JP4720836B2 - Solid-state imaging device - Google Patents

Description

  The present invention relates to a solid-state imaging device in which a plurality of unit pixels are arranged, and a signal from each unit pixel can be arbitrarily selected and read by address control. More specifically, the present invention relates to a solid-state imaging device in which a unit pixel is configured by a photoelectric conversion element and three transistors without having a selection transistor.

  An amplification type solid-state imaging device (APS; also called Active Pixel Sensor / gain cell), which is a kind of XY address type solid-state imaging device, is an active device (MOS transistor) such as a MOS structure in order to give the pixel itself an amplification function. ) To form a pixel. That is, signal charges (photoelectrons) accumulated in a photodiode which is a photoelectric conversion element are amplified by the active element and read out as image information.

  In this type of XY address type solid-state imaging device, for example, a plurality of pixel transistors are arranged in a two-dimensional matrix to form a pixel unit, and a signal charge corresponding to incident light for each line (row) or each pixel. Accumulation is started, and a current or voltage signal based on the accumulated signal charge is sequentially read out from each pixel by addressing.

<Conventional Unit Pixel Configuration; First Example>
FIG. 19A is a diagram illustrating a first example of a conventional unit pixel 3. The unit pixel 3 of the first example has a general-purpose four-transistor configuration as a CMOS sensor and has a well-known configuration.

  The unit pixel 3 of the first example includes a charge generation unit 32 having both a photoelectric conversion function for converting light into a charge and a charge storage function for storing the charge. A sensing element that detects potential changes in a read selection transistor 34 that is an example of a read unit (transfer gate unit / read gate unit), a reset transistor 36 that is an example of a reset gate unit, a vertical selection transistor 40, and a floating diffusion 38. An amplifying transistor 42 having a source follower configuration as an example.

  The read selection transistor 34 is driven by the transfer drive buffer 150 via a transfer wiring (read selection line) 55. The reset transistor 36 is driven by the reset driving buffer 152 via the reset wiring 56. The vertical selection transistor 40 is driven by the selection drive buffer 154 via the vertical selection line 52.

  In addition, the unit pixel 3 includes a pixel signal generation unit 5 having an FDA (Floating Diffusion Amp) configuration including a floating diffusion 38 which is an example of a charge injection unit having a function of a charge storage unit. The floating diffusion 38 is a diffusion layer having parasitic capacitance.

  The reset transistor 36 in the pixel signal generation unit 5 has a source connected to the floating diffusion 38 and a drain connected to the power supply VDD, and a reset pulse RST is input from the reset drive buffer 152 to the gate (reset gate RG).

  The vertical selection transistor 40 has a drain connected to the power supply VDD, a source connected to the drain of the amplification transistor 42, and a gate (in particular, a vertical selection gate SELV) connected to a vertical selection line 52. A vertical selection signal is applied to the vertical selection line 52. The amplification transistor 42 has a gate connected to the floating diffusion 38, a drain connected to the source of the vertical selection transistor 40, and a source connected to the vertical signal line 53 via the pixel line 51.

  In such a configuration, since the floating diffusion 38 is connected to the gate of the amplifying transistor 42, the amplifying transistor 42 sends a signal corresponding to the potential of the floating diffusion 38 (hereinafter referred to as FD potential) via the pixel line 51. To the vertical signal line 53. The reset transistor 36 resets the floating diffusion 38. The read selection transistor (transfer transistor) 34 transfers the signal charge generated by the charge generator 32 to the floating diffusion 38. A large number of pixels are connected to the vertical signal line 53. To select a pixel, the vertical selection transistor 40 is turned on only for the selected pixel. Then, only the selected pixel is connected to the vertical signal line 53, and the signal of the selected pixel is output to the vertical signal line 53.

  As described above, the unit pixel 3 is generally provided with a vertical selection transistor 40 for the purpose of selecting a pixel, and the unit pixel 3 in most current CMOS sensors has a selection transistor.

<Conventional Unit Pixel Configuration; Second Example>
On the other hand, as a technique for reducing the pixel size by reducing the area occupied by the transistor in the unit pixel 3, as shown in FIG. 19B, the unit pixel 3 is configured by a photoelectric conversion element and three transistors. One (hereinafter referred to as unit pixel 3 in the second example) has been proposed (see, for example, Patent Document 1).

Japanese Patent No. 2708455

  The unit pixel 3 of the second example corresponds to the charge generation unit 32 (for example, a photodiode) that generates a signal charge corresponding to the received light by performing photoelectric conversion, and the signal charge generated by the charge generation unit 32. The amplifying transistor 42 connected to the drain line (DRN) for amplifying the signal voltage to be reset and the reset transistor 36 for resetting the charge generation unit 32 are provided. In addition, a read selection transistor (transfer gate portion) 34 that is scanned from a vertical shift register (not shown) via a transfer wiring (TRF) 55 is provided between the charge generation portion 32 and the gate of the amplification transistor 42. Yes.

  The gate of the amplifying transistor 42 and the source of the reset transistor 36 are connected to the charge generation unit 32 via the read selection transistor 34, and the drain of the reset transistor 36 and the drain of the amplifying transistor 42 are connected to the drain line. The source of the amplifying transistor 42 is connected to the vertical signal line 53. The read selection transistor 34 is driven by the transfer drive buffer 150 via the transfer wiring 55. The reset transistor 36 is driven by the reset driving buffer 152 via the reset wiring 56. Both the transfer drive buffer 150 and the reset drive buffer 152 operate with a reference voltage of 0 V and a binary power supply voltage. In particular, the low level voltage supplied to the gate of the conventional read selection transistor 34 in this pixel is 0V.

  In the unit pixel 3 of the second example, as in the first example, the floating diffusion 38 is connected to the gate of the amplifying transistor 42, so that the amplifying transistor 42 outputs a signal corresponding to the potential of the floating diffusion 38. Output to the vertical signal line 53.

  In the reset transistor 36, a reset wiring (RST) 56 extends in the row direction, and a drain line (DRN) 57 is common to most pixels. The drain line 57 is driven by a drain drive buffer (hereinafter referred to as a DRN drive buffer) 140. The reset transistor 36 is driven by the reset drive buffer 152 and controls the potential of the floating diffusion 38. Here, in the technique described in Patent Document 1, the drain line 57 is separated in the row direction. However, since the drain line 57 has to pass a signal current of pixels for one row, it is actually in the column direction. Wiring is common to all rows so that current can be passed through.

  The signal charge generated by the charge generation unit 32 (photoelectric conversion element) is transferred to the floating diffusion 38 by the read selection transistor 34.

  Here, unlike the first example, the unit pixel 3 of the second example is not provided with the vertical selection transistor 40 connected in series with the amplifying transistor 42. A large number of pixels are connected to the vertical signal line 53, but the pixels are selected by controlling the FD potential instead of the selection transistor. Usually, the FD potential is set to low. When selecting a pixel, the signal of the selected pixel is output to the vertical signal line 53 by setting the FD potential of the selected pixel to high. Thereafter, the FD potential of the selected pixel is returned to low. This operation is performed simultaneously for one row of pixels.

  In order to control the FD potential in this way, 1) when the selected row FD potential is made high, the drain line 57 is made high, and the FD potential is made high through the reset transistor 36 of the selected row. 2) When the selected row FD potential is returned to low, the drain line 57 is set low and the FD potential is set low through the reset transistor 36 of the selected row.

  However, when the inventor of the present application prototyped a solid-state imaging device (device) composed of the second type of unit pixel 3, 1) the shading phenomenon is different between the peripheral pixel and the central pixel. In particular, the maximum charge amount (saturated electron number) that can be stored in the photoelectric conversion element is small in the center, and 2) the dynamic range is small.

  Regarding the above two problems, the inventor of the present application analyzed these phenomena and clarified the following.

  1) Since the drain line 57 is a wiring that extends over almost the entire area of the pixel portion, the potential of the well of the pixel portion (Well, which will be described below as a representative of the P-type well) fluctuates when the drain line 57 is driven. End up. A contact for applying a potential to the P-well is placed around the pixel portion. However, depending on whether the contact is near or far from this contact, the P-well swings differently and changes the characteristics of the pixel. In particular, when the drain line 57 is set to low, the P well is shaken negatively, so that the signal charge leaks from the charge generation unit 32 to the floating diffusion 38 and the P well. Since the center of the P well is far from the P well contact, the number of saturation electrons decreases at the center. This is called saturation shading.

