Secondary Actuator Research Articles

A proximate-time-optimal-control design and its application to a hard disk drive dual-stage actuator system

We propose a compensator-based strategy for design of a track-seeking and track-following control system for a dual-stage servo actuator in hard disk drives. A well-known decoupling structure is employed to disconnect the control of the primary voice coil motor (VCM) actuator from the loop for a secondary high-bandwidth actuator. The compensator is placed in the secondary loop and suitably combined with a saturation nonlinearity in order to obtain actuator signal boundedness. The design procedure consists of four steps: 1) design of an established nonlinear seek-settle-track following controller for the VCM; 2) design of a linear track following controller for the secondary actuator; 3) observer design; and 4) design of a compensator to retain global stability and to improve performance. The proposed control system improves performance of both long-span seeking (proximate-time-optimal controller) and short-span seeking. In addition, it achieves high-bandwidth track following performance. The experimental results show good track-following performance, and short-span/long-span-seeking performance with fast settling time. The overshoot during track seeking can be made negligible for a suitably tuned VCM-actuator control loop.

Active control of buzz-saw tones: Experimental results from a laboratory-scale, no-flow rig

This paper details the results of an experiment on a small scale no-flow rig to control a spinning mode with similar characteristics to the modes generated by buzz-saw noise. The spinning mode (m=−3, n=1) is generated by a mode synthesizer ring which has similar characteristics to that of a low frequency buzz-saw mode. A control system using seven microphones and seven control actuators is used to minimize the sum of the squared pressures at the duct wall using a filtered reference least mean squares (FXLMS) algorithm. It is shown that at certain frequencies the sound power transmitted to the duct termination is reduced by 14.5dB. The control mechanisms at different frequency ranges are identified and discussed with reference to computer simulations. It is shown that cut-off modes excited by the secondary actuator array have a significant effect on control performance at frequencies close to the cut-on frequency of the (m=−4, n=1) mode.

Practical Implementation of a Novel Anti-Windup Scheme in a HDD-Dual-Stage Servo-System

Two novel discrete anti-windup (AW) techniques are applied to a dual-stage actuator of an experimental hard disk drive system. The techniques, one low order, the other full order, employ convex l/sub 2/-performance constraints in combination with linear-matrix-inequality-optimization methods. It is shown that the AW compensators can improve the performance of the nominal dual-stage servo-system when the secondary actuator control signal saturates at its allowable design limits. Also, stability is achieved despite saturation of both the secondary actuator and the voice-coil-motor actuator. Practical results show that the performance of each AW compensator is superior to another well-known ad-hoc AW technique, the internal model control AW scheme. The main contribution of the paper is the application of theoretically rigorous AW methods to an industrially relevant servo system.

Stability evaluation of an accurately controllable multiple injection system

Aiming to further improve diesel engine fuel efficiency and exhaust emissions, we conducted tests of a newly devised rapidly operating secondary actuator that provides controllable multiple injection. Up to six shots per cycle were implemented under high-pressure injection. A high-speed camera and a flow rate metering system were used to obtain quantitative data regarding diesel spray and fuel flow dynamics. The advantages of the applied technique are very stable phasing, dwelling, duration, and dosing of multiple injection shots proved from data analysis.

