JP2014140212A - Solid state imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To produce accurate digital data while adopting a successive approximation technique in performing an analog-digital conversion process in multiple stages.SOLUTION: An AD converter provided includes a ramp waveform signal generation section 14 for generating a ramp voltage on the basis of a count signal from a counter 15, a signal conversion section 13, and a control section 18. The signal conversion section 13 includes a sample-and-hold circuit for holding an input signal voltage, a successive approximation capacitor group 16 for outputting a plurality of bias voltages in accordance with a connection combination of a predetermined number of capacitors having different capacitance values, and a comparison section 17 for comparing the signal voltage with one of the ramp voltage and the bias voltage. The control section 18 generates a digital signal of the signal voltage on the basis of the result of comparison in the comparison section 17 between the bias voltage and the signal voltage and the result of comparison in the comparison section 17 between the ramp voltage and the signal voltage, and acquires data for calibration of the successive approximation capacitor group 16 on the basis of the connection combination of capacitors and the ramp voltage.

Description

  The present invention relates to a solid-state imaging device.

  Conventionally, a plurality of types of analog-digital converters having different operation methods are known. As such an operation method, a so-called counter / ramp method or successive approximation method is known. In the counter ramp method, a ramp voltage whose voltage value changes with a constant slope is compared with an analog signal by a comparator, and a digital value of the analog signal is obtained from the number of counts until these two voltages match. In the successive approximation method, the comparator compares the reference voltage generated by the combination of multiple capacitors and the analog signal with the comparator, and asymptotically approximates the reference voltage to the analog signal, and the output of the comparator at that time is obtained as the digital value of the analog signal. It is done.

  There is also known an analog-digital converter that divides an analog signal into upper bits and lower bits and performs digital conversion in two stages (see, for example, Patent Document 1).

JP 2007-243324 A

  When combining a plurality of operation methods as in Patent Document 1, the successive approximation method is suitable for speeding up the analog-digital converter. However, the capacitance values of the capacitors used in the successive approximation method tend to vary due to manufacturing errors. As a result, if the voltage value of the bias voltage becomes inaccurate, there is a problem that accurate digital data cannot be obtained.

  The present invention has been made in view of the circumstances described above, and in an analog-digital converter that performs analog-digital conversion processing in a plurality of stages, solid-state imaging capable of obtaining accurate digital data while adopting a successive approximation method An object is to provide an apparatus.

In order to achieve the above object, the present invention provides the following means.
The present invention relates to a pixel portion in which a plurality of pixels are arranged in a matrix, a vertical shift register and a horizontal shift register that read a signal voltage from the plurality of pixels, and a signal read by the vertical shift register and the horizontal shift register A fixed imaging device comprising an analog-digital converter for digitally converting a voltage, wherein the analog-digital converter counts a clock signal in a predetermined direction and outputs a count signal, and a ramp based on the count signal A ramp waveform signal generator for generating a waveform signal; a plurality of signal converters; and a controller for controlling the signal converter. Each of the signal converters samples a voltage of an input signal. A sample-and-hold circuit that holds and a predetermined number of capacitors whose capacitance values are different every two times. A successive comparison capacitor group that outputs bias voltages of different levels by changing the connection of the capacitors, a selection voltage in which one of the voltage of the ramp waveform signal and the bias voltage is selected, and held by the sample hold circuit A voltage adding unit that outputs an addition voltage obtained by adding one of the voltage and a predetermined reference voltage; a voltage held by the sample hold circuit; and a non-addition target voltage that is the other of the predetermined reference voltage and the addition voltage A comparison unit, and the control unit uses the bias voltage as the selection voltage to add the addition voltage added by the voltage addition unit and the non-addition target voltage to the predetermined number of capacitors. In the connection combination, the first digital data generated based on the result of comparison by the comparison unit and the voltage of the ramp waveform signal are selected. Based on the second digital data generated based on the comparison result of the addition voltage added by the voltage addition unit as the voltage and the non-addition target voltage by the comparison unit, the voltage is held in the sample hold circuit. Digitally converting the voltage, acquiring calibration data of the successive approximation capacitor group based on a connection combination of the capacitors of the successive approximation capacitor group and a signal generated by the ramp waveform signal generation unit, and the pixel unit The vertical shift register and the horizontal shift register are formed on one semiconductor substrate, and other configurations excluding the pixel portion, the vertical shift register and the horizontal shift register are formed on another semiconductor substrate. Provided is a solid-state imaging device in which two semiconductor substrates are stacked in the thickness direction.

