WO1999031956A2

WO1999031956A2 - Automated servo gain adjustment using fourier transform

Info

Publication number: WO1999031956A2
Application number: PCT/US1998/026828
Authority: WO
Inventors: Leonardus J. Grassens; David L. Schell
Original assignee: Lsi Logic Corporation
Priority date: 1997-12-19
Filing date: 1998-12-16
Publication date: 1999-07-01
Also published as: WO1999031956A3

Abstract

Servo loop gain is automatically adjusted for control loops of precision tracking and focusing servos in an optical or magneto optical storage device (114). To adjust servo loop gain during operation, the servo loop (100) is modelled as a series of simple gain components (KPA, KACT, KOPT, KPRE A, KADC, KC, KDAC), with the total servo loop gain being the product of the gain components. A controller gain components (KC) is adjusted based on measured servo loop gain to keep the total servo loop gain constant. To measure the servo loop gain during operation, a sinusoidal signal (104) is injected into the servo loop (100) by addition to the normal servo control signal for a period of n cycles of the sinusoidal signal. The servo loop (100) response is measured for the last n-1 cycles of this injection period, with the Fourier Transform excitation frequency component of the measured response being computed to determine the response amplitude. The servo loop gain may then be computed from measured response amplitude and the known amplitude of the injected signal (104). If the computed gain value is within a range of expected values, the controller gain (KC) is adjusted to keep the total servo loop gain constant. Otherwise, a default gain value is employed to adjust the servo loop gain. In an alternative embodiment, the servo loop gain excluding the controller gain component (KC) is determined from two measurements of the system response to the injected sine wave (104).

Description

AUTOMATED SERVO GAIN ADJUSTMENT USING FOURIER TRANSFORM

1. Technical Field

The present invention relates generally to precision servo motors for optical or magneto optical devices and in particular to gain compensation for servo loops. Still more particularly, the present invention relates to compensating for gain variations in the server controller loop of a precision servo motor.

2. Description of the Related Art

Optical and magneto optical storage devices retrieve and/or store data by employing focused beams of electromagnetic radiation, typically at an optical frequency/wavelength, and measuring characteristics of the beam's reflection such as intensity or polarization. This requires the use of precision focusing servo motors to position the read/write head at a distance from the storage media surface corresponding to the focal length of the beam.

Information is stored on optical storage media in one or more tracks, which may form, for example, either a spiral from the center to the periphery of the storage media surface or a number of concentric circles on the storage media surface. Precision tracking servo motors are required to position the read/write head over the appropriate track or track portion of the storage media surface.

Both focusing and tracking servos in optical storage devices are controlled by actuators which receive signals generated by a controller based on error detected from beam reflection measurements. Typically a closed- loop control system, known as a servo loop, is employed to provide the necessary control signals. The servo loop includes a gain, or an amplitude ratio of transmitted signal to feedback signal, which must be controlled to accurately position the read/write head utilizing the focusing and tracking servo motors. These aspects of optical storage device implementation are well known in the art. Servo loops are subject to gain variations both from unit to unit for a given implementation and within a specific unit. Servo loop gain variances from unit to unit result from component tolerances and include pre- and post- amplifier gain variations caused by tolerances on resistors, variations of optical properties of the optical system such as wavefront quality and coupling efficiency, variations of actuator armature resistance, and variations of electromagnetic fields causing motor constant variation. Servo loop gain variances within a specific unit result from operating environment changes and include variations caused by loading a new storage disk, where disk characteristics such a reflectivity, pit depth, and pit width may vary from disk to disk and affect focus and tracking sensitivities, variations caused by changes in temperature, and variations caused by aging factors such as laser degradation or dust accumulation .

Servo loop gain variations are undesirable and may result in loop gain which is either too high or too low. These may lead to poor control and, in some cases, may cause servo loops to become unstable and cause oscillation. Adjustment of servo loop gain during manufacturing is insufficient to avoid gain loop variations during the lifetime of a device. Prior attempts to provide automatic, real time servo loop gain control have generally required a reference signal with a predetermined frequency central to the frequency band of the closed servo loop, together with a band pass filter and peak detection.

