U.S. patent application number 16/221839 was filed with the patent office on 2019-11-07 for write current switching in a data storage device using an effective footprint of a write pole.
The applicant listed for this patent is Seagate Technology LLC. Invention is credited to Ralph William Cross, Quan Li, Xiong Liu.
Application Number | 20190341075 16/221839 |
Document ID | / |
Family ID | 65032121 |
Filed Date | 2019-11-07 |
![](/patent/app/20190341075/US20190341075A1-20191107-D00000.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00001.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00002.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00003.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00004.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00005.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00006.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00007.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00008.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00009.png)
![](/patent/app/20190341075/US20190341075A1-20191107-D00010.png)
View All Diagrams
United States Patent
Application |
20190341075 |
Kind Code |
A1 |
Liu; Xiong ; et al. |
November 7, 2019 |
WRITE CURRENT SWITCHING IN A DATA STORAGE DEVICE USING AN EFFECTIVE
FOOTPRINT OF A WRITE POLE
Abstract
A sequence of symbols is generated to describe a set of write
data, the symbols having a length of nT, where T is a channel clock
rate and n is an integer over a predetermined range. Bi-directional
write currents are applied to a write pole to record the sequence
of symbols to a magnetic storage medium. The write pole has an
effective footprint with a downtrack length of mT, where m is an
integer. The write currents are switched between a first rail
current and a second rail current for alternating symbols, the
write currents further transitioning to an intermediate current
value for at least one channel clock period for symbols longer than
1T. Write currents are applied to the write pole when recording
symbols having a length longer than mT using the effective
footprint of the write pole as an interval.
Inventors: |
Liu; Xiong; (Singapore,
SG) ; Li; Quan; (Singapore, SG) ; Cross; Ralph
William; (Lyons, CO) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Seagate Technology LLC |
Cupertino |
CA |
US |
|
|
Family ID: |
65032121 |
Appl. No.: |
16/221839 |
Filed: |
December 17, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15971284 |
May 4, 2018 |
10192580 |
|
|
16221839 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G11B 5/012 20130101;
G11B 5/5539 20130101; G11B 5/09 20130101; G11B 5/5534 20130101;
G11B 20/10037 20130101; G11B 20/10222 20130101 |
International
Class: |
G11B 20/10 20060101
G11B020/10; G11B 5/012 20060101 G11B005/012; G11B 5/09 20060101
G11B005/09 |
Claims
1. A method, comprising: generating a sequence of symbols to
describe a set of write data, the symbols having a length of nT,
where T is a channel clock rate and n is an integer over a
predetermined range; applying bi-directional write currents to a
write pole to record the sequence of symbols to a magnetic storage
medium, the write pole having an effective footprint with a
downtrack length of mT, where m is an integer; and applying write
currents to the write pole when recording symbols having a length
longer than mT to the medium using the effective footprint of the
write pole as an interval.
2. The method of claim 1, comprising measuring the effective
footprint of the write pole.
3. The method of claim 1, wherein applying the write currents
comprises: applying write currents to the write pole according to a
first write scheme for a symbol having a length of 1T; applying
write currents to the write pole according to a second write scheme
for a symbol having a length of 2T to the effective footprint; and
applying write currents to the write pole according to a third
write scheme for a symbol having a length greater than the
effective footprint.
4. The method of claim 1, wherein an intermediate current value
between a first rail current and a second rail current is applied
at least to the last m-1 bits for each symbol having a length
longer than mT.
5. The method of claim 1, wherein, for each symbol having a length
from 2T to the effective write pole footprint: applying a write
current to the write pole only at a 1.sup.st T of the symbol; and
applying an intermediate current value between a first rail current
and a second rail current after the 1.sup.st T of the symbol.
6. The method of claim 1, wherein, for each symbol having a length
greater than the effective write pole footprint (EWPF): determining
a quotient by dividing the symbol length by the EWPF and rounding
down the quotient; determining a remainder by subtracting the EWPF
or a multiple of the EWPF from the symbol length; applying a write
current to the write pole at a 1.sup.st T of the symbol; and
applying a write current to the write pole at a predetermined T of
the symbol, wherein the predetermined T of the symbol is based on
the remainder and the quotient.
7. The method of claim 1, wherein, for each symbol having a length
greater than the effective write pole footprint (EWPF): determining
a quotient by dividing the symbol length by the EWPF and rounding
down the quotient; determining a remainder by subtracting the EWPF
or a multiple of the EWPF from the symbol length; applying a write
current (WC) plus an overshoot (OS) to the write pole at a 1.sup.st
T of the symbol; if the remainder=1, applying WC at a
(1+(quotient-1)*EWPF+remainder)T; and if the remainder>1,
applying WC+OS at a (1+(quotient-1)*EWPF+remainder)T.
8. The method of claim 1, wherein, for each symbol having a length
greater than the effective write pole footprint (EWPF): determining
a quotient by dividing the symbol length by the EWPF and rounding
down the quotient; determining a remainder by subtracting the EWPF
or a multiple of the EWPF from the symbol length; applying a write
current (WC) plus an overshoot (OS) to the write pole at a 1.sup.st
T of the symbol; and applying WC from (1+(quotient-1)*EWPF+1)T to
(1+(quotient-1)*EWPF+remainder)T.
9. The method of claim 1, wherein applying the bi-directional write
currents to record a selected symbol having a length greater than
the effective footprint further comprises: writing a selected
symbol to the medium by switching to a selected one of a first rail
current and a second rail current at a 1.sup.st T of the selected
symbol; applying an intermediate current value between the first
and second rail currents starting at a 2.sup.nd T of the selected
symbol; resuming application of the selected one of the first and
second rail currents after applying the intermediate current value;
and resuming application of the intermediate current value at an
ending portion of the selected symbol prior to switching to the
other of the first and second rail currents at a next symbol
boundary, the ending portion having a length equal to (m-1)T.
10. The method of claim 9, wherein the first rail current is
greater than 0 mA, the second rail current is less than 0 mA, and
the intermediate current value is nominally 0 mA.
11. An apparatus, comprising: a magnetic recording medium; a write
pole controllably positionable adjacent the medium to write data
thereto, the write pole having an effective footprint with a
downtrack length of mT, where m is an integer; a channel circuit
configured to generate a sequence of symbols having a length of nT,
where T is a channel clock rate and n is an integer over a
predetermined range; and a write driver configured to apply
bi-directional write currents to the write pole to record the
sequence of symbols to the medium, the writer driver further
configured to use the effective footprint of the write pole as an
interval for applying write currents to the write pole when
recording symbols having a length longer than mT to the medium.
12. The apparatus of claim 11, comprising circuitry configured to
measure the effective footprint of the write pole.
13. The apparatus of claim 11, wherein the writer driver is
configured to: apply write currents to the write pole according to
a first write scheme for a symbol having a length of 1T; apply
write currents to the write pole according to a second write scheme
for a symbol having a length of 2T to the effective footprint; and
apply write currents to the write pole according to a third write
scheme for a symbol having a length greater than the effective
footprint.
14. The apparatus of claim 11, wherein the writer driver is
configured to apply an intermediate current value between a first
rail current and a second rail current at least to the last m-1
bits for each symbol having a length longer than mT.
15. The apparatus of claim 11, wherein, for each symbol having a
length from 2T to the effective write pole footprint, the writer
driver is configured to: apply a write current to the write pole
only at a 1.sup.st T of the symbol; and apply an intermediate
current value between a first rail current and a second rail
current after the 1.sup.st T of the symbol.
16. The apparatus of claim 11, wherein, for each symbol having a
length greater than the effective write pole footprint (EWPF):
circuitry is configured to determine a quotient by dividing the
symbol length by the EWPF and rounding down the quotient, the
circuitry further configured to determine a remainder by
subtracting the EWPF or a multiple of the EWPF from the symbol
length; the writer driver is configured to apply a write current to
the write pole at a 1.sup.st T of the symbol; and the writer driver
is configured to apply a write current to the write pole at a
predetermined T of the symbol, wherein the predetermined T of the
symbol is based on the remainder and the quotient.
