Detecting back electromotive force voltage

Abstract

A Bemf detect circuit is configured for modeling Bemf error via calculation of time correlations associated with two or more different non-calibrated sense resistor gain values to derive a calibrated sense resistor gain value. A method is provided for calibrating a Bemf detector circuit by setting a sense resistor gain to a plurality of different gain values, observing Bemf sample values for each of the gain values, and deriving a calibrated gain value by non-iteratively modeling the relationship between the gain values and the associated Bemf sample values.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is an isometric view of a data storage device constructed in accordance with embodiments of the present invention.

FIG. 2 is a simplified functional diagram of the data storage device of FIG. 1.

FIG. 3 is a schematic representation of control circuitry used to control operation of the voice coil motor (VCM) of the data storage device of FIG. 1.

FIG. 4 diagrammatically illustrates stacked waveforms showing timing sequences of command current, sense resistor gain, and Bemf sampling strobes performed by the apparatus and method of the present embodiments.

FIGS. 5 and 6 diagrammatically compares the effect of the command current of FIG. 4 on V_tech and V_tech error in the presence of an actuator constant velocity disturbance and an actuator constant acceleration disturbance, respectively.

FIG. 7 is a model of expected V_tech error over a range of sense resistor gains that is constructed in accordance with embodiments of the present invention.

FIG. 8 is a model of expected V_tech error over a range of sense resistor gains that is constructed in accordance with alternative embodiments of the present invention.

FIG. 9 is a model of expected V_tech error over a range of sense resistor gains that is constructed in accordance with alternative embodiments of the present invention.

FIG. 10 is a flowchart of illustrative steps in a method for CALIBRATING K_sense in accordance with embodiments of the present invention.

DETAILED DESCRIPTION

Referring to the drawings in general, and more particularly to FIG. 1 that shows an isometric view of a data storage device 100 constructed in accordance with embodiments of the present invention. The device 100 is provided to show an exemplary environment in which preferred embodiments of the present invention can be advantageously practiced. It will be understood, however, that the embodiments of the present invention are not so limited.

The device 100 preferably includes a base 102 and a cover 104 (partially cutaway), which together provide a housing for a number of components. The components include a motor 106 to which is fixed one or more storage mediums 108 in rotation therewith. Adjacent the storage medium 108 is an actuator assembly 112 that pivots around a bearing assembly 114 through application of current to a voice coil motor (VCM) 122. In this way, controlled operation of the VCM 122 causes the actuator 112 to move radially across the storage medium 108.

The actuator assembly 112 includes an actuator arm 116 supporting a load arm 118 that, in turn, supports a head 120 (or “transducer”) at a distal end thereof in a data transfer relationship with the adjacent storage medium 108. Each storage medium 108 can be divided into data tracks, and the head 120 is positioned to retrieve data from and store data to the tracks.

Preferably, when the device is not in use, the heads 120 are moved (unloaded) to a ramp structure 124 located near an outermost periphery of the storage medium 108. The ramp structure 124 serves to provide respective surfaces on which the heads 120 can safely be disposed while the storage medium 108 is in a non-rotational state. When device I/O operation is desired, the motor 106 accelerates the storage medium 108 to a velocity sufficient to support the heads 120, and the heads 120 are moved (loaded) from the ramp structure 124 to supported positions adjacent the media 108.

While a ramp structure 124 has been provided for purposes of illustrating a preferred embodiment, such is not necessarily required. For example, embodiments that employ rotatable media and moveable heads as shown in FIG. 1 can alternatively utilize other support configurations such as a contact start stop (CSS) approach whereby the heads 120 are brought to rest upon texturized landing zones of the storage medium 108.