It would be desirable, therefore, to provide automatic, real time servo loop gain control with improved accuracy and the capability to employ any arbitrary frequency. It would further be advantageous for the gain control mechanism to employ small signal amplitudes with less system disturbance. It would further be desirable for the gain control mechanism to employ the Fourier Transform to reduce noise and improve accuracy. 3. Summary of the Invention

Servo loop gain is automatically adjusted for control loops of precision tracking and focusing servos in an optical or magneto optical storage device. To adjust servo loop gain during operation, the servo loop is modelled as a series of simple gain components, with the total servo loop gain being the product of the gain components. A controller gain components is adjusted based on measured servo loop gain to keep the total servo loop gain constant . To measure the servo loop gain during operation, a sinusoidal signal is injected into the servo loop by addition to the normal servo control signal for a period of n cycles of the sinusoidal signal. The servo loop response is measured for the last n- 1 cycles of this injection period, with the Fourier Transform excitation frequency component of the measured response being computed to determine the response amplitude. The servo loop gain may then be computed from measured response amplitude and the known amplitude of the injected signal. If the computed gain value is within a range of expected values, the controller gain is adjusted to keep the total servo loop gain constant. Otherwise, a default gain value is employed to adjust the servo loop gain. In an alternative embodiment, the servo loop gain excluding the controller gain component is determined from two measure- ments of the system response to the injected sine wave.

4. Brief Description of the Drawings

The novel features believed characteristic of the invention are set forth in the appended claims . The invention itself, however, as well as a preferred mode of use, further objectives and advantages thereof, will best be understood by reference to the following detailed description of an illustrative embodiment when read in conjunction with the accompanying drawings, wherein: Figure 1 depicts a block diagram of a servo loop with automatic gain adjustment in accordance with a preferred embodiment of the present invention; Figure 2 is a block diagram of a gain model employed in computing servo loop gain in accordance with a preferred embodiment of the present invention;

Figures 3A-3B depicts a simulation example of utilizing the Fourier Transform to filter noise within an automatic gain control scheme in accordance with a preferred embodiment of the present invention;

Figure 4 is a high level flowchart for a process of adjusting gain in accordance with a preferred embodiment of the present invention;

Figures 5A-5B depict examples of injected sine wave and measured response samples for automatic gain adjustment in accordance with a preferred embodiment of the present invention; Figures 6A-6B are an example of computing the Fourier

Transform excitation frequency on response samples for automatic gain control in accordance with a preferred embodiment of the present invention;

Figures 7A-7B depict gain computations for a gain adjustment process in accordance with the preferred embodiment of the present invention; and

Figure 8 depicts a block diagram of an alternative embodiment of a servo loop with automatic gain adjustment in accordance with a preferred embodiment of the present invention.