17. The apparatus of claim 11, wherein, for each symbol having a
length greater than the effective write pole footprint (EWPF):
circuitry is configured to determine a quotient by dividing the
symbol length by the EWPF and rounding down the quotient, the
circuitry further configured to determine a remainder by
subtracting the EWPF or a multiple of the EWPF from the symbol
length; the writer driver is configured to apply a write current
(WC) plus an overshoot (OS) to the write pole at a 1.sup.st T of
the symbol; if the remainder=1, the writer driver is configured to
apply WC at a (1+(quotient-1)*EWPF+remainder)T; and if the
remainder>1, the writer driver is configured to apply WC+OS at a
(1+(quotient-1)*EWPF+remainder)T.
18. The apparatus of claim 11, wherein, for each symbol having a
length greater than the effective write pole footprint (EWPF):
circuitry is configured to determine a quotient by dividing the
symbol length by the EWPF and rounding down the quotient, the
circuitry further configured to determine a remainder by
subtracting the EWPF or a multiple of the EWPF from the symbol
length; the writer driver is configured to apply a write current
(WC) plus an overshoot (OS) to the write pole at a 1.sup.st T of
the symbol; and the writer drive is configured to apply WC from
(1+(quotient-1)*EWPF+1)T to (1+(quotient-1)*EWPF+remainder)T.
19. The apparatus of claim 11, wherein the writer driver is
configured to: write a selected symbol to the medium by switching
to a selected one of a first rail current and a second rail current
at a 1.sup.st T of the selected symbol; apply an intermediate
current value between the first and second rail currents starting
at a 2.sup.nd T of the selected symbol; resume application of the
selected one of the first and second rail currents after applying
the intermediate current value; and resume application of the
intermediate current value at an ending portion of the selected
symbol prior to switching to the other of the first and second rail
currents at a next symbol boundary, the ending portion having a
length equal to (m-1)T.
20. The apparatus of claim 19, wherein the first rail current is
greater than 0 mA, the second rail current is less than 0 mA, and
the intermediate current value is nominally 0 mA.
Description
RELATED PATENT DOCUMENTS
[0001] This application is a continuation of U.S. application Ser.
No. 15/971,284, filed May 4, 2018, which is incorporated herein by
reference in its entirety.
SUMMARY
[0002] Embodiments are directed to a method comprising generating a
sequence of symbols to describe a set of write data, the symbols
having a length of nT, where T is a channel clock rate and n is an
integer over a predetermined range. The method comprises applying
bi-directional write currents to a write pole to record the
sequence of symbols to a magnetic storage medium. The write pole
has an effective footprint with a downtrack length of mT, where m
is an integer. The write currents are switched between a first rail
current and a second rail current for alternating symbols, the
write currents further transitioning to an intermediate current
value for at least one channel clock period for symbols longer than
1T. The method also comprises applying write currents to the write
pole when recording symbols having a length longer than mT using
the effective footprint of the write pole as an interval.
[0003] Embodiments are directed to an apparatus comprising a
magnetic recording medium and a write pole controllably
positionable adjacent the medium to write data thereto. The write
pole has an effective footprint with a downtrack length of mT,
where m is an integer. A channel circuit is configured to generate
a sequence of symbols having a length of nT, where T is a channel
clock rate and n is an integer over a predetermined range. A write
driver is configured to apply bi-directional write currents to the
write pole to record the sequence of symbols to the medium. The
write currents are switched between a first rail current and a
second rail current for alternating symbols. The write currents
also transition to an intermediate current value between the first
and second rail currents for at least one channel clock period for
symbols longer than 1T. The writer driver is further configured to
use the effective footprint of the write pole as an interval for
applying write currents to the write pole when recording symbols
having a length longer than mT to the medium.
[0004] The above summary is not intended to describe each
embodiment or every implementation. A more complete understanding
will become apparent and appreciated by referring to the following
detailed description and claims in conjunction with the
accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] FIG. 1 is a simplified functional block diagram of a data
storage device configured and operated in accordance with various
embodiments of the present disclosure;
[0006] FIG. 2 shows a data transducer adjacent a data recording
medium in accordance with some embodiments of the data storage
device of FIG. 1;
[0007] FIG. 3 represents a representative magnetization pattern for
the medium of FIG. 2;
[0008] FIG. 4 is a functional block diagram of the data transducer
(head) from FIG. 2 in conjunction with a read/write (R/W) channel
and a preamplifier/driver (preamp) circuit of a representative data
storage device in accordance with various embodiments;
[0009] FIG. 5 shows a write driver circuit of the preamp shown in
FIG. 4 in conjunction with a write coil of the transducer in
accordance with various embodiments;
[0010] FIGS. 6A and 6B show a schematic top plan view of a write
pole of the write coil of FIG. 5 positioned relative to a data
track, the write pole having an effective write pole footprint in
accordance with various embodiments;
[0011] FIG. 7 illustrates a method for measuring the effective
footprint of a write pole in accordance with various
embodiments;
[0012] FIG. 8 is a plot of correlation coefficient (R) for signal
patterns having varying length in connection with the methodology
of FIG. 7 for measuring the effective footprint of a write pole in
accordance with various embodiments;
[0013] FIG. 9 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments;
[0014] FIG. 10 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments;
[0015] FIG. 11 illustrates a representative write triplet in
accordance with various embodiments;
[0016] FIG. 12 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments;
[0017] FIG. 13 is a graphical representation of three different
write current command signals for writing data patterns with
different symbol length using the method of FIG. 12, wherein the
latter two signals are generated using different effective
footprints of a write pole in accordance with various
embodiments;
[0018] FIG. 14 is a graphical representation of three different
write current command signals for writing data patterns with
different symbol length using the method of FIG. 12, wherein the
latter two signals are generated using different effective
footprints of a write pole in accordance with various
embodiments;
[0019] FIG. 15 illustrates an alternative method for enhancing
write current switching efficiencies during data write operations
using the effective footprint of a write pole in accordance with
various embodiments; and
[0020] FIG. 16 is a graphical representation of three different
write current command signals for writing data patterns with
different symbol length using the alternative method of FIG. 15,
wherein the latter two signals are generated using different
effective footprints of a write pole in accordance with various
embodiments.
[0021] The figures are not necessarily to scale. Like numbers used
in the figures refer to like components. However, it will be
understood that the use of a number to refer to a component in a
given figure is not intended to limit the component in another
figure labeled with the same number.
DETAILED DESCRIPTION
[0022] The present disclosure is generally directed to data storage
systems, and more particularly to a method and apparatus for
enhancing write current switching efficiencies during data write
operations.
[0023] Data storage devices store and retrieve data from a host
device in a fast and efficient manner. Some data storage devices
employ rotatable recording media (discs) which are rotated at a
high rotational velocity. One or more data transducers (heads) are
movably positioned adjacent tracks defined on the disc surfaces to
write and read data to and from the media.
[0024] The data are often written in the form of symbols of nT bit
length, where T is a channel clock rate and n is an integer over a
selected range. For example, 1T refers to 1 bit of a particular
polarity, 2T refers to 2 bits with the same polarity, 3T refers to
3 bits with the same polarity, and so on. The range for the value n
can vary depending on the requirements of a given storage device
environment. Representative ranges can include symbol lengths of
from 1T to 8T, 3T to 16T, etc.
[0025] In magnetic recording discs, bi-directional write currents
are supplied to a write coil to write the symbol sequence to the
medium, with magnetic flux transitions (reversals) being supplied
at each symbol boundary. The magnetic flux transitions are
interpreted as a first logical value (e.g., logical 1), and a
second logical value (e.g., logical 0) is assigned to each of the
channel bits between adjacent symbol boundaries.
[0026] Run length limited (RLL) encoding (e.g. 8/9, 16/17, etc.) is
applied to establish the available symbol set by incorporating
rules with regard to how closely spaced and how far apart the
symbol boundaries (e.g., logical 1s) can be located in the recorded
sequence. For example, a symbol set with a minimum symbol length of
1T allows transitions (logical 1s) to be immediately adjacent one
another with no intervening logical Os therebetween (e.g., the
sequence "11" can be written to the medium). By contrast, a symbol
set with a minimum symbol length of 2T requires at least two (2)
logical Os be present between successive logical is (e.g., 1001 is
the closest two logical 1s can appear on the medium). A symbol set
with a maximum symbol size of 8T allows no more than eight (8)
logical Os between successive logical is (e.g., 100000001 is the
maximum distance between logical 1s), and so on.
