U.S. patent application number 15/858972 was filed with the patent office on 2018-05-03 for diagnostics in tmr sensors.
The applicant listed for this patent is International Business Machines Corporation. Invention is credited to Robert G. Biskeborn, Wlodzimierz S. Czarnecki, Icko E. T. Iben, Hugo E. Rothuizen.
Application Number | 20180120371 15/858972 |
Document ID | / |
Family ID | 60675575 |
Filed Date | 2018-05-03 |
United States Patent
Application |
20180120371 |
Kind Code |
A1 |
Biskeborn; Robert G. ; et
al. |
May 3, 2018 |
DIAGNOSTICS IN TMR SENSORS
Abstract
A computer-implemented method includes, by one or more
processors in electronic communication with a tunneling
magnetoresistive sensor, wherein the tunneling magnetoresistive
sensor is a component of a magnetic storage drive configured to
read magnetic data from a magnetic storage medium, detecting a
short across the tunneling magnetoresistive sensor, measuring a
change in resistance of the tunneling magnetoresistive sensor,
measuring a change in voltage amplitude for the tunneling
magnetoresistive sensor, and dividing said change in voltage
amplitude by said change in resistance to yield a ratio. The
computer-implemented method further includes, responsive to the
ratio being greater than a predetermined ratio threshold,
determining that the short is caused by a magnetic shunt. A
corresponding computer program product and computer system are also
disclosed.
Inventors: |
Biskeborn; Robert G.;
(Hollister, CA) ; Czarnecki; Wlodzimierz S.; (Palo
Alto, CA) ; Iben; Icko E. T.; (Santa Clara, CA)
; Rothuizen; Hugo E.; (Oberrieden, CH) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
International Business Machines Corporation |
Armonk |
NY |
US |
|
|
Family ID: |
60675575 |
Appl. No.: |
15/858972 |
Filed: |
December 29, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15656644 |
Jul 21, 2017 |
9915697 |
|
|
15858972 |
|
|
|
|
15193620 |
Jun 27, 2016 |
|
|
|
15656644 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01R 31/2829 20130101;
G01R 33/098 20130101; G01R 33/0035 20130101 |
International
Class: |
G01R 31/28 20060101
G01R031/28; G01R 33/09 20060101 G01R033/09; G01R 33/00 20060101
G01R033/00 |
Claims
1. A computer system, the computer system comprising: one or more
processing circuits; and one or more computer readable storage
media; wherein said computer readable storage media store
instructions for execution by said one or more processing circuits;
wherein said one or more processing circuits are in electronic
communication with a tunneling magnetoresistive sensor; wherein
said tunneling magnetoresistive sensor is a component of a magnetic
storage drive configured to read magnetic data from a magnetic
storage medium; said instructions for execution by said one or more
processing circuits comprising instructions to: detect a short
across said tunneling magnetoresistive sensor; measure a change in
resistance of said tunneling magnetoresistive sensor; measure a
change in voltage amplitude over a range for said tunneling
magnetoresistive sensor; divide said change in voltage amplitude by
said change in resistance to yield a ratio over said range;
responsive to both: said ratio being greater than said
predetermined ratio threshold where said change in resistance is
more positive than a predetermined resistance drop threshold, and
said ratio being less than said predetermined ratio threshold where
said change in resistance is more negative than said predetermined
resistance drop threshold, determine that said short is caused by a
magnetic shunt; responsive to said ratio being less than said
predetermined ratio threshold, determine that said short is caused
by a dielectric breakdown; responsive to determining that said
short is caused by said dielectric breakdown, limiting a bias
voltage across said tunneling magnetoresistive sensor to no greater
than a voltage limit, said voltage limit being a value effective
for protecting said dielectric breakdown from growing; and
responsive to determining that said short is caused by said
magnetic shunt, set a bias voltage across said tunneling
magnetoresistive sensor to a normal value, said normal value being
effective in the absence of said short; and, wherein: said
predetermined ratio threshold is 1.7; said change in resistance and
said change in voltage amplitude are both negative; wherein said
predetermined resistance drop threshold is -25%; said tunneling
magnetoresistance sensor is a component of a tape drive; said
normal value is a voltage that would have been deemed safe across a
tunnel junction of said tunnel magnetoresistance sensor in the
absence of said magnetic shunt; and said voltage limit is a voltage
less than what would have been deemed safe across said tunnel
junction of said tunnel magnetoresistance sensor in the absence of
said dielectric breakdown.
Description
BACKGROUND
[0001] The present invention relates generally to the field of
magnetic tape readers, and more particularly to diagnosing and
recovering from hardware failures in tunneling magnetoresistive
sensors.
[0002] Tunneling magnetoresistive ("TMR") sensors are
microelectronic devices that are characterized by a change in
electrical resistance in the presence or absence of a magnetic
field. Magnetic storage devices, such as magnetic tape drives and
hard disk drives, rely upon TMR sensors to read data from magnetic
media. Different regions of magnetic media correspond to bits of
data, each of which can apply either of two different magnetic
field states to the TMR sensor. The value of each bit, one or zero,
can be determined electronically by measuring resistance across the
TMR sensor such that one state may be characterized by a relatively
high resistance and the other characterized by a relatively low
resistance.
[0003] As with all microelectronic devices, TMR sensors experience
hardware failures of various kinds. In particular, TMR sensors are
often located at an air bearing surface, which exposes them to
various kinds of external damage. Using software, for example
controller firmware or driver software that operates a tape drive
or magnetic disk drive, engineers can diagnose and mitigate various
failures, thereby allowing devices to continue to function despite
some operational defect at the microelectronic device level. Also
with the aid of software, engineers can diagnose failures in
devices that have been rendered inoperable or subjected to failure
analysis.
SUMMARY
[0004] A computer-implemented method includes, by one or more
processors in electronic communication with a tunneling
magnetoresistive sensor, wherein the tunneling magnetoresistive
sensor is a component of a magnetic storage drive configured to
read magnetic data from a magnetic storage medium, detecting a
short across the tunneling magnetoresistive sensor, measuring a
change in resistance of the tunneling magnetoresistive sensor,
measuring a change in voltage amplitude for the tunneling
magnetoresistive sensor, and dividing said change in voltage
amplitude by said change in resistance to yield a ratio. The
computer-implemented method further includes, responsive to the
ratio being greater than a predetermined ratio threshold,
determining that the short is caused by a magnetic shunt. A
corresponding computer program product and computer system are also
disclosed.
[0005] In an aspect, the computer-implemented method further
includes, responsive to the ratio being less than the predetermined
ratio threshold, determining that the short is caused by at least
one of a dielectric breakdown and a nonmagnetic shunt.