5. Detailed Description

With reference now to the figures, and in particular with reference to Figure 1, a block diagram of a servo loop with automatic gain adjustment in accordance with a preferred embodiment of the present invention is depicted. Controller 102 performs necessary computations utilizing tracking and focusing error signals to control all actuators. The output signal of controller 102 is combined with a digitized sine wave input from signal generator 104. The digitized sine wave from signal generator 104 has a known amplitude 106. The combined output of controller 102 and signal generator 104 is converted to an analog signal by digital-to-analog (D/A) converter 108, the output of which is amplified by power amplifier 110 and then forwarded to actuator 112. Although only one actuator 112 is depicted in the exemplary embodiment, those skilled in the art will recognize that separate actuators for tracking and focusing may be desirable, with separate signal paths. Actuator 112 moves the read/write head (not shown) of the optical storage device including servo loop 100. The motion results in a particular alignment of the optical mechanics with respect to the storage disk in the opto mechanics/disk mechanism 114. The alignment of interest is vertical (height above the surface of the disk) for focusing and lateral (distance from the disk center) for tracking. Opto mechanics and disk mechanism 114 includes a focusing error detector (not shown) and/or a tracking error detector (not shown) , such that a measurement of radiation reflected from the disc surface provides a basis for determining focusing and/or tracking error. A focusing/tracking error signal is then passed by opto mechanics/disk mechanism 114 to an optional pre amplifier 116, and then converted to a digital signal by analog-to-digital (A/D) converter 118. The digital focusing/tracking error signal is received by controller 102, which employs the signal in computing subsequent control signals for actuator 112. Controller 102 includes digital signal processing capabilities, including the capability for computing a Fourier Transform. Servo loop 100 includes a gain which may be determined from an amplitude ratio of the measured signal A 120 with the known amplitude B 106 of the injected sine wave. To counteract the effect of gain variations, this ratio is computed by controller 102 and utilized to adjust the loop gain accordingly. Thus, in the present invention, while all servo loops are closed, a small sinusoidal signal or arbitrary frequency is injected in the system, the system response is measured, and the amplitude ratio or gain is computed. The signal injection and gain computations employ the Fourier Transform as described in greater detail below. Referring to Figure 2, a block diagram of a gain model employed in computing servo loop gain in accordance with a preferred embodiment of the present invention is illustrated. The servo loop is modelled as a number of gains corresponding to different components of the servo loop, and divided into a measured gain and an adaptive gain. The measured gain K_HEASURE includes the gain corresponding to the power amplifier, K_PA, the gain corresponding to the actuator, K_ACT, the gain corresponding to the opto mechanics and disk assembly, K_0PT, and the gain corresponding to the optional pre-amplifier , if any, K_PrβA. The measured gain is the product of these gain components:

^ EASURE ⁼ K_PA"K_ACT"K_0PT-Kp_{rβ A} . The components have a more complicated transfer function than a simple gain. The actuator, for instance, has a double integrator term within the transfer function. For the purposes of automatically adjusting the gain in real time, however, modelling the components as a series of simple gains is sufficient.

The adaptive gain, K-_XDXBT I includes the gain corresponding to the A/D converter, K_ωc, the gain corresponding to the controller, K_c, and the gain corresponding to the D/A converter, K_DAC, and is equal to the product of these components:

^ADAP_T ⁼ K_ADC-K_C"K_DAC .

The adaptive gain may correspond to a controller chip and may be adjusted by altering the controller gain component, K_c. The goal in adjusting the servo loop gain make the gain of the entire loop (i.e., the product of all terms) constant, regardless of any changes in any of the individual gain components. After the measured gain K_KSΑSm∑ is computed, the controller gain K_c is adjusted such that the product ■^■ADAPT"^MEASURE -^s constant . The measured gain K_UEASURE may be computed from the known amplitude of the injected sine and the measured amplitude of the response. The amplitude of the response is computed from the Discrete Fourier Transform (DFT) :

w) = x { n ) e ^'jwn= x ( n) (cos ( wn ) -jsin ( wn

where w is the normalized frequency in the range from 0 to 2π. For a periodic function, after proper scaling, the Discrete Fourier Transform simplifies to:

N-l 2πnk N- l

■ ∑ x { n) e N n cos ¹ 2πn* ^•jsin n=0 N

The Fourier Transform is capable of filtering out noise. Without any averaging and using a 20 point sinusoidal signal, the Fourier Transform computes the amplitude with only 6.6% error.

With reference now to Figures 3A and 3B, a simulation example of utilizing the Fourier Transform to filter noise within an automatic gain control scheme in accordance with a preferred embodiment of the present invention is depicted. Figure 3A depicts a table of 20 samples for a sinusoidal signal with a maximum amplitude (horizontal axis) of 1 as a function of time (vertical axis) , synthesized with noise added. Figure 3B depicts a Fourier Transform of the samples depicted in Figure 3A, with the real and imaginary Fourier Transform values (horizontal axis) of the excitation frequency displayed as a function of (vertical axis) . The X(0) value in Figure 3B contains the direct current (DC) offset of the signal sample, while X(l) value contains the amplitude of the excitation frequency (i.e., the amplitude of the original sine wave) . The phase is approximately -90°, indicating that the sample is a sine wave with phase almost equal to 0°. The effect of aliasing is visible where X(19) = -X(l) . The numerical algorithm for the full Fourier Transform of the signal sample is as follows:

for (k=0; k<N; k++)