To provide the requisite electrical conduction paths between the head 120 and device 100 control circuitry, the head 120 advantageously has a flex circuit that is routed on the actuator assembly 112 from the head 120, along the load arm assembly 118 and the actuator arm 116, and to a circuit portion 133 that is supported by a proximal end (sometimes referred to as “E block”) of the actuator assembly 112. The circuit portion 133 connects the head 120 flex circuit to another flex circuit 134 which passes through the base 102 to a printed circuit board (PCB) 138. An electrical connector 140 attached to the PCB 138 has a plurality of contacts 142 for connecting the device 100 to a mating connector (not shown), such as for placing the device 100 in communication with external control circuitry.

FIG. 2 provides a functional block diagram of the device 100 of FIG. 1. A programmable controller 148 provides top-level communications and control for the device 100. An interface (I/F) 150 facilitates I/O communications and transfers with a host device 152 with which the device 100 can be associated. A read/write (R/W) channel 154 conditions data to be written to the storage medium 108 from the host 152 during a write operation, and reconstructs data retrieved from the storage medium 108 for transfer to the host 152 during a retrieve operation.

A servo circuit 156 provides closed-loop positional control for the heads 120, such as by using servo data stored to the storage medium 108. The servo circuit 156 is preferably configured to carry out a number of control functions including track following operations whereby a selected head 120 follows a corresponding track on the storage medium 108, seek operations whereby a selected head 120 is moved from an initial track to a destination track, and head 120 load/unload operations whereby the heads 120 are moved away from or to the ramp structure 124.

Continuous velocity control of the actuator 112 can be accomplished using the servo data at times when the head 120 is adjacent the storage medium 108. However, when the head 120 is moved away from the storage medium 108, such as when the head 120 is moved on the ramp 124, the Bemf voltage of the VCM 122 can be used for continuous velocity control. It is the latter situation to which embodiments of the present invention are directed. However, it will also be noted that it can be advantageous to use Bemf control at times when the head 120 is adjacent the storage medium 108. For example, an error condition may occur preventing the data storage device 100 from successfully reading the servo information. Having both Bemf and servo control options available regardless of the head 120 position is preferable.

The servo circuit 156 applies the requisite currents to a coil 160 of the VCM 122 to carry out these various operations. The servo circuit 156 further preferably includes a spindle motor driver circuit 162 to controllably operate the spindle motor 106. Processing requirements to carry out these and other servo functions are preferably supplied by a DSP 164 or other processing device of the servo circuit 156, such as an ARM. The servo processing can alternatively be supplied by the top level controller 148 in a single processor environment.

FIG. 3 provides a generalized schematic representation of relevant portions of the servo circuit 156 of FIG. 2. A VCM control circuit 166 preferably utilizes an h-bridge driver circuit 168 to supply bi-directional currents through the VCM coil 160. Switching devices S1-S4 are numerically denoted at 170, 172, 174 and 176 and preferably comprise suitable field effect transistors (FETs). The switching devices S1-S4 are arranged between a V_C voltage source 178 and a reference line (ground) 180.

In this way, current can be passed through the coil 160 in a first direction through the selective energizing (induced source-drain conduction) of switches S1 and S4, and current can be passed through the coil 160 in a second opposite direction through the selective energizing of switches S2 and S3.

A current sense resistor 182 is preferably placed in series with the coil 160 as shown. The respective steady-state resistances of the coil 160 (R_M) and the sense resistor 182 (R_S) are both preferably on the order of a few ohms. The difference between lines 186 and 188 represents the overall voltage drop across the coil 160 and the sense resistor 182; the difference between lines 186 and 190 represents the voltage drop of the sense resistor 182; and the difference between lines 188 and 190 represents the voltage drop across the coil 160.

During I/O operation of the device 100, the servo circuit 156 operates to position the heads 120 either adjacent the rotating storage medium 108 or adjacent the load/unload ramp 124. The VCM control circuit 166 is commanded by the controller 148 to provide a specified current I_cmd (magnitude and direction) through the coil 160. At least in certain operational modes this current can be monitored using the relation I_M=V_S/R_S, where I_M is the actual current through both coil 160 and sense resistor 182, V_S is the voltage obtained from paths 186 and 190, and R_S is the known (or measured) resistance of the sense resistor 182.