  According to the present invention, the voltage held by the sample-and-hold circuit is partly digitally converted by the successive approximation method by comparing with a plurality of bias voltages generated by the successive approximation capacitor group by the comparison unit, and the ramp waveform By comparing the ramp waveform signal generated by the signal generator with the comparator, the other part is digitally converted by the counter ramp method. From these two digital data, a digital signal having the voltage of the signal input to the sample and hold circuit is obtained.

  In this case, by comparing the voltage generated by the ramp waveform signal generation unit with a voltage with a known voltage value and the bias voltage generated by the capacitance of the successive approximation capacitor group by the comparison unit, information on variation in the capacitance value of each capacitor can be obtained. Including calibration data is obtained. Accurate digital data can be obtained by compensating the digital data converted by the successive approximation method using the calibration data.

  In the above invention, the plurality of signal conversion units may be provided corresponding to the pixel columns of the pixel unit.

  According to the present invention, in an analog-digital converter that performs analog-digital conversion processing in a plurality of stages, there is an effect that accurate digital data can be obtained while adopting a successive approximation method.

It is a top view which shows the structure of the analog-digital converter which concerns on one Embodiment of this invention, and a solid-state imaging device provided with the same. It is a top view which shows the structure of the signal converter circuit of FIG. FIG. 3 is a plan view showing a configuration of a DAC circuit in FIG. 2. It is a figure explaining an example of operation | movement of the analog-digital converter of FIG. FIG. 3 is a diagram illustrating an example of an operation of high-order bit digital conversion by the signal conversion circuit of FIG. 2. FIG. 3 is a diagram illustrating another example of the operation of high-order bit digital conversion by the signal conversion circuit of FIG. 2. FIG. 4 is a plan view showing a configuration of a modified example of the DAC circuit of FIG. 3. It is a flowchart which shows the calibration operation | movement of the solid-state imaging device which concerns on the modification provided with the DAC circuit of FIG. It is a flowchart which shows the capacity | capacitance 26 calibration routine of the flowchart of FIG. It is a flowchart which shows the capacity | capacitance 25 calibration routine of the flowchart of FIG. It is a top view which shows the structure of the modification of the solid-state imaging device of FIG. 6 is a timing chart showing a modification of the operation of the solid-state imaging device of FIG. 1.

Hereinafter, an analog-to-digital (AD) converter 1 according to an embodiment of the present invention and a solid-state imaging device 100 including the same will be described with reference to the drawings.
As shown in FIG. 1, the solid-state imaging device 100 according to this embodiment includes a pixel array (pixel unit) 3 in which a plurality of pixels 2 are arranged in a matrix direction, and a vertical shift register that reads a signal voltage from the pixels 2. 4 and a horizontal shift register 5, an AD converter 1 for digitally converting a signal voltage read by these shift registers 4, 5, and a peripheral circuit provided around the AD converter 1. .

  Among these configurations, the pixel array 3 and the two shift registers 4 and 5 are formed on one semiconductor substrate 100a, and the other configurations are formed on another semiconductor substrate 100b. These semiconductor substrates 100a and 100b are stacked in the thickness direction. By doing in this way, compared with the case where all the circuits are formed on one semiconductor substrate, the area of the semiconductor substrates 100a and 100b can be reduced, and the device on which the solid-state imaging device 1 is mounted can be downsized. .

  Reference numeral 6 denotes a wiring for transmitting a signal from the horizontal shift register 5 on one semiconductor substrate 100a to the sample hold circuit 12 (described later) included in the AD converter 1 on the other semiconductor substrate 100b. The wiring 6 may be provided in the same number as the number of columns corresponding to each column of the pixel array 3. Alternatively, the wiring 6 may be provided corresponding to a plurality of columns, and signals of the pixels 2 in the plurality of columns may be transmitted through the same wiring 6.