{ X[k] .re = 0;

X[k] .im = 0; for (n=0; n<N; n++)

{

X[k] .re+= x[n] *cos (k*n*2*pi/N) ; X[k] .re+= -x [n] *sin (k*n*2*pi/N) ;

}

X[k] .re /= N/2;

X[k] .im /= N/2;

} where x = {x [0] , x [1] , . . . x[N-l]} are the signal samples as a function of time and X = {X [0] , X [1] , . . . X[N-1]} is the complex Fourier transform X[i] = X[i].re + j*X[i] .im.

Referring to Figure 4, a high level flowchart for a process of adjusting gain in accordance with a preferred embodiment of the present invention is illustrated. The process begins at step 402, which depicts injecting a sine wave into the closed servo loop for n cycles of the sine wave. Sinusoidal signal samples are read from a table and added to the value sent to the output D/A converter from the controller for the normal servo signal. The signal addition is repeated for n periods of the sine wave, where the length in number of cycles may vary.

At the same time as samples are being injected, the response of the servo loop is measured, as illustrated at step 404. For the first period after the injection of the sine wave begins, the servo loop response will include the transient response. Since only the steady state response is of interest, the first cycle is ignored. That is, samples are injected for n cycles, but measurements are only taken for the last n-l cycles. With reference to Figures 5A and 5B, examples of injected sine wave and measured response samples for automatic gain adjustment in accordance with a preferred embodiment of the present invention are depicted. Figure 5A depicts the injected signal (bottom trace) and measured tracking error signal (top trace) for 4 cycles of the injected sine wave, while Figure 5B depicts the injected signal (bottom trace) and measured tracking error signal

(top trace) for 10 cycles. The injected signal is a 1250 Hz sine wave, with 20 samples being employed and one sample injected every other servo clock cycle where the servo clock period is 20 μsec .

Referring again to Figure 4, since the measurements are periodic with the same frequency as the injected samples, effective and simple averaging may be obtained by repeated addition of the measurement samples. Thus, for twenty samples of a sine wave injected for n cycles, the corresponding response samples are taken as : n x [ i ] =∑ x [ i +20j ] , i=0, 1, 2, ... 20

At the end of injection and sampling, the process passes to step 406, which depicts computing the Fourier Transform of the response samples.

Referring to Figures 6A and 6B, an example of computing the Fourier Transform on response samples for automatic gain control in accordance with a preferred embodiment of the present invention is illustrated. The excitation signal samples are for a sine wave of 1250 Hz with 20 samples per cycle. The response samples, shown in Figure 6A, are measured and averaged for injection of the signal for 10 cycles. The Fourier Transform of the response samples, shown in Figure 6B, is then computed. In actual implementation, only the X[l] (excitation frequency) components of the Fourier Transform need be computed.

Referring again to Fi-gure 4, only the excitation frequency (X[l]) component of the Fourier Transform is computed, so that the algorithm described above simplifies to:

Re = 0; Im = 0; for (n=0; n<N; n++) ;

{

Re += x [n] *cos (k*n*2*pi/N) ;

Im += -x[n] *sin(k*n*2*pi/N) ;

} Re /= N/2;

Im /= N/2; Amplitude = sqrt ( Re*Re + Im*Im) ;

At the completion of sampling, the excitation frequency components are computed using such an algorithm.

Once the excitation frequency components of the Fourier Transform is computed for the measured response samples, the process passes to step 408, which illustrates checking that the servo loops are still closed. This helps to validate the measurements. If the servo loops are still closed, the process passes to step 410, which depicts computing the gain from the measured amplitude of the servo loop response and the known amplitude of the injected sine wave.

The process then passes to step 412, which illustrates a determination of whether the computed gain is within range . The gain computed from the measured values cannot vary by more than 6 dB (a factor of 2) from the expected nominal settings. This protects the unit from using gain values which cannot be correct and must have been the result of erroneous measurement conditions. If the computed gain is within range, the process proceeds to step 414, which depicts using the new (computed) gain value to adjust the controller gain in order to keep the total gain constant.

