The present invention relates to positioning of drive lines in a solid-state image pickup device.
In recent years, image pickup devices such as a digital camcorder and a digital camera applying a CMOS image sensor suitable for low power consumption and high speed readout have been generally available. A CMOS image sensor having a plurality of pixels each containing a photoelectric converting unit arranged in row and column directions is proposed which is configured as a global electronic shutter in which an exposure start and an exposure end are electronically controlled simultaneously in all of the pixels (International Publication No. WO 11/043432).
A device according to an aspect of the present invention having an image sensing region in which a plurality of pixels are arranged in a matrix form, each of the pixels having a converting unit, a first transistor configured to transfer electric carriers in the converting unit, an accumulating portion configured to accumulate electric carriers transferred from the first transistor, a second transistor configured to transfer electric carriers from the accumulating portion, a floating diffusion (hereinafter, called an FD) configured to accumulate electric carriers transferred from the second transistor, and a reset transistor configured to reset a potential of the FD includes gate drive lines for a plurality of pixel transistors configured to drive gates of the pixel transistors each including the first transistor, the second transistor, and the reset transistor, the gate drive lines extending in a direction of rows of the pixels in one wiring layer. In this case, a gate drive line for the first transistor among the gate drive lines for the pixel transistors is positioned in proximity to the converting unit in driving wiring of rows (n−1), rows (n), and rows (n+1) provided correspondingly to rows of the pixels.
Further features of the present invention will become apparent from the following description of exemplary embodiments with reference to the attached drawings.
A first embodiment of the present invention will be described with reference to
Referring to
Each of the pixels P1 is connected to a vertical output line Vout through a pixel output unit 22. A power supply 20, and a ground 21 are further provided therein. According to this embodiment, a configuration called a Vertical Overflow Drain (hereinafter, VOFD) is provided in which electric carriers are output from the photoelectric converting unit 1 to the semiconductor substrate 7 through an embedded layer 9.
A depleting inhibiting portion 5 is provided under the p-type electric carrier accumulating portion 3, and the depleting inhibiting portion 5 is made of a higher density semiconductor than that of the well 8.
Supply of pulses which turns on the first transfer transistor 14 to TX1 being a gate of the first transfer transistor 14 causes electric carriers in the photoelectric converting unit 1 to be transferred to the electric carrier accumulating portion 3.
Supply of pulses which turns on a second transfer transistor 15 to TX2 being a gate of the second transfer transistor 15 causes electric carriers accumulated in the electric carrier accumulating portion 3 to be transferred to the FD 6.
Next, with reference to
Referring to
At a time t0 in
Next, at a time t1, the levels of the gate drive lines pTX1(n−1), pTX1(n), and pTX1(n+1) for the first transfer transistors 14 are changed to an H level so that the first transfer transistors 14 are turned on. Thus, electrons are transferred to the electric carrier accumulating portion 3. After a lapse of a predetermined time period, the first transfer transistors 14 are turned off so that the transfer of electrons to the electric carrier accumulating portions 3 ends.
Because the electric carrier accumulating portions 3 are provided in this embodiment, signal electric carriers in the photoelectric converting units 1 in all pixels are simultaneously transferred to the electric carrier accumulating portions 3. This can implement an operation to be performed by a global electronic shutter which controls an exposure start and an exposure end simultaneously in all pixels through an electronic switch.
Next, at a time t2, the level of the substrate potential is changed to an H level so that a punch-through occurs between the photoelectric converting units 1 and the semiconductor substrate 7. Thus, electric carriers are output to the semiconductor substrate 7.
The period from the time t0 when VOFD is turned off to the time t1 when the first transfer transistors 14 are turned on may be set as required so that an image for an arbitrary accumulation time can be obtained.
The first transfer transistors may be turned on a plurality of number of times intermittently between the time t0 and the time t1. The turning on a plurality of number of times can reduce the signal electric carriers to be handled by one transfer operation and can facilitate the transfer operation. In a case with the turning on a plurality of number of times, the time t1 is a time for the last one of the plurality of ON operations.
Next, at times t3, t4, and t5, the level of the gate drive line pTX2 of the second transfer transistor 15 is changed to an H level line-sequentially so that the second transfer transistor 15 is turned on. Thus, signal electric carriers are transferred from the electric carrier accumulating portion 3 to the FD 6.