[0027] Increasing the data storage density along a track (such as
in terms of bits per inch, or BPI) can generally be achieved by
increasing the effective channel clock rate T. The channel clock
rate T can be increased by using a higher write frequency clock
and/or rotating the medium at a higher rotational velocity.
Generally, a higher channel clock rate T provides a shorter amount
of elapsed time between successive transitions on the medium as the
medium rotates adjacent the associated transducer.
[0028] There will generally be an upper limit to how short the
smallest symbols in the symbol set can be from an elapsed time and
distance standpoint. For example, the use of extremely short
symbols in the symbol set, such as 1T symbols, can provide
degradation in the reliability of the write operation if there is
insufficient time for the write driver to obtain a full reversal of
the write current through the coil during the period of time that
the 1T (or other short length) symbol is being written. Stated
another way, the shortest symbols in the set may not be "long"
enough from a time or distance standpoint to enable the system to
adequately magnetize the medium and store the desired magnetization
pattern with sufficient strength to ensure reliable recovery during
a subsequent read operation.
[0029] A related issue is that for longer symbol lengths (e.g., 5T,
8T, 12T, etc.), adjacent track erasure can arise, thereby limiting
track per inch (TPI) densities that can be achieved. That is,
longer symbol lengths may tend to provide radially "wider" symbols
as compared to shorter symbol lengths due to the extended
application of power from the write pole to the medium.
[0030] Thus, using a reduced frequency channel clock rate T to
accommodate short symbol writes may tend to increase adjacent track
interference and serve as an upper limit on achievable TPI
densities. The foregoing limitations are not limited to magnetic
recording, but can arise in other data recording systems as well
such as optical data recording systems that rely on changes in
optical detection levels to mark symbol boundaries.
[0031] Accordingly, various embodiments of the present disclosure
are generally directed to a method and apparatus for writing data
to a data recording medium. Various embodiments are directed to
measuring an effective footprint of a write pole (referred to
herein as the effective write pole footprint or EWPF). Various
embodiments are directed to applying write currents to a write pole
according to both the data pattern to be written and the effective
write pole footprint. Applying write currents to a write pole
according to both the data pattern to be written and the effective
write pole footprint advantageously reduces the risk of adjacent
track erasure and power consumption when writing relatively long
data patterns (e.g., symbols having a length>5T or 6T).
[0032] As explained below, various embodiments provide a write
driver circuit that supplies bi-directional write currents to a
write pole, such as but not limited to a magnetic coil. Data are
written in the form of a sequence of nT symbols over a selected
range, where T is a channel clock rate and n is an integer such as
from a minimum value for n=X to a maximum value for n=Y.
Alternating symbols are written using opposing rail currents of
selected respective magnitudes.
[0033] For symbols having a length>1T, an intermediate current
value (e.g., 0 mA) between the two rail current magnitudes is
applied to the write pole for one or more channel clock periods
prior to the next transition (symbol boundary). The number of
channel clock periods during which the intermediate current value
is applied is based on the length of the symbol and the effective
footprint of the write pole. The intermediate current can be a zero
current or a relatively small current. In this way, the current
switching time to initiate the writing of the next symbol can be
reduced. This can be achieved in a variety of ways, such as by
adding a zero status to the data stream or temporarily disabling
the write current at the end of each symbol.
[0034] The size of the effective write pole footprint corresponds
to the number of bits that are concurrently magnetized to the
desired magnetization orientation when a write current is applied
to the write pole. For example, a write pole with an effective
footprint of 6T will concurrently magnetize 6 bits to the desired
magnetization orientation when a write current is applied to the
write pole.
[0035] The effective write pole footprint and the length of the
symbol to be written are used to determine when and how long the
write current is to be applied to the write pole. According to some
embodiments, and as described in detail hereinbelow, the effective
write pole footprint is measured in terms of downtrack bit length,
expressed in terms of a downtrack length of mT symbol bits, where m
is an integer. For example, the effective footprint for a typical
write pole can be 5T, 6T, 7T or 8T long, for example. It will be
understood that the effective write pole footprint size can be
smaller or larger than this representative range. As will be
discussed below, the size of the effective write pole footprint
will vary from write pole to write pole (e.g., head to head).
Moreover, the size of the effective write pole footprint will vary
across the stroke as a function of skew between the outer diameter
(OD) and the inner diameter (ID) of the disc.
[0036] In accordance with various embodiments, the timing and
duration of write currents applied to the write pole when writing
data patterns of varying length can be implemented as follows:
[0037] Length of 1T: apply write current using a first write
scheme
[0038] Length of 2T to the EWPF: apply write current using a second
write scheme
[0039] Length>EWPF: apply write current using a third write
scheme
As can be seen above, different writing schemes are implemented
based on the relationship between the length of the data pattern to
be written and the effective write pole footprint.
[0040] According to some embodiments, the following writing schemes
can be implemented based on the effective write pole footprint and
the length T of the data pattern to be written. For a 1T data
pattern, the first scheme involves applying a write current for the
1.sup.st T. For a data pattern having a length from 2T to the EWPF,
a write current is applied only for the 1.sup.st T. No current (or
an intermediate current) is applied after the 1.sup.st T for a data
pattern having a length from 2T to the EWPF. For a data pattern
having a length>EWPF, the EWPF is used as an interval for
applying write currents, details of which are provided hereinbelow.
In general terms, for a data pattern having a length>EWPF, no
current (or an intermediate current) is applied for the last m-1
bits of the data pattern, where mT represents the effective write
pole footprint.
[0041] While it is contemplated that the rail current magnitudes
will be equal and opposite values, such as nominally .+-.25
milliamps, mA, such is not required. Depending on the type of
medium and the write characteristics of the write pole, the rail
current magnitudes can be any suitable values including values with
different magnitudes (e.g., +30 mA and -20 mA). Similarly, while a
zero current value can be used for the intermediate level, other
values of relatively small current can be applied, including
intermediate values that are selected based on factors such as the
size of a given symbol and/or the direction of the switching
current.
[0042] It has been found by the inventors that write power wave
shaping using the effective write pole footprint and intermediate
current values in accordance with various embodiments disclosed
herein can provide a substantial reduction of adjacent track
erasure effects and power consumption. Adjacent track erasure
generally involves the partial erasure of the data stored on an
adjacent track. Adjacent track erasure arises due to the
application of write current to a target track; the larger the
write current, generally the wider the adjacent track erasure, and
the longer the write current is applied, generally the wider the
adjacent track erasure. Since zero or small current is applied to
the writing of some bits, the adjacent track erasure can be
reduced. This in turn allows further reductions in track pitch and
higher TPI values.
[0043] The techniques disclosed herein can be applied to any number
of different forms of recording systems, including but not limited
to perpendicular magnetic recording (PMR), longitudinal magnetic
recording (LMR), heat assisted magnetic recording (HAMR), microwave
assisted magnetic recording (MAMR), two dimensional magnetic
recording (TDMR), shingled magnetic recording (SMR), interleaved
magnetic recording (IMR), etc. The write currents as discussed
herein can be supplied to magnetically responsive writer elements
(e.g., magnetic write coils) as well as other forms of write
devices such as, for example, a laser diode used in a HAMR system,
an optical or magneto/optical system, etc.
[0044] These and other features of various embodiments of the
present disclosure can be understood beginning with a review of
FIG. 1 which provides a simplified representation of a data storage
device 100 of the type used to store and retrieve user data from a
host device. The device 100 includes a controller (control circuit)
102 and a memory module 104. The controller 102 provides top level
communication and control functions as the device interfaces with
the host device. Data from the host device is transferred for
storage in the memory 104.
[0045] In some cases, the controller 102 can take the form of a
hardware or programmable processor with associated programming in a
memory location to carry out the requisite control functions. The
memory 104 can take any number of configurations to provide
non-volatile storage of data, including but not limited to magnetic
recording discs, optical recording discs, etc. The memory 104 may
include circuitry in the form of channel electronics,
preamplifier/driver stages, spindle and actuation motors, etc.
[0046] FIG. 2 shows an elevational representation of a data
transducer 110 of the data storage device 100 of FIG. 1 in
accordance with some embodiments. In FIG. 2, the storage device 100
is characterized as a hard disc drive (HDD), although such is
merely for purposes of providing a concrete example and is not
limiting. The techniques disclosed herein are applicable to a wide
variety of data storage devices including hybrid drives, optical
storage devices, magneto-optical storage devices, etc.