If either the loops were determined to be opened during measurement (step 408) or the computed gain value is too far from the expected nominal gain, the process proceeds instead to step 416, which illustrates using the default gain value for adjusting the total loop gain. Computed loop gain values are only utilized for gain adjustment if all conditions are acceptable (servo loop remains closed throughout injection and measurement and computed gain value within acceptable range) . Optionally, once the gain adjustment is complete, a final check for loop stability may be made and the old gain values restored if either servo loop (tracking or focusing) is unstable.

In actual implementation, total execution time required for the gain adjustment process depicted in Figure 4 takes

8.5 msec for focus plus 8.5 msec for tracking in the case of a 10 cycle sine wave sample injection and response measure- ment . For 20 cycle injection and measurement, the gain adjustment execution time is 33 msec.

With reference now to Figures 7A and 7B, gain computations for a gain adjustment process in accordance with the preferred embodiment of the present invention are depicted. Figures 7A and 7B illustrate tracking servo and focusing servo gain computations for a repeatability test in which both full and half amplitudes were employed for comparison.

Referring to Figure 8, a block diagram of an altern- ative embodiment of a servo loop with automatic gain adjustment in accordance with a preferred embodiment of the present invention is depicted. The embodiment depicted in Figure 1 employs the ratio of measured response amplitude A to known injected signal amplitude B for gain computations. However, the controller alters the amplitude of A for a given amplitude B through the controller action. In order to obtain a true measure of the servo loop gain independent of the controller gain, the response amplitude 802 at the output D/A converter 108 and the response amplitude 804 at the input of controller 102 are measured. This requires multiple measurements and multiple Fourier Transform computations to determine gain, but a more accurate determination of the servo loop gain without the controller gain component is provided. Gain adjustment by altering controller gain is thus simplified.

The present invention provides a powerful technique for adjusting optical and magneto optical storage device servos, making them more robust. Under the gain adjustment scheme of the present invention, the servo is more tolerant to disk changes, temperature variations, component variations within tolerances, and component aging. Additional hardware is not required and very little overhead, in terms of memory and execution time, is required.

The use of sinusoidal signals in the present invention is more accurate than systems employing peak detection or step response. Pulses employed in step response systems tend to saturate signal levels in various portions of the servo loop such as the power amplifier, and may also knock the servo off track or out of focus. Peak detection, where response signals are nonlinear or include noise, are far less accurate than the Fourier Transform. Smaller signal amplitudes may be employed in the present invention, resulting in less system disturbance and permitting measurements over a longer interval with more accuracy. An arbitrary injected signal frequency, best matched to an individual unit, may be employed for the present invention.

It is important to note that while the present invention has been described in the context of a fully functional system, those skilled in the art will appreciate that the mechanism of the present invention is capable of being distributed in the form of a computer readable medium of instructions in a variety of forms, and that the present invention applies equally regardless of the particular type of signal bearing media used to actually carry out the distribution. Examples of computer readable media include: nonvolatile, hard-coded type media such as read only memories (ROMs) or erasable, electrically programmable read only memories (EEPROMs) , recordable type media such as floppy disks, hard disk drives and CD-ROMs, and transmission type media such as digital and analog communication links.

The description of the preferred embodiment of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limit the invention in the form disclosed.

Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.

Claims

CLAIMS :What is claimed is :

1. A method of adjusting servo loop gain, comprising: adding a sinusoidal signal to a control signal for an actuator within the servo loop; measuring a response of the servo loop to the sinusoidal signal; computing a Fourier transform for the response; computing a gain of the servo loop from an amplitude of the Fourier transform; and adjusting a gain component for a controller within the servo loop to maintain the servo loop gain at a substantially constant value.

2. The method of claim 1, wherein the step of adding a sinusoidal signal to a control signal for an actuator within the servo loop further comprises : adding the sinusoidal signal to the control signal for n cycles of the sinusoidal signal.

3. The method of claim 1, wherein the step of adding a sinusoidal signal to a control signal for an actuator within the servo loop further comprises : adding samples read from a table for the sinusoidal signal to the control signal on alternating servo loop clock cycles .