  2) After a period for driving the pixels in the selected row and reading a signal (H invalid period), there is a period for sequentially outputting the signals to the outside (H valid period). In the H valid period, the drain line 57 is set high. In the case of driving, the FD potential gradually rises due to the leakage current of the reset transistor 36. For this reason, since the difference between the selected row and the non-selected row becomes small, the dynamic range is regulated here and becomes small.

  The above two problems and analysis results are all new matters that do not exist in the type of CMOS sensor in which pixels are selected by the vertical selection transistor 40.

  The present invention has been made in view of the above circumstances, and when using a device including a unit pixel having a three-transistor configuration, a driving technique capable of improving a reduction in dynamic range due to a leakage current of a reset transistor. The purpose is to provide.

  A solid-state imaging device according to the present invention is a solid-state imaging device including a unit pixel including a charge generation unit that generates a signal charge corresponding to received light and three transistors, and a signal in the charge storage unit The reset unit that resets the charge is configured by a depletion type transistor (corresponding to an improvement method by a fourth approach in an embodiment described later). In this case, it is desirable that the transistor of the reset unit be capable of setting the reset level of the charge storage unit to about the voltage level when the drain wiring is on.

  The present invention analyzes a problem in a unit pixel having a three-transistor configuration described in the prior art and, in detail, will be described in an embodiment described later, finds a solution to the problem (operation principle and its effect). That's what was done.

  The improvement of the saturation shading is made by the methods according to the first to third approaches described later, but the method by the fourth approach concerning the improvement of the dynamic range reduction, which is the point of the present invention, is a reset unit that constitutes a unit pixel. It was discovered that by making the reset transistor of the depletion type, the leakage current to the charge storage part such as floating diffusion can be suppressed, and thereby the dynamic range of the charge storage part can be expanded. It was made.

In addition, the improvement method by each of these first to fourth approaches is not limited to being applied alone, and can be applied in any combination.
That is, the transfer wiring connected to the transfer gate unit, which is connected in common with the other unit pixels, the transfer driving buffer for driving the transfer wiring, and the reset unit connected in common with the other unit pixels. The reset wiring connected, the reset driving buffer for driving the reset wiring, the drain wiring connected to the reset unit and the pixel signal generation unit connected in common with other unit pixels, and the drain wiring The drain drive buffer to be driven is included as a component of the unit pixel, and the transition time at the OFF time in the voltage waveform of the drain wiring driven by the drain drive buffer when the drive pulse is applied to the drain drive buffer is Reset wiring driven by reset driving buffer and transfer wiring driven by transfer driving buffer For any of the transition time during each off, it can be configured to be in and 10,000 times or less 5 times.
In addition, a transfer wiring connected to the transfer gate unit connected in common with other unit pixels, a transfer driving buffer for driving the transfer wiring, and a reset unit connected in common with other unit pixels The reset wiring connected, the reset driving buffer for driving the reset wiring, the drain wiring connected to the reset unit and the pixel signal generation unit connected in common with other unit pixels, and the drain wiring The drain drive buffer to be driven is included as a component of the unit pixel, and the transition time at the OFF time in the voltage waveform of the drain wiring driven by the drain drive buffer when the drive pulse is applied to the drain drive buffer is Reset wiring driven by reset driving buffer and transfer wiring driven by transfer driving buffer It can be configured to be longer than any of the transition time during each off.
In addition, a drain wiring connected to the reset unit and the pixel signal generation unit, which is connected in common with other unit pixels, is provided, and the off-voltage supplied to the transfer gate unit With respect to the master reference voltage that defines the reference voltage, a voltage value having a polarity opposite to the ON voltage supplied to the transfer gate portion can be used.
In addition, a transfer wiring connected to the transfer gate portion connected in common with other unit pixels, a reset wiring connected to a reset portion connected in common with other unit pixels, and other unit pixels Commonly connected drain wiring connected to the reset unit and the pixel signal generation unit, and a signal line connected to other unit pixels that receive the pixel signal generated by the pixel signal generation unit are provided. The pixel selection operation for outputting the pixel signal generated by the pixel signal generation unit to the signal line is performed by controlling the potential of the charge storage unit, and further receives a drive pulse for driving the drain wiring, The transition time at the OFF time in the voltage waveform when driving the drain wiring is the same as that at each OFF time in the voltage waveform when driving the reset wiring and the transfer wiring. It can be configured to include a waveform shaping unit that performs waveform shaping to be longer than any of the transfer time.

  According to the present invention, a method for solving the problem of reduction in dynamic range in a solid-state imaging device including unit pixels having a three-transistor configuration (an improvement method based on the fourth approach described above) The reset transistor is a depletion type. As a result, leakage current due to the reset transistor can be suppressed. As a result, the dynamic range of the charge storage unit can be expanded, and the dynamic range that has been limited by the conventional floating diffusion (charge storage unit) can be expanded.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following, a case where the present invention is applied to a CMOS image sensor, which is an example of an XY address type solid-state imaging device, will be described as an example. The CMOS image sensor will be described on the assumption that all pixels are made of NMOS.

<Configuration of solid-state imaging device>
FIG. 1 is a schematic configuration diagram of a CMOS solid-state imaging device according to an embodiment of the present invention. The solid-state imaging device 1 is applied as an electronic still camera that can capture a color image. For example, in a still image capturing mode, a mode for sequentially reading all pixels is set. .

  The solid-state imaging device 1 includes an imaging unit in which pixels including light receiving elements that output a signal corresponding to the amount of incident light are arranged in rows and columns (that is, in a two-dimensional matrix), and a signal output from each pixel is a voltage. It is a signal and is a column type in which a CDS (Correlated Double Sampling) processing function section is provided for each column. That is, as illustrated in FIG. 1A, the solid-state imaging device 1 includes a pixel unit (imaging unit) 10 in which a plurality of unit pixels 3 are arranged in rows and columns, and a drive provided outside the pixel unit 10. A control unit 7 and a CDS processing unit (column circuit) 26 are provided. As the drive control unit 7, for example, a horizontal scanning circuit 12 and a vertical scanning circuit 14 are provided.

  In FIG. 1A, some of the rows and columns are omitted for the sake of simplicity, but in reality, tens to thousands of pixels are arranged in each row and each column. In addition, a timing generator (an example of a read address control device) 20 that supplies a pulse signal at a predetermined timing to the horizontal scanning circuit 12, the vertical scanning circuit 14, and the CDS processing unit 26 is provided as another component of the drive control unit 7. It has been. Each element of these drive control units 7 is formed integrally with a pixel unit 10 in a semiconductor region such as single crystal silicon using a technique similar to a semiconductor integrated circuit manufacturing technique, and is a solid-state imaging which is an example of a semiconductor system It is configured as an element (imaging device). Each unit pixel 3 of the pixel unit 10 is connected to ground (GND) as a master reference voltage that defines the reference voltage of the entire device.

  The timing generator 20 may be provided as a separate semiconductor integrated circuit independently of other functional elements such as the pixel unit 10 and the horizontal scanning circuit 12. In this case, an imaging apparatus is constructed by the imaging device including the pixel unit 10 and the horizontal scanning circuit 12 and the timing generator 20. This imaging device may be provided as an imaging module in which peripheral signal processing circuits, power supply circuits, and the like are also incorporated.

  The unit pixel 3 is connected to a vertical scanning circuit 14 via a vertical control line 15 and a CDS processing unit 26 via a vertical signal line 19 for selecting a vertical column. Here, the vertical control line 15 indicates all the wiring that enters the pixel from the vertical scanning circuit 14. For example, in the pixel in FIG. 19B, when the transfer wiring 55 and the reset wiring 56 and the drain line enters from the vertical scanning circuit 14, the drain line is also included. The horizontal scanning circuit 12 and the vertical scanning circuit 14 are configured to include, for example, a decoder, and start a shift operation (scanning) in response to a driving pulse supplied from the timing generator 20. Therefore, the vertical control line 15 includes various pulse signals (for example, a reset pulse RST, a transfer pulse TRF, a DRN control pulse DRN, etc.) for driving the unit pixel 3.

