Dynamic Flip-Flop Latch Using Weak Keeper to Cut Leakage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dynamic flip-flops suffer from issues such as glitches and leakage currents, leading to erroneous data signals and storage loss, while maintaining the advantage of reduced circuit layout area and production cost.
Innovation Solution
The dynamic flip-flop design incorporates a transmission gate, two inverters, a pull-up transistor, and a pull-down transistor, utilizing a tri-state inverter to reduce leakage currents and a weak keeper circuit to prevent data collisions, allowing for high-speed operation without storage loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a conventional dynamic flip-flop architecture is used, then the circuit layout area is reduced and production cost is lowered, but leakage currents occur causing erroneous data signals and storage loss
Solution Approach 1:
The flip-flop is divided into master and slave sections with separate clock phases. The master latch captures data on one clock phase while the slave latch holds data on the other phase, segmenting the data path to prevent leakage-induced errors while maintaining compact area.
Solution Approach 2:
The patent employs dynamic clocked latches where transistors are periodically activated and deactivated based on clock signals. This dynamic operation allows the circuit to maintain data while controlling leakage currents through timed transistor switching, resolving the contradiction between area efficiency and reliability.
2Speed
If transmission gates and inverters are used to reduce input signal load for high-speed operation, then operation speed is improved, but glitches and leakage currents occur leading to erroneous data output
Solution Approach 1:
The patent uses periodic clock signals to control the master and slave latches alternately. During the first clock phase, the master latch is active and slave is held; during the second phase, roles reverse. This periodic action enables high-speed operation while preventing glitches and leakage from corrupting data by ensuring proper timing isolation.
Solution Approach 2:
The slave latch acts as an intermediary between the master latch and the output. It receives data from the master latch during one clock phase and transfers it to the output during the next phase, mediating the data flow to prevent direct exposure to leakage currents and glitches while maintaining high-speed operation.
3Device complexity
If dynamic flip-flop architecture is used instead of static, then circuit area and production cost are reduced, but leakage currents cause storage loss
Solution Approach 1:
The patent pre-charges and pre-discharges specific nodes during defined clock phases to prepare the latches for data capture. This preliminary action ensures that when data is latched, the circuit is in the correct state, preventing leakage currents from causing storage loss while maintaining the simplified dynamic architecture.
Solution Approach 2:
The patent changes the operational parameters of transistors dynamically through clock-controlled gating. By periodically activating and deactivating transistor channels based on clock phases, the circuit achieves low leakage during hold phases while maintaining functionality during active phases, resolving the contradiction between simplicity and energy loss.
Data Source
AI summary
A dynamic flip flop is provided. The dynamic flip-flop comprises a transmission gate, a first inverter, a second inverter, a pull-up transistor and a pull-down transistor. The pull-up transistor and the pull-down transistor constitute a feedback inverter, and the feedback inverter is configured as a weak keeper circuit compared to the first inverter serving as a tri-state inverter. Therefore, the dynamic flip-flop can be such that makes a master latch to use the tri-state inverter for capturing data in order to reduce electric leakage. In addition, the dynamic flip-flop can also be such that makes a slave latch to use the weak keeper circuit for storing data, thereby avoiding floating point to drive the output.