The VCM control circuit 166, in turn, has circuitry that continuously provides a voltage V_tach that is proportional to the Bemf of the VCM 122. This information can be used by the controller 148 in ascertaining the velocity of the actuator 112 at times when the servo information is unavailable, such as when the heads 120 are being moved on the ramp structure 124. The circuitry must effectively cancel the voltage drop associated with the coil 160 and the sense resistor 182 so that a calculated V_tach error is entirely the result of Bemf error.

The cancellation circuitry in some embodiments as illustrated in FIG. 3 connects line 186 to inverting sides of operational amplifiers K_BEMF 200 and K_sense 202. Line 188 is connected to the non-inverting side of K_BEMF 200, and line 190 is connected to the non-inverting side of K_sense 202. K_sense is calibrated such that:

I _M(R_M+R_S)K_BEMF=I_MR_SK_sense   (1)

When K_sense is thus properly calibrated there is no error voltage imparted to V_tach by the voltage drop across the VCM 122 and sense resistor 182. That is, with a correctly calibrated K_sense no time correlation exists between a commanded I_cmd and the resulting V_tach. In this case, the following relationship exists:

$\begin{array}{cc}{V}_{\mathrm{tach}}=\mathrm{Bemf}\left(K_{\mathrm{BEMP}}\right)+L_M\left(\frac{I_M}{t}\right)K_{\mathrm{BEMF}}& \left(2\right)\end{array}$

Under steady state conditions this relationship is simplified to:

V _tach =Bemf(K_BEMF)   (3)

According to this relationship, the degree of V_tach error is directly related to the degree to which K_sense is incorrectly calibrated. The present embodiments model this expectation of Bemf error by employing a real time error calculation of time correlations associated with two or more different non-calibrated K_sense values, in order to derive the calibrated K_sense value. For example, observed error resulting from a known condition wherein K_sense is incorrectly calibrated can be defined by obtaining a Bemf sample during a bipolar command current cycle:

$\begin{array}{cc}{V}_{\mathrm{tach}}\mathrm{error}=\stackrel{t_n}{\sum _{t_0}}I_{\mathrm{cmd}}V_{\mathrm{tach}}& \left(4\right)\end{array}$

By performing this calculation for two or more different selected K_sense values, the observed Bemf samples can be used to construct the model of the V_tach error values over a range of non-calibrated K_sense values. Curve-fitting techniques can then be applied to the model to non-iteratively select the calibrated K_sense value associated with a V_tach error value of zero.

For example, FIG. 4 has a number of stacked waveforms illustrating a method of calibrating K_sense in accordance with embodiments of the present invention. The top waveform is a bipolar current representing I_cmd. Four current cycles are shown, the first current cycle beginning at time t₀ and ending at t₁, the other three current cycles beginning at the end of the previous current cycle and ending, respectively, at t₂, t₃, and t₄.

It will be noted in the illustrative embodiments of FIG. 4 for I_cmd that at time t₃ an ending of one bipolar current cycle is of the same polarity (in this case negative) as a beginning of the next bipolar current cycle (“dual bipolar cycles”). FIG. 5 is simulation data showing responses Of V_tach (equation 3) and V_tach error (equation 4) in relation to I_cmd and in the presence of an actuator constant velocity disturbance (or DC offset) 196. FIG. 6 is similar data but demonstrating responses in the presence of an actuator constant acceleration disturbance 198. In both cases it will be noted that by making I_cmd both bipolar and of dual opposing bipolar cycles, V_tach error (equation 4) converges to zero within the interval from t2 to t4, just two bipolar cycles of I_cmd. This insensitivity to constant velocity and constant acceleration disturbances demonstrates the robustness of the calibration procedure of the present embodiments.