ANALYSIS AND COMPARISON OF FLEXURE STRUCTURES USED IN 3-D ELECTROSTATIC MICROACTUATORS FOR HARD DISK DRIVES

International Journal of Computational Engineering ScienceVol. 04, No. 02, pp. 227-230 (2003) Actuators and SensorsNo AccessANALYSIS AND COMPARISON OF FLEXURE STRUCTURES USED IN 3-D ELECTROSTATIC MICROACTUATORS FOR HARD DISK DRIVESBANGTAO CHEN, JIANMIN MIAO, and HONG ZHUBANGTAO CHENMicromachines Centre, School of Mechanical and Production Engineering, Nanyang Technological University, Singapore 639798, Singapore Search for more papers by this author , JIANMIN MIAOMicromachines Centre, School of Mechanical and Production Engineering, Nanyang Technological University, Singapore 639798, Singapore Search for more papers by this author , and HONG ZHUMicromachines Centre, School of Mechanical and Production Engineering, Nanyang Technological University, Singapore 639798, Singapore Search for more papers by this author https://doi.org/10.1142/S1465876303000958Cited by:0 PreviousNext AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsRecommend to Library ShareShare onFacebookTwitterLinked InRedditEmail AbstractTo fulfill high-bandwidth and high-accuracy positioning requirements of hard disk drives (HDD), a novel slider-driven 3-dimensional electrostatic microactuator using dual-stage servo system was proposed in this paper. The microactuator is assembled between the suspension and the slider as the secondary actuator, and it is featured with the high aspect ratio 3-D silicon structure, high positioning bandwidth, low driving voltage and good reliability. The electrostatic microactuator has the parallel comb drives and mounting planes in the perpendicular planes, which greatly contributes to the large electrostatic force. The design and selection of the microflexures are crucial to the natural frequency and reliability of the microactuator. The flexure should have high stiffness effects and stiffness ratio. Four different flexure structures have been studied: straight-plate flexure, folded-plate flexure, curvy-plate flexure, and interlapped-L flexure. By analytical and numerical analysis, the interlapped-L flexure presents with higher stiffness effects and better actuation reliability, secondly is the straight-plate flexure. With ANSYS simulation, a servo bandwidth of over 2 kHz and a stroke of ±0.5 μm of the actuator can be obtained by utilizing interlapped-L flexure under a low driving voltage of 30V. It well satisfies the designed objective of the 3-D microactuators.Keywords:3-dimensionalelectrostaticmicroactuatorflexureinterlapped-L References T. Imamuraet al., IEEE Transactions on Mechatronics 3(3), 166 (1998). Crossref, Google ScholarS. Koganezawaet al., IEEE Transactions on Magnetics 3(4), 1910 (1998). Google Scholar W. C. Tang, Electrostatic comb drive for resonant sensor and actuator applications, Ph.D. Thesis, University of California at Berkeley, 1990 . Google Scholar G. K. Fedder, Simulation of Microelectromechanical System, Ph.D. thesis, University of California, Berkeley, 1994 . Google Scholar FiguresReferencesRelatedDetails Recommended Vol. 04, No. 02 Metrics History Keywords3-dimensionalelectrostaticmicroactuatorflexureinterlapped-LPDF download

HDD dual-stage servo-controller design using a μ-analysis tool

A simple tool for assessment of the closed-loop performance of a linear standard dual-stage controller for track following in hard-disc drives (HDD) is presented. For this reason, a μ-control analysis problem is posed so that the uncertainty of the voice-coil-motor (VCM) actuator and of the secondary actuator can be considered while raising a performance constraint for the sensitivity magnitude. The algebraic formula for the resulting μ-control analysis problem is readily calculated and allows the computation of the amplitude of a worst case sensitivity frequency response for explicit evaluation of the prospective robust control performance. The validity of the approach is verified for a practical example using a HDD-system with piezoelectric secondary actuator.

Frequency response evaluation method for microactuators in dual-stage hard disk servo systems

Much of the interest in dual-stage microactuation for hard disk drives centers on control system design methodology, error rejection improvements, and optimal division of the control effort between the coarse and fine actuators. The authors, instead, place the emphasis on evaluating the secondary actuator mechanics from the point of view of the control system. The effective performance of the mechanical plant can be better assessed when combined with a "target" compensator than it can as a raw mechanical transfer function or frequency response function alone.

Dealing with Secondary Actuator Saturation in Dual-Stage HDD Servo

This paper proposes a modification in the decoupled master-slave dual-stage hard disk drive (HDD) servo control configuration to enhance its stability when the secondary actuator is saturated. This method utilizes the special structure of the decoupled master-slave dual-stage control design and uses a nonlinear secondary actuator model instead of the usual linear one to predict its output and decouple the dual-actuator loop. This technique relaxes the displacement range requirements on the secondary actuator, and also greatly improves the stability of the servo system during the actuator saturation.