[0047] The data transducer 110 is controllably positioned adjacent
a magnetic recording medium (disc) 112 using a flexible suspension
(flexure) member 114. In some cases, an air bearing surface (ABS)
may be formed on a slider portion of the transducer to maintain
stable aerodynamic flight of the transducer using fluidic
atmospheric currents established by the high speed rotation of the
disc 112.
[0048] The data transducer 110 (also referred to as a "head")
includes a number of operative elements including a read (R)
element 116 and a write (W) element 118. The read element may take
the form of a magnetoresistive (MR) sensor, and the write pole may
take the form of a perpendicular magnetic writing coil. Other forms
for these elements can be used as desired. Additional operative
elements can be incorporated into the transducer 110 such as a heat
assisted magnetic recording (HAMR) system, a fly height adjustment
(FHA) mechanism, contact sensors, etc.
[0049] FIG. 3 shows a magnetization pattern 120 that is written by
the write pole 118 to a recording layer of the medium 112 from FIG.
2 during a write operation. A perpendicular magnetic recording
pattern is shown, although such is merely exemplary and is not
limiting. The perpendicular magnetization direction is vertical, or
perpendicular, to the top surface of the medium 112, and
constitutes a sequence of symbols of alternating magnetic
orientation.
[0050] The pattern 120 is written as a sequence of symbols with
lengths nT where T is a channel clock rate at a selected frequency
and n is an integer which ranges over a selected interval set from
a minimum value X to a maximum value Y. For purposes of the present
discussion, the encoding scheme is contemplated as providing
symbols of from 1T to 8T in length. Other encoding schemes can be
used. FIG. 3 shows an exemplary symbol sequence of 5T, 2T, 8T, 4T,
1T and 6T symbols that have been written to the medium 112. For the
pattern 120, the 5T symbol is interpreted as the bit pattern 11111,
the 2T symbol is interpreted as the bit pattern 00, and so on. The
number of arrows representing each symbol is not significant other
than to denote an exemplary magnetic orientation for that
particular region of the medium 112.
[0051] FIG. 4 shows the transducer (head) 110 of FIG. 2 in
conjunction with a read/write (R/E) channel circuit 130 and a
preamplifier/driver circuit (preamp) 132. The channel 130 can be
realized in a number of different hardware or programmable
processor configurations, including SOC (system on chip) integrated
circuit devices, programmable devices that use programming in
memory to execute program steps, state machines, hardwired logic
gates, transistors, etc. Regardless of form, the channel circuit
includes encoding circuitry used during write operations to
transition input write data to a sequence of symbols. The channel
130 further includes decoding circuitry used during read operations
that reconstructs the originally written data from a recovered bit
sequence corresponding to the originally written symbols. The
preamp 132 includes write driver and read amplification and
conditioning circuitry to interface with the transducer 110.
[0052] During a read operation, a readback signal is generated by
the read sensor 116 (FIG. 2). The pulses are used to adjust a
variable clock oscillator (VCO) or similar circuit in the channel
130 to establish a readback clock that provides search windows at
each T interval. In this way, the bit sequence shown in FIG. 3 can
be recovered by the channel and decoded to provide the originally
stored user data sequence.
[0053] FIG. 5 shows a write driver circuit 140 of the preamp 132 of
FIG. 4. The write driver circuit 140 can take any number of
suitable forms, including an H-bridge circuit made up of power
MOSFETs (metal oxide semiconductor field effect transistors)
connected in an H-configuration, an operational amplifier circuit,
a digital to analog converter (DAC) circuit, etc. Responsive to an
input symbol sequence, the write driver 140 supplies bi-directional
write currents to a write coil 142 of the write pole 118 (FIG. 2)
to write magnetization patterns such as represented in FIG. 3.
Except as modified below, the write currents nominally switch
direction at each symbol boundary between a maximum rail current
I.sub.MAX from current source 144 and a minimum rail current
I.sub.MIN from current source 146. The rail currents can vary, but
exemplary values may be .+-.25 milliamps, mA, etc. Any suitable
current values can be used, including asymmetric values (e.g., +30
mA and -20 mA). Intermediate currents, such as zero currents, can
be generated as a current between the maximum I.sub.MAX and minimum
I.sub.MIN rail currents.
[0054] Rail voltage sources can be used by the write driver in lieu
of the current sources represented in FIG. 5, but write currents
will still be applied through the coil to effect the desired
magnetization fields to magnetize the medium. Therefore, the
present discussion will describe the write driver in terms of
applied write currents. This applies to other forms of write poles
as well since even if voltages are applied, currents will flow
through the write pole.
[0055] FIGS. 6A and 6B show a schematic top plan view of a main
pole footprint 600 of the write coil 142 (FIG. 5) positioned
relative to a data track 610. The main pole footprint 600 is the
magnetic field contour generated by a write operation. The actual
configuration of the main pole footprint 600 will depend on the
construction of the storage device, so the geometry shown in FIGS.
6A and 6B is merely exemplary and is not necessarily limiting. The
main pole footprint 600 includes a central aperture 602, where the
magnetic field is slightly weaker than at other regions of the main
pole footprint 600. As such, the central aperture 602 may only
magnetically flip a certain percentage (e.g., 90%) of the media
grain under the central aperture 602, though it is not completely
flipped (shown as incomplete area 624). Because the percentage of
the media grain flipped under the central aperture 602 is high,
this media grain can be considered fully flipped for practical
purposes. It is noted that small incomplete areas 624 near the
sides of the leading edge of the main pole footprint 600 do not
significantly affect the written-on media and read back
waveform.
[0056] Embodiments of the disclosure are directed to determining
the effective size (e.g., downtrack length) of the main pole
footprint 600. For example, the effective write pole footprint of
the write coil characterized in FIGS. 6A and 6B may be determined
to be 6T or 7T in length. Determining the effective write pole
footprint is important, since writing data patterns to the media is
based on both the length of the data pattern and the effective
write pole footprint according to embodiments of the
disclosure.
[0057] When writing data patterns to track 610 using a conventional
approach, the write current is applied continuously over the length
of the data pattern, as is shown in FIG. 6B. For example, to write
a 7T data pattern using a conventional approach, the write current
is applied seven times longer than when writing a 1T data pattern.
This conventional writing approach has three primary drawbacks when
applying the write current continuously for a long data pattern.
The first involves high adjacent track erasure. The second involves
high power consumption. The third involves a strong background
signal.
[0058] For example, using a conventional writing approach, write
current is continuously applied for the entire length of the data
pattern. After writing the last bit (e.g., +1), the next m bits
(e.g., 6 bits) in the downtrack direction also become +1 due to the
effective write pole footprint (of length mT, such as 6T or 7T).
The write pole would need to flip all the media grain magnetized as
+1 when writing -1, for example, which is indicative of a strong
background signal. According to a writing approach of the present
disclosure, because no write current is applied after the first two
bits (assuming and EWPF of 6T or 7T), for example, the background
will be random with both +1 and -1, which makes it much easier for
the write pole to flip the media grain to -1 (e.g., a weak
background signal).
[0059] Individual bits 612 along the track 610 correspond to the
channel clock rate T and are represented by rectangular boxes. The
bits along the track 610 have different statuses during the write
operation as shown. Completed bits (e.g., bits that have been
successfully written by the EWPF 604 of the write pole 600) are
denoted at 620. Actively switched bits (e.g., written bits that are
being switched by the EWPF 604 of the write pole 600) are denoted
at 622 and are shown as a dot-pattern for reference. It can seen in
FIGS. 6A and 6B that, of the switched bits 622, those nearest the
trailing edge of the write pole 600 are fully switched bits. Those
bits under the central aperture 602 and those nearest the leading
edge of the write pole 600 are partially switched bits, shown as an
incomplete area 624. As will be recognized, a bit may include many
magnetic grains in the recording layer. An incomplete bit is one in
which not all of the grains have been completely switched from
positive to negative or negative to positive since the bit is not
fully covered by the EWPF 604 of the write pole 600. However, it is
not necessary for all bits to be completed bits for long T data
patterns, because no write current is applied after the first two
bits (assuming an EWPF of 6T or 7T). Old bits are denoted at 628
and represent old data previously written during a previous write
operation and which are about to be overwritten by the present
write operation. Each of the bits 612 will nominally have a
selected magnetization (e.g., into or out of the page) based on the
direction of write current. Flux transitions will occur at certain
bit boundaries in the manner described above (see e.g., FIG.
