Patent Application
20090164145
This invention relates to testing fluid dynamic bearings in spindle motors for use in a hard disk drive.
Today, a typical hard disk drive includes a spindle motor rotating a spindle, and a hub to which one or more disks are attached, creating rotating disk surfaces upon which information is stored for access by the hard disk drive. For many years, most hard disk drives used ball bearing spindle motors. These motors are becoming obsolete due to imperfections in the roundness of the bearings transmitted through the metal-to-metal interface of the spindle shaft to adversely affect the writing of tracks on the rotating disk surfaces.
A new class of spindle motors known as fluid dynamic bearing motors are likely to replace the ball bearing motors, mainly due to a fundamental improvement in the spindle design using a different kind of bearing between its spindle shaft, and the motor. One side of the bearing couples to the spindle shaft, and the other couples to the motor. The two sides interface through a very thin pool of fluid lubricant, which removes the metal-to-metal interface, consequently minimizing the effect of manufacturing imperfections in either side of the bearing.
While the fluid dynamic bearing spindle motor is better than the ball bearing spindle motor, its use in hard disk drives is not without problems. One problem these motors have relates to how to make efficient, and accurate parts level mechanical performance estimates. Today, the central method of evaluating performance is in an assembly often including several other components of the hard disk drive, often the disk(s), and actuator assembly, itself including the head gimbal assemblies, and voice coil motor. These are very complex assemblies, which make isolating the dynamic response contributed by the fluid dynamic bearing of the spindle motor very difficult, time consuming, and expensive.
Test apparatus, and methods are needed which can provide efficient, accurate performance estimates of fluid dynamic bearings used in spindle motors for hard disk drives.
Embodiments of the invention include a test stand used to evaluate the dynamic performance of a Fluid Dynamic Bearing spindle motor includes a spindle motor coupled through a Fluid Dynamic Bearing (FDB) to a spindle shaft and to a hub. The hub includes an air target and a displacement plate. The test stand rigidly configures the spindle motor, at least one air nozzle, and at least one displacement sensor, so that an air jet from the air nozzle can strike the air target and the displacement plate engages the displacement sensor. The spindle motor is stimulated to rotate the hub. The air jet displaces the rotating hub as measured by the displacement sensor interacting with the displacement plate to create a displacement reading. The displacement readings are analyzed to create an estimate of the dynamics of the FDB. This provides a way to test FDB spindle motors inexpensively and accurately.
The air jet may strike the air target at a constant air pressure to create a radial displacement or eccentricity for estimating the stiffness of the FDB, and/or the air jet may strike the air target with a time varying air pressure to create time varying displacements measured by a succession of displacement readings to estimate the mechanical damping of the FDB.
This invention relates to testing Fluid Dynamic Bearing (FDB) spindle motors for use in a hard disk drive, in particular, a test stand for evaluating the dynamic performance of the FDB using at least one air nozzle, and a displacement sensor, and inventive methods to estimate the dynamics of the FDB as well as loading, and unloading FDB spindle motors from the test stand.
Referring to the drawings more particularly by reference numbers,
The displacement sensor 34 is preferably a non-contact sensor. The non-contact sensor may further be a capacitive probe sensor, and/or a laser displacement sensor. In certain situations, the laser displacement sensor may be preferred. In certain embodiments of the invention, in particular during calibration, use of more than one displacement sensor may be preferred. After calibration, certain embodiments may prefer just one displacement sensor.
A processor 70 stimulates 78 the spindle motor 2, rotating the spindle shaft 6, and the hub 8 via the FDB 4. An air source 60 provides the air nozzle 30 with the air jet 32, which strikes the air target 10, and displaces the hub 8 as measured by the displacement sensor 34 interacting with the displacement plate 12. The processor controls 72 the air source, and receives 76 at least one displacement reading 84 from the displacement sensor, from which a stiffness estimate 86, and/or a damping estimate 88 are created.
Providing the air jet 32 is often through an air hose 62 connecting the air source 60 to the air nozzle 30. The air source may include an air compressor, and/or a pressurized air reservoir, and/or a pressure throttle. The processor 70 may direct 72 the air source based upon at least one air source setting 80. These settings may cause the air jet to have an essentially constant air pressure, or one that varies over time. A time varying air pressure setting may cause the air jet to have a periodic air pressure.
The air jet 32 may strike the air target 10 at a constant air pressure that creates a radial displacement or eccentricity as a displacement reading 84. The displacement reading may be analyzed to create a stiffness estimate 86 of the FDB 4. The air source setting 80 may depend upon range of spindle stiffness to be tested.
The air jet 32 may strike the air target 10 with a time varying air pressure to create time varying hub displacements. These displacements of the hub 8 are measured by a succession of displacement readings 84. The readings may be analyzed to estimate the mechanical damping of the FDB 4, known herein as a damping estimate 88.