A simulation and control design environment for single-stage and dual-stage hard disk drives

We describe an integrated tool for the design and evaluation of the performance of different control algorithms and strategies, applied to hard disk drives (HDDs) with single-stage actuators and dual-stage actuators. The tool is composed of a set of procedures for the computer-aided design of servocontrollers and a HDD simulator. In order to perform realistic simulations, each part of the HDD is described with high level of detail. As for the mechanics, the usual inertia plus resonance model of the voice coil motor (VCM) has been enriched with nonlinear friction modeling and, in case of DSAs, with the dynamic coupling between primary and secondary actuator. The nonlinear friction model has been tuned on experimental data, using an experimental system presented in the paper. As for the electronics, the VCM driver model is included and quantizations in the position error signal measurement and computation are explicitly considered. Also, in order to evaluate the tracking performance of different servocontrollers, repetitive run outs and nonrepetitive run outs are included. Such disturbances have been obtained from commercial HDDs, with a procedure that is detailed in the paper. The simulator is then validated by comparing simulation and experimental results, both in open-loop and closed-loop conditions, thus, confirming the effectiveness of the developed tool.

DUAL-STAGE ACTUATOR DESIGN FOR HDD SYSTEMS VIA CNF CONTROL

Dual-stage actuator design for hard disk drives via composite nonlinear feedback control is presented, which consists of a primary actuator, the voice coil motor and a secondary actuator, the suspension piezoelectric microactuator. The first is effective in both track seeking and following stages to produce fast rising time and low overshoot and the latter works in track following stage only to help the primary actuator. The implementations show that the dual-stage actuator does yield better performance compared to the single primary actuator.

Deflection characteristics of a trapezoidal multilayer in-plane bending piezoelectric actuator.

A novel multilayer split-morph actuator has been designed and fabricated using the thick film screen-printing technology. Deflection characteristics of the split-morph actuator have been investigated by theoretical analysis and experimental measurement. The results indicate that the tip displacement is inversely proportional to the thickness of each piezoelectric layer, but is independent of the number of layers and the total thickness of the actuator. The displacement/voltage sensitivity of the trapezoidal actuator is larger than that of the rectangular design, assuming both have the same width of clamped end and the same thickness of the piezoelectric layers. The maximum displacement/voltage sensitivity of 0.157 micron/V was obtained with a split-morph actuator with 30-micron thick layers. The proposed actuator is a promising candidate for the secondary fine-tuning actuator of a dual stage head-positioning servo system in high density hard disk drives.

Trial production and experiments of linear actuator with HTS bulk secondary

To investigate characteristics of a linear actuator with high-temperature superconducting (HTS) bulk secondary, a short secondary type linear actuator has been designed and constructed. The actuator is comprised of an YBCO bulk secondary (mover) and copper windings with iron core (primary). A zero-field-cooled bulk located at the center of the actuator plays a role of generating thrust, while four field-cooled (trapped field) bulks are used for levitation and guidance of the secondary. The authors measured the starting thrust force and magnetic field distribution in the air gap. They developed a simulation program based on the finite element method (FEM) taking the voltage-current (E-J) characteristics of the bulk into consideration to investigate electromagnetic behaviors within the bulk exposed to a time-varying magnetic field. Agreement between experiment and simulation is good, and it validates the simulation program and the presented assumptions in the numerical approach, using the simulation program, they investigated the dependency of n-value and critical current density of the bulk material on the magnetic flux density in the air gap and the starting thrust force. Supercurrent density within the bulk is a key factor for the magnetic flux density and the thrust force.

Piezo-based milliactuator on a partially etched suspension

This paper describes a piezo-based moving-suspension secondary actuator (called milliactuator) which provides increased servo bandwidth and reduced track-misregistration in a hard disk drive. The milliactuator comprises a pair of piezo-elements driving a relatively thick loadbeam that is partially etched in selected areas to form a honeycomb like structure. Partial etching allows more freedom in optimizing mass and stiffness to enhance the milliactuator dynamics. For a 18 mm (swage to dimple) milliactuator head gimbal assembly (HGA) with load/unload tab, we were able to achieve a first torsion frequency (T1) of 6 kHz. In a prototype drive populated with milliactuators, a new dual stage servo algorithm resulted in 60% higher bandwidth (0 dB open-loop crossover) compared to without the milliactuator. Further modifications to the milliactuator along with improvements in sampling rate and computational time could allow up to 100% higher servo bandwidth. The benefits and limitations of milliactuators are also discussed. The milliactuator provides a low cost, enhanced performance solution for high track densities.

