Disk Head Oscillation Detection Using Self-Learned Spectra
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices struggle to detect high-frequency vertical oscillations of heads, leading to incomplete write recovery and potential unrecoverable data sectors due to noise from head oscillation, which current methods fail to address effectively.
Innovation Solution
Implementing a data storage device with control circuitry that generates an expected oscillation spectrum for disk bands and compares it with actual oscillation spectra from read signals to detect abnormal vertical oscillations, enabling remediation actions such as fly height adjustment, data reallocation, or head depopulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional oscillation detection methods are used, then device complexity is reduced, but measurement precision of high-frequency head oscillations deteriorates
Solution Approach 1:
The patent replaces conventional mechanical oscillation detection methods with a signal processing-based approach. The system uses read signals from data sectors to extract oscillation information through spectral analysis, substituting direct mechanical sensing with electrical signal analysis. This enables high-frequency oscillation detection without adding complex mechanical detection components.
Solution Approach 2:
The patent introduces an intermediary signal processing mechanism that uses data read signals as a mediator to detect head oscillations. Instead of directly measuring head position, the system analyzes the intermediary read signals to infer oscillation characteristics through spectral analysis, enabling indirect but precise measurement of head dynamics.
2Reliability
If no oscillation detection is implemented, then device complexity remains low, but reliability of write operations deteriorates due to unrecoverable data sectors
Solution Approach 1:
The patent implements a feedback mechanism where oscillation detection results from read signals are fed back to control write operations. The system continuously monitors head oscillations during data reading and uses this feedback to prevent or correct oscillation-related write errors, improving write reliability through closed-loop control without requiring complex external monitoring systems.
Solution Approach 2:
The system uses its own read signals to detect oscillations that affect write operations, enabling self-diagnosis and self-correction. The data storage device monitors its own head dynamics using signals already present in its operational cycle, eliminating the need for separate detection systems and reducing overall device complexity while improving reliability.
3Measurement precision
If oscillation detection is performed continuously, then measurement precision of head position is improved, but use of energy increases
Solution Approach 1:
The patent employs periodic oscillation detection by analyzing read signals at specific intervals during normal data operations. Instead of continuous monitoring, the system performs spectral analysis periodically when data is being read or written, maintaining measurement precision for oscillation detection while significantly reducing energy consumption compared to continuous monitoring approaches.
Data Source
AI summary
Various illustrative aspects are directed to a data storage device, method, and one or more processing devices that are configured to: generate an expected oscillation spectrum associated with a band of the selected disk; detect oscillation abnormality of the selected head based on comparing an actual oscillation spectrum associated with a data sector in the band to the expected oscillation spectrum associated with the band, wherein the data sector includes data written by the selected head; and perform a remediation action in response to detecting oscillation abnormality of the selected head.