3).
[0060] From FIGS. 6A and 6B, it can be seen that the size of the
EWPF 604 of the write pole 600 is such that multiple bits (e.g.,
the switched bits 622) are switched concurrently. As illustrated,
switching the first bit at the trailing edge of the EWPF 604 causes
concurrent switching of the next seven bits. Accordingly, and
assuming a sufficiently large EWPF, the arrangement of FIGS. 6A and
6B may allow the intermediate value of current to be applied for up
to seven clock periods (seven bits 612) prior to each transition on
the basis that these bits will have already been magnetized to the
correct magnetization orientation. For longer symbol lengths, such
as a 12T symbol, the EWPF 604 is used as an interval for applying
one or more additional write currents, which is discussed in
greater detail hereinbelow. Such longer symbol lengths can also
have one or more intermediate bits 612 with the intermediate
current value (e.g., 0 mA) since these bits will also have been
already magnetized to the correct orientation. Using the effective
write pole footprint as an interval and intermediate current values
during the writing of relatively longer symbols provides a number
of beneficial effects, including reduced power and heat
dissipation, reduced background signal, reduced coil saturation,
minimized adjacent track erasure, etc.
[0061] FIG. 7 illustrates a method for measuring the effective
footprint of a write pole in accordance with various embodiments.
The method shown in FIG. 7 involves writing 702 a signal pattern
having a length of nT to a number of first sectors of a magnetic
storage disc, where n is an integer and the write current is
continuously applied. In some approaches, the length of the signal
pattern can be lower than or greater than the expected length of
the effective footprint of the write pole. For example, it may be
known in the factory that a write pole of a particular head design
has an effective footprint that generally varies between 5 and 7T.
As such, the signal pattern of block 702 can have a length of 3 or
4T or, alternatively, a length of 8 or 9T. In other approaches, the
length of the signal pattern used in block 702 can be the same as
the expected effective write pole footprint (e.g., 6T) of a given
head design.
[0062] The method involves collecting 704 a first readback signal
from the first sectors, which is preferably converted from analog
form to digital form. The amplitude and shape of the first readback
signal represents a first writeability metric. The method shown in
FIG. 7 also involves writing 706 a number of second sectors with
the same nT signal pattern (same length as in blocks 702, 704) by
only applying write current to the 1.sup.st T. No write current is
applied after the 1.sup.st T. The method further involves
collecting 708 a second readback signal from the second sectors,
which is preferably converted from analog form to digital form. The
amplitude and shape of the second readback signal represents a
second writeability metric.
[0063] The method involves calculating 710 a correlation
coefficient (R) using the first and second writeability metrics. It
is noted that if both the amplitude and shape of the first and
second readback signals are the same, then the signal pattern
written to the first and second sectors is the same, resulting in a
correlation coefficient (R) of 1. If the amplitude and/or shape of
the first and second readback signals differ, then the signal
pattern written to the first and second sectors is different,
resulting in a correlation coefficient (R) of less than 1. The
correlation coefficient (R) is compared to a threshold (e.g.,
0.98). If the correlation coefficient (R) is greater than the
threshold, the effective footprint of the write pole is longer than
the nT signal pattern used to generate the signal pattern. If the
correlation coefficient (R) is not greater than the threshold, the
effective footprint of the write pole is shorter than the nT signal
pattern used to generate the signal pattern.
[0064] Referring to block 712, if the correlation coefficient (R)
is greater than the threshold, the processes of blocks 702-710 are
repeated by increasing the pattern length by 1T until the
correlation coefficient (R) is less than the threshold. When the
correlation coefficient (R) is less than the threshold at block
714, it is determined at block 716 that the previous pattern length
represents the effective write pole footprint. For example, the
second to last iteration of block 714 may result in increasing the
signal pattern length from 7T to 8T. It is assumed that at 7T, the
correlation coefficient (R) is greater than the threshold. At 8T
(last iteration), however, it may be determined that the
correlation coefficient (R) is less than the threshold. As such,
the effective write pole footprint determined at block 716 is the
previous pattern length, 7T. The processes at block 714 seek to
increase the signal pattern length until the signal pattern length
produces a correlation coefficient (R) that is less than the
threshold.
[0065] Referring again to block 712, if the correlation coefficient
(R) is not greater than the threshold, the processes of blocks
702-710 are repeated by reducing the signal pattern length by 1T
until the correlation coefficient (R) is greater than the
threshold. When the correlation coefficient (R) is greater than the
threshold, it is determined at block 720 that the current pattern
length represents the effective write pole footprint. For example,
the last iteration of block 718 may result in reducing the signal
pattern length from 7T to 6T. It is assumed that at 7T, the
correlation coefficient (R) is less than the threshold, and that at
6T the correlation coefficient (R) is greater than the threshold.
As such, the effective write pole footprint determined at block 720
is the current pattern length, 6T. The processes at block 718 seek
to reduce the signal pattern length until the signal pattern length
produces a correlation coefficient (R) that is greater than the
threshold. It is noted that the computations performed in FIG. 7
may be implemented by the preamp 132, the write control circuit
220, or other processor, controller or logic circuit of the R/W
channel 130 (see FIG. 4).
[0066] FIG. 8 is a plot 802 of correlation coefficient (R) for
signal patterns having a length ranging from 6T to 16T. The
correlation coefficient (R) is calculated for first and second
writeability metrics as described in block 702-710 of FIG. 7. It
can be seen in FIG. 8 that the correlation coefficient (R) exceeds
a threshold of 0.98 for signal patterns having a length of 6T, 7T,
and 8T. At a length of 9T, however, the correlation coefficient (R)
falls below the threshold of 0.98. By applying the EWPF measurement
methodology shown in FIG. 7 to the data illustrated in FIG. 8, it
can be seen that the effective write pole footprint for the head
under test has a length of 8T.
[0067] As was previously discussed, the effective write pole
footprint will vary across the stroke as a function of skew between
the outer diameter (OD) and the inner diameter (ID) of the disc.
According to various embodiments, the effective write pole
footprint can be measured on a per-head basis and across the stroke
between the OD and ID of the disc. For example, the effective write
pole footprint can be measured for pre-selected zones from the OD
to the ID of the disc which takes into consideration the effect of
skew angle. Measuring the EWPF for pre-selected zones reduces test
time. Also, there may be little difference in the EWPF among
adjacent zones, such that many zones can share one EWPF. For
example, one EWPF can be used for all data zones rather than
pre-selected zones in some embodiments. Each of these measurements
can be stored in a memory of the HDD and recalled when writing data
to the disc.
[0068] According to some embodiments, 10 zones with equal track
intervals can be selected between the OD and the ID of the disc.
The effective write pole footprint can be measured for each of the
selected zones based on the methodology shown in FIG. 7. Curve
fitting can be used to obtain the effective write pole footprint
for each of the zones, the result of which can be rounded down to
the nearest bit number. For example, the effective write pole
footprint at zone 102 may be calculated from curve fitting as 5.8.
Rounding down to the nearest bit number from 5.8 results in an
effective write pole footprint of 5. Accordingly, an effective
write pole footprint of 5 is used for writing to zone 102. This
procedure can be used for determining the effective write pole
footprint for each head and each zone of the disc.
[0069] FIG. 9 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments. The method shown in FIG. 9 involves providing 902 a
measure (mT, where m is an integer) of the effective footprint of
the write pole. The EWPF measurement may be obtained using the
methodology shown in FIG. 7 for a particular head and zone on the
disc. The method also involves generating 904 a sequence of symbols
to describe a set of write data. Each of the symbols has a length
of nT. The method further involves applying 906 write currents to
the write pole according to both the symbol pattern and the
effective footprint of the write pole to record the sequence of
symbols to a magnetic storage medium. According to some
embodiments, applying a write current comprises applying a write
current and an overshoot for the 1.sup.st T of each symbol.