4. The method of claim 1, wherein the step of measuring a response of the servo loop to the sinusoidal signal further comprises : measuring the response of the servo loop to the sinusoidal signal for n-l cycles of the sinusoidal signal.

5. The method of claim 1, wherein the step of computing a Fourier transform for the response further comprises : computing the Fourier transform for the response at an excitation frequency.

6. The method of claim 1, wherein the step of computing a gain of the servo loop from an amplitude of the Fourier transform further comprises: computing the gain of the servo loop from a ratio of an amplitude of the Fourier transform at an excitation frequency to a known amplitude of the sinusoidal signal.

7. The method of claim 1, wherein the step of adjusting a gain component for a controller within the servo loop to maintain the servo loop gain at a substantially constant value further comprises : adjusting the controller gain component so that a product of the controller gain component and a gain component for a remainder of the servo loop is substantially constant .

8. The method of claim 1, further comprising: averaging a plurality of samples for the measured response of the servo loop to the sinusoidal signal .

9. The method of claim 1, further comprising: verifying that the servo loop remains closed during the addition of the sinusoidal signal and the measurement of the servo loop response; responsive to determining that the servo loop was not closed during the addition of the sinusoidal signal and the measurement of the servo loop response, employing a default gain to adjust the controller gain component.

10. The method of claim 1, further comprising: comparing the computed gain to a range of expected gains for the servo loop; responsive to determining that the computed gain is within the range of expected gains, employing the computed gain to adjust the controller gain component; and responsive to determining that the computed gain is not within the range of expected gains, employing a default gain to adjust the controller gain component.

11. The method of claim 1, further comprising: measuring the control signal received by the actuator; and computing a Fourier transform for the measured control signal , wherein the step of computing a gain of the servo loop from an amplitude of the Fourier transform includes computing the servo loop gain from a ratio of an amplitude of the Fourier transform at an excitation frequency of the measured response to an amplitude of the Fourier transform at the excitation frequency of the measured control signal.

12. An apparatus for adjusting servo loop gain, comprising: a servo loop including: an actuator moving a servo in response to a control signal; a light source and a detector moving in tandem with the servo, the light source emitting a focused light beam and the detector measuring a reflection of the light beam; and a controller receiving an error signal from the detector and providing the control signal to the actuator, wherein the servo loop adds a sinusoidal signal to the control signal for the actuator, wherein the detector measures a response of the servo loop to the sinusoidal signal, and wherein the controller computes a Fourier transform for the measured response, computes a gain of the servo loop from an amplitude of the Fourier transform, and adjusts a gain component for the controller within the servo loop to maintain the servo loop gain at a substantially constant value .

13. The apparatus of claim 12, wherein the servo loop further comprises : a signal adder adding the sinusoidal signal to the control signal for the actuator.

14. The apparatus of claim 12, wherein the servo loop further comprises : a digital-to-analog signal converter converting the control signal from the controller to an analog control signal for the actuator.

15. The apparatus of claim 12, wherein the servo loop further comprises: an analog-to-digital signal converter converting the measured response from the detector to a digital measured response signal for the controller.

16. The apparatus of claim 12, wherein the servo loop adds the sinusoidal signal to the control signal for n cycles of the sinusoidal signal .

17. The apparatus of claim 12, wherein the servo loop adds samples read from a table for the sinusoidal signal to the control signal on alternating servo loop clock cycles.

18. The apparatus of claim 12, wherein the detector measures the response of the servo loop to the sinusoidal signal for n-l cycles of the sinusoidal signal.

19. The apparatus of claim 12, wherein the controller computes the Fourier transform for the response at an excitation frequency.

20. The apparatus of claim 12, wherein the controller computes the gain of the servo loop from a ratio of the amplitude of the Fourier transform at an excitation frequency to a known amplitude of the sinusoidal signal .

21. The apparatus of claim 12, wherein the controller adjusts the controller gain component so that a product of the controller gain component and a gain component for a remainder of the servo loop is substantially constant .