Modal control of beam flexural vibration

An active control system was developed to control the flexural vibrations of a beam with a modal filtering with only one secondary actuator. Segmented piezoelectric actuators and sensors were used for driving and sensing the bending beam vibrations. The primary actuator was fed by a broadband random disturbance signal in order to excite the first five modes of the structure. However, only the second to fifth modes were controlled. The control algorithm was implemented on a DSP board and the input and output signals were filtered using high order low pass filters. These filters, implemented on the DSP board avoid the degrading effect on the control performances of the higher order modes and which are not controlled. The modal filtering was achieved by computing. To this end, it is based on a previous identification procedure. This latter models, in one step, the dynamics of the structure and also the transfer function of the electronic circuits of the controller. The identified filtered modes were then used to compute the gain matrix using a LQR technique (linear quadratic regulator). Simulations of the active control were carried out and practical implementation of the control algorithms was performed. Experimental and simulation results were then compared and discussed.

ELECTROPNEUMATIC TRANSDUCERS AS SECONDARY ACTUATORS FOR ACTIVE NOISE CONTROL PART III: EXPERIMENTAL CONTROL IN DUCTS WITH THE SUBSONIC SOURCE

The possibility of performing active control of periodic noise propagating in ducts using a subsonic electropneumatic acoustic generator as secondary source is investigated. The subsonic generator has been studied both theoretically and experimentally in two companion papers, and this sound source was shown to be highly efficient, but non-linear. The non-linear behaviour of the source decreases as the acoustic pressure at its output is reduced, however, as in the case when the source is used as a secondary actuator in an efficient active control system and thus the source is well suited to such applications. Residual non-linearities of the subsonic source are shown to be due to the mechanical design of the actuator. An harmonic controller is discussed which accounts for the residual non-linear behaviour of the subsonic source. Experiments carried out with a manual version of this controller, that controls the first five harmonic components of the signal driving the subsonic source, reveal that it is efficient in controlling periodic primary sound fields. The implementation of a fully coupled harmonic controller is, however, shown to require large processing capacities. A theoretical analysis of a simplified version of the harmonic controller—the decentralized harmonic controller—is carried out, and simple conditions are established under which the decentralized controller offers similar performances to the fully coupled harmonic controller. These conditions are shown to be satisfied when using the subsonic source as a secondary actuator. The non-linear behaviour of the subsonic source can also cause a further problem, since the error surface experienced by the control system may exhibit local minima. It was found that the likelihood of this happening was much reduced is only the fundamental component of the harmonic controller was adapted initially, and then the other harmonics were changed in a second phase of adaptation. The implementation of a decentralized harmonic controller is considered, using a dual channel signal processing board. A purely linear model for the plant under control is shown to be accurate enough for modelling the system under control and to ensure convergence of the controller. Experiments with the automatic controller reveal that attenuations measured at the monitor microphone are around 25 dB, for sinusoidal primary sound fields.

ELECTROPNEUMATIC TRANSDUCERS AS SECONDARY ACTUATORS FOR ACTIVE NOISE CONTROL, PART I: THEORETICAL ANALYSIS

A theoretical analysis is presented of an acoustic source which operates by the release of compressed air through an aperture, the area of which is made to vary with time. A distinction is made between sonic and subsonic compressed air sources (depending on the Mach number at the throat), and in both cases simple equations are derived that describe the acoustical characteristics of the device. The theory is also developed with a view to using the source as a secondary actuator in an active noise control system, and the pneumatic efficiency and linearity of compressed air sources are discussed. Although the sonic source has a high internal acoustic impedance and its output is almost linearly dependent on the aperture opening, it is shown to be very inefficient. The subsonic source can be much more efficient than the sonic one but the output is generally no longer a linear function of aperture opening. A simple method of linearizing such a source is discussed. A numerical comparison between compressed air sources and electrodynamic loudspeakers shows that the former offer a useful alternative for active noise control, especially when the secondary source has to act in an extreme environment. The subsonic compressed air source is particularly useful when efficiency is a major issue. Details and results of the experimental evaluation of a subsonic compressed air source built at the Laboratory of Acoustics of the Faculté Polytechnique de Mons (Belgium) are presented in companion paper (Part II).