The solid-state imaging device 100 outputs the signal voltage of the pixel 2 to the outside as a digital signal by the following operation.
First, the vertical shift register 4 selects the pixels 2 in the first row in the pixel array 3. The horizontal shift register 5 receives a signal voltage from the pixels 2 in the first row selected by the vertical shift register 4 and outputs the signal voltage to the AD converter 1. The digital signal of the pixels 2 for one row converted from the signal voltage by the AD converter 1 is stored in the memory circuit 7 and then converted from the parallel signal to the serial signal by the parallel-serial conversion circuit 8 and serial digital output such as LVDS. The signal is output from the circuit 9 to an external circuit (not shown). Thereby, the signal voltage of the pixels 2 in the first row is output to the outside as a digital signal.

The solid-state imaging device 100 outputs the signal voltage of the pixel 2 for each row as a digital signal to the outside while shifting the row of the pixel array 3 selected by the vertical shift register 4 one row at a time.
In the figure, reference numeral 10 denotes a power supply circuit, and reference numeral 11 denotes a pulse generation circuit that generates a pulse for driving each circuit.

Next, the configuration and operation of the AD converter 1 according to this embodiment will be described in detail.
The AD converter 1 has a configuration in which the upper bits of the signal voltage from the pixel 2 are digitally converted by a successive approximation method and a configuration in which the lower bits are digitally converted by using a counter / ramp method. Specifically, the AD converter 1 includes a plurality of sample and hold circuits (signal conversion units) 12 and signal conversion circuits (signal conversion units) 13 provided corresponding to the respective columns of the pixel array 3, and these signals. A ramp waveform generation circuit (ramp waveform signal generation unit) 14 that supplies a ramp voltage (ramp waveform signal) to the conversion circuit 13 and a counter circuit (counter) 15 that counts clock signals are provided.

  Although FIG. 1 shows a configuration in which one sample hold circuit 12 and one signal conversion circuit 13 are provided for each column of the pixel array 3, instead of this, the sample hold circuit 12 is provided. One signal conversion circuit 13 may be provided for each of a plurality of columns of the pixel array 3.

  FIG. 2 is a plan view showing the configuration of each signal conversion circuit 13. As shown in FIG. 2, the signal conversion circuit 13 includes a DAC circuit (successive comparison capacitor group, voltage addition unit) 16, a comparator (comparison unit) 17, and a control circuit (control unit) 18. The input end of each signal conversion circuit 13 is connected to the corresponding sample and hold circuit 12 via a switch SWin.

Comparator 17, the output voltage Vdac from DAC circuit 16 is input to the inverting terminal, the reference voltage _{V 0} is input to the non-inverting terminal. The reference voltage _{V 0} is, for example, 0V. Comparator 17, the two voltage Vdac, compares the magnitude of _{V 0.} Result of the comparison, the output of the comparator 17, when the voltage Vdac smaller than the reference voltage _{V 0} when "high" and the voltage Vdac greater than the reference voltage _{V 0} becomes "low". The signal level output from the comparator 17 is latched by the latch circuit 19. The latch circuit 19 outputs the latched signal level to the control circuit 18 and the memory circuit 7 provided in the subsequent stage of each signal conversion circuit 13.

The counter circuit 15 starts counting a clock signal generated by a clock circuit (not shown) based on an instruction from the control circuit 18.
The ramp waveform generation circuit 14 generates a ramp voltage Vramp that decreases in voltage with a constant slope in the time axis direction in synchronization with the start of the count operation of the counter circuit 15 and outputs the ramp voltage Vramp to the DAC circuit 16.

  FIG. 3 is a plan view showing the configuration of the DAC circuit 16. FIG. 3 illustrates a configuration for generating 5-bit digital data. In this configuration, the DAC circuit 16 includes seven capacitors 20 to 26 and switches SW0 to SW6 provided corresponding to the capacitors 20 to 26, respectively. The capacitance values of the capacitors 20 to 26 are C, C, 2C, 4C, 8C, 16C, and 32C (C is a positive constant), where C is the smallest capacitance value of the capacitors 20 and 21.

  The upper electrodes of the capacitors 20 to 26 are connected in parallel to the switch SWin and the inverting input terminal of the comparator 17. The lower electrode of one capacitor 20 having the smallest capacitance value C is connected to the lamp input terminal to which the ramp voltage Vramp is input from the ramp waveform generation circuit 14 or the ground voltage GND by switching the switch SW0. The lower electrodes of the other capacitors 21 to 27 are connected to the reference voltage Vref or the ground voltage GND by switching the switches SW1 to SW6. Switching of the switches SW0 to SW6 is controlled by the control circuit 18.