  The CDS processing unit 26 as a column circuit is provided for each column, receives a signal of pixels for one row, and processes the signal. For example, on the basis of two sample pulses such as the sample pulse SHP and the sample pulse SHD given from the timing generator 20, the signal level immediately after the pixel reset is applied to the voltage mode pixel signal input via the vertical signal line 19 ( (Noise level) and signal level are processed. As a result, noise signal components called fixed pattern noise (FPN) and reset noise are removed. Note that an AGC (Auto Gain Control) circuit, an ADC (Analog Digital Converter) circuit, or the like may be provided in the same semiconductor region as the CDS processing unit 26 as necessary after the CDS processing unit 26.

  The horizontal scanning circuit 12 defines a horizontal readout column (selects an individual column circuit in the CDS processing unit 26), and a CDS processing unit 26 according to a readout address defined by the horizontal decoder 12a. And a horizontal drive circuit 12b for guiding each signal to the horizontal signal line 18. The vertical scanning circuit 14 defines a vertical readout row (selects a row of the pixel unit 10), and controls the unit pixel 3 on the readout address (row direction) defined by the vertical decoder 14a. And a vertical drive circuit 14b that drives the line by supplying pulses. Note that the vertical decoder 14a selects a row for electronic shutter, in addition to a row from which a signal is read. The timing generator 20 outputs a horizontal address signal to the horizontal decoder 12a and a vertical address signal to the vertical decoder 14a, and each decoder 12a, 14a receives it and selects a corresponding row or column.

  The voltage signal processed by the CDS processing unit 26 is transmitted to the horizontal signal line 18 via a horizontal selection switch (not shown) driven by a horizontal selection signal from the horizontal scanning circuit 12, and further input to the output buffer 28. After that, it is supplied to the external circuit 100 as the imaging signal S0. That is, in the column-type solid-state imaging device 1, the output signal (voltage signal) from the unit pixel 3 is output in the order of the vertical signal line 19 → CDS processing unit 26 → horizontal signal line 18 → output buffer 28. The drive is such that the pixel output signals for one row are sent in parallel to the CDS processing unit 26 via the vertical signal line 19, and the signals after the CDS processing are serially output via the horizontal signal line 18. The vertical control line 15 controls selection of each row.

  As long as each vertical column or horizontal column can be driven, each pulse signal is arranged in the row direction or the column direction with respect to the unit pixel 3, that is, a driving clock line for applying the pulse signal. The physical wiring method is free.

  As will be described later, it is preferable to supply a negative voltage to the pixel through the vertical drive circuit 14b. For this reason, a negative voltage generation circuit may be mounted. Of course, a negative voltage may be supplied from the outside without mounting this.

  As the external circuit 100 of the solid-state imaging device 1, a circuit configuration corresponding to each photographing mode is adopted. For example, as shown in FIG. 1B, an A / D (Analog to Digital) conversion unit 110 that converts an analog imaging signal S0 output from the output buffer 28 into digital imaging data D0, and A / D conversion A digital signal processor (DSP) 130 that performs digital signal processing based on the imaging data D0 digitized by the unit 110. The digital signal processing unit 130 performs, for example, color separation processing to generate image data RGB representing each image of R (red), G (green), and B (blue), and outputs other signals to the image data RGB. Processing is performed to generate image data D2 for monitor output. In addition, the digital signal processing unit 130 includes a functional unit that performs signal compression processing for storing imaging data in a recording medium.

  The external circuit 100 also includes a D / A (Digital to Analog) converter 136 that converts the image data D2 digitally processed by the digital signal processor 130 into an analog image signal S1. The image signal S1 output from the D / A converter 136 is sent to a display device such as a liquid crystal monitor (not shown). The operator can perform various operations while viewing the display image of the display device.

  The details of the unit pixel 3 are omitted, but the unit pixel 3 has the same configuration as that of the three-transistor configuration shown in FIG. The drain line 57 is common to most of the pixels of the pixel portion 10 and extends in the column direction and is common at the end of the pixel portion 10 or is a lattice-like shape having holes on the charge generation portion 32. Wiring. There may be some pixels such as dummy pixels that have separate drain lines 57. Further, although not shown in the figure, a wiring and a contact for supplying a potential of the P well are provided around the pixel portion 10.

  Since the drain line 57 is connected to most or all of the pixels, when the drain line 57 is swung low, the potential of the P well of the pixel unit 10 fluctuates, and the fluctuation width and time are different between the periphery and the center. The charge leaking from the charge generation unit 32 increases in the center, and the saturation signal charge in the center decreases. That is, as described in the section of the prior art, the first problem that the characteristics are different between the peripheral pixel and the central pixel is presented as it is.

  2-10 is a figure explaining the 1st method of the said 1st problem and its countermeasure approach. First, FIG. 2 is a diagram for specifically explaining the first problem (saturation shading phenomenon). As a device, a CMOS sensor conforming to the VGA standard of about 300,000 pixels (640 × 480 pixels) was used. The unit pixel 3 has the three-transistor configuration shown in the second example of the prior art, and the pixel pitch is 4.1 μm. VGA is an abbreviation for “Video Graphics Array” and defines a graphics mode and display resolution.

  The power supply voltage supplied to the prototype device is 3.0 V, and the clock frequency is 6 MHz (frame rate 13.3 fps). The prototype device can change the low level of the transfer gate drive voltage (hereinafter also referred to as the transfer gate low level), and the low level potential (here, 0 V) of the drain line 57 is applied from the outside of the device. It has a supply terminal (DRN drive buffer ground side wiring terminal) DRNL. The other driving for the unit pixel 3 is performed at 0 V (grounding; GND) and the power supply voltage (3.0 V).

  As a measurement method of saturation shading, a signal obtained by the output buffer 28 is observed with a waveform monitor such as an oscilloscope while irradiating a light amount sufficiently saturated by the charge generation unit, and a peripheral monitor The difference from the central part was measured as the shading amount. As shown in FIG. 2, in the prototype device used for verification, a large difference is seen between the peripheral portion and the central portion. And it turns out that the signal output in a center part is smaller than the signal output in a peripheral part.

  FIG. 3 is a diagram showing a measurement circuit for examining the voltage change of the drain line 57. The control resistor 146 is inserted between the ground side wiring terminal DRNL and GND of the DRN drive buffer 140 of the prototype device, and the voltage of the control resistor 146 is measured. The voltage source 149 is set to 0V. The DRN drive buffer 140 is provided in the vertical drive circuit 14b of the vertical scanning circuit 14.

  When a DRN control pulse (pulse-shaped DRN control signal) is input to the DRN drive buffer 140 (not shown) of the prototype device, the voltage waveform measured by the control resistor 146 reflects the current waveform flowing through the DRN drive buffer 140. The voltage waveform in the drain line 57 is also represented. The control resistor 146 is closely related to the second example of the improvement method based on the first approach described later, and the voltage source 149 is closely related to the improvement method based on the second approach described later.

  FIG. 4 is a diagram illustrating the relationship between the resistance value and the saturation shading when the control resistor 146 is inserted between the ground side wiring terminals DRNL and GND and the DRN control pulse is blunted. In the figure, the low level voltage of the transfer gate is indicated by Vtl. The resistance value used for the measurement is 1,10,47,150,330,680,1000 (unit is Ω) according to the E12 series.

  As shown in FIG. 4A, in the case of Vtl = −0.6 V, the change in the saturation shading amount is small in the range of 1Ω to 10Ω, and there is a sign of change in the saturation shading amount from about 10Ω. A big change is seen. That is, if the resistance value of the control resistor 146 is smaller than about 10Ω, it does not affect the current device. If it is about 10Ω or more, the effect of reducing the saturation shading can be obtained, and if it is 50Ω or more, a significant effect can be obtained.

  Further, as shown in FIG. 4B, when Vtl = −1V, a large change is seen in the saturation shading amount even in the range of 1Ω to 10Ω, the smallest in the range of 50Ω to 200Ω, and the higher ( The saturation shading amount tends to increase slightly (for example, up to about 200Ω to 1000Ω). That is, when the resistance value of the control resistor 146 is about 10Ω or more, a significant effect of reducing the saturation shading is seen, and about 50 to 200Ω is considered best.

  FIG. 5 is a diagram showing the results shown in FIG. 4 in relation to the fall time of the voltage waveform (transition time at the OFF time) in the control resistor 146 and saturation shading. The drive pulse shape in the CMOS sensor is generally set to fall time and rise time (on-time transition time) of several ns (for example, 1 to 3 ns) for any of the transfer wiring 55, reset wiring 56, and drain line 57. Below. Therefore, if the fall time and the rise time of the voltage waveform appearing at the control resistor 146 are about several ns or less, it can be considered that the device is generally driven under normal conditions.