[0070] FIG. 10 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments. The method shown in FIG. 10 involves providing 1002 a
measure (mT) of the effective footprint of the write pole, such as
by use of the methodology shown in FIG. 7 for a particular head and
zone on the disc. The method also involves generating 1004 a
sequence of symbols to describe a set of write data. Each of the
symbols has a length of nT. According to the embodiment shown in
FIG. 10, the application of write currents to the write pole is
dependent on the length of the data pattern relative to the length
of the effective write pole footprint. For a 1T data pattern, write
current is applied 1006 for the 1.sup.st T. For a data pattern
having a length from 2T to the effective write pole footprint (mT),
write current is applied 1008 only for the 1.sup.st T. For example,
no current (or the intermediate current) is applied after the
1.sup.st T. For a data pattern that is longer than the effective
write pole footprint (mT), write current is applied 1010 for the
1.sup.st T and the effective write pole footprint (mT) is used as
an interval for applying a subsequent write current(s). FIG. 12,
for example, illustrates a representative approach for using the
effective write pole footprint as an interval for applying a
subsequent write current when writing data patterns longer than the
EWPF. According to some embodiments, applying a write current
comprises applying a write current and an overshoot for the
1.sup.st T of each symbol.
[0071] According to various embodiments, writing data patterns of
various lengths involves writing so-called "write triplets" to the
disc. FIG. 11 illustrates a representative write triplet. The write
triplet shown in FIG. 11 includes three components: an overshoot
duration (OSD), an overshoot amplitude (OSA), and a write current
(I.sub.WRT). In general, a write triplet is written at each
transition (symbol boundary). A write triplet is also written at
each positive or negative excursion from the intermediate (e.g.,
zero) current value for symbols longer than the EWPF. In some
cases, a write triplet is written at the positive and negative
excursions from the intermediate (e.g., zero) current value for
symbols longer than the EWPF. Representative examples of writing
data patterns including write triplets are provided in FIGS.
13-15.
[0072] FIG. 12 illustrates a method for enhancing write current
switching efficiencies during data write operations using the
effective footprint of a write pole in accordance with various
embodiments. The method shown in FIG. 12 involves providing 1202 a
measure (mT) of the effective footprint of the write pole, such as
by use of the methodology shown in FIG. 7 for a particular head and
zone on the disc. The method also involves generating 1204 a
sequence of symbols to describe a set of write data. Each of the
symbols has a length of nT. According to the embodiment shown in
FIG. 12, the application of write currents to the write pole is
dependent on the length of the data pattern relative to the length
of the effective write pole footprint. For a 1T data pattern, write
current (WC) plus overshoot (OS) (hereinafter referred to as WC+OS)
is applied 1206 for the 1.sup.st T. For a data pattern having a
length from 2T to the effective write pole footprint (mT, such as
5T), WC+OS is applied 1208 only for the 1.sup.st T. For example, no
current (or the intermediate current) is applied after the 1.sup.st
T. For a data pattern (e.g., 8T) that is longer than the effective
write pole footprint (mT, such as 5T) 1210, WC+OS is applied for
the 1.sup.st T, and the effective write pole footprint (mT) is used
as an interval for applying 1212 one or more subsequent write
currents in the following manner.
[0073] A count is made to determine 1214 the number of remainder
bits after applying the interval of mT (or multiples of the
interval mT). A Quotient is also calculated 1214 by dividing the
data pattern length by the EWPF, and rounding down. If the
remainder=1, write current (WC) is applied 1216 at the
(1+(Quotient-1)*EWPF+remainder)T, where Quotient=(symbol
length)/EWPF rounded down. If the remainder>1, WC+OS is applied
1218 at the (1+(Quotient-1)*EWPF+remainder)T. It is noted that the
remainder and the Quotient may be determined by the preamp 132, the
write control circuit 220, or other processor, controller or logic
circuit of the R/W channel 130 (see FIG. 4).
[0074] For example, assume that mT for the EWPF is 5T and the
symbol length is 6T. In this example, the remainder is 1 (6T-5T)
and the Quotient is 1 (6/5=1.2 rounded down to 1). Since the
remainder of 1=1, write current (WC) is applied 1216 at the
(1+(1-1)5+1)T=2T. As such, write current (WC) is applied at the
2.sup.nd T. In this scenario, WC+OS is applied at the 1.sup.st T
and WC is applied at the 2.sup.nd T. Notably, applying WC at the
2.sup.nd T results in switching the last five bits of the 6T
symbol, which is equivalent to the EWPF of 5T.
[0075] By way of further example, assume that mT for the EWPF is 5T
and the symbol length is 8T. In this example, the remainder is 3
(8T-5T) and the Quotient is 1 (8/5=1.6 rounded down to 1). Since
the remainder of 3 is greater than 1, WC+OS is applied 1218 at the
(1+(1-1)5+1)T=4T. As such, WC+OS is applied at the 4.sup.th T. This
scenario is illustrated in FIG. 13 for the 8T symbol, where the
EWPF=5T. As is shown in FIG. 13, WC+OS is applied at the 1.sup.st T
and the 4.sup.th T. Notably, applying WC+OS at the 4.sup.th T
results in switching the last five bits of the 8T symbol, which is
equivalent to the EWPF of 5T.
[0076] FIG. 13 is a graphical representation of three different
signals 1302, 1304, and 1306, wherein the latter two signals 1304
and 1306 are write current command signals generated in accordance
with various embodiments. The signal 1302 represents a data pattern
comprising symbols with different lengths to be written to a disc.
The signal 1304 represents a write current command signal generated
by a writer having an effective write pole footprint of 5T in
accordance with various embodiments. The signal 1306 represents a
write current command signal generated by a writer having an
effective write pole footprint of 6T in accordance with various
embodiments. Data can be written to a disc according to the
methodologies illustrated in FIGS. 9, 10, and 12 using the write
current command signals 1304 and 1306.
[0077] The signals 1302, 1304, 1306 are plotted against an elapsed
time x-axis and a combined current magnitude y-axis. The signals
1304 and 1306 can take the form of extended frequency modulated
(EFM) signals with alternating levels from -1 to +1 and signal
transitions T(1) through T(6) at symbol boundaries. Seven (7)
symbols having lengths of 1T, 3T, 8T, 5T, 2T, 11T, and 2T are shown
for purposes of illustration, although other symbol lengths can be
used as desired.
[0078] Using a conventional writing approach for writing the signal
1302, a full rail-to-rail current switching operation would be
required to transition between each adjacent pairs of symbols. For
example, positive-to-negative current switching transitions in
signal 1302 are denoted at the transitions T(1), T(3), and T(5) as
the system commences writing the 3T, 5T, and 11T symbols,
respectively. Negative-to-positive current switching transitions
are denoted at the transitions T(1), T(2), T(4), and T(6) to
signify the writing of the 1T, 8T, and 2T symbols and the symbol
(2T) that immediately follows the 11T symbol.
[0079] While operable, it has been found that switching the write
current between the I.sub.MIN and I.sub.MAX current rails can
require a relatively significant amount of rise time and settle
time as the current direction is switched over the full range
between the respective rails (e.g., from -25 mA to +25 mA and vice
versa). This can provide an upper limit to the smallest achievable
symbol size based on the non-instantaneous response characteristics
of the circuit.
[0080] Accordingly, various embodiments configure the storage
device 100 to provide intermediate current values, or levels,
immediately after the 1.sup.st T of each symbol longer than 1T. For
symbols longer than the effective write pole footprint,
intermediate current levels are provided for at least the last m-1
bits, where mT is the effective write pole footprint. The
intermediate (reduced or zero) current values are between the
respective rail current values levels I.sub.MAX and I.sub.MIN. In
FIG. 13 (and FIGS. 14-15), the intermediate current level is 0 mA.
The intermediate current level can be any suitable value between
the respective rail currents. In some embodiments, the I.sub.MAX
and I.sub.MIN values are nominally about .+-.25 mA and the
intermediate value is nominally about 0 mA. The EFM signals 1304,
1306 can thus be considered as a tri-state signal with three
logical values -1 (I.sub.MIN), 0 (intermediate value), and +1
(I.sub.MAX).
[0081] The intermediate current values at the end of each symbol
reduce the overall current swing necessary for the next symbol
boundary. As can be observed from FIG. 13, the transition at T(2)
for writing signal 1302 would require nominally a full 50 mA
current swing, while the same transition for signals 1304, 1306
only requires about half that, or about 25 mA. Using one or more
intermediate values as represented in FIG. 13 can provide
significantly faster current switching and higher data recording
densities. For example, empirical testing has determined that the
rise time can be consistently reduced by about 35% or more using an
intermediate current value as depicted in FIG. 13 (and FIGS.