Next, the operation of the AD converter 1 configured as described above will be described using an example shown in FIG.
The signal voltage Vin of the pixels 2 for one row read out by the horizontal shift register 5 is input to the sample hold circuit 12 of the corresponding column and held in the sample hold circuit 12. Next, when the switch SWin is closed, each sample and hold circuit 12 outputs the signal voltage Vin to the signal conversion circuit 13 in the corresponding column in the subsequent stage. At this time, the switches SW0 to SW6 are switched to the ground voltage GND side by the control circuit 18. As a result, the output voltage Vdac of the DAC circuit 16 becomes equal to the signal voltage Vin. Thereafter, the switch SWin is opened, whereby the output voltage Vdac of the DAC circuit 16 is held at Vin.

Next, the control circuit 18 switches the SW1~SW6 based on the signals received from the latch circuit 19, by comparing the reference voltage _{V 0} and voltage Vdac by the comparator 17, the upper 5 bits of the signal voltage Vin Digitally convert.

Specifically, first, the switch SW6 of the capacitor 26 having the largest capacitance value 32C is switched to the reference voltage Vref side. As a result, the output voltage Vdac of the DAC circuit 16 becomes Vin−Vref / 2. Comparator 17 compares the output voltage Vdac and the reference voltage _{V 0.} In FIG. 4, Vdac> V ₀ , that is, Vin <Vref / 2 + V ₀ , the output signal of the comparator 17 is “low”, and “0” is obtained as the digital value of the most significant bit. The latch circuit 19 latches this signal level and outputs it to the control circuit 18 and the memory circuit 7.

Next, the control circuit 18 receives the “low” input from the latch circuit 19, keeps the switch SW 6 connected to the reference voltage Vref side, and switches the switch SW 5 to the reference voltage Vref side. As a result, the output voltage Vdac of the DAC circuit 16 becomes Vin− (Vref / 2 + Vref / 4). Comparator 17 compares the output voltage Vdac and the reference voltage _{V 0.} In FIG. 4, Vdac <V ₀ , that is, Vin <Vref / 2 + Vref / 4 + V ₀ , the output signal of the comparator is “high”, and “1” is obtained as the digital value of the upper second bit. The latch circuit 19 latches the signal level and outputs it to the control circuit 18 and the memory circuit 7.

Next, upon receiving “high” input from the latch circuit 19, the control circuit 18 switches the switch SW5 to the ground voltage GND side and switches the switch SW4 to the reference voltage Vref side. As a result, the output voltage Vdac of the DAC circuit 16 becomes Vin− (Vref / 2 + Vref / 8). Comparator 17 compares the output voltage Vdac and the reference voltage _{V 0.} In FIG. 4, Vdac> V ₀ , that is, Vin <Vref / 4 + Vref / 8 + V ₀ , the output signal of the comparator 17 is “low”, and “0” is obtained as the digital value of the upper third bit.

  As described above, the control circuit 18 changes the combination of the switches SW1 to SW6 to be switched to the reference voltage Vref side according to the comparison result by the immediately previous comparator 17, thereby changing the bias voltage generated by the capacitance to Vref / 2. Vref / 2 + Vref / 4, Vref / 2 + Vref / 8,... Are changed asymptotically to the signal voltage Vin. Then, “0” or “1”, which is a comparison result between the signal voltage Vin and each bias voltage, is used as the digital value of the bit to obtain the digital value up to the fifth bit.

Here, as shown in FIGS. 5 and 6, the control circuit 18 stores the digital code of the DAC circuit 16 when the output of the comparator 17 is “0” (see the black line in the figure). Then, after the comparison by the comparator 17 for the upper 5 bits is completed, the output voltage Vdac corresponding to the last stored digital code is generated once again. Accordingly, the digital converting operation by successive approximations, so that the output voltage Vdac is completed at a greater state than the reference voltage V _0. Note that FIGS. 5 and 6 illustrate a case where 4 bits are digitally converted.

Next, the control circuit 18 digitally converts the lower bits of the signal voltage Vin by the counter / ramp method. Specifically, the control circuit 18 switches the switch SW0 of the capacitor 20 having the smallest capacitance value C to the lamp input terminal side. At the same time, the control circuit 18 instructs the counter circuit 15 to start counting the clock signal. The ramp waveform generation circuit 14 receives the start of the count operation of the counter circuit 15, sets the initial value to the reference voltage Vref, and outputs a ramp voltage Vramp that decreases with a constant slope. The comparator 17 compares the voltage Vdac input to the inverting input terminal with the reference voltage V _0, and outputs “low” while Vdac> Vref.