  As shown in FIG. 5A, when Vtl = −0.6 V, the saturation shading amount changes until the fall time 10 ns corresponding to the range of 1Ω to 10Ω (normally about 3 to 10 times or more). There is a sign of a change in the saturation shading amount from about 10 ns corresponding to about 10Ω, and a large change is seen at 40 ns or more corresponding to about 50Ω. That is, focusing on the fall time, if it is smaller than about 10 ns, it does not affect the current device. Further, if it is about 10 ns or more, the effect of reducing the saturation shading can be obtained, and if 40 ns or more, a significant effect can be obtained. This effect continues until a fall time of 10,000 ns (normally about 3000 to 10,000 times or less).

  In addition, as shown in FIG. 5B, when Vtl = −1V, a large change is seen in the saturation shading amount even at 10 ns to 40 ns, and the fall time is 40 ns (usually about 13 to 20 times) or more. In particular, the saturation shading amount is the smallest in a range of 170 to 600 to 1000 ns (about 56 to 1000 times the normal value) corresponding to a resistance value of about 50 to 200 Ω, and more than that (for example, about 1000 ns to 5000 ns) ; About 330 to 5000 times normal), the saturation shading amount tends to slightly increase. That is, it is possible to improve the saturation shading by dulling the DRN voltage, and a significant effect of reducing the saturation shading when the fall time is about 40 ns or more is seen, and about 170 to 600 ns (for example, the normal 56 (About 600 times) is considered the best.

  Thus, although the range in which the effect appears depends on the low level voltage Vtl, the transition time at the off time (in this example, the fall time) is approximately 3 to 10 (on average, 5 times the normal time). Saturation shading can be improved by dulling the DRN voltage in the range of about) to 10,000 (10,000) times, more preferably in the range of about 50 to 600 times.

  FIG. 6 to FIG. 10 are diagrams showing the results of reproducing the swing of the P well by simulation. Each is shown for each value of the control resistor 146. Note that the simulation result is shown here because it was difficult to actually measure the swing of the P well. The waveform lines W1 to W4 in each figure are at the respective device positions shown in each figure. Further, as shown in FIG. 6, the waveform line of SEL_0 indicates that of the DRN control pulse, and the waveform line of VSS_D is at the terminal measured in the experiment.

  As shown in the figure, when the value of the control resistor 146 is increased, the fall time of the DRN voltage in the drain line 57 is increased, the fluctuation of the P well is reduced, and the difference between the central portion and the peripheral portion is also reduced. . In other words, increasing the value of the control resistor 146 or increasing the fall time of the DRN voltage leads to the uniform characteristics of the peripheral pixel and the central pixel, which improves saturation shading. It can be seen that the effect is high.

  In the configuration of the present embodiment, based on the above analysis result, a configuration for improving the saturation shading by dulling the DRN voltage is adopted as an improvement method based on the first approach for solving the first problem (saturation shading phenomenon). . Specifically, a mechanism for controlling the fall time when the drain line 57 is swung low and improving the saturation shading phenomenon is provided. This mechanism will be briefly described. First, when the drain line 57 is swung to a low level, a driving method is adopted in which the fall time is lengthened and slowly lowered. As a result, the fluctuation width of the potential of the P well can be reduced, or the P well potential difference between the periphery and the center of the pixel portion 10 can be reduced. In the configuration of the present embodiment, the fall time is significantly (intentionally) longer than that in the normal driving method.

  As a definition method when “making the fall time significantly longer than in the normal drive method”, a method of specifying the ratio (multiple) of the drive pulse fall time in the normal drive method, A method defined by the ratio of the fall time in correspondence with the number of pixels (more specifically, the driving cycle), or the potential difference between the peripheral portion and the central portion of the P well is not more than a predetermined level (a level at which image quality deterioration is not noticeable). Various definition methods are conceivable, such as a method of defining as a period of time.

  Further, when the ratio is specified by the ratio (multiple) of the drive pulse fall time in the normal drive method, it is not limited to the comparison with the fall time of the DRN voltage in its own normal drive, but in comparison with other drive pulses. You may prescribe. For example, the size of the buffer for driving each wiring may be determined so that the falling time of the DRN voltage is longer than the falling time of the transfer wiring and the reset wiring by a predetermined multiple or more.

  The operation of returning the selected pixel to the non-selected state is performed by setting the DRN control pulse to the low level within the blanking period. When the ratio is specified by the ratio of the fall time in correspondence with the driving cycle, the method of defining the maximum value becomes a problem. For example, the maximum value is defined by the low level period of the DRN control pulse, and is actually within this range. It is recommended to specify the fall time of. In the CMOS sensor of this experiment, the low level period of the DRN control pulse (that is, the off period with respect to the drain line 57) is set to about 600 ns.

  Note that setting the fall time to be equal to or longer than the off period with respect to the drain line 57 is not excluded, and in this experiment, the fall time of 600 ns or more is obtained from the extrapolation curve of the measurement data. In this case, it is required to reach a low voltage enough to return the selected pixel to the non-selected state.

  In any case, the problem that the characteristics are different between the peripheral pixel and the central pixel and the cause that is caused by the P-well potential difference are discovered. The improvement method based on the approach is characterized in that the fall time is set to a level at which image quality deterioration (saturation shading phenomenon) due to the P-well potential difference is not noticeable.

  For example, based on the results shown in FIGS. 2 to 10, a falling time that is 10 times or more longer than each falling time of the transfer pulse TRF and the reset pulse RST which are other pulses of the pixel unit 10 is given. For example, the pulse shape in the other part of the CMOS sensor has a fall time of about several ns or less, and this is set so that the DRN voltage at the drain line 57 is 40 ns (nanoseconds) or more. This 40 ns is a period of about half of the pixel clock period when an image is output at 30 frames / second from a VGA (about 300,000 pixels) CMOS sensor. Here, the case of a VGA-compliant CMOS sensor is shown, but it is considered that even a display sensor with other display resolutions may have a period longer than about half of the pixel clock cycle.

  If the display resolution, that is, the total number of pixels is different, the absolute amount of the fall time is naturally different accordingly. Here, the fall time may be a general definition, that is, a transition time from 90 to 10 where the high level is 100 and the low level is 0. Hereinafter, a specific example of the improvement technique based on the first approach for solving the above-described saturation shading problem will be described.

<Improvement method by the first approach; First example>
FIG. 11 is a diagram illustrating a first example of a method (fall time control method) for controlling the fall time of the drive voltage applied to the drain line 57 in accordance with the improvement method based on the first approach. Here, FIG. 11A is a conceptual diagram focusing on a circuit for driving the drain line 57, and FIG. 11B shows a detailed example of a DRN drive buffer (hereinafter also simply referred to as a buffer) 140 for driving the drain line 57. FIG. 11C is a diagram showing an example of drive timing.

  As shown in FIG. 11A, a drain line 57 extends in the column direction corresponding to each column of the pixel portion 10 and is connected to an output terminal of a DRN drive buffer (hereinafter also simply referred to as a buffer) 140 at the lower end. ing. The buffer 140 is in each column, and a control pulse (DRN control pulse) for driving the drain line 57 is applied from the outside of the pixel unit 10. In response to this, each buffer 140 performs the same drive with respect to the drain line 57 of each column. That is, the drain line 57 in each column is common to all the pixels.

  As shown in FIG. 11B, the buffer 140 of the drain line 57 is configured using two CMOS inverters 142 and 144. Each of the inverters 142 and 144 is composed of an NMOS transistor indicated by symbol a and a PMOS transistor indicated by symbol b. Each transistor is collectively referred to as a buffer transistor. Here, normally, the final stage inverter 144 connected to the drain line 57 increases the W / L ratio (W: gate width, L: gate length) of the buffer transistor so that both the rise time and fall time are Try not to be long. For example, as shown in FIG. 11A, a buffer 140 is provided in each column of drain lines 57. In the case of the number of pixels VGA class, the fall time is set to several ns or less. The W / L ratio of the final stage NMOS transistor 144a is set to, for example, about 5 to 10 / 0.6 (10 / 0.6 in a typical example).