The displacement readings 84 may preferably be generated with a regular sampling rate. The sampling rate may preferably be approximately 256 samples per second, although many other sampling rates may be usable. In certain embodiments of the invention, the time varying air pressure may further be periodic. A resonant displacement in tune with the period of the air pressure may be used to estimate the mechanical damping of the FDB.
The processor 70 may include at least one instance 90 of a controller 92 as shown in
The processor 70 may preferably present 74 a motor speed setting 82 to a motor controller 64 to provide the stimulus 78 to the spindle motor 2, thereby controlling the rotational rate of the hub 8. The motor speed setting may preferably cause the spindle motor to rotate at approximately one of the following rotation rates: 7200 Revolutions Per Minute (RPM), 5400 RPM, 4800 RPM, or 3600 RPM. In other embodiments of the invention, the motor speed setting may cause the spindle motor to rotate at other rotational rates, which may be greater than 7200 RPM. The approximation by the spindle motor of the rotational rate may be within two percent, further within one percent, further within one half percent of the rotational rate. Being within a tolerance of the rotational rate may be calculated by the average root mean square of the variation from the rotational rate over a time interval or may be calculated a statistical measure, such as the standard deviation or variance.
The processor 70 may receive 76 the displacement reading 84 via an interface 66 communicating 68 with the displacement sensor 34, which is engaged with the displacement plate 12 of the hub 8 responding to the air jet 32 striking the air target 10, while the hub 8 is rotated by the spindle motor 2. The displacement sensor may include a stress-strain sensor, which may incorporate one or more piezoelectric devices.
Some of the following figures show flowcharts of at least one embodiment of the method, which may include arrows signifying a flow of control, and sometimes data, supporting various implementations of the method. These include a program operation, or program thread, executing upon the computer 94, and/or a state transition in the finite state machine. The operation of starting a flowchart refers entering a subroutine or a macro instruction sequence in the computer, and/or directing a state transition in the finite state machine, possibly while pushing a return state. The operation of termination in a flowchart refers completion of those operations, which may result in a subroutine return in the computer, and/or popping of a previously stored state in the finite state machine. The operation of terminating a flowchart is denoted by an oval with the word “Exit” in it.
One model that may be preferred assumes that the damping is linear so that the dynamics of the FDB 4 may be approximated by the following equation:
{right arrow over (F)}={right arrow over (F)}0+K{right arrow over (e)}+C{right arrow over (e)}1 (1)
Where {right arrow over (F)} is the applied force that is air pressure times the effective hub area of the air jet 32 striking the air target 10 of the hub 8. {right arrow over (F0)} is a bias force that is normally zero. K is the static stiffness, and is preferably measured as force over distance, or Newtons per meter, and may preferably be a version of the stiffness estimate 86. {right arrow over (e)} is the displacement produced by the air jet striking the air target as measured by the interaction between the displacement plate 12, and the displacement sensor 34, and reported as the displacement reading 84. C is the damping coefficient, and is preferably measured as force over velocity, or Newtons*seconds per meter, and may preferably be a version of the damping estimate 88. {right arrow over (e′)} is the rate of change in time calculated from a succession of the displacement readings. {right arrow over (e′)} may be seen as the velocity of the hub 8 toward the displacement sensor 34 from the center of the spindle shaft 6.
To estimate the stiffness estimate 86, K, the air pressure may preferably be kept constant, causing the rotating hub 8 to shift, and maintain a particular offset, making {right arrow over (e′)}=0.
Preferably, once K has been estimated, applying a linearly increasing but sufficiently slowly varying air pressure through the air jets 32 can be used to estimate {right arrow over (e′)}. The damping estimate 88, C may be analyzed from the estimate of {right arrow over (e′)} using the above equation. The analysis may use a least squares best fit approach. It may be preferred to slowly vary the air pressure so that the above equation is applicable, as K is often found to be a static stiffness.
Loading the FDB spindle motor onto the test stand 50 may preferably be performed in one of the following ways. The assembly 36 with the air nozzles 30, and displacement sensor 34 may be removed from the base 20. The FDB spindle motor may be mounted so that the spindle motor 2 is secured to the base. The assembly may be attached as shown in
Unloading the test stand 50 of the tested FDB spindle motor 2 may preferably operate by essentially reversing these steps: the displacement sensor 34 is disengaged from the displacement plate 12. The assembly 36 detached from the base 20., and the FDB spindle motor detached from the base for its removal.
In certain refinements of the invention, ambient conditions, for instance the ambient temperature, atmospheric humidity, and possibly the ambient air pressure about the FDB spindle motor 2 may measured, and/or controlled. The FDB spindle motor, and the relevant components of the test stand 50 may be enclosed in an environmental chamber.
The preceding embodiments provide examples of the invention, and are not meant to constrain the scope of the following claims.