14-15).
[0082] FIG. 13 shows the application of the write current switching
methodology of FIG. 12 using write current command signal 1304 and
a write pole having an effective footprint of 5T. For the 1T data
pattern, WC+OS is applied for 1T to reach the +1 current level,
returning to the 0 current level after the 1.sup.st T. The next
symbol to be written has a length of 3T. Because the 3T symbol has
a length between 2T and the EWPF of 5T, WC+OS is applied only for
the 1.sup.st T to reach the -1 current level, returning to the 0
current level after the 1.sup.st T. The write current remains at
the 0 current level for the next 2T of the 3T symbol. The next
symbol to be written has a length of 8T. Because the 8T symbol has
a length greater than the EWPF of 5T, WC+OS is applied for the
1.sup.st T to reach the +1 current value, returning to the 0
current level after the 1.sup.st T. A remainder is calculated as
8T-5T=3T, and a Quotient is calculated as (8/5 rounded down)=1.
Since the remainder of 3 is greater than 1, WC+OS is applied at the
4.sup.th T (the (1+(1-1)*5+3)T)) of the 8T symbol. It can be seen
in FIG. 13 that applying WC+OS at the 4.sup.th T results in
switching the last five bits of the 8T symbol, which is equivalent
to the EWPF of 5T.
[0083] The next symbol to be written has a length of 5T. Because
the symbol 5T has a length equal to that of the EWPF, WC+OS is
applied only for the 1.sup.st T to reach the -1 current level,
returning to the 0 current level after the 1.sup.st T. The write
current remains at the 0 current level for the next 4T of the 5T
symbol. The next symbol to be written has a length of 2T. Because
the symbol 2T has a length from 2T to the EWPF, WC+OS is applied
only for the 1.sup.st T to reach the +1 current level, returning to
the 0 current level after the 1.sup.st T. The next symbol to be
written has a length of 11T, which is longer than the EWPF of 5T.
Because the 11T symbol has a length greater than the EWPF, WC+OS is
applied for the 1.sup.st T to reach the -1 current value, returning
to the 0 current level after the 1.sup.st T. A such, the first five
bits (equal to the size of the EWPF) of the 11T symbol are
concurrently switched by application of WC+OS at the 1.sup.st T.
The 11T symbol has a length that includes two intervals of the EWPF
(e.g., 11T-5T-5T=1T), leaving a remainder of 1T. As such, WC+OS is
applied at the 6.sup.th T (1T+5T). With a remainder=1 and a
Quotient=2 (11/5 rounded down), WC is applied at (1+(2-1)5+1)T=7T.
As a result, WC+OS is applied at the 6.sup.th T to reach the -1
current level. WC is applied at the 7.sup.Th T at the -1 current
level, which switches the remaining bits of the 11T symbol, such
that the last 5 bits are switched (equal to the size of the
EWPF).
[0084] FIG. 13 also shows the application of the write current
switching methodology of FIG. 12 using write current command signal
1306 and a write pole having an effective write pole footprint of
6T. For the 1T data pattern, WC+OS is applied for 1T to reach the
+1 current level, returning to the 0 current level after the
1.sup.st T. The next symbol to be written has a length of 3T.
Because the 3T symbol has a length between 2T and the EWPF of 6T,
WC+OS is applied only for the 1.sup.st T to reach the -1 current
level, returning to the 0 current level after the 1.sup.st T. The
write current remains at the 0 current level for the next 2T of the
3T symbol. The next symbol to be written has a length of 8T.
Because the 8T symbol has a length greater than the EWPF of 6T,
WC+OS is applied for the 1.sup.st T to reach the +1 current,
returning to the 0 current level after the 1.sup.st T. A remainder
is calculated as 8T-6T=2T, and a Quotient is calculated as (8/6
rounded down)=1. Since the remainder of 2 is greater than 1, WC+OS
is applied at the 3.sup.rd T (the (1+(1-1)*5+2) of the 8T symbol.
It can be seen in FIG. 13 that applying WC+OS at the 3.sup.rd T
results in switching the last six bits of the 8T symbol, which is
equivalent to the EWPF of 6T.
[0085] The next symbol to be written has a length of 5T. Because
the symbol 5T has a length less than that of the EWPF, WC+OS is
applied only for the 1.sup.st T to reach the -1 current level,
returning to the 0 current level after the 1.sup.st T. The next
symbol to be written has a length of 2T. Because the symbol 2T has
a length from 2T to the EWPF, WC+OS is applied only for the
1.sup.st T to reach the +1 current level, returning to the 0
current level after the 1.sup.st T. The next symbol to be written
has a length of 11T, which is longer than the EWPF of 6T. Because
the 11T symbol has a length greater than the EWPF, WC+OS is applied
for the 1.sup.st T to reach the -1 current value, returning to the
0 current level after the 1.sup.st T. A such, the first six bits
(equal to the size of the EWPF) of the 11T symbol are concurrently
switched by application of WC+OS at the 1.sup.st T. A remainder is
calculated as 11T-6T=5T, and a Quotient is calculated as (11/6
rounded down)=1. Since the remainder of 5 is greater than 1, WC+OS
is applied at the 6.sup.th T (1+(1-1)*5+5) of the 11T symbol. It
can be seen in FIG. 13 that applying WC+OS at the 6.sup.th T
results in switching the last six bits of the 11T symbol, which is
equivalent to the EWPF of 6T.
[0086] FIG. 14 shows the application of the write current switching
methodology of FIG. 12 using write current command signal 1404 and
a write pole having an effective write pole footprint of 5T. The
signal 1402 represents a data pattern comprising symbols with
different lengths to be written to a disc. In FIG. 14, the first
five symbols (1T, 3T, 8T, 5T, 2T) are the same as those shown in
FIG. 13. In FIG. 14, the sixth symbol has a length of 12T. Applying
the methodology of FIG. 12 for write current command signal 1404,
the write current is switched in the same way as that shown in FIG.
13 for the first five symbols (1T, 3T, 8T, 5T, 2T) when using write
current command signal 1304.
[0087] The next symbol to be written has a length of 12T, which is
longer than the EWPF of 5T. Because the 12T symbol has a length
greater than the EWPF, WC+OS is applied for the 1.sup.st T to reach
the -1 current value, returning to the 0 current level after the
1.sup.st T. A such, the first five bits (equal to the size of the
EWPF) of the 12T symbol are concurrently switched by application of
WC+OS at the 1.sup.st T. The 12T symbol has a length that includes
two intervals of the EWPF (e.g., 12T-5T-5T=2T), leaving a remainder
of 2T. As such, WC+OS is applied at the 6.sup.th T (1T+5T). With a
remainder=2 and a Quotient=2 (12/5 rounded down), WC+OS is applied
at the 8.sup.th T ((1+(2-1)5+2)T) to reach the -1 current level,
returning to the 0 current level after the 8.sup.th T. Applying
WC+OS at the 8.sup.Th T at the -1 current level switches the
remaining bits of the 12T symbol, such that the last 5 bits are
switched (equal to the size of the EWPF).
[0088] FIG. 14 also shows the application of the write current
switching methodology of FIG. 12 using write current command signal
1406 and a write pole having an effective write pole footprint of
6T. Applying the methodology of FIG. 12 for write current command
signal 1406, the write current is switched in the same way as that
shown in FIG. 13 for the first five symbols (1T, 3T, 8T, 5T, 2T)
when using write current command signal 1306.
[0089] The next symbol to be written has a length of 12T, which is
longer than the EWPF of 6T. Because the 12T symbol has a length
greater than the EWPF, WC+OS is applied for the 1.sup.st T to reach
the -1 current value, returning to the 0 current level after the
1.sup.st T. A such, the first six bits (equal to the size of the
EWPF) of the 12T symbol are concurrently switched by application of
WC+OS at the 1.sup.st T. The 12T symbol has a length that includes
two intervals of the EWPF (e.g., 12T-6T-6T=0T), leaving a remainder
of 0T. Because the remainder=0, the last of the multiple intervals
of the EWPF is not used to generate a write current. WC+OS is
applied at the 7.sup.th T (1T+6T) to reach the -1 current level,
returning to the 0 current level after the 7.sup.th T. Applying
WC+OS at the 7.sup.th T at the -1 current level switches the
remaining bits of the 12T symbol, such that the last 6 bits are
switched (equal to the size of the EWPF).
