Hard Disk Head Deceleration State Estimator Re-qualification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face issues with servo faults during head positioning, leading to increased acoustic noise, poor seek settle, command time-outs, and off-track writes due to sudden deceleration to zero velocity and errors in state estimator initialization.
Innovation Solution
Implementing a model-based just-in-time (JIT) control to decelerate the head while re-qualifying the state estimator before reaching zero velocity, switching to closed-loop control for stable seek operations, and using a more arcuate deceleration and acceleration trajectory to minimize resonance excitation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the head is quickly decelerated to zero velocity using minimum-time seek control when a servo fault is detected, then the servo fault recovery time is reduced, but acoustic noise increases and seek settle performance deteriorates
Solution Approach 1:
The patent applies dynamics by transitioning from static minimum-time deceleration to dynamic model-based deceleration control. The servo controller uses a dynamic model of the actuator system to adjust deceleration parameters in real-time, optimizing the balance between recovery speed and noise generation. This allows the system to adaptively modify deceleration profiles based on current operating conditions, reducing acoustic noise while maintaining acceptable recovery times.
Solution Approach 2:
The patent changes the deceleration parameters from fixed minimum-time values to model-based optimized values. By using a dynamic model of the actuator system, the controller can adjust deceleration magnitude, rate of change, and timing parameters to minimize resonance excitation and acoustic noise while still achieving timely fault recovery. This parameter optimization resolves the contradiction between speed and noise.
2Loss of time
If the head is quickly decelerated to zero velocity using minimum-time seek control when a servo fault is detected, then the servo fault recovery time is reduced, but seek settle performance deteriorates
Solution Approach 1:
The patent uses dynamic model-based control to optimize the deceleration profile, ensuring smooth transitions that maintain system stability. By incorporating the actuator's dynamic characteristics into the control algorithm, the system can decelerate efficiently while avoiding conditions that lead to poor seek settle performance, thus maintaining reliability during fault recovery.
Solution Approach 2:
The patent applies beforehand cushioning by using the dynamic model to predict and prepare for potential instability during deceleration. The model-based control anticipates resonance conditions and adjusts deceleration parameters in advance to cushion against adverse effects, ensuring smooth seek settle performance even during rapid fault recovery operations.
3Speed
If a significant error in the estimated servo states is used to initialize the double integrator model-based open loop control, then the state estimator initialization speed is improved, but a high-speed runaway condition may occur which can damage the head
Solution Approach 1:
The patent implements feedback by continuously monitoring servo sector data during the deceleration process to update and refine state estimates. This feedback mechanism allows the system to detect and correct errors in real-time, preventing significant estimation errors from causing high-speed runaway conditions while maintaining fast initialization through efficient use of available data.
Solution Approach 2:
The patent applies preliminary action by using available servo sector information before deceleration to pre-initialize the state estimator with reasonable estimates. This preliminary initialization, combined with subsequent feedback during deceleration, allows fast startup while preventing dangerous estimation errors through continuous verification and correction.
Data Source
AI summary
A data storage device is disclosed comprising a disk comprising a plurality of tracks defined by servo sectors, a head, and a servo controller configured to servo the head over the disk based on an estimated state generated by a state estimator. The servo controller is configured to execute a seek operation to seek the head over the disk and recover from a servo fault during the seek operation by generating an initial state estimate of the head at the beginning of the servo fault, and decelerating the head open-loop using a model-based deceleration control in response to the initial state estimate. While decelerating the head, the state estimator is re-qualified before the head reaches zero velocity, and after re-qualifying the state estimator and before the head reaches zero velocity, a seek operation is executed to seek the head to a target track closed-loop using the state estimator.