Returning to FIG. 4, below the I_cmd waveform is a K_sense waveform which is set at a significantly higher than nominal value, K₁, during the first current cycle of I_cmd, and which is then set at a significantly lower than nominal value, K₂, during the second current cycle. K_sense is then set to a target (calibrated) value, K_T, during the third and fourth current cycles. The value of K_T results from the calibration operations performed during the first and second current cycles.

Below the K_sense waveform are three strobe window waveforms that illustrate the intervals over which each Bemf sample is obtained. That is, Bemf sample, is obtained by summing instantaneous values of a first strobe window (within the positive polarity portion of I_cmd) with instantaneous values of a second strobe window (within the negative polarity portion of I_cmd) during the first interval while K_sense is set to K₁. Similarly, Bemf sample₂ is obtained during the second interval while K_sense is set to K₂.

FIG. 7 graphically shows illustrative embodiments whereby a curve fitting technique is used on two observed Bemf samples and corresponding K_sense values to construct a mathematical model 206 of the expected values therebetween and therebeyond. This model is used to derive the calibrated value, K_T, associated with a Bemf sample value of zero. It will be noted that the model 206 employs linear interpolation between the data pairs. However, FIG. 8 illustrates equivalent alternative embodiments employing a non-linear interpolation technique to construct the model 208. Furthermore, FIG. 8 illustrates other equivalent alternative embodiments whereby a curve fitting technique is used to extrapolate beyond the data pairs (either linearly or nonlinearly) to construct the model 210.

Returning to FIG. 4, as mentioned above K_sense is set to the newly derived calibrated value, K_T, during the third and fourth current cycles of I_cmd. The Bemf sample₃ is obtained during these cycles and compared to a predetermined threshold value in order to ensure that a robust calibration has been achieved. If Bemf sample₃ exceeds the threshold, then the calibration procedure is repeated until the threshold is satisfied.

FIG. 10 is a flowchart of illustrative steps for performing a CALIBRATING K_sense method 250 in accordance with embodiments of the present invention. The method begins by the controller 148 commanding a bipolar current I_cmd in block 252. Control then passes to block 254 where the VCM control circuit 166 sets K_sense to the first gain K₁ value (see FIG. 4). The Bemf sample₁ is determined during the first current cycle in block 256. Control then passes to block 258 where the VCM control circuit 166 sets K_sense to the second gain K₂. The Bemf sample₂ is determined during the second current cycle in block 260.

In block 262 the gain values and Bemf samples are used to construct a model of the expected Bemf error over a range of non-calibrated K_sense values. Based on the model, block 264 derives the calibrated sense resistor gain value, K_T, from the Bemf error value of zero.

In block 266 the VCM control circuit 166 sets K_sense to K_T, and then in block 268 the Bemf sample₃ is determined. The Bemf sample₃ is compared to a preselected threshold in decision block 270. If the determination of block 270 is yes, then the method ends. Otherwise, control passes to block 272 where the values for I_cmd, K₁, and K₂ can be adjusted as desired and then the calibration procedure is repeated beginning at block 252.

Finally, in some embodiments the circuit and associated method described above reside in the data storage device 100 as illustrated in FIG. 1 to provide continuous velocity control of the actuator 112. Accordingly, the data storage device is characterized by a servo control system 156 having a Bemf detector circuit 166 and means for calibrating the Bemf detector circuit. For purposes of this detailed description and the appended claims, the meaning of “means for calibrating” expressly contemplates embodiments whereby a real time error calculation of the time correlations associated with Bemf error from a non-calibrated K_sense condition is performed to non-iteratively derive the calibrated K_sense value. The term “means for calibrating” expressly does not contemplate previously attempted solutions requiring an iterative convergence analysis, or that in some way restricts the movement of the actuator during calibration in order to force the Bemf to a zero value.