[0090] FIG. 15 illustrates an alternative method for enhancing
write current switching efficiencies during data write operations
using the effective footprint of a write pole in accordance with
various embodiments. The method shown in FIG. 15 involves providing
1502 a measure (mT) of the effective footprint of the write pole,
such as by use of the methodology shown in FIG. 7 for a particular
head and zone on the disc. The method also involves generating 1504
a sequence of symbols to describe a set of write data. Each of the
symbols has a length of nT. According to the embodiment shown in
FIG. 15, the application of write currents to the write pole is
dependent on the length of the data pattern relative to the length
of the effective write pole footprint. For a 1T data pattern, WC+OS
is applied 1506 for the 1.sup.st T. For a data pattern having a
length from 2T to the effective write pole footprint, WC+OS is
applied 1508 only for the 1.sup.st T. For example, no current (or
the intermediate current) is applied after the 1.sup.st T. For a
data pattern that is longer than the effective write pole footprint
1510, WC+OS is applied for the 1.sup.st T, and the effective write
pole footprint is used as an interval for applying 1512 one or more
subsequent write currents in the following manner.
[0091] A count is made to determine 1514 the number of remainder
bits after applying the interval of mT (or multiples of the
interval mT). A Quotient is also calculated 1514 by dividing the
data pattern length by the EWPF, and rounding down. A write current
(WC) is applied 1516 from the (1+(Quotient-1)*EWPF+1)T to the
(1+(Quotient-1)*EWPF+remainder)T. For example, assume that mT for
the EWPF is 5T and the symbol length is 12T. In this example, the
remainder is 2 (12T-5T-5T) and the Quotient is 2 (12/5=2.4 rounded
down). The write current (WC) is applied 1516 from the 7.sup.th T
((1+(2-1)*5+1)T) to the 8.sup.th T ((1+(2-1)*5+2)T). As such, WC+OS
is applied at the 1.sup.st T and the 6.sup.th T (1T+5T), and WC is
applied from the 7.sup.th T to the 8.sup.th T.
[0092] FIG. 16 shows the application of the alternative write
current switching method of FIG. 15 using write current command
signal 1604 and a write pole having an effective write pole
footprint of 5T. FIG. 16 also shows the application of the
alternative write current switching method of FIG. 15 using write
current command signal 1606 and a write pole having an effective
write pole footprint of 6T. The signal 1602 represents a data
pattern comprising symbols with different lengths to be written to
a disc. For data patterns less than or equal to the EWPF, the
alternative write current switching method of FIG. 15 implements
the same rules as the method illustrated in FIG. 12. The two
methods differ when the data pattern is longer than the EWPF. As
such, the following discussion is directed to writing the 8T and
12T data patterns shown in FIG. 16.
[0093] For the 8T data pattern shown in FIG. 16 and an EWPF of 5T,
WC+OS is applied for the 1.sup.st T to reach the +1 current level,
and the EWPF is used as an interval for applying subsequent write
currents. The remainder=3 (8T-5T) and the Quotient=1 (8/5=1.6
rounded down). The write current (WC) is applied from the 2.sup.nd
T ((1+(1-1)*5+1)T) to the 4.sup.th T ((1+(1-1)*5+3)T), returning to
the 0 current level after the 4.sup.th T. Applying WC at the
4.sup.Th T at the +1 current level switches the remaining bits of
the 8T symbol, such that the last 5 bits are switched (equal to the
size of the EWPF).
[0094] For the 12T data pattern shown in FIG. 16 and an EWPF of 5T,
WC+OS is applied for the 1.sup.st T to reach the -1 current level,
returning to the 0 current level after the 1.sup.st T. As such, the
first five bits (equal to the size of the EWPF) of the 12T symbol
are concurrently switched by application of WC+OS at the 1.sup.st
T. The EWPF is used as an interval for applying subsequent write
currents. The 12T symbol has a length that includes two intervals
of the EWPF (e.g., 12T-5T-5T=2T), leaving a remainder of 2T. As
such, WC+OS is applied at the 6.sup.th T (1T+5T) to reach the -1
current level. With a remainder=2 and a Quotient=2 (12/5=2.4
rounded down), the -1 write current (WC) is applied from the
7.sup.th T ((1+(2-1)*5+1)T) to the 8.sup.th T (1+(2-1)*5+2),
returning to the 0 current level after the 8.sup.th T. Applying WC
at the 8.sup.Th T at the -1 current level switches the remaining
bits of the 12T symbol, such that the last 5 bits are switched
(equal to the size of the EWPF).
[0095] For the 8T data pattern shown in FIG. 16 and an EWPF of 6T,
WC+OS is applied at the 1.sup.st T to reach the +1 current level,
and the EWPF is used as an interval for applying subsequent write
currents. The remainder=2 (8T-6T) and the Quotient=1 (8/6=1.33
rounded down). The write current (WC) is applied from the 2.sup.nd
T ((1+(1-1)*5+1)T) to the 3.sup.rd T ((1+(1-1)*5+2)T), returning to
the 0 current level after the 3.sup.rd T. Applying WC at the
3.sup.rd T at the +1 current level switches the remaining bits of
the 8T symbol, such that the last 6 bits are switched (equal to the
size of the EWPF).
[0096] For the 12T data pattern shown in FIG. 16 and an EWPF of 6T,
WC+OS is applied for the 1.sup.st T to reach the -1 current level,
returning to the 0 current level after the 1.sup.st T. As such, the
first six bits (equal to the size of the EWPF) of the 12T symbol
are concurrently switched by application of WC+OS at the 1.sup.st
T. The EWPF is used as an interval for applying subsequent write
currents. The 12T symbol has a length that includes two intervals
of the EWPF (e.g., 12T-6T-6T=0T), leaving a remainder of 0T. As
such, WC+OS is applied at the 7.sup.th T (1T+6T) to reach the -1
current level, which switches the remaining bits of the 12T symbol,
such that the last 6 bits are switched (equal to the size of the
EWPF).
[0097] With reference again to FIG. 4, the R/W channel 140 includes
a write control circuit 220 configured to provide write command
signals to the write driver 140 (FIG. 5) so that the write driver
outputs respective rail current and intermediate values through the
write coil. In some embodiments, a tri-state EFM signal (see
signals 1304, 1306, 1404, 1406, 1504, 1506 in FIGS. 13-15) is
generated so that the intermediate values are received by the write
driver as commands for zero (or some other suitable intermediate
value) of current. Other control mechanisms can be utilized, such
as enable/disable signals which are provided in addition to a
"conventional" EFM signal with conventional full-rail transitions.
The enable/disable signals can operate to temporarily disengage
further outputting of current by the write driver.
[0098] The write control circuit 220 can be realized in hardware or
software, or can involve functionality supplied by the controller
102 (FIG. 1) as required. In some cases, the circuit 220 analyzes
the generated symbol sequence in the input encoded data stream and
interjects intermediate values as required to form a modified
symbol sequence which is then output to the write driver. In other
embodiments, conventional input data sequences can be supplied to
the preamp 144 (FIG. 5) and the preamp can be provided with
internal circuitry that performs these functions. The length of the
applied intermediate value (e.g., one bit, multiple bits) can be
the same for all symbol lengths>1T, or the length can vary for
different lengths of symbols.
[0099] While various embodiments have been presented in the context
of rotatable magnetic recording media, other forms of storage media
can be utilized as well, including optical media, magneto-optical
recording media, HAMR media, microwave assisted magnetic recording
(MAMR) media, multi-dimensional media, etc.
[0100] Systems, devices or methods disclosed herein may include one
or more of the features structures, methods, or combination thereof
described herein. For example, a device or method may be
implemented to include one or more of the features and/or processes
above. It is intended that such device or method need not include
all of the features and/or processes described herein, but may be
implemented to include selected features and/or processes that
provide useful structures and/or functionality.
[0101] Various modifications and additions can be made to the
disclosed embodiments discussed above. Accordingly, the scope of
the present disclosure should not be limited by the particular
embodiments described above, but should be defined only by the
claims set forth below and equivalents thereof.
* * * * *