[0006] In an aspect, the computer-implemented method further
includes, responsive to determining that the short is caused by the
magnetic shunt, setting a bias voltage across the tunneling
magnetoresistive sensor to a normal value, wherein the normal value
is effective in the absence of the short.
[0007] In an aspect, the computer-implemented method further
includes, responsive to determining that the short is caused by the
dielectric breakdown, limiting a bias voltage across the tunneling
magnetoresistive sensor to no greater than a voltage limit, wherein
the voltage limit is a value effective for protecting the
dielectric breakdown from growing.
[0008] In an aspect, another computer-implemented method includes,
by one or more processors in electronic communication with a
tunneling magnetoresistive sensor, responsive to detecting an
operational anomaly in the tunneling magnetoresistive sensor,
measuring a first resistance change in the presence of a positive
bias current, measuring a second resistance change in the presence
of a negative bias current. The computer-implemented method further
includes. responsive to at least one of the first resistance change
and the second resistance change being more positive than expected
based on a device geometry for the tunneling magnetoresistive
sensor, returning a probable determination that the operational
anomaly is caused by a short across the tunneling magnetoresistive
sensor. A corresponding computer program product and computer
system are also disclosed.
[0009] In an aspect, another computer-implemented method includes,
by one or more processors in electronic communication with a
tunneling magnetoresistive sensor, wherein the tunneling
magnetoresistive sensor is a component of a magnetic tape drive
configured to read magnetic data from a magnetic tape, detecting a
short across the tunneling magnetoresistive sensor
contemporaneously with the tape running across the tunneling
magnetoresistive sensor, and measuring a voltage amplitude across
the tunneling magnetoresistive sensor as a function of a fractional
current through the tunneling magnetoresistive sensor to yield a
voltage amplitude data set. The computer-implemented method further
includes, responsive to the voltage amplitude data set fitting a
power law, wherein the power law comprises an exponent, and wherein
the exponent is greater than an exponent threshold, determining
that the short is caused by a magnetic shunt. A corresponding
computer program product and computer system are also
disclosed.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] FIG. 1 is a functional block diagram illustrating an
operational environment for a TMR diagnostic program, in accordance
with at least one embodiment of the invention.
[0011] FIG. 2 is a flowchart depicting operational steps for a TMR
diagnostic program, in accordance with at least one embodiment of
the invention.
[0012] FIG. 3 is a flowchart diagram depicting alternative
operational steps for a TMR diagnostic program, in accordance with
at least one embodiment of the present invention.
[0013] FIG. 4 is a flowchart diagram depicting alternative
operational steps for a TMR diagnostic program, in accordance with
at least one embodiment of the present invention.
[0014] FIG. 5A is a plot of resistance versus time in an experiment
demonstrating resistance drop in a TMR sensor that has experienced
dielectric breakdown as compared with several that have not.
[0015] FIG. 5B is a plot of change in voltage amplitude versus
change in resistance for a TMR sensor that has experienced
dielectric breakdown through a pin-hole.
[0016] FIG. 6A is a plot of the change in voltage amplitude at a
constant current versus change in resistance for TMR sensors with
shorts due to dielectric breakdown, wherein a first group of TMR
sensors experienced dielectric breakdown from Electrical Over
Stress (EOS) pulses.
[0017] FIG. 6B is a plot of the change in voltage amplitude at a
constant current versus change in resistance for TMR sensors with
shorts due to dielectric breakdown, wherein a second group of TMR
sensors experienced dielectric breakdown from EOS pulses.
[0018] FIG. 7 is a plot of voltage amplitude versus fractional
current for TMR sensors that have experienced dielectric breakdown
from EOS pulses.
[0019] FIG. 8A is an atomic force microscopy image at 25 .mu.m of a
tape reader that has experienced a lapping scratch.
[0020] FIG. 8B is an atomic force microscopy image at 5 .mu.m of a
tape reader that has experienced a lapping scratch.
[0021] FIG. 8C is a magnetic force microscopy image at 5 .mu.m of a
tape reader that has experienced a lapping scratch.
[0022] FIG. 9A is a plot of resistance versus time for a group of
TMR sensors, one of which has experienced a lapping scratch.
[0023] FIG. 9B is a plot of voltage amplitude versus time for a
group of TMR sensors, one of which has experienced a lapping
scratch.
[0024] FIG. 10A is a plot of change in voltage amplitude at
constant current versus change in resistance due to a short
occurring while running against tape for -40%<.DELTA.R<0.
[0025] FIG. 10B is a plot of change in voltage amplitude at
constant current versus change in resistance due to a short
occurring while running against tape for -60%<.DELTA.R<0.
[0026] FIG. 11A is a plot of fractional current in a first TMR
sensor when a short occurs while tape is running.
[0027] FIG. 11B is a plot of fractional current in a second TMR
sensor when a short occurs while tape is running.
[0028] FIG. 12A is a plot of resistance versus bias current for a
TMR sensor prior to a short.
[0029] FIG. 12B is a plot of resistance versus bias current for a
TMR sensor after a short.
[0030] FIG. 13 is a plot of change in amplitude versus change in
resistance across the TMR sensor for a first group with shorts due
to lapping scratches and a second group with shorts due to
dielectric breakdown.
[0031] FIG. 14A is plot of a distribution of .DELTA.Amp/.DELTA.R
from shorts due to dielectric breakdown with
.DELTA.R.ltoreq.50%.
[0032] FIG. 14B is plot of a distribution of .DELTA.Amp/.DELTA.R
from shorts due to dielectric breakdown with .DELTA.R.ltoreq.50%
and from lapping scratches with .DELTA.R.ltoreq.20%.
[0033] FIG. 14C is plot of a distribution of .DELTA.Amp/.DELTA.R
from shorts due to dielectric breakdown with .DELTA.R.ltoreq.50%
and from lapping scratches with .DELTA.R.ltoreq.65%.
[0034] FIG. 15A is a side profile view schematic depiction of a TMR
sensor as considered in magnetic modeling with respect to at least
one embodiment of the invention.
[0035] FIG. 15B is a plan view schematic depiction of a TMR sensor
as considered in magnetic modeling, with respect to at least one
embodiment of the present invention.
[0036] FIG. 16A is a schematic depiction of a model for magnetic
shielding of a shunt.
[0037] FIG. 16B is a theoretical plot of magnetic field strength
versos material depth for shielded and normal shunts.