It is to be understood that even though numerous characteristics and advantages of various embodiments of the present invention have been set forth in the foregoing description, together with details of the structure and function of various embodiments of the invention, this detailed description is illustrative only, and changes may be made in detail, especially in matters of structure and arrangements of parts within the principles of the present invention to the full extent indicated by the broad general meaning of the terms in which the appended claims are expressed. For example, the particular elements may vary depending on the devices in which they are incorporated or the particular environment in which they are used without departing from the spirit and scope of the present invention.

In addition, although the illustrative embodiments described herein are directed to a data storage system, it will be appreciated by those skilled in the art that the claimed subject matter is not so limited and various other electronic devices can utilize the embodiments of the present invention without departing from the spirit and scope of the claimed invention.

Claims

1. An apparatus comprising a Bemf detect circuit configured for modeling Bemf error via observed time correlations associated with setting two or more different non-calibrated gain values on a voltage drop across a sense resistor, the non-calibrated gain values set independently of the sense resistor value and set during a different current cycle of the same command current, to derive a calibrated gain value.

2. The apparatus of claim 1 wherein the circuit is configured to curve-fit two or more Bemf sample values in modeling the Bemf error.

3. The apparatus of claim 2 wherein the circuit is configured to interpolate between first and second Bemf sample values to derive the calibrated sense resistor gain value.

4. The apparatus of claim 2 wherein the circuit is configured to extrapolate from first and second Bemf sample values to derive the calibrated sense resistor gain value.

5. The apparatus of claim 2 wherein the circuit is configured to linearly model the relationship from two Bemf sample values.

6. The apparatus of claim 2 wherein the circuit is configured to non-linearly model the relationship from more than two Bemf sample values.

7. The apparatus of claim 2 wherein the circuit is configured to evaluate each Bemf sample value during a bipolar current cycle.

8. The apparatus of claim 7 wherein the circuit is configured to evaluate first and second Bemf sample values during adjacent bipolar current cycles, wherein a trailing end of one of the adjacent bipolar current cycles is of the same polarity as a leading end of the other of the adjacent bipolar current cycles.

9. The apparatus of claim 1 wherein the circuit is configured to verify the calibrated sense resistor gain value by comparing a Bemf sample value associated with the calibrated sense resistor gain value to a predetermined threshold.

10. A method for calibrating a Bemf detector circuit comprising: setting a gain value on a voltage drop across a sense resistor to each of a plurality of different gain values, each gain value set independently of the sense resistor value and set during respective different current cycles of the same command current;observing Bemf sample values for each of the gain values; andderiving a calibrated gain value by non-iteratively modeling the relationship between the gain values and the associated Bemf sample values.

11. The method of claim 10 wherein the setting step is characterized by setting the sense resistor gain to a plurality of non-calibrated gain values.

12. The method of claim 11 wherein the observing step is characterized by summing a plurality of instantaneous Bemf readings during a bipolar current cycle.

13. The method of claim 12 wherein the observing step is characterized by summing first and second Bemf readings during adjacent bipolar current cycles, wherein a trailing end of one of the adjacent bipolar current cycles is of the same polarity as a leading end of the other of the adjacent bipolar current cycles.

14. The method of claim 12 wherein the deriving step is characterized by curve fitting two or more Bemf sample values.

15. The method of claim 14 wherein the deriving step is characterized by interpolating between first and second Bemf sample values.

16. The method of claim 14 wherein the deriving step is characterized by extrapolating from first and second Bemf sample values.

17. The apparatus of claim 10 wherein the deriving step is characterized by linearly modeling the relationship between the gain values and the associated Bemf sample values.

18. The method of claim 10 wherein the deriving step is characterized by non-linearly modeling the relationship between the gain values and the associated Bemf sample values.

19. The method of claim 10 wherein the deriving step is characterized by verifying the calibrated sense resistor gain value by comparing a Bemf sample value associated with the calibrated sense resistor gain value to a predetermined threshold.

20. A data storage device comprising: a servo control system having a Bemf detector for velocity control of an actuator; andmeans for calibrating the Bemf detector during a non-zero velocity condition of the actuator.