A conventional method for a CMOS image sensor is applicable as a signal transfer method for the FD 6 and subsequent stages. In other words, signals are output to the vertical signal lines via the source follower transistor 17, the row selection transistor 18, and the pixel output unit 22. A signal of a noise component may be output to the vertical signal line before the second transfer transistor 15 is turned on. Though the row selection transistor 18 is provided in
The gate drive lines here are made of conductors to transmit drive pulses illustrated in
With a conventional line-sequential shutter instead of a global electronic shutter, differences in accumulation timing within a screen may not be a significant problem in image quality because there are relatively large differences in accumulation timing within the screen. However, with a global electronic shutter, because a difference in accumulation timing for each row may not occur easily, the degradation of image quality due to the difference in accumulation timing caused by a propagation delay may be significant within an image sensing region. The ratio of a difference in accumulation timing to an accumulation time increases as the accumulation time decreases. Thus, the difference becomes more significant, which may possibly be one factor which prevents reduction of the accumulation time.
The accumulation timing in a global electronic shutter is controlled by the first transfer transistors 14 as described above. Thus, the parasitic capacitance of the gate drive line pTX1 for the first transfer transistor 14 may be reduced to reduce the propagation delays so that the difference in accumulation timing can be reduced within the image sensing region. Such an influence of the parasitic capacitance may significantly occur in a case where gate drive lines for a plurality of pixel transistors are provided within one wiring layer and the gate drive lines are positioned closely.
Positioning of pTX1
Next, with reference to
Referring to
The gate drive line pRES(n) for the reset transistor 16 is only provided in proximity of the gate drive line pTX1(n) and far away from the gate drive line pTX2(n+1) for the second transfer transistor 15. In this arrangement, pTX1(n) can have a lower parasitic capacitance compared with an arrangement in which gate drive lines for pixel transistors are closely provided on both sides of pTX1(n).
It is assumed here that the total of a wiring distance between a gate drive line for a first transfer transistor and gate drive lines for pixel transistors provided on both sides of the gate drive line for the first transfer transistor in the driving wiring of the rows (n−1), the rows (n), and rows (n+1) is equal to DTX1Total.
Similarly, the totals of a wiring distance between a gate drive line for other pixel transistors excluding the first transfer transistor and gate drive lines for the pixel transistors provided on both sides of the gate drive line are equal to DTX2Total for a second transfer transistor, DRESTotal for a reset transistor, and DSELTotal for a row selection transistor.
In this case, DTX1Total is a value higher than all of DTX2Total, DRESTotal, and DSELTotal.
Referring to
Here, in the driving wiring of the rows (n−1), rows (n), and rows (n+1), DTX1Total is equal to D1+D3+D6+D8+D9. On the other hand DRESTotal is equal to D1+D2+D7+D8+D9+D10. DSELTotal is equal to D2+D4+D5+D7+D10+D11.
Because D3 and D6 are three times or more of the interval between other lines, for example, DTX1Total is equal to a value higher than DRESTotal and DSELTotal in rows (n−1), rows (n), and rows (n+1). It should be noted that the wiring distance between drive lines is defined herein by a distance between an end portion of pTX1(n) and an end portion of pRES(n), for example, as illustrated in
According to the configuration as described above, the parasitic capacitance occurring in the gate drive line pTX1 can be reduced, and a smaller propagation delay as a result can reduce the differences in accumulation timing.
The positioning of drive lines as described above can be applied not only to front side illumination (FSI) but also back side illumination (BSI). Also for back side illumination, it is assumed that drive lines for pixel transistors are placed correspondingly to rows of pixels. Thus, the drive line positioning example as described above may be applied in that case.
Positioning of pTX1 and pTX2
According to this embodiment, the gate drive line pTX1 and the gate drive line pTX2 are not positioned in proximity of each other. The expression “not positioned in proximity of each other” refers to a positional relationship in which pTX2 is not placed on both sides of pTX1. Alternatively, even in a case where pTX2 is placed on one side of pTX1, the distance between pTX1 and pTX2 is three times or more, for example, of the distance between pTX1 and a drive line placed on the other side of pTX1.
Problems involved in a case where the gate drive line pTX1 and the gate drive line pTX2 are placed in proximity of each other will be described below.
When the gate drive line pTX1 and the gate drive line pTX2 have capacitive coupling, the potential of the gate drive line pTX2 may have a fluctuation when the level of the gate drive line pTX1 changes. For example, at the time t1 illustrated in
In some cases, the gate drive line pTX2 may be set to a negative potential during a period when the second transfer transistor 15 being an NMOS has an OFF state. Thus, holes are excited in a channel part so that re-coupling between electrons and holes can inhibit dark current even when electrons occur. If the gate drive line pTX1 and the gate drive line pTX2 have capacitive coupling here, the potential of the gate drive line pTX2 changes to be higher and the hole excitation can be insufficient when the level of the gate drive line pTX1 is changed from an L level to an H level. As a result, the dark current inhibition may possibly become insufficient.