The counter circuit 15 continues to count the number of clock signals while “low” is input from the latch circuit 19, and stops counting when the input from the latch circuit 19 changes to “high”. The count number of the clock signal counted by the counter circuit 15 is obtained as digital data of lower bits of the signal voltage Vin. The counter circuit 15 outputs the count number to the memory circuit 7.
The digital signal of the signal voltage Vin is obtained by combining the upper bit digital data and the lower bit digital data generated in the above two steps.

  As described above, according to the present embodiment, by adopting a sequential conversion method suitable for high-speed operation and adopting a counter / ramp method having a simple circuit configuration for lower-order digital conversion, the area and consumption of the entire circuit are reduced. The number of capacitors 20 to 26 having a relatively large power ratio can be reduced, and the circuit can be reduced in size and power consumption.

In addition, by making the last voltage Vdac output from the DAC circuit 16 larger than the reference voltage V ₀ in the digital conversion of the higher-order bits by the successive approximation method, the digital conversion by the counter-ramp method to be performed over time is performed over time. The ramp voltage Vramp, which decreases to a certain level, reliably crosses the reference voltage V ₀ . As a result, the lower-order bit digital conversion operation can be reliably executed.

  Next, the calibration operation of the AD converter 1 according to this embodiment will be described. The calibration operation may be performed every time the AD converter 1 performs digital conversion of the signal voltage Vin, and timing such as when the power of the solid-state imaging device 100 is turned on or an environment such as temperature changes. May be executed.

  First, a calibration voltage Vcal having a known voltage value is applied from the ramp waveform generation circuit 14 to the lower electrode plate of the capacitor 20 via the switch SW0 switched to the lamp input terminal side. At this time, the lower plates of the other capacitors 21 to 26 are connected to the ground voltage GND by the switches SW1 to SW6. As a result, the voltage Vdac = −Vcal is input to the inverting input terminal of the comparator 17. The calibration voltage Vcal is digitally converted in the same way as the digital conversion of the upper bits of the signal voltage Vin by the successive approximation method described above. Similarly, a plurality of calibration voltages Vcal having different voltage values are digitally converted.

  The digital signal of the calibration voltage Vcal obtained by digital conversion is used as calibration data for capacitors 20 to 26. That is, when the capacitance values of the capacitors 20 to 26 vary due to manufacturing errors, the digital signal obtained by actually digitally converting the calibration voltage Vcal is different from the digital signal theoretically calculated from the calibration voltage Vcal. . Therefore, a table in which these two digital signals are associated with each other is stored in a storage device (not shown) provided outside the solid-state imaging device 100, for example. As a result, the high-order bit digital signal obtained by the AD converter 1 can be compensated to be a correct digital signal based on the table.

  As described above, according to the present embodiment, the ramp waveform generation circuit 14 used for digital conversion of the lower bits of the signal voltage Vin is also used to generate the calibration voltage Vcal, thereby generating a calibration voltage that has been separately provided conventionally. The structure for doing so becomes unnecessary. Thereby, further miniaturization can be achieved.

In this embodiment, in the low-order bit digital conversion, the ramp voltage Vramp whose voltage value decreases with time is used. Instead, a ramp voltage whose voltage value increases with time is used. It is good as well. In this case, the digital conversion of upper bits may be so DAC circuit 16 is smaller than the voltage Vdac reference voltage V ₀ to be output finally. That is, the control circuit 18 stores the digital code of the DAC circuit 16 when the output of the comparator 17 is “high”, and after the comparison by the comparator 17 for the upper 5 bits is completed, the last stored digital code The voltage Vdac corresponding to can be generated once more.

  Further, calibration may be performed by a circuit form as shown in FIG. In this embodiment, as shown in the flowchart of FIG. 8, when the calibration operation is started, the switch SWinput is switched to the “2” side and the switch SWcal is switched to the “2” side (step S1), and the largest capacitance value 32C is set. Calibration is carried out in order from the capacitor 26 having (Steps S2 to S7).

