Image Sensor Pixels with P-Channel Source Followers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reduced pixel size in solid-state image sensors leads to increased noise and reduced performance due to thermal kTC-reset noise and random telegraph signal noise, necessitating a design that minimizes transistors and metal wire interconnections while maintaining low noise performance.
Innovation Solution
A compact image sensor pixel design with a shared floating diffusion node, utilizing a p-channel MOSFET source follower transistor and an n-channel MOSFET reset transistor, eliminates the need for a row addressing transistor, reducing transistor count and noise, and incorporates a mini n-well region to enhance charge storage capacity and dynamic range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If pixel size is reduced to submicron dimensions, then manufacturing cost decreases and integration density increases, but noise performance deteriorates due to increased thermal kTC-reset noise and random telegraph signal noise
Solution Approach 1:
The invention extracts and removes the row addressing transistor from the pixel circuit, eliminating the source of random telegraph signal noise associated with this transistor. By taking out this unnecessary component, the patent reduces noise while maintaining pixel functionality at submicron dimensions.
Solution Approach 2:
The patent merges multiple functions into fewer transistors. The source follower transistor is configured to perform both the buffering function and the row addressing function that were previously separated. This merging reduces the total transistor count and eliminates the noise contribution from the separate row addressing transistor.
2Device complexity
If the number of transistors per pixel is reduced, then device complexity and manufacturing cost decrease, but noise performance worsens due to increased dominance of random telegraph signal noise in smaller transistors
Solution Approach 1:
The invention extracts and removes the row addressing transistor from the pixel circuit, eliminating the source of random telegraph signal noise associated with this transistor. By taking out this unnecessary component, the patent reduces noise while maintaining pixel functionality at submicron dimensions.
Solution Approach 2:
The patent eliminates the row addressing transistor entirely, using a disposable approach where this component is removed from the circuit. This reduction in transistor count simplifies the device and reduces manufacturing complexity while the remaining transistors are designed to minimize noise impact.
3Area of moving object
If pixel components are reduced in size, then area efficiency increases, but noise performance deteriorates due to increased random telegraph signal noise in smaller transistors
Solution Approach 1:
The invention extracts and removes the row addressing transistor from the pixel circuit, eliminating the source of random telegraph signal noise associated with this transistor. By taking out this unnecessary component, the patent reduces noise while maintaining pixel functionality at submicron dimensions.
Solution Approach 2:
The patent merges multiple functions into fewer transistors. The source follower transistor is configured to perform both the buffering function and the row addressing function that were previously separated. This merging reduces the total transistor count and eliminates the noise contribution from the separate row addressing transistor.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This design achieves improved noise performance, increased charge storage capacity, and dynamic range by minimizing transistor count and noise, while maintaining robust operation and efficient charge transfer, even at submicron pixel sizes.
Implementation Method 1
Typical image sensors sense light by converting impinging photons into electrons (or holes) that are integrated (collected) in sensor pixels
Data Source
AI summary
An image sensor may include image sensor pixels formed on a substrate. Each pixel may have a photodiode, a floating diffusion node, and charge transfer gate. The pixel may include an n-type doped well region and a p-channel MOS source follower transistor formed within the n-well region. An n-channel MOS reset transistor may be coupled between the floating diffusion region and a bias voltage column line and may have a drain terminal that overlaps with the n-well region. If desired, the pixel may include a p-channel JFET source follower transistor formed within the floating diffusion region on the substrate and an n-channel MOSFET reset transistor coupled to the floating diffusion. The polarities of the doping in the substrate on which the pixels are formed may be reversed. The pixel may be formed without row select transistors to increase photodiode area and charge storage capacity.