  On the other hand, in the configuration of the first fall time control method, the fall time is positively (intentionally) by making the W / L ratio of the buffer transistor smaller than that of the normal (conventional) configuration. )Lengthen. In particular, the fall time is intentionally increased without increasing the W / L ratio of the NMOS transistor 144a. For example, in comparison with the above (the number of pixels VGA class in the configuration of FIG. 11A), it may be set to about 1 / 0.6 to 1/20.

  That is, it is set to be significantly smaller than that of the conventional configuration. For example, in the previous example, the normal ratio may be set to about 1/10 to 1/320. Of course, in an example, this may be set to a range of at least 1/5 to 1/500.

  In this case, it is defined by comparison with the W / L ratio of the final stage of the buffer in the conventional configuration. However, the rise time of the transistors that drive the transfer wiring (read selection line) 55 and the reset wiring 56 in the unit pixel 3 Since the fall time is set to several ns or less, the above-mentioned numerical relationship can be said to be the same in comparison with these W / L ratios. That is, the W / L ratio of the transistor connected to the drain line 57 is any of the W / L ratio of the transistor connected to the transfer wiring 55 and the W / L ratio of the transistor connected to the reset wiring 56. Is preferably set in the range of 1/5 to 1/500, more preferably in the range of 1/10 to 1/320. It is impossible to make the W / L ratio at the final stage of the buffer such a small value in a normal design.

  As a result, as shown in FIG. 11C, the fall time of the DRN control pulse applied to the buffer 140 is about several ns or less, but the DRN voltage output from the buffer 140 and drives the drain line 57. This pulse shape has a fall time of 40 ns or more. As a result, the problem that the number of saturated electrons is small at the center of the pixel unit 10 can be solved, the P-well potential difference can be reduced to a practical level, and image quality deterioration called saturation shading can be improved.

  The rising side does not affect the number of saturated electrons. However, it is similar to the falling in that the P-well is shaken and the difference between the periphery and the center is different. Although not confirmed in the prototype of the present inventor, when there is a low-voltage N-type diffusion layer in the pixel, it becomes a forward bias with the P well, and electrons are injected into the P well. May enter the charge generation unit 32. Therefore, if time permits, it is preferable to make the PMOS of the final stage of the buffer (ie, the inverter 144) small so that the rise also becomes slow. However, since the signal current of the pixel flows when the drain line 57 is high, it is necessary to prevent the voltage from decreasing so as to cause a problem.

<Improvement method by the first approach; modification of the first example>
FIG. 12 is a diagram for explaining a modification of the technique for realizing the fall time control method of the first example. Here, FIG. 12A is a conceptual diagram focusing on a circuit that drives the drain line 57, FIG. 12B is a diagram showing a detailed example of the DRN drive buffer 140 that drives the drain line 57, and FIG. ) Is a chart in which the W / L ratio in the fall time control method of the first example is arranged in comparison with the conventional example.

  As shown in FIG. 12A, this modification is characterized in that the drain line 57 is driven from the lateral direction of the pixel portion 10. The left and right ends of the pixel unit 10 are connected to the output terminal of the buffer 140. The drain line 57 is a grid-like wiring with a hole on the photodiode (charge generation unit 32). The buffer 140 is provided in each row, and the buffer 140 performs the same drive with respect to the drain line 57 of each row by a DRN control pulse from the outside of the pixel unit 10. Here, only the low-side power supply wiring in the final stage of the buffer, that is, the source terminal of the NMOS transistor 144b is explicitly drawn out, and this wiring is a GND wiring as shown in FIG.

  As described above, in the configuration in which the drain line 57 is driven from the lateral direction and the fall time is set to several ns or less in the case of the number of pixels VGA class, conventionally, the W / L ratio of the NMOS transistor 144b in the final stage of the buffer is Similarly to the configuration in which the buffer 140 is provided in each column, for example, it is set to about 5 to 10 / 0.6 (6 / 0.6 in a typical example).

  On the other hand, in this modification, the W / L ratio of the NMOS transistor 144b is set to about 1/1 to 1/20. That is, it may be set to a range of about 1/10 to 1/200 with respect to the conventional configuration (normal ratio). Of course, for example, this may be set to at least about 1/5 to 1/300.

  It is impossible to make the W / L ratio at the final stage of the buffer such a small value in a normal design. By doing so, as shown in FIG. 11C, the pulse shape for driving the drain line 57 can have a fall time of 40 ns or more and is similar to the configuration in which the buffer 140 is provided in each column. You can enjoy the effect.

  The fall time control method of the first example is not limited to the configuration shown in FIGS. 11A and 12A, and the drain line 57 on the entire pixel portion is connected to one DRN drive buffer 140. A driving structure can also be adopted. Such a configuration is not practically adopted in a normal design, but can be adopted in the first example. In this case, if a conventional mechanism is adopted in which the fall time is several ns or less, the W / L ratio of the buffer final stage NMOS is set to, for example, about 5000 / 0.6. On the other hand, when the mechanism of the first example is adopted, the W / L ratio of the NMOS transistor 144b is about 500 / 0.6 to 2 / 0.6 (the normal ratio is in the range of 1/10 to 1/2500). The fall time is set to 40 ns or more.

  As described above, according to the fall time control method of the first example, the W / L ratio of the transistor that constitutes the buffer that drives the drain line 57 is set smaller than the normal (conventional) configuration. I did it. As a result, the fall time of the drive voltage can be lengthened positively (intentionally). As a result, even in a three-transistor type pixel structure without a selection transistor, it is possible to prevent a decrease in the number of saturated electrons in the central portion of the pixel portion due to shaking of the well. As a result, the image quality degradation caused by the P-well potential difference can be reduced to an inconspicuous level in practice, and the image quality is improved.

<Improvement method by the first approach; second example>
FIG. 13 is a diagram for explaining a second example of the fall time control method. Here, FIG. 13A is a conceptual diagram focusing on the circuit that drives the drain line 57, and FIGS. 13B, 13C, and 13D show a modification of the second example. FIG.

  In the second example, the drain line 57, the low-side power supply wiring (off-side reference wiring) of the buffer 140 that drives the drain line 57, and the low-level voltage source (off-side with respect to the drain line 57) are driven. This is characterized in that a resistor element for limiting the drive current is inserted between the reference power source (including GND) that regulates the voltage).

  The basic configuration of the drive circuit is the same as that of the modified example of the first example shown in FIG. The difference is that the low-side power supply wiring at the final stage of the buffer, that is, the source terminal of the NMOS transistor 144b of the inverter 144, which is explicitly drawn out, is not directly connected to the GND wiring, but the control resistor 146 is connected to the GND wiring. Connect through.

  In the drawing, for the sake of convenience, the row power supply wiring at the final stage is linearly extended for each of the buffers 140 in one vertical column. Although not shown, the detailed example of the DRN drive buffer 140 that drives the drain line 57 is the same as the modified example of the first example shown in FIG. The technique of the second example can be similarly applied to the configuration shown in FIG. 11A and the configuration in which the drain line on the entire pixel portion is driven by one DRN drive buffer.

  According to the method of the second example, even if the W / L ratio of the buffer transistor is increased as usual, by using this control resistor 146, the drain line 57 is set low as in the method of the first example. The fall time when swinging to can be extended. Therefore, similar to the technique of the first example, the effect of improving the saturation shading can be enjoyed.

  In the method of adjusting only the W / L ratio of the transistor as in the method of the first example, if the W / L ratio is determined at the time of designing, the correction is not easily performed. On the other hand, in the second method, the resistance value can be changed only by changing one manufacturing mask. Alternatively, as shown in FIG. 13B, a configuration in which a plurality of resistance elements are provided in advance and the resistance elements are selected by an internal program (may be selected in any combination) (resistance switching circuit) Can also be taken. In this case, the resistance value can be changed very easily. As a matter of course, the control resistor 146 and the resistance switching circuit may be provided outside the device.

  As can be seen from the device analysis shown in FIGS. 2 to 10, according to experiments, when a control resistor 146 having a resistance value of about 50Ω to 200Ω is used as the control resistor 146, In this way, it is possible to obtain a good result with no problem in the operation speed while preventing the decrease in the number of saturated electrons. If the modification shown in FIG. 13B is applied, a suitable value can be found under actual device conditions, and the suitable resistance value can be set, which is convenient.

