Pipe Register Circuit for Semiconductor Memory

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional semiconductor memory apparatuses experience reduced operation speed due to delays in data output, particularly caused by the data pipe register, which increases the area required for a large number of data input/output pads and results in inefficiencies in data transfer.

Innovation Solution

A pipe register circuit is introduced, featuring an address storage section to temporarily store address signals and an address output control section that generates control signals to manage the output, reducing delays and optimizing data transfer by reflecting CAS latency, thereby enhancing operation speed and reducing area requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If a data pipe register is provided for each data input/output pad to manage data output delays, then CAS latency can be controlled, but the area of the semiconductor memory apparatus increases significantly

Engineering Contradiction:
Improvedata output delayVSAvoidarea of semiconductor memory apparatus
Core Design Contradiction:
Loss of timeVSArea of stationary object

Solution Approach 1:

The patent merges multiple data pipe registers into a single shared pipe register that serves all data input/output pads. Instead of having separate pipe registers for each pad, one pipe register is shared among all pads, significantly reducing the area occupied by pipe registers while still enabling controlled data output timing through the address output control section.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The single pipe register is designed to serve multiple functions by managing data output for multiple data input/output pads simultaneously. The address output control section generates control signals that coordinate the shared pipe register's operation across different pads, allowing one register to perform the work of many individual registers.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Loss of time

If multiple data pipe registers are provided for each data input/output pad, then data output timing can be precisely controlled, but the operation speed of the semiconductor memory apparatus is reduced

Engineering Contradiction:
Improvedata output timing controlVSAvoidoperation speed of semiconductor memory apparatus
Core Design Contradiction:
Loss of timeVSSpeed

Solution Approach 1:

By merging multiple pipe registers into one shared register, the patent eliminates the sequential delays that would occur if data had to pass through multiple separate registers. This reduces the overall data output path length and enables faster data transfer while maintaining timing control through the address output control section.

Inventive Principle:
Principle #5Merging (Combining)

3Ease of manufacture

If a separate data pipe register is provided for each data input/output pad, then data transfer control is simplified, but the number of components increases leading to higher complexity

Engineering Contradiction:
Improvedata transfer control simplicityVSAvoidnumber of pipe register components
Core Design Contradiction:
Ease of manufactureVSDevice complexity

Solution Approach 1:

The patent combines multiple pipe register components into a single shared pipe register, dramatically reducing the number of components needed. The address output control section provides the necessary control functionality that would otherwise require separate control circuits for each pipe register, simplifying the overall device structure.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS8929172B2Pipe register circuit and semiconductor memory apparatus having the same
Publication Date: 2015.01.06 SK HYNIX INC
  • US8929172B2 patent drawing
  • US8929172B2 patent drawing
  • US8929172B2 patent drawing

AI summary

A pipe register circuit includes an address storage section configured to temporarily and sequentially store address signals input from an external in correspondence with a read command signal input together with the address signals, and an address output control section configured to generate an address output control signal for allowing the address signals stored in the address storage section to be output in correspondence with CAS latency, and output the address output control signal to the address storage section.