[0038] FIG. 17A is a plot of the drop in magnetic field inside of a
TMR sensor with width of magnetic shunt having a thickness of 1 nm,
2 nm, 3 nm, 6 nm and 10 nm.
[0039] FIG. 17B is a plot of the drop in magnetic field inside of a
TMR sensor versus permeability of the magnetic shunt for shunt
widths of 200 nm and 500 nm.
[0040] FIG. 17C is a plot of the drop in magnetic field inside of a
TMR sensor versus the depth of the magnetic shunt for shunt widths
of 200 nm and 500 nm.
[0041] FIG. 18 is a block diagram depicting various logical
elements for a computer system capable of executing program
instructions, in accordance with at least one embodiment of the
present invention.
DETAILED DESCRIPTION
[0042] Referring now to the invention in more detail, FIG. 1
depicts an operational environment for a TMR diagnostic program
101, in accordance with at least one embodiment of the present
invention. Broadly, the TMR diagnostic program 101 may be
understood as a method to recover drive performance following
shoring of a TMR sensor by distinguishing the shorting mechanism
and setting the bias voltage accordingly. In the depicted
embodiment, a magnetic storage control computer 100 is embedded in
a magnetic tape drive. In alternative embodiments the magnetic
storage control computer 100 may be embedded in a magnetic hard or
floppy disk drive or any other magnetic storage medium for which
TMR sensors may be applied as a reading device. In the depicted
embodiment, the magnetic storage medium 115 (i.e., a tape, disk, or
other magnetic data storage material) is proximate to a TMR sensor
109. In general, the TMR (a tunneling magnetoresistive sensor) may
be understood as a component of a magnetic storage drive configured
to read magnetic data from a magnetic storage medium.
[0043] More particularly, with respect to various embodiments, a
TMR sensor is a multi-layer device which include three significant
layers, among others: a pinned layer, a tunnel junction, and a free
layer. The resistance of the TMR depends on the alignment of the
magnetization of the pinned layer versus the free layer. When the
magnetization of the pinned layer and free layer are aligned, the
resistance is lowest. Correspondingly, when the pinned layer and
free layer are anti-aligned, the resistance is highest. To
fabricate a TMR with a reasonable resistance, the tunnel junction
is very thin. For read heads configured to read from tape or hard
disks, the tunnel junction thickness may be .about.1 nm or lower.
Because of the thin tunnel junction layer, shorts across the tunnel
junction can occur, wherein current is shunted through the short
rather than through the TMR layer.
[0044] The inventors have observed and/or recognized three
different shorting mechanisms: (1) dielectric breakdown, wherein
the dielectric material of the TMR sensor collapses in a narrow
region (i.e., a pin-hole), as with an electric arc through air; (2)
lapping scratches, wherein the lead material is bridged across the
TMR sensor, and (3) tape scratches, which occur when running tape,
and wherein the tape carries some form of particulate material
across the TMR sensor, which results in a scratch across the read
element, and, potentially, a conductive short across the thin
tunneling barrier of the TMR sensor, in view of the high contact
pressure of the tape across tape bearing surface. The inventors
have observed and/or recognized that, in all three cases, if the
voltage across the tunnel junction is maintained constant after the
short has occurred, then the voltage amplitude of the TMR sensor in
response from a magnetic field from the tape medial should revert
to its value prior to the short. It should be noted that voltage
amplitude may be understood as the peak-to-peak change in the
resistance of the TMR sensor multiplied by the applied current to
the TMR sensor. Thus, in the case of a conductive short in parallel
with a TMR sensor, the voltage amplitude will be reduced, as
described below. Thus, the inventors have observed and/or
recognized, the TMR can still function after the short. However,
the inventors have observed and/or recognized that a problem with
the short from dielectric breakdown is that high levels of current
passing through the small shorting pillar will likely cause the
size of the pillar to grow, which in turn causes the shorting
resistance to decrease without bound until the drive can no longer
supply sufficient current to operate. The inventors have further
observed and/or recognized that his problem is not present in
shorts that occur due to lapping scratches or tape scratches.
[0045] Thus, the inventors have further observed and/or recognized
that a means of distinguishing between the different sources of
shorting can be employed to enable the continued use of TMR sensors
that have shorted. The inventors have further observed and/or
recognized that the previously known means of distinguishing
between shorts due to dielectric breakdown and shorts due to
lapping or tape scratches include only failure analysis. In failure
analysis, a scratch is visible in an atomic force microscopy image
of the air bearing surface in which the TMR sensor is embedded
(e.g., FIGS. 8A-8B), similarly, under atomic force microscopy
imaging, a TMR having experienced dielectric breakdown will not
have a surface scratch. The inventors have further observed and/or
recognized that there dielectric breakdown can occur at the tape
bearing surface of the TMR, however such an event would be visible
as a dot in an atomic force microscopy or scanning electron
microscopy image, and the imaged shape would be visibly and
diagnostically distinct from that of a scratch. In any event,
failure analysis of TMR sensors requires the TMR sensor that has
experienced a short to be taken out of service and physically
brought to a laboratory setting where it can be examined with the
aforementioned microscopy technologies and other imaging
techniques. Thus, the inventors have further observed and/or
recognized, a diagnostic method based solely on in situ electrical
measurements of the TMR sensor would permit control software to
keep the TMR sensor in service after detecting a short. It will be
understood that the aforementioned problems solved by some
embodiments of the invention and/or advantages over the prior art
exhibited by some embodiments of the invention are not intended as
limitations on the invention as claimed, and any particular
advantage need not necessarily be present in all embodiments.
[0046] Referring still to FIG. 1, in the depicted embodiment, the
magnetic storage medium 115 produces a magnetic field 113
(designated by the magnetic field strength symbol {right arrow over
(H)}). The magnetic field 113 changes in direction and/or magnitude
with the data encoded upon that region of the magnetic storage
medium 109 that is proximate to the TMR sensor 109. As the magnetic
field 113 changes, the magnetization of the free layer in the TMR
sensor 109 rotates, and the resistance of the TMR sensor 109
correspondingly changes. The TMR sensor 109 may be understood as a
resistor 109A, the resistance of which may be measured via a
digital measurement unit 104. The change in the TMR resistance with
applied field times the bias current through the TMR sensor gives a
voltage amplitude. It will be understood that FIG. 1 is a schematic
representation only and that it is not the intent of the Applicant
to suggest that the TMR sensor 109 literally contains a resistor
109A, but rather that the TMR sensor 109 has the property of a
measurable resistance represented by a notional resistor 109A. The
digital measurement 104 unit may be understood to include any
circuit elements, sensors, logic gates, firmware, etc. that enable
electrical measurements of the TMR sensor 109.
