Dynamic magnetic information storage or retrieval – Checking record characteristics or modifying recording...
Reexamination Certificate
2000-10-02
2002-07-02
Holder, Regina N. (Department: 2651)
Dynamic magnetic information storage or retrieval
Checking record characteristics or modifying recording...
C360S046000
Reexamination Certificate
active
06414806
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and system for detecting and compensating for thermal asperities in magnetic disk drives, and more particularly to a method and system for detecting a thermal asperity in real time and compensating for the thermal asperity.
2. Description of Prior Art
Data read from a disk is susceptible to errors caused by media defects producing thermal asperities (TA's) in the read channel electronics. For example, in disk drives using magnetoresistive (MR) or giant magnetoresistive (GMR) heads which take advantage of the resistivity of MR material, a DC bias current induces voltage across the sensor (MR stripe of the head) which varies according to changes in the sensor resistance. When a head strikes a particle or magnetic media defect, the temperature of the head can increase by more than 100° C., the temperature rise is a TA. Due to the temperature coefficient of resistivity of the head (about 0.02% per degree C), a TA can cause a significant voltage transient, or DC baseline shift, saturating the analog read signal. Reading data from the disk is then either not possible or has a large bit error rate (BER) for about 0.5-5 microseconds (&mgr;s), until the head temperature returns to an acceptable range. Typically, in order to recover the unreadable data, the channel controller needs to initiate a re-read operation.
One approach to TA detection involves detection at the analog-to-digital converter (ADC) output of the channel input circuit. Because the ADC output is typically late in the signal chain, more of the electronic circuits in the channel become saturated before the TA is detected. Delayed detection can make compensation ineffective if the error cannot be contained to a relatively small amount of data, typically that amount which an error correction code (ECC) can recover.
In non-programmable systems having a moveable AC coupling pole at the variable gain amplifier (VGA) input, the frequency of the pole is simply elevated when a TA is detected and decays at some future time. This can result in increased TA transient duration because the remaining DC offset can decay slowly if the pole is moved back abruptly. In a system wherein a filter is applied prior to threshold detection in the digital domain and there is non-linear rectification to improve detection reliability, the movement of the AC coupling pole at the VGA input is adjustable only in two steps, “high squelch” and “low squelch”. This does not allow for flexibility in the position (frequency) to which the AC pole is moved in response to a TA, or the return to a normal position.
In systems which work within the arm electronics module prior to the VGA input, a TA transient is detected following low-pass filtering and “dead-zone” rectification. An analog signal is generated which moves an AC coupling pole in the Arm Electronics Module in a continuous fashion. This is less desirable because the signal amplitude at that point in the signal chain has not been acted upon by the automatic gain control (AGC) loop. Therefore, the normal amplitude of data signals can vary widely, and therefore, needs to be considered when setting the TA detection threshold. This makes reliable TA detection more difficult. Filtering prior to TA detection is simple first order, however, the first order filter provides less effective discrimination between TA transients and normal signals, and thus TA detection is less reliable. Further, there is no formal detection threshold besides that provided by the dead zone of the “dead-zone” rectifier. The analog compensation circuit must be adjusted with external pins on the package of the arm electronics module.
In systems where a TA transient is detected at the output of the channel continuous-time filter, the AC coupling pole frequency at the VGA input is elevated in response to the TA. The AC coupling pole is allowed to gradually return to normal following the TA transient. Filtering prior to TA detection is only by way of the channel continuous-time filter; there is no dedicated low-pass filter for the TA detection circuit. This provides a less effective means of discriminating between the TA transient and normal data signals, because the cutoff frequency of the continuous-time filter is typically set at 0.3 to 0.6 of the channel data rate. Accordingly, reliable TA detection can be difficult.
Therefore, a need exists for a method of detecting a thermal asperity in magnetic disk drives in real time and compensating for the asperity at the channel input, thereby shortening the length of error caused by the TA.
SUMMARY OF THE INVENTION
A method according to one embodiment of the present invention is disclosed for detecting a thermal asperity. The method amplifies a data signal having a thermal asperity, detects the thermal asperity in the data signal and adjusts a variable resistor prior to amplification for compensating for the thermal asperity.
The method adjusts the variable resistor by setting a detection threshold with a threshold comparator and transmitting an event signal to the variable resistor for adjusting a pole frequency. The pole frequency decays at a predetermined frequency ratio at each step. The steps are taken at fixed intervals of about two to about eight bytes of data. The predetermined frequency ratio can be expressed as:
R
=
exp
[
ln
(
f
H
)
-
ln
(
f
N
)
n
]
where f
H
is a highest frequency to which the pole is adjusted after the thermal asperity is detected, f
N
is a pole frequency prior to the adjustment, and n is the number of steps.
Detecting includes discriminating between the thermal asperity and the data signal using a low-pass filter. The detection occurs within 1.2 bytes of the data signal. The low-pass filter includes a third order Chebyshev filter with about ½ dB of passband ripple and a −3 dB cutoff frequency of about 5% of a data rate.
According to another embodiment of the present invention, the method also includes filtering the data signal with a continuous-time filter, converting the data signal using an analog-to-digital converter, applying the data signal to a gain control for adjusting a gain of the data signal to a constant signal amplitude, and error correcting the data signal after conversion.
In yet another embodiment of the present invention, a system for detecting a thermal asperity within a data signal is presented, including a variable gain amplifier for amplifying the data signal, a thermal asperity detector for detecting the thermal asperity within the amplified signal and outputting an event signal, and a programmable control unit for receiving the event signal and adjusting a variable resistor accordingly, the variable resistor located prior to the variable gain amplifier in a channel.
The system also includes a continuous-time filter for preventing distortion of the data signal, an analog-to-digital converter for converting the data signal to a digital data signal, and an automatic gain control for adjusting the gain of the variable gain amplifier.
The thermal asperity detector according to the present invention, includes a loss-pass filter for discriminating between the thermal asperity signal and the data signal, a non-linear rectifier for selectively amplifying large amplitude portions of the data signal, and a threshold comparator for setting a detection threshold and outputting the event signal.
The non-linear rectifier includes a differential amplifier for transmitting a V
1
+ signal and a V
1
− signal to a first differential source-coupled pair, and a Vcm signal to a second differential source-coupled pair, and a second differential amplifier for accepting an output of said first and second differential source-coupled pair and transmitting a signal to the threshold comparator. The differential source-coupled pairs further include a pair of source degeneration resistors.
The variable resistor according to the present embodiment includes at least one fixed resistor, and a set of field effect transistors (FETs)
Gowda Sudhir Muniswamy
Reynolds Scott Kevin
F. Chau & Associates LLP
Holder Regina N.
International Business Machines - Corporation
LandOfFree
Method for thermal asperity detection and compensation in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for thermal asperity detection and compensation in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for thermal asperity detection and compensation in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2876703