  Note that the method of the second example is not limited to the configuration shown in FIG. 13, that is, the device shown in FIG. 12A, but can be applied to the device shown in FIG. is there. Further, the control resistor 146 is not limited to be inserted between the control resistor 146 and the GND, but may be inserted in each buffer 140. In this case, as shown in FIG. 13C, a configuration that is inserted between the ground-side wiring terminal of each buffer 140 and GND, a configuration that is inserted in the output side of each buffer 140 as shown in FIG. A combination of these can be taken. The configuration of FIG. 13D is effective when not only the falling but also the rising is gentle. The configuration placed on the source side of each buffer 140 is one in which the control resistor 146 is distributed to each buffer 140 and is substantially equivalent to the configuration shown in FIG.

<Improvement method by the first approach; third example>
FIG. 14 is a diagram for explaining a third example of the fall time control method. Here, FIG. 14A is a conceptual diagram focusing on a circuit for driving the drain line 57, and FIG. 14B is a diagram showing an example of drive timing.

  The third example defines the drain line 57, the low-side power supply wiring (off-side reference wiring) and the low-level voltage source (off-side voltage with respect to the drain line 57) of the buffer 140 that drives the drain line 57 (particularly the last stage of the buffer). This is characterized in that a current source for defining a drive current is inserted between the power source and the reference power source (including GND). Specifically, the control resistor 146 used in the method of the second example is replaced with a current source 148. In this configuration, the low level voltage source (corresponding to the voltage source 149 shown in FIG. 3) is replaced with GND. The technique of the third example can be similarly applied to the configuration shown in FIG. 11A and the configuration in which the drain line on the entire pixel portion is driven by one DRN drive buffer.

  The fall time of the drain line 57 can be controlled by the current value controlled by the current source 148. The current source 148 may include only one N-type transistor or may be configured to control the current with a current mirror. In short, any current source may be used as long as the flowing current can be maintained substantially constant. Various configurations can be applied. By adjusting the driving current value, the above-mentioned conditions can be satisfied, and an optimum state with a small amount of saturation shading can be set. If the set current value is variable, the configuration is more preferable. The constant current source is a normal one as described above, and a constant current cannot be passed in the vicinity of 0V, and the curve becomes loose in FIG. 14B and settles to 0V.

  In the methods of the first example and the second example described above, as shown in FIG. 11C, the DRN potential (the output voltage of the buffer 140) suddenly drops at the beginning of the fall. On the other hand, according to the technique of the third example, it is possible to suppress (control) the sudden drop of the DRN potential over the entire falling period. Therefore, similarly to the techniques of the first and second examples, the fall time when the drain line 57 is swung low can be extended, and the effect of improving the saturation shading can be enjoyed.

<Improvement method by the second approach>
Next, a second approach for improving saturation shading will be described from a different aspect from the improvement method according to the first approach shown in the first to third examples.

  FIG. 15 is a diagram showing the relationship between the transfer gate low level and saturation shading. The measurement condition is that the resistance value of the control resistor 146 is 0Ω (the control resistor 146 is not provided and the ground side wiring terminal DRNL is connected to GND).

  As shown in FIG. 15A, it can be seen that the absolute value of saturation shading decreases when the transfer gate low level is about −0.7 V or less. Further, as shown in FIG. 15B, the shading amount with respect to the edge saturation signal, that is, the shading ratio, becomes smaller when the transfer gate low level is made negative, and is constant at about −0.8V. I understand.

  In the improvement method by the second approach, paying attention to this point, the transfer gate low level is negative in order to form a potential barrier against charge leakage from the charge generation unit 32 to the floating diffusion 38 (charge storage unit). The voltage that can be set was used.

  FIG. 16 is a diagram for explaining an improvement method based on the second approach. The transfer drive buffer 150 already described in FIG. 19B has a level shifter 160 and an output buffer 161, and outputs an input pulse having a low level of GND as a pulse having a low level of negative voltage. This negative voltage is supplied from the built-in negative voltage generation circuit 162. The negative voltage generation circuit 162 may be a general charge pump circuit. Of course, a negative voltage may be supplied from the outside without including the negative voltage generation circuit 162.

  By making the low level (Vtl) of the transfer transistor gate voltage negative, a decrease in the number of saturated electrons (saturation shading) at the center of the pixel portion can be suppressed. This is because, by making the low level voltage Vtl negative, a potential barrier against charge leakage from the charge generation unit 32 to the floating diffusion 38 can be increased. Note that the maximum value on the minus side is set so that the device does not break down.

  As shown in FIG. 15, according to an experiment, the ratio of shading to the saturation signal amount is reduced by making the low level voltage Vtl negative. This method can be operated independently of the improvement method by the first approach for dulling the DRN voltage shown in the first to third examples. As can be seen from FIG. 15, if the set voltage value is variable, the configuration is more preferable.

  The relationship between the low level voltage Vtl and the saturation shading shown in FIG. 15 shows the effect of the low level voltage Vtl without dulling the drain line 57. In this figure, the absolute value of the shading amount is −0.7V or less and smaller than 0V. Below −0.8 V, both the saturation signal amount and the shading amount are constant. This is because when the voltage is −0.8 V or less, a channel of a hole having a polarity opposite to that of the signal charge is generated at the Si-oxide film interface (Si semiconductor interface) constituting the unit pixel 3, and the low level voltage Vtl is applied to the channel. This is because the hole state of the channel only changes, and the bulk state does not change even if it is lowered. Such a phenomenon is called a pinning phenomenon. Therefore, based on the above experiment, it is desirable to set the output voltage of the voltage source 149 to about −0.7V or less. More preferably, it is set to a value (for example, about −0.8 V) or less sufficient to generate a hole channel at the semiconductor interface.

  The inventor of the present application has proposed a technology for making the low level voltage Vtl a negative voltage for the unit pixel 3 having a four-transistor configuration including a selection transistor in Japanese Patent Application No. 2001-6657. This point is common to the technique described in the fourth example. However, while the technique in Japanese Patent Application No. 2001-6657 is intended to reduce dark current, the purpose of the technique of the fourth example is for the unit pixel 3 having a three-transistor configuration. This is intended to suppress the phenomenon that the saturation voltage decreases at the center, and the mutual purpose is different. That is, the phenomenon targeted by the technique of the fourth example is that the unit pixel 3 does not include a selection transistor connected in series with the amplifying transistor 42, and the pixel is selected through the reset transistor 36 by changing the DRN potential. It is unique. According to the configuration of the fourth example, the saturation shading problem peculiar to the three-transistor configuration can be suppressed by making the low level voltage Vtl of the unit pixel 3 negative.

<Improvement method by the third approach>
Next, an improvement method based on the third approach will be described. The third approach is characterized in that the unit pixel 3 is configured to have a wiring for fixing the well potential. Specifically, the well potential is fixed by using a vice wiring and a contact (well contact) for applying a potential to the P well.

  FIG. 17 is a diagram for explaining an improvement technique based on the third approach. Here, a conceptual diagram focusing on the unit pixel 3 is shown. As shown in FIG. 17, for each unit pixel 3, a P well bias line 59 that applies a potential to the P well is disposed in parallel with the vertical signal line 53 in the pixel. For each unit pixel 3, a P well contact (hereinafter also referred to as a well contact) 59a, which is an example of a contact portion for connecting the P well bias line 59 and the well, is provided at a predetermined position of the P well bias line 59. With this structure, it is possible to suppress the fluctuation width and time of the P-well potential, and the characteristics of the peripheral pixel and the central pixel can be made uniform. In other words, using the P well contact 59a to reduce the influence of P well fluctuation is highly effective in improving the saturation shading.

  The mechanism based on the third approach is preferably executed instead of taking the measures of the first and second approaches. Of course, you may combine with the improvement method by the 1st or 2nd approach.

  In the case of a four-transistor pixel having a selection transistor, which may have a large pixel, a well capacitor may be placed in the pixel. However, even if there is no well capacitor, it is clear that there is no major problem because most CMOS sensors currently announced and commercialized do not have a well capacitor in the pixel. Of course, it does not have the mechanism of the improvement method by the first approach.

  However, the three-transistor unit pixel 3 in which the selection transistor is omitted omits the selection transistor for the purpose of reducing the pixel size, and as shown in FIG. It goes against reducing the pixel size. For this reason, considering a normal design approach, it is difficult to select a configuration in which a well capacitor is applied to a three-transistor configuration.

