for display Diagram showing display when voltage corresponding to black gradation is applied to low gradation display section The figure which showed the structure of the signal required for the output stage and the output stage when making it possible to select whether to apply the precharge voltage for each frame The figure which showed the pattern of the pixel which applies a precharge in the case of performing precharge once in 3 frames The figure which showed the pattern of the pixel which applies the precharge in the case of performing precharge once in 2 frames The figure which showed an example of the relationship between the input gradation and the ratio of the frame which precharges The figure which showed the circuit at the time of preparing two or more precharge power supplies from the structure of FIG. In the configuration of FIG. 33, the two precharge voltages and the output voltage can be changed for each frame. The figure which showed the relationship (current voltage characteristic of a pixel transistor) of the source signal line current with respect to the power supply voltage for precharge of FIG. The figure which showed the composition of the output stage which can change the output voltage for every plural precharge voltage and frames The figure which showed the application pattern with respect to the gradation of the some pre-charge voltage of FIG. A diagram showing an example of an output signal line configured to select and output either current output according to gradation or voltage output according to gradation Diagram showing the display when the precharge is not performed and when it is performed on a screen with a low lighting rate Diagram showing the display when the precharge is not performed and when it is performed on the screen with a high lighting rate The figure which showed the flowchart which determines whether the precharge voltage in this invention is output The figure which showed the precharge determination signal generation part for implement | achieving the precharge application system of this invention The figure which showed the operation | movement of the FRC register for determining whether it precharges for every flame | frame The figure which showed the display device using the pixel composition of the current mirror form The configuration of the output stage when the current corresponding to the gradation and the voltage corresponding to the gradation can be selected and output within one horizontal period or can be output in order in time. Figure The figure which showed the circuit example of the precharge voltage application determination part of FIG. 45 shows a relationship between a precharge determination signal and a precharge pulse when outputting only current, outputting only voltage, or outputting current after voltage output in one horizontal scanning period in the output stage having the configuration of FIG. The figure which showed the wiring of EL power supply line in FIG. 6 or FIG. The figure which showed the pattern of the pixel which precharges in each frame at the time of performing precharge twice in 3 frames Diagram showing the relationship between gradation and precharged frame ratio The figure which showed the difference in the ratio of the frame to precharge with respect to the gradation in three different lighting rates by a lighting rate setting signal and a FRC precharge setting signal. A diagram showing the relationship between gradation and output current variation when the transistor size of the transistor group corresponding to the current output of the upper 3 bits is halved Diagram showing the arrangement of transistor groups in the current output section A diagram showing the relationship between the channel size and variation of the transistors that make up each transistor group The figure which showed the relationship between the gradation and the output current variation when the transistor sizes of the transistor groups corresponding to the current output of the upper 3 bits are varied. For an 8-bit video input, the magnitude relationship of the current between the lower 2 bits and the upper 6 bits is adjusted by the transistor channel width, and in each bit, the current is changed according to the number of transistors. Diagram showing gradation versus output current characteristics when gradation inversion occurs For 8-bit video input, the magnitude relationship of the current between the lower 2 bits and the upper 6 bits is adjusted by the transistor channel width, and the most significant bit in the output stage configuration in which the current is changed according to the number of transistors in each bit The figure which showed the composition which can add the current source further to the current source corresponding to The figure which showed the electric current difference of the gradation 127 and the gradation 128 The figure which showed the relationship with the gradation 127 electric current when the electric current of the gradation 128 decreased by dispersion | variation Diagram showing gradation-current characteristics when gradation inversion disappears The figure which showed the current output stage with the highest bit current source current raising function when using the raising signal line 61 is a functional block diagram of the driver IC when the current output stage of FIG. 61 is used. The figure which showed the relationship between the gradation 127 current when the lower 2-bit output current increased by 20% and the gradation 128 current after raising The figure which showed the current output stage with the highest bit current source current raising function when using the raising signal line Diagram showing whether gradation inversion occurs due to luminance difference and output variation between several gradations The figure which showed the relationship of the electric current value and dispersion | variation in the gradations 3 and 4 The figure which showed the relationship between the electric current value and the dispersion | variation in the gradations 131 and 132 The figure which showed the current output stage with the highest bit current source current raising function when using the raising function by the raising signal line and laser processing The figure which showed the structure of the driver IC which can control whether it raises for every output with a raising signal The figure which showed the block diagram of driver IC in the case of using a current copier circuit for an output stage The figure which showed the example of a circuit which realizes a digital-analog conversion part The figure which showed the wiring of the gradation reference current signal when a plurality of driver ICs are connected The figure which showed the circuit of the electric current holding means The figure which showed that the drain current of the node 742 and the drive transistor 731 changes by the gate signal line 741 The figure which showed the drain current-gate voltage characteristic of the drive transistor Figure showing the difference in drain current due to "punch-through" when transistors with different mobilities are used for the drive transistors of each output The figure which showed the electric current holding means at the time of inserting one transistor in order to reduce "penetration" in a current copier circuit The figure which showed the circuit of the gradation reference current generation part 77 shows waveforms of two gate signal lines in FIG. The figure which showed the circuit of the gradation reference current generation part Diagram showing the reference current generator The figure which showed the circuit of the digital analog conversion part which includes the enable signal The figure which showed the relationship between the timing pulse in one horizontal scanning period, a chip enable signal, a select signal, and a gradation current signal The figure which showed the current-voltage characteristic of the transistor from which W / L differs The figure which showed the structural example of the display panel at the time of using the source driver which transfers a video signal and a precharge flag at low amplitude and high speed, and has a 