Electricity: measuring and testing – Magnetic – Magnetic information storage element testing
Reexamination Certificate
2000-02-25
2002-04-02
Snow, Walter (Department: 2862)
Electricity: measuring and testing
Magnetic
Magnetic information storage element testing
C360S031000, C360S039000, C360S137000
Reexamination Certificate
active
06366081
ABSTRACT:
INTRODUCTION
The present invention relates generally to the detection of media defects during media production, which is a necessary industrial quality control step in the production process. In the magnetic media industry, media has to undergo glide, burnish and certification before being graded and eventually qualified. The present invention provides an improved testing method which potentially speeds media production.
BACKGROUND OF THE INVENTION
In media certification, there are tests for Missing Pulse, Extra Pulse, Bit Shift and Modulation. All these tests are run, based on user defined thresholds as a percentage of Track Average Amplitude (TAA). It is therefore necessary to determine Track Average Amplitude before thresholds can be set and tests conducted. Production testing apparatus typically obtain the Track Average Amplitude for a particular track using formulae. This saves time in the derivation of the Track Average Amplitude for every track but the value derived from formulae doe not reflect the true Track Average Amplitude, known as the ‘true TAA’ of the track. Parametric testing apparatus samples through each track and computes the average of all bytes on that track before using the computed value to compare with every byte on that track. This is ‘true TAA’, but the method requires a very large amount of memory for storing all of the data for the respective track, a high sampling rate to keep up with disk speed and a lengthy processing time.
A number of different ways are known for determining Track Average Amplitude, depending on the method and type of testing and preferably allow achievement of a minimum ‘production time’. The following prior art specifications describe different types of testing and the corresponding manner of determining Track Average Amplitude.
U.S. Pat. No. 4,746,995 describes a sequencing operation whereby all tracks are written first and then all tracks are read back to obtain a calculated signal. This has the advantage over writing and reading on a track by track basis in that several drives can be processed simultaneously using the same electronics for one drive. It has the same processing speed as that which uses four carriages on one drive, each doing its write-read in the same sequence. However, the calculated signal is compared with a known standard which means that Track Average Amplitude is not obtained in real time and is not true for every track.
U.S. Pat. No. 5,124,849 discloses writing a series of signal pulses and reading back the pulses using two channels one of which is delayed by a multiple of the pattern's period. A comparator circuit called the CYCOMP detects amplitude deviation between the two signals; another called the Bandreject circuit detects phase deviation, both of which are indicative of aberrations and are compared with a preset error threshold. The preset error threshold is a percentage of the prevailing value of peak values of the read back signal generated by a peak following means. To collect peak values before establishing the prevailing reference value, one additional revolution for every track is required if a ‘true TAA’ value is used and this is at the expense of production time.
U.S. Pat. No. 5,280,395 describes a method and an apparatus whereby the mean average value computed based on an average level value for a plurality of tracks prior to the processing track is used as a reference value for comparison with subsequent read back values to detect defects. The process includes a read/write means to write and read back test signals; parametric measurement means for measuring an average-level value for each track based on read back signals: memory means for storing the average-level value for a plurality of tracks prior to the processing track. Arithmetic means for computing an average average-level value based on the average-level values for the plurality of tracks: defect detection means for detecting defects based on the average average-level value compared against read back signal of the processing track. This method gives a more reliable Track Average Amplitude in the form of average average-level value but at the expense of huge processing time. Processing time includes the time to compute average level value for every track prior to the processing track; store the values in memory; compute the average average-value and compare with the read back signal of the processing track.
U.S. Pat. No. 5,532,586 describes an Extra Pulse testing method. An erasing means erases all information on the disk in one revolution. A testing means tests every track for defects. A writing means writes data on the first track before erasure. A reading means read data from the first track before erasure. A data making means calculate from read back signal the reference value to be used for defects detection for the first track. In short there is the initializing process to generate reference data; erasing process to erase all information on disk in one revolution and testing process to test every track for defects. The advantage is that only one revolution is required to erase all information on disk. Again there is no provision for using ‘true TAA’ for every processing track.
U.S. Pat. No. 4,929,894 describes a method for performing Missing Pulse and Extra Pulse tests simultaneously. This method is intended to enhance throughput for disk drive quality control testing. This method involves a delay means to receive a periodic input signal and to delay that signal for a period equal to one period of the input signal. A difference means determines the difference in amplitudes between the delayed and undelayed input signal. A comparator means compares the difference in amplitudes with a reference signal and generates an error signal should there be difference in amplitudes exceeding the reference threshold. The Missing pulse test is performed at the same time as the undelayed signal is read back and testing time is thus shortened.
To summarise the prior art, the following methods for determining the Track Average Amplitude are described. The method of U.S. Pat. No. 4,746,995 uses a known standard and compares read back signal of every track with the known value. The method of U.S. Pat. No. 5,532,586 performs an initializing process for one track and computes the Track Average Amplitude for that track. That same Track Average Amplitude will be adjusted accordingly for subsequent tracks on the basis of a known relationship defining approximate changes in Track Average Amplitude changes with track number. The method of U.S. Pat. No. 5,280,395 describes the calculation of the average Track Average Amplitude for a plurality of tracks prior to the processing track. The method of U.S. Pat. No. 5,124,849 describes the calculation of a prevailing reference value and there is no account of how that value is computed. The above prior art does not show a method that uses ‘true TAA’ for every track in the testing nor does it describe a method in which errors are first detected, based on ‘estimated TAA’, herein called ‘potential errors’ following which the ‘potential errors’ are verified based on ‘true TAA’, the verified errors herein called ‘true errors’.
SUMMARY OF THE INVENTION
According to a first aspect, the present invention provides a method of detecting defects in a selected track of a media element of a mass storage device including the steps of:
a) determining an estimated Track Average Amplitude which approximates the true Track Average Amplitude of a read signal from a track under test of the media element;
b) during first revolution of the media element, writing onto the track under test, a stream of data bits of a predetermined pattern;
c) during a second revolution of the media element, reading back the stream of data bits written to the track under test during the first revolution;
d) comparing the read signal amplitude of each bit of the stream of data read back from the track under test during the second revolution with the estimated Track Average Amplitude;
e) generating an error signal when the amplitude of a bit
Liew Thomas Yun Fook
Loh Teck Ee
Silva Udaya Ahangama W.
Tan Seng Ghee
Cowan Liebowitz & Latman P.C.
Data Storage Institute
Dippert William H.
Snow Walter
LandOfFree
Method and apparatus for high throughput media defect... 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 and apparatus for high throughput media defect..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for high throughput media defect... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2887580