[0047] Referring still to the embodiment of FIG. 1, in addition to
resistance, the digital measurement unit 104 may apply a bias
current to the TMR sensor 109. The bias current may be in either
direction across the TMR sensor 109 or it may alternate. The
digital measurement unit 104 may measure the output voltage or
voltage amplitude as a result of applying a defined bias current,
which can include, selectively, direct current in either direction
and alternating current over a range of frequencies. Similarly, the
digital measurement unit 104 may measure resistance simultaneously
with the application of a bias current so as to measure resistance
as a function of changes in the bias current. Further, the digital
measurement unit 104 may measure resistance repetitiously or
continuously such that various effects can be observed as a
function of changes in resistance.
[0048] Referring still to the embodiment of FIG. 1, the magnetic
storage control computer 104 is in electronic digital communication
with the digital measurement unit 104. For example, the digital
measurement unit 104 may include one or more analog-digital
converter circuits whereby the electrical properties of the TMR
sensor 109 are made accessible as digital values to software
operating on the magnetic storage control computer 100, such as the
TMR diagnostic program 101. The TMR diagnostic program 101 may be
understood as performing various electrical measurements of the TMR
sensor 109 by accessing the digital measurement unit.
[0049] More generally, the electrical measurements of the TMR
sensor 109 may be understood in terms of several relationships,
which the inventors have identified. For a tunnel junction with a
resistance R.sub.mro and a parallel short with a resistance
R.sub.s, current is diverted through R.sub.s, and overall
resistance R for the tunnel junction is:
R = R mro R s ( R s + R mro ) Equation 1 ##EQU00001##
[0050] The change in resistance .DELTA.R.sub.TMR for the same
tunnel junction is:
.DELTA. R TMR = 100 % ( R - R mro ) R mro = - 100 % R mro ( R s + R
mro ) Equation 2 ##EQU00002##
[0051] For a constant current of I.sub.mro applied to the TMR
sensor, the current through the tunnel junction, I.sub.mr, is
decreased by the ratio,
R R mro : ##EQU00003##
I mr = V R mro = I mro R R mro = I mro R s ( R s + R mro ) Equation
3 ##EQU00004##
[0052] When a magnetic field of strength H.sub.f is applied to a
TMR sensor, the change in resistance, .DELTA.R.sub.TMRH, is
proportional to H.sub.f and is a fraction F.sub.TMR of the
resistance of the TMR sensor:
.DELTA.R.sub.TMRH=F.sub.TMRH.sub.fR.sub.mro Equation 4
[0053] The voltage amplitude Amp is the current times the
resistance change, Amp.sub.o is the voltage amplitude at a current
of I.sub.mro prior to the short:
Amp = I mr F TMR H f R mro = I mro F TMR H f R mro R R mro = Amp o
R R mro Equation 5 ##EQU00005##
[0054] Thus, under constant current, the change in voltage
amplitude .DELTA.Amp is:
.DELTA. Amp = 100 % ( Amp - Amp o ) Amp o = .DELTA. R Equation 6
##EQU00006##
[0055] Thus, the change in amplitude, expressed as a percentage
(or, as a ratio, if the factor of 100% is removed), if equal to the
change in resistance, also expressed as a percentage (or, as a
ratio, if the factor of 100% is removed), and thus:
.DELTA. Amp .DELTA. R = 1 Equation 7 ##EQU00007##
[0056] Equation 7 provides a theoretical prediction for the ratio
of the change in voltage amplitude to the change in resistance,
however the inventors have observed, by experiment, that the
theoretical value is not realized in practice. FIG. 5A is a plot of
TMR resistance versus time for five TMR sensors, one of which has
suffered dielectric breakdown during a constant voltage stress
experiment. FIG. 5A shows how the short continues to grow for the
TMR sensor which has experienced dielectric breakdown, as evidenced
by the change in resistance continuing to drop over time. The
inventors have further observed and/or recognized that, if the
voltage stress on the TMR sensor with a dielectric breakdown is
decreased sufficiently, then the short will stop growing and the
amplitude will remain stable.
[0057] FIG. 5B is a plot of .DELTA.Amp versus .DELTA.R.sub.TMR with
the predicted line of Equation 7. The cluster of low values on the
left shows how, after dielectric breakdown occurs, if the voltage
remains high across the shorted TMR sensor, then the size of the
shorting pillar grows and the amplitude continues to drop. This
leads to failure of the TMR sensor. In FIG. 5B, the dotted line
shows the ideal Equation 7, with a slope of 1, and is the value
determined when fitting the data when forcing
.DELTA. R .DELTA. Amp = 0 ##EQU00008##
at R=0. The lighter line shows a linear regression of the data
points with a slope of 1.56.
[0058] FIG. 6A is a plot of the change in voltage amplitude at a
constant current versus change in resistance for TMR sensors with
shorts due to dielectric breakdown for group of 20 TMR sensors that
experienced dielectric breakdown from EOS pulses over a short time
(.about.25 ns-100 ns). The slope (i.e., average) of
.DELTA.Amp/.DELTA.R.sub.TMR was 1.27.+-.0.05. FIG. 6B is a similar
plot for three TMRs that experienced dielectric breakdown from EOS
over a long time (.about.10 s-days). The slope of
.DELTA.Amp/.DELTA.R.sub.TMR was 1.38.+-.0.08. These results are 27%
and 38% higher, for FIGS. 6A and 6B, respectively, than for the
ideal case of Equation 7. The inventors hypothesize that the higher
slopes are explained by and/or due to the loss in area in the TMR
due to the short.
[0059] FIG. 7 is a plot of amplitude versus fractional current
through the same TMR sensors whose measurements are presented in
FIG. 6B. For the ideal case of a shorting resistor in parallel with
the TMR sensor, the slope of Amplitude (Amp) versus current through
the TMR (ITMR) is expected to be linear with a slope of unity (as
in Equation 7). Data for all three shorted TMR sensors was fit with
a linear equation with a slope of 1.38.+-.0.09%/% and a zero
intercept of -0.28.+-.0.09%/%, as shown.
[0060] In all of the cases studied by the inventors, the dielectric
breakdown shorts generated by EOS pulses resulted in amplitude
drops which are similar to, but higher than expected from an ideal
parallel short model. The amplitude versus the effective current
through the shorted TMR sensor can be fit with a linear equation
with a slope of between about 1 to 1.6%/%. The change in amplitude
versus change in resistance can be fit with a slope of 1.3%/%
within a range of .+-.0.3%/%.