In order to prevent this, the gate drive line pTX1 and the gate drive line pTX2 are not positioned in proximity of each other according to this embodiment. Particularly, referring to
As understood from
The gate drive line pSEL may be positioned in proximity of the gate drive line pTX1 instead of the gate drive line pRES. This is because, even when the row selection transistor has an OFF state when the first transfer transistor is turned on, no transfer path for signal electric carriers exists if the constant current source has an OFF state and no variations in signals occur within the image sensing region.
Alternatively, a power supply line or a ground line instead of the gate drive line pRES may be positioned in proximity of the gate drive line pTX1 to inhibit a potential fluctuation. This is because such a power supply line and a ground line are not directly associated with the transfer path for signal electric carriers and no variations in signals occur within the image sensing region.
According to this embodiment, the FD 6 is provided in each pixel. However, the FD 6 may be shared by a plurality of pixels. Also in this case, the gate drive line pTX1 for the first transfer transistor may be positioned as described above to acquire the same effect. Sharing the FD 6 by a plurality of pixels can reduce the number of drive lines so that the gate drive line pTX1 for the first transfer transistor can be provided far away from other drive lines and the parasitic capacitance can thus be reduced.
A second embodiment of the present invention will be described with reference to
This embodiment is different from the first embodiment in that an overflow transistor is separately provided without applying a VOFD configuration. In other words, as illustrated in
Next, at a time t1, the level of the gate drive line pTX1 for the first transfer transistor 14 is changed to an H level, and the first transfer transistor 14 is turned ON. Thus, electrons are transferred to the electric carrier accumulating portion 3. After a lapse of a predetermined time period, the first transfer transistor 14 is turned off so that the transfer of electrons to the electric carrier accumulating portion 3 ends.
Next, at a time t2, the level of the gate drive line pOFG is changed to an H level so that the overflow transistor 19 is turned on. Thus, electric carriers can be output from the photoelectric converting unit 1 to the power supply 20 which is an overflow drain.
The period from the time t0 when the change of the level of the gate drive line pOFG to the L level to the time t1 when the first transfer transistor 14 is turned on may be set as required so that an image for an arbitrary accumulation time can be obtained.
Here, in rows (n−1), rows (n), rows (n+1), DTX1Total is equal to D1+D3+D8+D10+D11. On the other hand, DRESTotal is equal to D1+D2+D9+D10+D11. DSELTotal is equal to D4+D5+D6+D7+D12. DOFGTotal is equal to D2+D4+D7+D9.
Because D3 and D8 are three times or more of the interval between other lines, for example, DTX1Total is equal to a value higher than DRESTotal, DSELTotal, and DOFGTotal in the rows (n−1), rows (n), rows (n+1). According to the configuration, the parasitic capacitance occurring in the gate drive line pTX1 can be reduced, and a smaller propagation delay as a result can reduce the differences in accumulation timing.
This wiring arrangement can be expressed as that pTX1 and pTX2 are not positioned in proximity of each other. Furthermore, the distance between pTX1 and pTX2 in rows (n−1), rows (n), and rows (n+1) can be expressed as a maximum wiring distance of wiring distances between gate drive lines for pixel transistors.
Positioning of pTX1 and pOFG
A problem may possibly arise when the gate drive line pTX1 and the gate drive line pOFG have capacitive coupling. In other words, at a time t1 in
Accordingly, as illustrated in
This embodiment applies overflow transistors while the first embodiment applies a VOFD configuration. Application of the deeper photoelectric converting unit 1 in order to increase the sensitivity on a longer wavelength side makes output of electric carriers difficult with the VOFD configuration. However, also in this case, overflow transistors may be applied so that electric carriers can be output. In the VOFD configuration, a punch-through occurs between the photoelectric converting unit 1 and the semiconductor substrate 7. Thus, the embedded layer 9 may be required to be made of a low density semiconductor to some extent. On the other hand, with application of overflow transistors, the embedded layer 9 may be made of a high density semiconductor. This may improve the sensitivity of the photoelectric converting unit 1 and improve the efficiency of transfer to the electric carrier accumulating portion 3.