Nanometer positioning of a linear motion stage under static loads

A standard lead-screw-driven linear translation table was fitted with a secondary voice coil actuator, which can apply forces of /spl plusmn/0.5 N on the translating stage. The DC servomotor-driven lead screw is capable of positioning the table to a nominal precision of /spl plusmn/0.5 /spl mu/m. The secondary actuator then positions the table to /spl plusmn/1 mm. Both actuators utilize closed-loop feedback algorithms. This configuration has significant advantages over the more common piggyback arrangement, which often uses piezoelectric actuators. The advantages include preservation of the original load surface and position sensors and an easier to control actuator. Its ability to operate under static load and a model of the microdynamic behavior are addressed in this paper. When static loads of up to 36 N were applied to the translating stage, the secondary actuator could still displace the table to /spl plusmn/500 nm and control the position down to its zero-load precision. The microdynamic behavior in this region is nonlinear and was modeled using the modified Dahl model of friction. The key parameters in the model were identified using a 2 Hz sinusoidal input force with varying amplitudes from the secondary actuator. For a constant loading force, the parameters did not vary significantly with different excitation forces. The friction parameter relating the initial slope of the friction force versus displacement curve did change with different loading forces. Also, for smaller input signals causing displacements of less than 200 nm, the friction parameters would tend to drift, depending on input magnitudes.

Microfabricated electrostatic actuators for hard disk drives

A dual-stage servo system which utilizes a high-bandwidth secondary actuator has been proposed for magnetic hard disk drives. Microfabricated actuators are promising candidates for this secondary actuator, since they offer the benefits of extremely small size and weight and may be batch fabricated for reduced production cost. This paper presents the design of an electrostatic microactuator which produces sufficient output force to move a 1.6-mg picoslider from a conventional disk drive at frequencies above 2 kHz. A linear second-order actuation model is developed and fit to experimental data. The prototype device is shown to have an open-loop resonance frequency of 532 Hz when loaded with the picoslider, and a DC-voltage-to-position gain of 0.12 /spl mu/m/V. A closed-loop bandwidth of 2.5 kHz is achieved using a simple PD controller.

DEFLECTION CHARACTERISTICS OF SPLIT ELECTRODE INPLANE BENDING PIEZOELECTRIC ACTUATORS WITH VARYING WIDTH

The cantilever split-morph, a new type of piezoelectric actuator, have been designed for dual-stage head positioning servo system of high density hard disk drives. In this actuator configuration, a PZT ceramic plate, with one end mechanically clamped and the other free, was poled along its thickness and divided into two symmetric parts along length by splitting electrodes. Driving voltages are preferably so arranged that they make one of the two parts expand by the “31”action while the other part, i.e., the symmetric part, contracts. This will result in a bending motion of the actuator along the direction of its width. The driving voltage-tip displacement characteristics of the actuator have been investigated by theoretical analysis, finite element modeling and experiment. Theoretical analysis indicated that the electric field induced stress is much smaller than stretch strength of piezoelectric ceramic. Tip displacement measured by Michelson interferometer is a little larger than the calculated result. This is mainly due to the increase of piezoelectric coefficient d31 at high driving field. Tip displacement of 1 to 2μm can be achieved at 20 to 50V driving voltage for the split-morph whose size is similar to that of the existing suspension. As to 25 kTPI (track per inch) high track density hard disk drives, this actuating stroke can cover at least one track width and meet the primary requirements of dual-stage head positioning servo system for displacement of the secondary actuator.

ACTIVE POWER MINIMIZATION AND POWER ABSORPTION IN A PLATE WITH FORCE AND MOMENT EXCITATION

This paper presents a comparison of the active control strategies of minimizing the total power supplied to a plate and maximizing the power absorbed by the secondary source. Force and moment excitations of infinite and finite plates are considered. For an infinite plate analytic solutions can be obtained for the total power supplied to the plate by the primary and secondary actuator arrays when using the two control strategies. Minimizing the power supplied by a primary force or moment with a secondary force or moment can produce large attenuation provided that the two sources are close together compared with a flexural wavelength. Minimizing the power supplied by a primary force with a secondary moment can also give attenuation of up to 5 dB when the spacing between the sources is about 0·3 times the flexural wavelength. In contrast to the acoustic case, for the infinite plate the total power supplied is generally reduced when the power absorbed by the secondary source is maximized. On a finite panel, however, maximizing the power absorbed by the secondary source can significantly increase the total power supplied. The strategy of minimizing the total power supplied can give considerably larger values of attenuation for finite rather than infinite plates, particularly at higher frequencies. This is because the structural modes can be actively controlled. A particularly efficient secondary actuator for total power minimisation appears to be an independently adjustable, but collocated, force and moment pair.