[0061] In addition to dielectric breakdown, lapping and tape
scratches are also a cause of shorts in TMR sensors. FIG. 8A is an
atomic force microscopy image at 25 .mu.m of a tape reader that has
experienced periodic scratches from a particle on the tape. The
spacing between the scratches is the physical distance the tape is
stepped over between wraps of the tape (also known in the art as
"track pitch"). FIG. 8B is an atomic force microscopy image at 5
.mu.m of the same tape reader. FIG. 8C is a magnetic force
microscopy image at 5 .mu.m of the same tape reader. In the
experiment shown, the scratches were 15.8 nm deep, 2.02 .mu.m
spacing between, and 0.57 .mu.m wide. As shown, the scratch
features demonstrate how a particulate embedded in the tape can be
dragged across the TMR sensor at the air bearing surface, causing a
short.
[0062] FIG. 9A is a plot of resistance versus time for a group of
TMR sensors, one of which suffered a short due to a particulate on
the tape scratching the surface of the TMR sensor and dragging
metal material across the tunnel junction, causing a short in
parallel with the bulk of the TMR. FIG. 9B is a plot of the
2T-Amplitude from a read-back signal from magnetic transitions
written on the tape. 2T is a fundamental period of data density for
a given tape. The resistance dropped by -15% while the Amplitude
dropped by 43%, or a ratio of
.DELTA. Amp .DELTA. R TMR ##EQU00009##
of 2.9. Two important observations are: (1) The ratio of
.DELTA. Amp .DELTA. R TMR ##EQU00010##
of 2.9 is almost 3 times the expected value for a simple parallel
short, and (2) the resistance did not continue to drop in this
case, so the voltage can be increased to yield higher output.
[0063] FIGS. 10A and 10B are a plots of the change in voltage
amplitude at a constant current versus change in resistance for TMR
sensors with shorts due to scratches across the tunnel junction
from tape wear. For FIG. 10A, the relevant range is
-40%<.DELTA.R<0. For FIG. 10B, the relevant range is
-60%<.DELTA.R<0%.The observed change in resistance due to the
short was between -10% and -25%. The inventors concluded that the
reason for the range in .DELTA.R.sub.TMR is because the TMR studied
suffered multiple scratches, with the resistance dropping with each
additional scratch. The inventors observe that the slope,
.DELTA. Amp .DELTA. R TMR , ##EQU00011##
is 1.74 for these parts, is significantly higher than the predicted
value of 1 for a simple parallel short, and larger than the values
measured from dielectric breakdown. The inventors further observe
that the results are essentially identical for 2T and 8T
amplitudes, and thus they conclude that the additional amplitude
drop is not due to Wallace spacing losses. Further, for
.DELTA.R>-40% (i.e., for Amplitude losses of less than 40%), the
drop in voltage amplitude with change in resistance is higher than
calculated from a simple parallel resistance shunt, and for
.DELTA.R<-40%, (i.e., for Amplitude losses of greater than 40%),
the slope in
.DELTA. Amp .DELTA. R ##EQU00012##
decreases.
[0064] FIGS. 11A and 11B are plots of normalized 2T and 8T
amplitudes versus fractional current through the shorted TMR
sensor, where the shorts are due to scratches on tape, as in FIG.
10. The data is fit to a power law, with an exponent .beta. of 2.0
and 2.3:
I TMR = I 0 R R TMR 0 Equation 8 Amp Norm = ( I TMR I 0 ) .beta. =
I TMR Norm .beta. Equation 9 ##EQU00013##
[0065] In Equations 8 and 9 above, R is the resistance with the
short, R.sub.TMR is the initial resistance of the TMR sensor,
Amp.sub.Norm is the amplitude normalized to the TMR's pre-short
value, I.sub.0 is the bias current applied to the sensor in
measuring the amplitude, and ITMR is the current flowing through
the TMR sensor in the presence of the short. ITMR.sub.Norm is the
current through the TMR sensor normalized to I.sub.0. In the ideal
case, .beta. is 1.0.
[0066] The slope of Amp.sub.Norm versus ITMR.sub.Norm varies with
ITMR.sub.Norm:
.DELTA. Amp Norm .DELTA. TMR Norm = .beta. I .beta. - 1 Equation 10
##EQU00014##
[0067] The slope is maximum at ITMR.sub.Norm=1, and equal to 1. For
the parts studied by the inventors, the slope is between 2 and 3
for ITMR.sub.Norm greater than about 0.8, whereas the slope is 1
for the ideal model of a shorting resistor.
[0068] FIG. 12A is a plot of resistance versus bias current for a
TMR sensor prior to a short. As shown, the resistance decreases
with bias current for both positive and negative polarity, which is
the expected behavior for a TMR device, and TMR sensors without a
short have been observed to decrease resistance with increasing
voltage across the TMR sensor. Resistance is therefore expected to
decrease monotonically for at least one polarity. FIG. 12B shows
resistance increasing after a short for both polarities. The
increase in resistance is understood by the inventors to be due to
joule heating because of the principle that the resistance of a
metal increases with its temperature. Therefore, the increase in
resistance indicates that a thin short brides across the tunnel
junction and is heating up due to the large current density through
the short. While the resistance of a tunnel junction will decrease
with increasing voltage across it, the resistance of a metal will
increase. The inventors have observed and/or recognized that this
technique of determining a short may be applicable if the history
of a given TMR is not known (e.g., if it is installed in a newly
manufactured and unused read head), and therefore it is not
possible to take a difference between a current and a previously
measured resistance value. The inventors have further observed
and/or recognized that the aforementioned method may also be
applicable in distinguishing between a drop in resistance due to a
magnetically induced change and a physical short. If a short is
detected by a drop in measured resistance, but the change in
resistance versus bias current is normal, then it is likely a
magnetically induced change in the sensor and not physically
induced short such as a scratch or dielectric breakdown.
[0069] FIG. 13 compares the properties of shorts from drive
scratching with those from dielectric breakdown. FIG. 13 is a plot
of change in voltage amplitude versus change in TMR resistance for
a first group of TMR sensors with shorts that occurred in a drive
due to scratches and a second group of TMR sensors that
intentionally experienced dielectric breakdown. As shown,
dielectric breakdowns, denoted by the white squares, are shown to
have a linear profile over a range of tested .DELTA.R, while the
scratches, denoted as black dots, have a nonlinear profile in
accordance with the above-described equations. The white squares
are short from the drive of FIGS. 9A and 9B. Also shown is a fit to
shorts due to scratches using the concept of shielding at the ABS
due to magnetic material in the shunt, as discussed below. The
dotted line fits the change in amplitude versus resistance for a
resistive shunt combined with a magnetic shunt.
