Subword Driver Transistor Sharing for Memory Die Size Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in semiconductor memory devices is to reduce the number of transistors in subword drivers while maintaining efficient memory cell access, which is essential for minimizing die size and improving layout design.
Innovation Solution
The solution involves coupling subword drivers in cascade connections, sharing common transistors between adjacent drivers, which reduces the average number of transistors per subword driver from three to 2.5, thereby optimizing the layout and reducing the memory die size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the number of transistors in subword drivers is reduced, then the die size is reduced, but the reliability may be compromised due to row hammer issues
Solution Approach 1:
Adjacent subword drivers share common transistors (specifically, the nth transistor of one driver is shared with the nth transistor of the adjacent driver), merging functional elements to reduce total transistor count while maintaining individual driver operation. This reduces die size without completely sacrificing reliability through the sharing mechanism.
Solution Approach 2:
The patent implements preliminary protection mechanisms by configuring the shared transistor structure to inherently mitigate row hammer effects. The cascade connection and shared transistor architecture are designed beforehand to prevent certain types of interference before they can cause failures, addressing reliability concerns proactively.
2Quantity of substance
If subword drivers are coupled in cascade connections with shared transistors, then the average number of transistors per driver is reduced from three to 2.5, but the layout complexity increases
Solution Approach 1:
Adjacent subword drivers share common transistors, merging previously separate components into a unified structure. This reduces the total quantity of transistors from three per driver to 2.5 average per driver, directly addressing the reduction goal while the shared structure simplifies the overall layout.
Solution Approach 2:
The shared transistors serve multiple functions simultaneously - they are the nth transistor of one subword driver and the nth transistor of the adjacent subword driver. This multi-functionality reduces the total component count and simplifies the layout by eliminating redundant elements.
3Use of energy by moving object
If the number of transistors is reduced in subword drivers, then power consumption is reduced, but the ability to maintain efficient memory cell access may be compromised
Solution Approach 1:
By merging adjacent subword drivers through shared transistors, the patent reduces the total transistor count and associated power consumption. The cascade connection ensures that memory cell access efficiency is maintained through proper signal propagation across the shared transistor boundary.
Solution Approach 2:
The cascade connection configuration allows dynamic control of the shared transistors based on which subword driver is active. The structure adapts its operation mode to maintain efficient memory cell access while consuming less power through the reduced transistor count.
Data Source
AI summary
In some examples, a subword driver block of a memory device includes a plurality of active regions of a first type and a plurality of active regions of a second type adjacent to the plurality of active regions of the first type. The subword driver block further includes a plurality of first gate electrodes overlapping with the plurality of active regions of the first type to form a plurality of first transistors, and a plurality of second gate electrodes overlapping with the plurality of active regions of the first type to form a plurality of second transistors. Each of the second transistors is shared by a first subword driver and a second subword driver. Each of the second transistors may include a drain and a source respectively coupled to a first and second word line, which are driven by the first subword driver and the second subword driver, respectively.