  FIG. 9 shows the capacity 26 calibration routine S2 of FIG. As shown in FIG. 9, first, the switch SW6 is connected to the ground voltage GND, and the switches SW0 to SW5 are connected to the reference voltage Vref (step S21). The voltage V261 input to the inverting terminal of the comparator 17 at this time is obtained by inputting the calibration voltage Vcal having a known voltage value from the ramp waveform generating circuit 14 from the switch SWcal (step S22).

  Next, the switch SW6 is connected to the reference voltage Vref, and the switches SW0 to SW5 are connected to the ground voltage GND (step S23). The voltage V262 input to the inverting terminal of the comparator 17 at this time is obtained by inputting the calibration voltage Vcal having a known voltage value from the ramp waveform generating circuit 14 from the switch SWcal (step S24).

Next, the calibration data of the capacitor 26 is calculated from the voltages 261 and 262 acquired in steps S22 and S24 (step S25). Specifically, assuming that the sum of the capacitance values of the capacitors 20 to 26 is Ctot, the capacitance value of the capacitor 26 is Ctot / 2 + ΔC26, and the sum of the capacitance values of the capacitors 20 to 25 is Ctot / 2−ΔC26, the voltages V261 and V262 Is represented by the following equations.
V261 = VrefX (Ctot / 2−ΔC26) / Ctot
V262 = VrefX (Ctot / 2 + ΔC26) / Ctot
Thereby, the difference between the voltage V261 and the voltage V262 is as follows.
V262-V261 = VrefX (2XΔC26 / Ctot)
Therefore, the conversion error due to the manufacturing error of the capacitor 26 is obtained as (V262−V261) / 2, and this is used as calibration data at the time of conversion by the capacitor 26.

  Subsequently, the capacity 25 is calibrated by the capacity 25 calibration routine S3 of FIG. As shown in FIG. 10, first, the switches SW6 and SW5 are connected to the ground voltage GND, and the switches SW0 to SW4 are connected to the reference voltage Vref (step S31). The voltage V251 input to the inverting terminal of the comparator 17 at this time is obtained by inputting the calibration voltage Vcal having a known voltage value from the ramp waveform generating circuit 14 from the switch SWcal (step S32).

  Next, the switch SW5 is connected to the reference voltage Vref, and the switches SW0 to SW4 are connected to the ground voltage GND (step S33). The voltage V252 input to the inverting terminal of the comparator 17 at this time is obtained by inputting the calibration voltage Vcal having a known voltage value from the ramp waveform generating circuit 14 from the switch SWcal (step S34).

Next, the calibration data of the capacitor 25 is calculated from the voltages 251 and 252 acquired in steps S32 and S34 (step S35). Specifically, when the capacitance value of the capacitor 25 is Ctot / 4−ΔC26 / 2 + ΔC25 and the sum of the capacitance values of the capacitors 20 to 24 is Ctot / 4−ΔC26 / 2−ΔC25, the voltages V251 and V252 are expressed by the following equations. Each is represented.
V251 = VrefX (3XCtot / 4−ΔC26 / 2−ΔC25) / Ctot
V252 = VrefX (3XCtot / 4−ΔC26 / 2 + ΔC25) / Ctot

Thereby, the difference between the voltage V251 and the voltage V252 is as follows.
V252−V251 = VrefX (2XΔC25 / Ctot)
Therefore, the conversion error due to the manufacturing error of the capacitor 25 is obtained as (V252−V251) / 2, and this is used as calibration data at the time of conversion by the capacitor 25.
Similarly, the capacitors 24 to 21 are processed in the same capacity as the capacitors 25 and 26 to obtain calibration data, and then the switch SWinput is switched to the “1” side and the switch SWcal is switched to the “1” side to perform calibration. The process ends (step S8).

In the present embodiment, the circuit constituting the solid-state imaging device 100 is formed separately on the two semiconductor substrates 100a and 100b. Instead, as shown in FIG. It is good also as being formed in one semiconductor substrate 100c.
In the example shown in FIG. 11, the horizontal shift register 5 is omitted, and the signal voltage Vin from the pixel 2 of each column is directly input to the sample and hold circuit 12.

  In the present embodiment, only the operation of digitally converting the signal voltage Vin of the pixel 2 has been described, but the noise voltage of the pixel 2 may be digitally converted prior to digital conversion of the signal voltage Vin.