[0070] FIGS. 14A, 14B, and 14C demonstrate the diagnostic value of
the
.DELTA. Amp .DELTA. R ##EQU00015##
ratio. FIG. 14A shows the distribution of
.DELTA. Amp .DELTA. R ##EQU00016##
from shorts due to dielectric breakdown taken from FIG. 13. FIG.
14B introduces
.DELTA. Amp .DELTA. R ##EQU00017##
for scratches, which is generally and diagnostically greater than
for dielectric breakdown, where the data is from FIG. 13 for parts
with .DELTA.R<10%. FIG. 14C shows the distribution of
.DELTA. Amp .DELTA. R ##EQU00018##
from shorts due to dielectric breakdown and from Scratches with
.DELTA.Amp.ltoreq.65%, where the data is again taken from the data
in FIG. 13. 93% of the parts with dielectric breakdown have a
.DELTA. Amp .DELTA. R < 1.7 , ##EQU00019##
and 100% of the scratches had a
.DELTA. Amp .DELTA. R > 1.7 . ##EQU00020##
Thus the value of
.DELTA. Amp .DELTA. R ##EQU00021##
is diagnostic for determining whether a short is from a scratch or
dielectric breakdown.
[0071] The inventors have further studied a magnetic model related
to the present invention. FIGS. 15A and 15B show the geometry of
the modeled tape reader 1500. FIG. 15A is a side profile view of
the modeled TMR sensor. FIG. 15B is a plan view of the same device.
Conductive shields 1501 and 1502 are opposed across a gap, which is
bridged by a shunt 1506. This attempts to model a tape scratch
and/or lapping scratch wherein conductive material is dragged
across the reader layers. The free layer 1504 sits in between the
shields and provides its properties into the model. FIG. 16A models
the shields 1501 and 1502 as R.sub.0 and R.sub.s. FIG. 16B
describes the expected H field behavior for a shielded and
unshielded reader. The shield effects in the depicted model are as
follows:
R = R 0 R s ( R s + R 0 ) Equation 11 .DELTA. R = ( R - R 0 ) R 0 =
- R 0 R s + R 0 Equation 12 I mr = V R 0 = I b 0 R R 0 Equation 13
Signal : .DELTA. R amp = A s R 0 H F 2 Equation 14 Amp = I mr
.DELTA. R amp = I mr A s R 0 = I b 0 H F 2 A s R = ( H F H N ) Amp
0 R R 0 Equation 15 .DELTA. Amp = ( Amp - Amp 0 ) Amp 0 = ( H F H N
) ( 1 + .DELTA. R ) - 1 Equation 16 ##EQU00022##
[0072] In the case of a simple short (i.e., no magnetic shielding,
H.sub.F =H.sub.N, and no magnetic shielding occurs:
If H F = H N : .DELTA. Amp .DELTA. R = 1 Equation 17
##EQU00023##
[0073] In the case where a magnetic shielding is present, H.sub.F
=Hs <H.sub.N, magnetic shielding occurs. Thus, the signal is
decreased even further as compared with a simple electrical
short.
If H F = H S < H N : .DELTA. Amp = ( 1 - ( H S H N ) ) + ( H S H
N ) .DELTA. R = ( 1 - H S H N ) + H S H N .DELTA. R Equation 18
##EQU00024##
[0074] The above analytical model includes a both an electrical
short and a decrease in the magnetic field reaching the sensor due
to a magnetic shielding effect. This shows that, if the magnetic
field is somehow decreased due to magnetic shielding, then with
shielding, the ratio
.DELTA. Amp .DELTA. R ##EQU00025##
will be larger (as per Equation 18) than in the case of a simple
conductive shunt (as per Equation 17).
[0075] Micromagnetic calculations were then done using a finite
element model (FEM) including a magnetic shunt across the shields
1501 to 1502.
[0076] FIG. 17A shows the drop in magnetic field inside a TMR
sensor versus the width of the magnetic shunt for shunt thickness
of 1 nm, 2 nm, 3 nm, 6 nm, and 10 nm. It should be noted that the
sensor is 1500 nm wide, which is the same value used for the
sensors for the experiments on scratching and dielectric breakdown.
The magnetic field in the sensor drops essentially linearly with
width of the magnetic shunt. FIG. 17B shows the drop in magnetic
field inside the TMR sensor versus the depth (thickness) of the
magnetic shunt for shunt widths of 200 nm and 500 nm. Note that the
magnetic field loss saturates at around a 3 nm thick magnetic
shunt. This model explains very well the drop in amplitude versus
shunt resistance shown in FIG. 13. FIG. 17C shows a plot of the
magnetic field inside TMR sensor versus permeability of the
magnetic shunt for shunt widths of 200 nm and 500 nm.
[0077] More generally, a short due to a scratch from a particulate
on the tape is expected to result in a drop in amplitude and the
ratio of
.DELTA. Amp .DELTA. R ##EQU00026##
is significantly higher than the value of 1 expected for a simple
parallel short across the tunnel junction which shunts current away
from the TMR sensor and through the short. By contrast, a short due
to a dielectric breakdown across the tunnel junction is expected to
result in a drop in amplitude, wherein
.DELTA. Amp .DELTA. R ##EQU00027##
is close to the value of 1 expected for a simple parallel short.
For a TMR sensor in a tape drive, when
.DELTA. Amp .DELTA. R .gtoreq. dAdr Limit , ##EQU00028##
then the short is most likely cause by a scratch from a particle on
the tape. When
.DELTA. Amp .DELTA. R < dAdr Limit , ##EQU00029##
then the short is most likely due to dielectric breakdown. A first
choice for the limit, dAdR.sub.Limit, is 1.7. A second choice of is
dAdR.sub.Limit 2.0. For values of
.DELTA. Amp .DELTA. R ##EQU00030##
below 1.7, a magnetic shunt is unlikely. For values of
.DELTA. Amp .DELTA. R ##EQU00031##
between 1.7 and 2.0, dielectric breakdown is unlikely. In the case
of a scratch, the voltage across the TMR sensor can be increased to
the limits for unshorted TMRs, while in the case of dielectric
breakdown, the voltage limit should be set low enough to avoid
growth of the shorting pillar.
[0078] The drop in amplitude resulting from shorts across the
tunnel junction at the air bearing surface causes by scratches at
the air bearing surface from running tape drops more rapidly than
expected from a parallel short. For .DELTA.R.sub.TMR above about
-20%, the measured slope of .DELTA.Amp versus resistance were all
greater than 1.7, and several were between 2 and 3. Also, a plot of
amplitude versus fractional current through the TMR sensor with a
parallel short were fit with a power law with an exponent between 2
and 2.3 rather than a linear curve as expected for the ideal
case.