22. The apparatus of claim 12, wherein the controller averages a plurality of samples for the measured response of the servo loop to the sinusoidal signal.

23. The apparatus of claim 12, wherein: the controller verifies that the servo loop remained closed during the addition of the sinusoidal signal and the measurement of the servo loop response, and the controller, responsive to determining that the servo loop was not closed during the addition of the sinusoidal signal and the measurement of the servo loop response, employs a default gain to adjust the controller gain component .

24. The apparatus of claim 12, wherein: the controller compares the computed gain to a range of expected gains for the servo loop, the controller, responsive to determining that the computed gain is within the range of expected gains, employs the computed gain to adjust the controller gain component, and the controller, responsive to determining that the computed gain is not within the range of expected gains, employs a default gain to adjust the controller gain component.

- li

25. A computer program product, comprising: a computer usable medium; instructions within the computer usable medium for adding a sinusoidal signal to a control signal for an actuator within a servo loop; instructions within the computer usable medium for measuring a response of the servo loop to the sinusoidal signal ; instructions within the computer usable medium for computing a Fourier transform for the response; instructions within the computer usable medium for computing a gain of the servo loop from an amplitude of the Fourier transform; and instructions within the computer usable medium for adjusting a gain component for a controller within the servo loop to maintain the servo loop gain at a substantially constant value .

26. The computer program product of claim 25, wherein the instructions for adding a sinusoidal signal to a control signal for an actuator within the servo loop further comprise : instructions for adding the sinusoidal signal to the control signal for n cycles of the sinusoidal signal.

27. The computer program product of claim 25, wherein the instructions for adding a sinusoidal signal to a control signal for an actuator within the servo loop further comprise : instructions for adding samples read from a table for the sinusoidal signal to the control signal on selected servo loop clock cycles.

28. The computer program product of claim 25, wherein the instructions for measuring a response of the servo loop to the sinusoidal signal further comprise: instructions for measuring the response of the servo loop to the sinusoidal signal for n-l cycles of the sinusoidal signal .

29. The computer program product of claim 25, wherein the instructions for computing a Fourier transform for the response further comprise: instructions for computing the Fourier transform for the response at an excitation frequency.

30. The computer program product of claim 25, wherein the instructions for computing a gain of the servo loop from an amplitude of the Fourier transform further comprise: instructions for computing the gain of the servo loop from a ratio of the amplitude of the Fourier transform at an excitation frequency to a known amplitude of the sinusoidal signal .

31. The computer program product of claim 25, wherein the instructions for adjusting a gain component for a controller within the servo loop to maintain the servo loop gain at a substantially constant value further comprise: instructions for adjusting the controller gain component so that a product of the controller gain component and a gain component for a remainder of the servo loop is substantially constant.

32. The computer program product of claim 25, further comprising: instructions within the computer usable medium for averaging a plurality of samples for the measured response of the servo loop to the sinusoidal signal.

33. The computer program product of claim 25, further comprising: instructions within the computer usable medium for verifying that the servo loop remains closed during the addition of the sinusoidal signal and the measurement of the servo loop response; instructions within the computer usable medium, responsive to determining that the servo loop was not closed during the addition of the sinusoidal signal and the measurement of the servo loop response, for employing a default gain to adjust the controller gain component.

34. The computer program product of claim 25, further comprising: instructions within the computer usable medium for comparing the computed gain to a range of expected gains for the servo loop; instructions within the computer usable medium, responsive to determining that the computed gain is within the range of expected gains, for employing the computed gain to adjust the controller gain component; and instructions within the computer usable medium, responsive to determining that the computed gain is not within the range of expected gains, for employing a default gain to adjust the controller gain component.

35. The computer program product of claim 25, further comprising: instructions within the computer usable medium for measuring the control signal received by the actuator; and instructions within the computer usable medium for computing a Fourier transform for the measured control signal, wherein the instructions for computing a gain of the servo loop from an amplitude of the Fourier transform include instructions for computing the servo loop gain from a ratio of an amplitude of the Fourier transform at an excitation frequency of the measured response to an amplitude of the Fourier transform at the excitation freqeuncy of the measured control signal .