1-bit command line for electronic volume setting and precharge period setting A diagram showing an example of a transmission pattern when high-speed transmission is performed using the same signal line for the precharge flag and the video signal line Diagram showing timing chart of command line The figure which showed the circuit structure of the precharge voltage conversion part which produces | generates the precharge voltage according to a gradation 85 is an internal block diagram of the source driver used in FIG. The figure which showed the example of the transfer of the precharge determination signal sent in synchronization with the relationship between the current voltage output corresponding to gradation data, and gradation data The figure which showed each transfer pattern example in the case of inputting a reference current setting and precharge application period setting signal to the same signal line as a video signal line The figure which showed the relationship between the period which transfers data within one horizontal scanning period, and the blanking period Diagram showing the internal configuration of the source driver when the video signal line and the reference current and precharge period setting signal line are shared Diagram showing wiring between driver ICs when using a source driver with gate driver control line output The figure which showed the data transfer method in embodiment of this invention A diagram showing an example of data transfer within one horizontal scanning period Diagram showing each signal line waveform after separating gradation data, precharge inversion signal and gate driver control line from video signal line inside source driver Diagram showing internal configuration of source driver with gate driver control line output function The figure which showed the pre-charge voltage generation part of FIG. The figure which showed the precharge voltage selection and application determination part of FIG. The figure which showed the input-output relationship of the decoding part 1001 in FIG. The figure which showed the relationship between the source signal line current when using the pixel circuit of FIG. 6, and the source signal line voltage A diagram in which a current source for supplying a current through a current precharge line is provided in the current output stage in addition to a current source corresponding to a gradation. The figure which showed the mode of a change when a source signal line current changes from 10 nA to 0 nA The figure which showed the mode of a change when a source signal line current changes from 0 nA to 10 nA 104 and 105 show the change in the current-voltage characteristics of the source signal line. Diagram showing how the source signal line current changes when current precharge is performed The figure which showed the time change of the source driver output when outputting the electric current 10 times the predetermined electric current at the beginning of the horizontal scanning period The figure which showed the structure of the source driver for implement | achieving current output like FIG. Diagram showing the configuration of the reference current generator and current output stage of the source driver that supports multi-color output Diagram showing source driver precharge current output configuration (precharge reference current generator, precharge current output stage) that supports multi-color output The figure which showed the composition of the source driver which enabled output of precharge current and precharge voltage to the source signal line 112 shows the internal configuration of the precharge current voltage output stage of FIG. The figure which showed the relationship between the input of the determination signal decoding part 1131 of FIG. 113, and the state of switches 1132 to 1135 The figure which showed the flowchart which outputs the precharge flag 862 input into a source driver The figure which showed the precharge flag production | generation part and the transmission part to a source driver The figure which showed the structure of the source driver which can perform a current precharge by selecting one period in a voltage precharge and several different periods The figure which showed the circuit of the current output part 1171 which has the function to perform an electric current precharge The figure which showed the relationship of the input / output signal of the pulse selection part 1175. 119 shows precharge pulses 1174, 451, precharge determination line 984, and output temporal changes when the pulse selection unit is operated based on FIG. The figure which showed the input signal format of the driver IC which comprised the structure of FIG. The figure which showed the circuit of the current output part 1171 which has the function to perform an electric current precharge Diagram showing the relationship between display gradation and required precharge current output period Diagram showing current change when using current precharge The figure which showed the mode of the change of the source signal line current in case each of a precharge voltage and a precharge current is output in each horizontal scanning period FIG. 5 shows a change in source signal line current when the precharge voltage application period 1251 and the precharge current output period 1252 are not provided when the source signal line current does not change over a plurality of horizontal scanning periods. The figure which showed the example of the display pattern which may change when the source signal line outputs the same current continuously, and may change 127 shows a change in source signal line current when the present invention is used in FIG. A diagram showing a determination method for generating a period in which a precharge voltage or precharge current is output only when there is a change in the current of the source signal line The figure which showed that the relationship between the drain current of the drive transistor 62 and gate voltage changes with temperature. A diagram showing a configuration in which a voltage that varies depending on temperature is input to the precharge voltage generator using a resistance element and a temperature compensation element outside the source driver The figure which showed the example of change of the precharge voltage when changing the precharge voltage by temperature FIG. 132 shows a change in drain current of the transistor 62 with respect to temperature when the precharge voltage is output as shown in FIG. The figure which showed the circuit block which applies the pre-charge voltage to the pixel circuit when the temperature compensation element is provided outside The figure which showed the circuit block which changes the value of the electronic volume for precharge voltage generation with temperature by the command control from a controller using the data of temperature detection means The figure which showed the relationship of the electronic volume output voltage with respect to temperature in the circuit structure of FIG. FIG. 136 is a graph showing changes due to the temperature of the transistor 62 when the precharge voltage is controlled according to the relationship between temperature and electron volume in FIG.

Explanation of symbols

11 Video data 12 Data line 13 Address 14 Data after distribution 15 Clock 16 Start pulse

Claims (2)

A current output type semiconductor circuit having a function of outputting a current corresponding to a desired gradation to a display panel and a function of outputting a voltage from the same terminal,
A current output type semiconductor circuit comprising a voltage generation unit including a temperature compensation element and at least one resistance element for outputting the voltage. When a current corresponding to a desired gradation is output to the display panel, a function of outputting a voltage from the same terminal,
A current output type semiconductor circuit having an electronic volume circuit for determining the voltage output value,
A method of driving a current output type semiconductor circuit, wherein a set value of the electronic volume circuit is changed according to temperature.

Images