[0079] Referring now to the embodiment depicted in FIG. 2, FIG. 2
is a flowchart diagram for a TMR diagnostic program 101, in
accordance with at least one embodiment of the present invention.
In the depicted embodiment, at step 200, the TMR diagnostic program
101 detects a short across the TMR sensor 109. The TMR diagnostic
program 101 may detect a short by monitoring resistance across the
TMR sensor 109 (measurement of resistance may generally include a
measurement with respect to the same device with a known resistance
at an earlier time) and observing a sudden drop in resistance,
which indicates that a new electric pathway (i.e., a short) has
been made in the partially dielectric material of the TMR sensor
109. At step 210, the TMR diagnostic program 101 measures the
change in resistance of the TMR sensor 109. The change in
resistance is computed by taking the difference between the
measured resistance after the short and the original resistance
before the short. Normally, the change in resistance will have a
negative sign, since the shunt normally causes resistance to
drop.
[0080] Referring still to the embodiment depicted in FIG. 2, at
step 220, the TMR diagnostic program 101 measures the change
voltage amplitude for the TMR sensor 109. The change in voltage
amplitude is computed by taking the difference between the measured
voltage amplitude after the short and the original voltage
amplitude before the short. Normally, the change in voltage
amplitude will have a negative sign, as shown in FIG. 13.
[0081] Referring still to the embodiment depicted in FIG. 2, at
step 240, the TMR diagnostic program divides the change in voltage
amplitude by the change in resistance to yield a ratio. As in
Equation 7, the theoretical expected value of the ratio is 1,
however, as described above, the observed ratio is greater than
one, and the inventors have observed and/or recognized that the
value of the ratio is, at least in part, diagnostic as to the cause
of failure in
[0082] TMR sensors.
[0083] At decision block 260, the TMR diagnostic program 101 tests
whether the ratio is greater than a limit. More generally, in the
depicted embodiment, responsive to the ratio being greater than a
predetermined threshold (decision block 260, YES branch), the TMR
diagnostic program 101 determines, at step 280, that the short is
caused by a magnetic shunt, for example by a tape scratch and/or
lapping scratch. An exemplary value for the predetermined ratio is
1.7, wherein the inventors have observed and/or recognized that the
method is diagnostic for some TMR sensors, in at least one
embodiment of the invention. In alternative embodiments, functional
values for the predetermined ratio for other TMR configurations,
for example having different geometries, can be determined by
engineers without undue experimentation, according to the models
described herein. In addition to the ratio, the TMR diagnostic
program 101 may require that the change in resistance and the
change in voltage amplitude both be negative. By contrast, in the
depicted embodiment, responsive to the ratio being less than the
predetermined ratio threshold (decision block 260, NO branch), or,
additionally, responsive to the either the change in resistance or
the change in voltage amplitude being positive, the TMR diagnostic
program 101 determines at step 285 that the short is caused by at
least one of a dielectric breakdown and a nonmagnetic shunt.
[0084] Stated differently, in the embodiment depicted in in FIG. 2,
for a TMR sensor used in a tape drive to read magnetic data from
the tape, measure the change in the resistance, .DELTA.R.sub.TMR,
and the change in the amplitude, .DELTA.Amp, of a TMR sensor used
in a tape drive from their initial values, and take the ratio,
.DELTA. Amp .DELTA. R TMR , ##EQU00032##
and if .DELTA.Amp and .DELTA.R.sub.TMR are both negative and their
ratio is larger than a given value, dAdR.sub.Limit, then the
shorting mechanism is a magnetic shunt from a scratch, and if the
ratio is less than dAdR.sub.Limit, then the short is due to
dielectric breakdown or a non-magnetic shunt, where an exemplary
value of dAdr.sub.Limit is 1.7.
[0085] Referring still to the embodiment depicted in FIG. 2,
responsive to determining that the short is caused by a dielectric
breakdown (step 285), the TMR diagnostic program 101, at step 295,
limits a bias voltage across the TMR sensor to no greater than a
voltage limit. The voltage limit is set at a value that is
effective for protecting the dielectric breakdown from growing. An
exemplary voltage limit for various contemplated embodiments is
between 175 mV and 200 mV.
[0086] Referring still to the embodiment depicted in FIG. 2,
responsive to determining that said short is caused by a magnetic
shunt (step 280), the TMR diagnostic program 101, at step 290, sets
a bias voltage across the TMR sensor to a normal value, which is
effective in the absence of the short. An exemplary normal value
for contemplated embodiments is between 200 mV and 300 mV.
[0087] Stated differently, if the cause of the drop in amplitude
and resistance is determined to be due to a magnetic shunt, then
the bias voltage across the TMR tunnel junction may be set to a
value as would have been deemed safe (for example, by one of skill
in the art for a particular embodiment) for a TMR tunnel junction
prior to the development of the magnetic shunt. Similarly, if the
cause of the drop in amplitude and resistance is determined to be
due to a dielectric breakdown, then the bias voltage across the TMR
tunnel junction may be set to a value which is lower than was
previously deemed safe for a TMR sensor which had not undergone
dielectric breakdown, and which is safe for a TMR sensor of the
area and thickness of the TMR that has undergone dielectric
breakdown.
[0088] In alternative embodiments, the TMR diagnostic program 101
measures the change in voltage amplitude as per step 220 over a
range of values of the change in resistance. Correspondingly, at
step 240, the TMR diagnostic program 101 determines the ratio over
the range. Thus, in such embodiments, responsive to both: (i) the
ratio is greater than the predetermined ratio threshold where the
change in resistance than a predetermined drop threshold; and (ii)
the ratio is less than the predetermined ratio threshold where the
change in resistance is more negative than the predetermined
resistance drop threshold, the TMR diagnostic program 101
determines that the short is caused by the magnetic shunt. An
exemplary value for the predetermined drop threshold is -25%. In
physical terms, if the voltage amplitude fails to continue to drop
as fast as resistance across the TMR sensor, then it is possible to
conclude that the shunt is not getting worse, and is therefore a
magnetic shunt, and not a dielectric breakdown or nonmagnetic
shunt.