  However, as shown in FIG. 17, by providing the pixel with a well contact 59a, a phenomenon peculiar to the three-transistor type in which the selection transistor is omitted, such as a decrease in the number of saturated electrons at the center of the pixel portion, can be prevented. In addition, the improvement method by the third approach has a great effect in that the area can be smaller than that of providing the selection transistor.

  In the illustrated example, a P-well contact 59a is prepared for each unit pixel 3 in order to prevent image unevenness due to well voltage unevenness. If this event can be tolerated, not only every pixel but also every few pixels, the places where the P-well contacts 59a are arranged may be scattered.

<Improvement method by the fourth approach>
Next, the improvement method by the 4th approach is demonstrated. The fourth approach is characterized in that the reduction of the dynamic range, which is peculiar to the three-transistor type having no selection transistor, is eliminated by making the reset transistor 36 constituting the unit pixel 3 into a depletion type. The circuit configuration itself of the unit pixel 3 may be the same as that applied in each approach described above, and only the element structure used as the reset transistor 36 is different.

  First, the problem of dynamic range reduction will be described. If the drain line 57 is kept low, electrons may leak from the drain line 57 to the charge generation unit 32 (photoelectric conversion element) via the floating diffusion 38, resulting in noise. For this reason, the drain line 57 is kept high during the horizontal effective period that occupies most of the time. However, even if the reset transistor 36 is turned off at this time, the drain line from the floating diffusion 38 due to the leakage current of the reset transistor 36. Electrons escape to 57 and the potential of the floating diffusion 38 rises. In particular, during low-speed operation, it has been confirmed that the potential of the floating diffusion 38 increases from 100 mV to 400 mV from the beginning in one frame.

  As the unit pixel 3, a three-transistor type having no selection transistor uses a property that a pixel having the highest potential of the floating diffusion 38 is selected from a large number of pixels connected to the vertical signal line 53. In the selected pixel, the floating diffusion 38 is reset to a high level, and then the signal charge (photoelectrons) of the charge generation unit 32 is transferred to the floating diffusion 38. At this time, the FD potential swings to the lower side.

  Therefore, when the FD potential of the non-selected pixel is increased, the potential difference from the selected pixel is reduced and the dynamic range cannot be obtained. This phenomenon of dynamic range reduction is a phenomenon peculiar to a three-transistor type pixel having no selection transistor. In the improvement method by the fourth approach, the reset transistor 36 is made to be a depletion type in order to avoid this dynamic range reduction.

  FIG. 18 is a diagram for explaining an improvement technique based on the fourth approach. Here, FIG. 18A is a timing chart of drive pulses. FIG. 18B and FIG. 18C are voltage potential diagrams.

  In the operation of the selected pixel, the floating diffusion 38 is set to a high level by the first reset pulse (RST). Next, signal charges are introduced into the floating diffusion 38 by a transfer pulse (TRF), and the potential of the floating diffusion 38 is lowered. At this time, the FD potential of the selected pixel is higher than the FD potential of other pixels connected to the same vertical signal line 53, which is a condition for reading. Thereafter, when the drain line 57 is set to low and a reset pulse (RST) is applied, the floating diffusion 38 returns to low.

  FIG. 18B and FIG. 18C are potential diagrams thereof. FIG. 18B shows the case where the reset transistor 36 (indicated by “RST” in the figure) is not a depletion type, and the initial low level is determined by the low level of the drain line 57. Then, during the non-selection period of one frame, electrons gradually leak from the floating diffusion 38 (indicated by “FD” in the figure) to the drain line 57 (indicated by “DRN” in the figure), causing a voltage increase. On the other hand, the high level of the floating diffusion 38 in the selected pixel is determined by the channel voltage Vch (ON) when the reset transistor 36 is turned ON. Specifically, it is a slightly low value. From that state, it receives photoelectrons and swings low. Therefore, the dynamic range of the floating diffusion 38 is as shown in the figure excluding the margin.

  FIG. 18C shows a case where the reset transistor 36 is a depletion type. The channel voltage Vch moves downward in the figure as the threshold value is low. However, if the amplitude of the reset pulse RST is the same, the amplitude ΔVch is the same as in FIG. The initial low level is determined by the channel voltage Vch (off) when the reset transistor 36 is turned off. In FIG. 18B, there is a threshold difference between Vch (off) and the initial low level, but there is no difference in FIG. 18C.

  Although electrons gradually leak from this state, even if the reset transistor 36 is turned off when the pixel in the next row is selected and the drain line 57 swings low, it is a depletion type. The floating diffusion 38 of the previous pixel returns to the initial low level again. Each time the row advances, the potential of the floating diffusion 38 returns to the initial low level each time. Therefore, the potential rise is small even after one frame.

  For these reasons, as can be seen in FIG. 18C, when the reset transistor 36 is a depletion type, the dynamic range of the floating diffusion 38 can be expanded, and the unit of the three-transistor configuration without the vertical selection transistor 40. The problem of the dynamic range reduction peculiar to the pixel 3 can be improved.

  Incidentally, by using this margin, the degree of depletion can be made deeper and Vch (ON) can be set to be higher than the high level of the drain line 57. That is, the reset transistor 36 is of a depletion type that is deep enough to reset the floating diffusion 38 to the high level of the DRN voltage when the power supply voltage is applied to the gate. Thereafter, when the reset transistor 36 is turned off, the high level of the floating diffusion 38 is reduced by the contribution of feedthrough or the like from the high level of the DRN voltage. In this case, there is an advantage that the high level of the floating diffusion 38 is determined by the high level of the drain line 57 and the threshold variation of the reset transistor 36 is not applied to the pixel output.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. Various changes or improvements can be added to the above-described embodiment without departing from the gist of the invention, and embodiments to which such changes or improvements are added are also included in the technical scope of the present invention.

  Further, the above embodiments do not limit the invention according to the claims (claims), and all combinations of features described in the embodiments are not necessarily essential to the solution means of the invention. Absent. The embodiments described above include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. Even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, as long as an effect is obtained, a configuration from which these some constituent requirements are deleted can be extracted as an invention.

  For example, in the improvement method based on the first approach, when the supply of the drive pulse having the normal pulse shape input to the DRN drive buffer 140 is received, the drain wiring voltage is turned off. A configuration example that can increase the transition time is shown. As a specific means for that purpose, a method such as optimizing the W / L ratio of the transistors constituting the driving buffer, or optimizing the operating current during driving using a control resistor or a current source, etc. Had been applied.

  However, the present invention is not limited to this configuration, and various control methods and structures can be used as long as the voltage transition time of the drain wiring can be increased, and these are also included in the technical idea of the present invention. .

  For example, the device side uses the same three-transistor unit pixel as the conventional one, but the drive signal input to the DRN drive buffer 140 is not in a pulse shape but the waveform of itself. It is good also as a structure which dulls and inputs so that the satisfy | filled conditions may be satisfied. For this purpose, it is preferable to provide a waveform shaping circuit between the timing generator that generates a pulse-shaped drive signal and the device (vertical scanning circuit 14 in the previous example) so that the pulse satisfies the above-described conditions. Thus, as described in the above embodiment, the characteristics of the peripheral pixels and the central pixels can be made uniform, and the saturation shading amount can be reduced.

  In the above-described embodiment, the sensor composed of unit pixels composed of NMOS has been described as an example. However, the present invention is not limited to this, and the potential relationship is also inverted (the positive / negative of the potential is reversed) for a pixel composed of PMOS. By conversely, it is possible to enjoy the same operations and effects as described in the above embodiment.

  In the above-described embodiment, the pixel of one photodiode and three transistors has been described as an example. However, the present invention is not limited to this, and a reset transistor is used for two photodiodes and two readout selection transistors. The same applies to pixels that operate in principle, such as sharing amplifying transistors one by one.

  In the above, we focused on saturation shading, which was the most experimentally affected. However, if the well potential fluctuates differently in the center and the periphery of the pixel portion, characteristics other than the saturation signal amount also have shading. That is obvious. Dulling the fall time or the rise time for driving the drain line reduces the fluctuation of the potential of the well and brings it closer to the uniform, so that the shading phenomenon other than the saturation signal amount is also improved.

