Radiant energy – Photocells; circuits and apparatus – Photocell controlled circuit
Reexamination Certificate
1999-09-15
2002-10-01
Pyo, Kevin (Department: 2878)
Radiant energy
Photocells; circuits and apparatus
Photocell controlled circuit
C250S214100, C257S251000, C348S295000, C377S057000
Reexamination Certificate
active
06459077
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to TDI photodiode sensor. In particular, the invention relates to use of a bucket brigade architecture for the TDI sensor to achieve CMOS process compatible TDI sensor.
2. Description Of Related Art
Today's Time Delay and Integrate (TDI) sensors are implemented in Charge Coupled Device (CCD) technology. CCDs provide high charge transfer efficiency (CTE). However, CCD fabrication requires many steps and tight controls on the processes, and in general, it is more expensive than other technologies, such as bucket brigade charge transfer devices. Bucket brigade devices are NMOS devices that can be fabricated using standard CMOS processes. However, without efficient charge transfer, it is not possible to design and fabricate advanced TDI sensors. Known bucket brigade devices suffer from poor charge transfer efficiency relative to the efficiency needed in modern TDI sensors.
Bucket brigade charge transfer devices have been known since the 1970s. In U.S. Pat. No. 3,683,193 to Weimer, filed Oct. 26, 1970, incorporated herein by reference, a bucket brigade charge transfer device used as a frame transfer device has been integrated with an array of photodiodes to achieve a bucket brigade sensor. In
FIG. 1
, plural n type diffusions (D
1
through D
6
) are formed in a p type substrate, each diffusion having the shape of the letter “T”. An insulator film is formed over the entire substrate so as to have thick and thin regions as described below. Plural metal straps are formed over the insulator film so as to overlay the diffusions at alternating large and small areas, short channels being disposed beneath a metal strap and between adjacent large and small areas.
In
FIG. 2
(a sectional view longitudinally through the metal strap that carries CLK
2
), the channel between diffusions D
2
and D
3
is covered by a thick insulator, and the channel between diffusions D
4
and D
5
is covered by a thick insulator. The thick insulator diminishes the electric field in the underlying channel so that the channel does not conduct charge even when a positive clock signal (e.g.,+5 to +15 volts) is applied to the metal strap. In contrast, the channel between diffusions D
3
and D
4
is covered by a thin insulator, and the channel between diffusions D
5
and D
6
is covered by a thin insulator so that the channel conducts charge when a positive clock signal (e.g.,+5 to +15 volts) is applied to the metal strap. In this way, the bucket brigade device becomes a one way device transferring charge from input IN to output OUT (FIG.
1
).
Each large area forms a large capacitance between the metal strap and the underlying diffusion. Each small area forms a relatively smaller capacitance between the metal strap and the underlying diffusion. When a positive clock signal (e.g.,+5 to +15 volts) is applied to a metal strap, the channels under the metal strap (where the insulator film is thin) become conductive (i.e., forms a surface channel MOS type transistor in an “on” state between the two adjacent diffusions). The source is characterized by a capacitance between the source and the metal strap. The drain is characterized by a capacitance between the drain and the metal strap. All other things being equal, charge will distribute between the source and drain (i.e., the adjacent diffusion regions) in proportion to their respective capacitances. When the clock signal returns to ground and the surface channels between the adjacent diffusions becomes non-conductive, a greater quantity of the total charge will be trapped in the diffusions with the greater capacitance. A greater capacitance difference will provide a greater charge transfer efficiency.
By alternately clocking CLK
1
and CLK
2
(not overlapping), charge is moved from input IN to output OUT (FIG.
1
). Clocking CLK
2
will move charge Q from diffusion D
1
through MOS transistor M
1
into diffusion D
2
. Then, clocking CLK
1
will move charge Q from diffusion D
2
through MOS transistor M
2
into diffusion D
3
, but charge Q will not back transfer through the channel under CLK
1
from diffusion D
2
to diffusion D
1
since this channel is covered by a thick insulator. Then, clocking CLK
2
again will move charge Q from diffusion D
3
through MOS transistor M
3
into diffusion D
4
, but charge Q will not back transfer through the channel under CLK
2
from diffusion D
3
to diffusion D
2
since this channel is covered by a thick insulator.
U.S. Pat. No. 3,683,193 to Weimer also teaches that the diffusion areas form a PN junction with the substrate so as to act as reverse biased photodiodes when this structure is illuminated from the back. When illuminated from the back, the substrate must be thinned (to about 0.5 mils thickness) to a thickness comparable to the diffusion range of photocarriers. When illuminated from the front (i.e., the metalized side) the substrate may be thicker (e.g., 10 mils or so), but the metal straps must be either narrow or sufficiently thin to be semi-transparent to the illumination.
SUMMARY OF THE INVENTION
It is an object to the present invention to provide a TDI bucket brigade structure that is adapted to standard single poly CMOS processes. It is another object of the present invention to provide a TDI sensor with good blue and UV response. It is a further object of the present invention to provide a TDI sensor low lag time and high charge transfer efficiency.
These and other objects are achieved in a TDI sensor that includes a bias charge voltage circuit, a reset voltage circuit, a bucket brigade column having a plurality of nodes, and a plurality of pinned photodiodes. A cathode of each photodiode is formed integral with a corresponding node of the bucket brigade column. The bucket brigade column is coupled between the bias charge voltage circuit at an initial node and the reset voltage circuit at a final node. The bucket brigade column includes a plurality of first phase clock conductors, and a plurality of second phase clock conductors, and the first and second phase clock conductors are interdigitated and formed of poly-crystalline silicon. The TDI sensor is formed in a substrate of a first conductivity type, and a cathode of each pinned photodiode is formed of a second conductivity type, and each pinned photodiode includes a pinning layer of the first conductivity type.
REFERENCES:
patent: 3683193 (1972-08-01), Weimer
patent: 3811055 (1974-05-01), Weimer
patent: 3856989 (1974-12-01), Weimer
patent: 4001878 (1977-01-01), Weimer
patent: 4117514 (1978-09-01), Terui et al.
patent: 4264930 (1981-04-01), White
patent: 4405857 (1983-09-01), Hirai et al.
patent: 4479201 (1984-10-01), Domingo
patent: 4779124 (1988-10-01), Hynecek
patent: 5315114 (1994-05-01), Kansy et al.
patent: 5323186 (1994-06-01), Chow
patent: 6215139 (2001-04-01), Iida et al.
Dalsa Inc.
Pyo Kevin
LandOfFree
Bucket brigade TDI photodiode sensor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Bucket brigade TDI photodiode sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bucket brigade TDI photodiode sensor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2990563