Memory Controller Temperature Calibration Without E-Fuses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional temperature sensor calibration in memory controller chips requires testing at multiple temperatures, which is costly, time-consuming, and necessitates the use of e-fuses for data storage, increasing manufacturing complexity and cost.
Innovation Solution
A method for calibrating temperature sensors in a memory controller chip that involves measuring the sensor output at a single temperature, such as room temperature, and using a generic characteristic of the sensor type to calculate temperature readings, eliminating the need for e-fuses and allowing calibration after system assembly, with data stored in a separate nonvolatile memory chip.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temperature sensor calibration is performed at multiple temperatures using conventional methods, then measurement precision is improved, but manufacturing cost and time increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-characterizing temperature sensors during manufacturing to establish a generic characteristic (slope value) that represents the sensor's response. This pre-established characteristic allows field calibration to be performed at a single temperature rather than requiring multiple temperature points, significantly reducing calibration time while maintaining accuracy through the use of the pre-determined sensor behavior model.
Solution Approach 2:
The patent changes the calibration approach from using multiple temperature parameters to using a single temperature parameter combined with a pre-determined generic characteristic parameter. By storing the sensor output at one temperature and applying the generic characteristic (slope) to calculate temperatures across the operating range, the method reduces the number of measurement parameters needed while maintaining calibration precision.
2Measurement precision
If temperature sensor calibration is performed at multiple temperatures, then measurement precision is improved, but device complexity increases due to e-fuse requirements
Solution Approach 1:
The patent extracts the complex e-fuse calibration mechanism and replaces it with a simpler approach using standard nonvolatile memory storage. Instead of requiring specialized e-fuse hardware for storing calibration data, the method stores the single-temperature sensor output and generic characteristic in regular memory, eliminating the need for additional hardware complexity while maintaining calibration accuracy.
Solution Approach 2:
The patent replaces expensive, complex e-fuse technology with cheaper, more readily available standard memory components. By using conventional nonvolatile memory to store calibration data instead of requiring specialized e-fuse hardware, the solution reduces device complexity and manufacturing cost while achieving the same calibration functionality.
3Measurement precision
If e-fuses are used for calibration data storage, then measurement precision is maintained, but manufacturing cost increases
Solution Approach 1:
The patent replaces expensive e-fuse technology with cheaper standard memory components. By storing calibration data in conventional nonvolatile memory rather than requiring specialized e-fuse hardware, the solution significantly reduces manufacturing cost while maintaining the integrity and reliability of calibration data storage.
4Measurement precision
If calibration is performed before system assembly, then measurement precision is ensured, but productivity decreases due to rework requirements
Solution Approach 1:
The patent applies preliminary action by pre-characterizing sensors during manufacturing to capture their generic characteristics. This pre-work enables post-assembly calibration using only single-temperature measurements, eliminating the need for complex multi-temperature testing after assembly and allowing efficient calibration to be performed later without requiring system disassembly or rework.
Solution Approach 2:
The patent segments the calibration process into two distinct phases: pre-manufacturing sensor characterization (to establish generic characteristics) and post-assembly field calibration (using stored characteristics plus single-temperature measurement). This segmentation allows each phase to be optimized independently, improving overall manufacturing efficiency while maintaining calibration accuracy.
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 approach reduces manufacturing costs and time by eliminating the need for e-fuses and multiple temperature testing, while providing accurate temperature calibration and monitoring, enabling efficient heat management in memory controller chips.
Implementation Method 1
a temperature sensor in the memory controller chip providing an output that is dependent on temperature
Data Source
AI summary
A nonvolatile memory system includes a memory controller chip with at least one temperature sensor that is individually calibrated, at a single temperature, after the nonvolatile memory system is assembled, so that the calibration data is stored outside the memory controller chip, in a nonvolatile memory chip, thus obviating the need for components to store calibration data in the memory controller chip.