On the other hand, because application of a VOFD configuration eliminates the necessity for the gate OFG and gate drive line pOFG for an overflow transistor, the empty region may be allocated to other components. For example, the intervals between drive lines can be increased while keeping the same size of the photoelectric converting unit 1. Thus, the parasitic capacitance of the gate drive line pTX1 for the first transfer transistor can be further reduced compared with the first embodiment. The area of the photoelectric converting unit 1 may be increased so that the saturation charge quantity of the photoelectric converting unit 1 can be increased, and its sensitivity can be increased.
A third embodiment of the present invention will be described with reference to
Referring to
The accumulation timing with a global electronic shutter can be controlled also by an input of drive pulses to a gate OFG of the overflow transistor 19, as described above. Therefore, the parasitic capacitance of the gate drive line pOFG for the overflow transistor 19 can be reduced, and a propagation delay occurring therein can be reduced. As a result, the difference in accumulation timing can be reduced.
According to this embodiment, the gate drive line pOFG and the gate drive line pTX2 are not positioned in proximity. If they are positioned in proximity, the saturation signal amount in the electric carrier accumulating portion 3 can be reduced, which may cause variations in saturation signal amount in the electric carrier accumulating portion 3 within the image sensing region.
In some cases, the gate drive line pTX2 may be set to a negative potential during a period when the gate drive line pTX2 for the second transfer transistor 15 being an NMOS has an OFF state. Thus, holes are excited in a channel part, and re-coupling of electrons and holes is thus caused, which can inhibit dark current. If the gate drive line pOFG and the gate drive line pTX2 have capacitive coupling here, the potential of the gate drive line pTX2 changes to be higher at a time point when the level of the gate drive line pOFG is changed from an L level to an H level. As a result, the inhibition of dark current may possibly in sufficient.
Accordingly, a gate drive line pRES for a reset transistor is positioned between the gate drive line pOFG and the gate drive line pTX2 in one driving wiring region. A gate drive line pSEL for a row selection transistor, a power supply line, a ground line or the like may be positioned instead of the gate drive line pRES.
In this case, in driving wiring of rows (n−1), rows (n), and rows (n+1), DTX1Total is equal to D1+D3+D8+D10+D11+D13. On the other hand, DRESTotal is equal to D1+D2+D9+D10+D11+D12. DSELTotal is equal to D4+D5+D6+D7+D14. Furthermore, DTX2Total is equal to D2+D4+D7+D9+D12+D14.
Because D3, D8, and D13 are three times or more of the interval between other lines, for example, DTX1Total is a value higher than DRESTotal, DSELTotal, and DTX2Total in rows (n−1), rows (n), and rows (n+1).
For the reason above, pTX1 and pOFG are not positioned in proximity as illustrated in
Furthermore, pOFG is positioned in proximity to the photoelectric converting unit 1.
In addition, as described above, the wiring distance between pTX1 and pOFG is a maximum wiring distance of wiring distances between gate drive lines for pixel transistors.
A fourth embodiment of the present invention will be described with reference to
A propagation delay of each drive pulse transmitted through a conductor can be expressed by a product of a parasitic capacitance and an electric resistance of the conductor. Thus, the electric resistance may be reduced to obtain the same effect as the reduction of the parasitic capacitance. In other words, the line width of the gate drive line pOFG and the line width of the gate drive line pTX1 may be increased to reduce the propagation delays and thus reduce the differences in accumulation timing.
Having described that according to this embodiment, the line widths of both of the gate drive line pOFG and the gate drive line pTX1 are increased, one of them may only be increased. Alternatively, one of them may further be increased.
Such a configuration with drive lines having increased line widths may require reduction of the width of a wiring open region where no drive lines are placed or and reduction of the intervals between drive lines while keeping the width of the wiring open region. For that, by examining the effect of the reduction of parasitic capacitance and the effect of the reduction of electric resistance, one of the configurations with a greater effect may be selected.
While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the following claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.
Number | Date | Country | Kind |
---|---|---|---|
2015-027924 | Feb 2015 | JP | national |
This application is Continuation of U.S. application Ser. No. 15/498,197, filed Apr. 26, 2017, which is a Continuation of U.S. application Ser. No. 15/040,727, filed Feb. 10, 2016, now becomes U.S. Pat. No. 9,666,633, issued on May 30, 2017, which claims priority from Japanese Patent Application No. 2015-027924, filed Feb. 16, 2015, which are hereby incorporated by reference herein in their entireties.
Number | Date | Country | |
---|---|---|---|
Parent | 15498197 | Apr 2017 | US |
Child | 15674419 | US | |
Parent | 15040727 | Feb 2016 | US |
Child | 15498197 | US |