1 is a schematic configuration diagram of a CMOS solid-state imaging device according to an embodiment of the present invention. It is a figure explaining a saturation shading phenomenon. It is a figure which shows the measurement circuit for investigating the voltage change of a drain line. It is a figure which shows the relationship between resistance value and saturation shading when a DRN control pulse is blunted. FIG. 5 is a diagram showing the result shown in FIG. 4 in relation to the fall time of the voltage waveform in the control resistor and the saturation shading. It is a figure which shows the result of having reproduced the fluctuation of P well by simulation (control resistance 146 = 0ohm). It is a figure which shows the result of reproducing the fluctuation of P well by simulation (control resistance 146 = 10Ω). It is a figure which shows the result of having reproduced the fluctuation of P well by simulation (control resistance 146 = 150 ohms). It is a figure which shows the result of having reproduced the fluctuation of P well by simulation (control resistance 146 = 330 ohms). It is a figure which shows the result of having reproduced the fluctuation of P well by simulation (control resistance 146 = 680 ohms). It is a figure explaining the 1st example of the fall time control method according to the improvement technique by the 1st approach. It is a figure explaining the modification of the method of implement | achieving the fall time control method of a 1st example. It is a figure explaining the 2nd example of the fall time control method according to the improvement technique by the 1st approach. It is a figure explaining the 3rd example of the fall time control method according to the improvement technique by the 1st approach. It is a figure which shows the relationship between a transfer gate low level and saturation shading. It is a figure explaining the improvement method by a 2nd approach. It is a figure explaining the improvement technique by a 3rd approach. It is a figure explaining the improvement method by a 4th approach. It is a figure which shows the structural example of the unit pixel in the conventional CMOS sensor.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Solid-state imaging device, 3 ... Unit pixel, 5 ... Pixel signal generation part, 7 ... Drive control part, 10 ... Pixel part, 100 ... External circuit, 110 ... A / D conversion part, 12 ... Horizontal scanning circuit, 12a ... Horizontal decoder, 12b ... Horizontal drive circuit, 14 ... Vertical scanning circuit, 14a ... Vertical decoder, 14b ... Vertical drive circuit, 15 ... Vertical control line, 20 ... Timing generator, 26 ... CDS processor, 28 ... Output buffer, 32 ... Charge generation unit 34 ... Read selection transistor 36 ... Reset transistor 38 ... Floating diffusion 40 ... Vertical selection transistor 42 ... Amplification transistor 51 ... Pixel line 52 ... Vertical selection line 53 ... Vertical signal line 55 ... Transfer wiring, 56 ... Reset wiring, 57 ... Drain line, 59 ... P well bias line, 59a ... P well contact, DESCRIPTION OF SYMBOLS 30 ... Digital signal processing part, 136 ... D / A conversion part, 140 ... DRN drive buffer, 146 ... Control resistance, 148 ... Current source, 149 ... Voltage source, 150 ... Transfer drive buffer, 152 ... Reset drive buffer, 154 ... Selection drive buffer, 160 ... level shifter, 161 ... output buffer, 162 ... negative voltage generation circuit

Claims (5)

A charge generator that generates a signal charge corresponding to the received light;
A charge storage section for storing the charge generated by the charge generation section;
A transfer gate unit disposed between the charge generation unit and the charge storage unit to transfer the signal charge generated by the charge generation unit to the charge storage unit;
A pixel signal generation unit that generates a pixel signal corresponding to the signal charge stored in the charge storage unit;
A reset unit configured by a depletion type transistor for resetting the signal charge in the charge storage unit, and as a component of a unit pixel,
A transfer wiring connected to the transfer gate unit, connected in common with other unit pixels;
A transfer driving buffer for driving the transfer wiring;
A reset wiring connected to the reset unit connected in common with other unit pixels;
A reset driving buffer for driving the reset wiring;
A drain line connected to the reset unit and the pixel signal generation unit, which is connected in common with other unit pixels;
A drain driving buffer for driving the drain wiring;
Wherein receiving the pixel signal generated by the pixel signal generating section, and a connected signal line provided in common with other unit pixels,
The pixel selection operation for outputting the pixel signal generated by the pixel signal generation unit to the signal line is performed by controlling the potential of the charge storage unit ,
The transition time at the time of OFF in the voltage waveform of the drain wiring driven by the drain driving buffer when a driving pulse is applied to the drain driving buffer, the reset wiring driven by the reset driving buffer and the transfer driving buffer A solid-state imaging device configured to be not less than 5 times and not more than 10,000 times with respect to any transition time when each of the transfer wirings driven by is turned off . A charge generator that generates a signal charge corresponding to the received light;
A charge storage section for storing the charge generated by the charge generation section;
A transfer gate unit disposed between the charge generation unit and the charge storage unit and transferring the signal charge generated by the charge generation unit to the charge storage unit;
A pixel signal generation unit that generates a pixel signal corresponding to the signal charge stored in the charge storage unit;
A reset unit configured by a depletion type transistor for resetting the signal charge in the charge storage unit, and as a component of a unit pixel,
A transfer wiring connected to the transfer gate unit, connected in common with other unit pixels;
A transfer driving buffer for driving the transfer wiring;
A reset wiring connected to the reset unit connected in common with other unit pixels;
A reset driving buffer for driving the reset wiring;
A drain line connected to the reset unit and the pixel signal generation unit, which is connected in common with other unit pixels;
A drain driving buffer for driving the drain wiring;
Wherein receiving the pixel signal generated by the pixel signal generating section, and a connected signal line provided in common with other unit pixels,
The pixel selection operation for outputting the pixel signal generated by the pixel signal generation unit to the signal line is performed by controlling the potential of the charge storage unit ,
The transition time at the time of OFF in the voltage waveform of the drain wiring driven by the drain driving buffer when a driving pulse is applied to the drain driving buffer, the reset wiring driven by the reset driving buffer and the transfer driving buffer A solid-state imaging device configured to be longer than any of the transition times when each of the transfer wirings driven by is turned off . A charge generator that generates a signal charge corresponding to the received light;
A charge storage section for storing the charge generated by the charge generation section;
A transfer gate unit disposed between the charge generation unit and the charge storage unit and transferring the signal charge generated by the charge generation unit to the charge storage unit;
A pixel signal generation unit that generates a pixel signal corresponding to the signal charge stored in the charge storage unit;
A reset unit configured by a depletion type transistor for resetting the signal charge in the charge storage unit, and as a component of a unit pixel,
A signal line that receives the pixel signal generated by the pixel signal generation unit and is connected in common with other unit pixels is provided,
The pixel selection operation for outputting the pixel signal generated by the pixel signal generation unit to the signal line is performed by controlling the potential of the charge storage unit ,
A drain line connected to the reset unit and the pixel signal generation unit, connected in common with other unit pixels, is provided,
The off voltage supplied to the transfer gate unit is a voltage value having a polarity opposite to the on voltage supplied to the transfer gate unit with respect to the master reference voltage that defines the overall reference voltage of the unit pixel.
Solid-state imaging device. A charge generation unit that generates a signal charge corresponding to the received light; a charge storage unit that stores the charge generated by the charge generation unit; and the charge generation unit disposed between the charge generation unit and the charge storage unit. A transfer gate unit that transfers the signal charge generated by a charge generation unit to the charge storage unit; a pixel signal generation unit that generates a pixel signal corresponding to the signal charge stored in the charge storage unit; The transfer gate unit including a reset unit configured by a depletion type transistor that resets the signal charge in the charge storage unit as a component of the unit pixel and connected in common with other unit pixels Transfer wiring connected to the same, connected in common with other unit pixels, reset wiring connected to the reset unit, connected in common with other unit pixels A drain line connected to the reset unit and the pixel signal generation unit, and a signal line that receives the pixel signal generated by the pixel signal generation unit and is connected in common with other unit pixels are provided. A solid-state imaging device in which a pixel selection operation for outputting the pixel signal generated by the pixel signal generation unit to the signal line is performed by controlling a potential of the charge storage unit;
Upon receiving a drive pulse for driving the drain wiring, the transition time at the OFF time in the voltage waveform when driving the drain wiring is the time at each OFF time in the voltage waveform when driving the reset wiring and the transfer wiring. A solid-state imaging device including a waveform shaping unit that performs waveform shaping to be longer than any of the transition times . A drain wiring connected to the reset unit and the pixel signal generation unit, connected in common with other unit pixels,
The transistor of the reset unit can be set to a voltage level when the drain wiring is on while the transistor of the reset unit is on.
The solid-state imaging device according to any one of claims 1 to 4 .

Images