  When the signal voltage Vin of the pixel 2 is reset to zero, a noise voltage is generated along with the reset operation, and is stored in the pixel 2 after the noise voltage is reset. Therefore, the signal voltage Vin generated by receiving light next by the pixel 2 is superimposed on the noise voltage, and the sum thereof is read out by the horizontal shift register 5. Therefore, it is possible to read out the noise voltage after resetting the pixel 2 and then read out the signal voltage Vin generated by receiving light from the pixel 2 together with the noise voltage, and calculate the signal voltage Vin from the difference therebetween.

  Here, the solid-state imaging device 100 may operate as shown in FIG. That is, among the m columns of pixels 2 arranged in the pixel array 3, the voltage output operation from the horizontal shift register 5 to the sample hold circuit 12, the digital conversion operation, and the digital Perform signal output operation. Then, the remaining pixels 2 in the m / 2 + 1 to m-th columns are operated one step later than the pixels 2 in the 1-2 / m-th columns.

In the figure, R is a period during which the voltage of the pixel 2 is reset to zero, N is a period during which the noise voltage of the pixel 2 after reset is read out to the horizontal shift register 5, and T is a signal charge accumulated by the photodiode included in the pixel 2. A period S during which the voltage is converted into the voltage Vin indicates a period during which the signal voltage Vin and noise voltage of the pixel 2 are read out to the horizontal shift register 5.
In this way, by dividing the pixels 2 in half by columns and processing with time differences, the processing can be performed without interruption.

1 Analog-digital converter 2 Pixel 3 Pixel array (pixel unit)
4 vertical shift register 5 horizontal shift register 6 wiring 7 memory circuit 8 parallel serial conversion circuit 9 serial digital output circuit 10 power supply circuit 11 pulse generation circuit 12 sample hold circuit (signal conversion unit)
13 Signal conversion circuit (signal conversion unit)
14 Ramp waveform generation circuit (ramp waveform signal generator)
15 Counter circuit (counter)
16 DAC circuit (successive comparison capacitor group, voltage adder)
17 Comparator (Comparator)
18 Control circuit (control unit)
19 latch circuit 20 to 26 volume 100 solid-state imaging device 100a, 100b, 100c semiconductor substrate SWin, SW0 to switch Vin signal voltage Vref reference voltage V ₀ reference voltage Vramp ramp voltage (ramp waveform signal)

Claims (1)

A pixel portion in which a plurality of pixels are arranged in a matrix;
A vertical shift register and a horizontal shift register for reading signal voltages from the plurality of pixels;
A fixed imaging device comprising an analog-to-digital converter for digitally converting a signal voltage read by the vertical shift register and the horizontal shift register,
The analog-digital converter is
A counter that counts a clock signal in a predetermined direction and outputs a count signal;
A ramp waveform signal generator for generating a ramp waveform signal based on the count signal;
A plurality of signal converters;
A control unit for controlling the signal conversion unit,
Each of the signal converters is
A sample hold circuit that samples and holds the voltage of the input signal; and
A successive-approximation capacitor group having a predetermined number of capacitances having different capacitance values every two times, and outputting bias voltages of different levels by changing the connection of these capacitors;
A voltage adding unit that outputs an addition voltage obtained by adding one of the voltage of the ramp waveform signal and the bias voltage selected and one of the voltage held by the sample hold circuit and a predetermined reference voltage;
A comparison unit that compares the voltage held by the sample hold circuit and the non-addition target voltage, which is the other of the predetermined reference voltage, and the addition voltage;
The control unit is
Generated based on a result of comparing the added voltage added by the voltage adding unit with the bias voltage as the selection voltage and the non-addition target voltage by the comparing unit in the predetermined number of connection combinations of the capacitors. The first digital data is generated based on the result of comparing the added voltage added by the voltage adding unit with the voltage of the ramp waveform signal as the selection voltage and the non-addition target voltage by the comparing unit. 2 and digitally converting the voltage held in the sample and hold circuit based on the digital data of 2;
Obtaining calibration data for the successive approximation capacitor group based on the connection combination of the capacitors of the successive approximation capacitor group and a signal generated by the ramp waveform signal generator;
The pixel portion, the vertical shift register and the horizontal shift register are formed on one semiconductor substrate,
Other configurations excluding the pixel portion, the vertical shift register, and the horizontal shift register are formed on another semiconductor substrate,
A solid-state imaging device in which these two semiconductor substrates are stacked in the thickness direction.

Images