Active pixel sensor (APS) imaging devices are described in U.S. Pat. No. 5,471,515. These imaging devices include an array of pixel cells that convert light energy into electrical signals. Each pixel includes a photodetector and one or more active transistors formed of MOS technology. The transistors typically provide amplification, readout control and reset control, in addition to producing the electrical signal output from the cell.
APS devices may be fabricating using CMOS technology. An APS sensor, including a pixel array and readout structure, may be provided on the same integrated circuit (IC) chip. The gain stage of the APS sensor may consume a considerable amount of the chip area. However, reducing the size of this area may sacrifice signal gain in the APS sensor.
The readout structure of an APS typically includes a source follower transistor. Use of such a transistor may itself limit the dynamic range of the sensor when reading out the pixel and provide a voltage gain less than unity.
An active pixel sensor (APS) according to an embodiment includes an array of individually addressable APS cells, each of which include a row select transistor and an output transistor, and a pixel readout structure that includes an amplifier. The amplifier structure includes the row select transistor and output transistor of each pixel. The amplifier may provide a gain of one or higher.
According to an embodiment, the amplifier has a two branched structure. One branch includes a transistor connected to VDD, and the output transistor and the row select transistor for each pixel in a column. The other branch includes three transistors in series, the first connected to VDD, and the third connected to the drain of a source follower transistor connected to ground. The row select transistor for each pixel in the column is also connected to the source follower transistor. The transistors in the branches connected to VDD may be p-type MOSFETS and the other transistors may be n-type MOSFETS.
According to another embodiment, a transistor may be connected between the pixel output transistor and the p-type transistor in each pixel to protect the output transistor from voltage charges in the column line.
According to another embodiment, the amplifier may be provided with a gain selector to optionally change the gain in the amplifier. The gain selector may comprise two p-type transistors connected between the source of the p-type transistor in each branch and VDD: a gain transistor and a gain-enable transistor. The gain transistor, connected to VDD, may be self-biased, and the gain-enable transistor may be controlled between ON and OFF states by a gain voltage source. When the gain-enable transistor is ON, the gain transistor and other transistor at VDD are effectively in parallel, thereby altering the gain of the amplifier.
The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.
Like reference symbols in the various drawings indicate like elements.
An embodiment is shown in FIG. 1. The readout structure 102 of an active pixel sensor (APS) 100 includes an amplifier circuit 104. The APS includes an array 106 of independently addressable pixels arranged in n rows and m columns.
The drain 130 of each output transistor 118 in the column is connected to the source 132 of a p-type transistor 134 (denoted M3). The drain 136 of transistor 134 is connected to a system drain voltage VDD, and the gate 138 of transistor 134 is connected to its source. The source of each row select transistor 120 in the column is connected to the drain 140 of an n-type source follower transistor 142, which has a gate 144 connected to a bias voltage, Vbias, and a source 146 connected to ground.
When a pixel is being read out, the p-type transistor 134, pixel output transistor 118, and pixel row enable transistor 120 form one branch of the amplifier circuit 104. A second branch of the amplifier circuit includes a p-type transistor 150 (denoted M4) and two n-type transistors 152, 154 (denoted M2 and M7, respectively). The drain 156 of p-type transistor 150 may be connected to VDD and its gate 158 connected to its source 160. The drain of transistor 152 is connected to the source of transistor 150 and the gate 162 is connected to a load voltage Vn. The drain of transistor 154 is connected to the source of transistor 152, its gate 164 is connected to VDD and its source is connected to the drain of source-follower transistor 142. Transistor 154 in the second branch may be provided for symmetry with row enable transistor 120 in the pixel being read out.
When the pixel is read out, the row enable voltage is set HIGH, and row select transistor 120 and transistor 154 are essentially shorted out. The relationship of the remaining transistors in the amplifier circuit 104 may be described as shown in FIG. 2. The gain of the amplifier is:
A0=gm1*(rds2//rs4)=//rds2>>rs4//≈gm1*rs4=gm1/gm1/gm4=//μn≈3*μp≈(3*(W/L)1/(W/L)4)1/2 (1)
When Vp equals Vn the output is the same and these outputs are:
Vout+=Vout−=VDD−Vgs4=VDD−Veff−|Vtp|=VDD−(Ibias/μp*Cox*(W/L)4)1/2−|Vtp| (2)
A simple first order transfer function of a dominant pole op amp can be described by:
Av(S)=A0/(l+s/ω) (3)
Av(S)=gm1*(rout//1/sCL)=/for midband frequencies CL dominates/=gm1/sCL=>/|Av(jωta)|=1/=> (4)
ωta=gm1/CL=>/ω<<ωta/; and (5)
A0*ω=ωta=gm1/CL=> (6)
tswitch≅CL=> (7)
The dimensions of the various transistors may be adjusted to reduce noise. For example, the length of transistor 150 (M4) may be decreased to decrease flicker noise. Input noise may be independent of the width of transistor 150. That width hence may be widened to maximize signal swing at the output. Also, increasing the length of the pixel output transistor 118 may increase noise, whereas increasing the width of the transistor 118 may reduce flicker noise, thermal noise, and white noise.
The gain of the amplifier 104 according to the embodiment may be set to unity or higher. The response of a simulated sensor according to the embodiment shown in
According to an embodiment shown in
The branched amplifier according to an embodiment may reduce noise from the substrate because it provides a differential output. The amplifier may also improve the dynamic range of the sensor when reading out a pixel because loss from the source follower may be reduced with the addition amplifier structure.
A number of embodiments of the invention have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the invention. Accordingly, other embodiments are within the scope of the following claims.
This application claims the benefit of the U.S. Provisional Application No. 60/139,348 entitled NOVEL IDEA FOR A NEW READOUT STRUCTURE OF APS filed on Jun. 15, 1999.
Number | Name | Date | Kind |
---|---|---|---|
5471515 | Fossum et al. | Nov 1995 | A |
5920345 | Sauer | Jul 1999 | A |
6130423 | Brehmer et al. | Oct 2000 | A |
6493030 | Kozlowski et al. | Dec 2002 | B1 |
Number | Date | Country | |
---|---|---|---|
60139348 | Jun 1999 | US |