[0089] Referring now to FIG. 3, FIG. 3 is a flowchart diagram of a
TMR diagnostic program 101, in accordance with at least one
embodiment of the present invention. For the depicted embodiment,
the method may be understood to be performed by one or more
processors in electronic communication with a TMR sensor. In one
embodiment, the TMR sensor is installed in the read head of a
magnetic tape drive. In another embodiment, the TMR sensor is
installed in the read head of a hard disk drive. More generally,
the TMR sensor may be a component of a magnetic storage drive
configured to read magnetic data from a magnetic storage medium. In
the depicted embodiment, at step 300, the TMR diagnostic program
101 detects an operational anomaly. An operational anomaly may
include a general drop in resistance (according to any type of
measurement) across the TMR sensor, as compared with an earlier
time for the same TMR sensor or a TMR sensor of similar geometry.
At a higher level, an operational anomaly may be detected in read
errors or reduced performance of the reading device of which the
TMR sensor is a component. For example, in a tape or hard disk
drive, automated error correction may require multiple rereadings
of the same data due to the operational anomaly, which can manifest
in repetitive drive activity and delayed access times to data on
disk or tape.
[0090] With reference to FIGS. 12A and 12B, the TMR diagnostic
program 101 may, responsive to detecting the operational anomaly,
detect a short using positive and negative bias currents. In a TMR
sensor undamaged by a short, as per FIG. 12A, resistance is
expected to decrease in the presence of a bias current with either
polarity, however each polarity may behave differently with
resistance dropping in differing amounts depending on the polarity
of the bias current. By contrast, FIG. 12B shows resistance
increasing with bias current in the presence of a short by joule
heating, according to the principle that resistance in ordinary
conductive materials increases with temperature. FIG. 12B shows an
extreme case where an actual increase in resistance is observed
with bias current, however more generally, a change in resistance
that, even if negative, is more positive than expected is
diagnostic of a short. The expected change in resistance can be
modeled for a given device geometry. In the context of an
implemented device in a tape drive, the expected device properties,
including the expected resistance change under bias current of
varying magnitude and polarity, can be known at design time and
made available as parameters to the TMR diagnostic program 101.
[0091] More particularly, the TMR diagnostic program 101 may
diagnose a short using bias currents by applying a first positive
bias current across the TMR sensor, measuring a first positive
resistance in the presence of the first positive bias current,
applying a second positive bias current across the TMR sensor,
measuring a second positive resistance change in the presence of
the second positive bias current, determining a positive resistance
change based on the first positive bias current and the second
positive bias current (e.g., by computing the difference between
them), applying a first negative bias current across the TMR
sensor, measuring a first negative resistance change in the
presence of said first negative bias current, applying a second
negative bias current across the TMR sensor, measuring a second
negative resistance in the presence of the second negative bias
current, determining a negative resistance change based on the
first negative bias current and the second negative bias current
(e.g., by computing the difference between them), and, responsive
to at least one of the positive resistance change and the negative
resistance change being more positive than a predetermined short
detection limit, determining that a short has been detected.
[0092] The predetermined short detection limit may be set based on
the device properties and the desired conservativeness, according
to engineering considerations. The first positive bias current and
the second negative bias current may be set at a sufficiently low
magnitude to effectively not affect measured resistance for said
tunneling magnetoresistive sensor. The second positive bias current
and second negative bias current may be set at a sufficiently
greater magnitude than said first positive bias current to affect
measured resistance for said tunneling magnetoresistive sensor.
Still more particularly, the second positive bias current and
second negative bias current may be of a magnitude that is less
than an operational limit based on the device geometry and at least
sufficiently great to consistently affect measured resistance for
said tunneling magnetoresistive sensor.
[0093] At step 310, the TMR diagnostic program 101 applies a
positive bias current across the TMR sensor and measures a first
resistance change in the presence of the positive bias current. At
step 320, the TMR diagnostic program 101 applies a negative bias
current across the TMR sensor and measures a second resistance
change in the presence of the negative bias current. In the context
of the embodiment depicted in FIG. 3, "positive" and "negative"
currents may be understood as currents in opposing directions, with
an arbitrary selection of the direction in which the currents are
opposed. Similarly, the invention may be applied to any two
opposing directions of current regardless of whether current is
modeled as a flow of negative electrons or as an opposing flow of
positive current. Additionally, a resistance change may be
generally understood as the difference between a pre-short
measurement of the resistance of the TMR sensor and a measurement
in the presence of the short and/or one or more other
conditions.
[0094] Referring still to the embodiment depicted in FIG. 3, at
decision block 330, the TMR diagnostic program 101 determines the
sign of the change in resistance for the first and second changes
in resistance. The sign of the change in resistance may be
understood such that a positive change in resistance denotes an
increase in resistance. The underlying physical mechanism for some
shorts is that, where a thin electrically conductive bridge exists
across the tunnel junction layer, applying a bias current in at
least one direction will cause the thin bridge to heat up, which
increases resistance. In the embodiment depicted in FIG. 3, at
decision bock 340 (Decision block 330 YES branch), responsive to
either the first resistance change being relatively positive or the
second resistance change being relatively positive, at step 370
(decision block 340, NO branch) the TMR diagnostic program 101
returns a probable determination that the operational anomaly is
caused by a short, such as an electrically conductive bridge (e.g.,
a pin-hole dielectric breakdown) across the tunnel junction layer
of the tunneling magnetoresistive sensor.
[0095] Referring still to the embodiment depicted in FIG. 3, at
decision block 340 (decision block 330, YES branch), the TMR
diagnostic program 101 determines whether both of the first
resistance change and the second resistance change are positive,
which denotes a resistance increase for both polarities across the
TMR sensor. Responsive to both the first resistance change being
positive and the second resistance change being positive, at step
365, returning a definite determination that said short is caused
by the electrically conductive bridge. As used herein, a probable
determination may be understood generally as being less certain
than a definite determination, and a definite determination may be
understood as arbitrarily certain up to effectively absolute
certainty. Thus, a probable determination implies relatively
conservative engineering assumptions (whatever those may be for a
particular embodiment) regarding protecting the TMR from further
dielectric breakdown as compared with a definite determination.
[0096] Referring still to the embodiment of FIG. 3, stated
differently, if the resistance of a TMR sensor's tunnel junction
increases with bias current for either positive or negative
current, then the TMR sensor most likely has a short. If the
resistance of a TMR sensor's tunnel junction increases with bias
current for both positive and negative current, then the TMR sensor
conclusively has an electrically conductive short.
[0097] Referring still to the embodiment depicted in FIG. 3, at
decision block 350 (decision block 340, NO branch), the TMR
diagnostic program 101 tests whether the magnitude of the change in
a measured value R.sub.COLD is greater than zero (where magnitude
is expressed using the absolute value pipe symbol, given as
|.DELTA.R.sub.COLD|). R.sub.COLD may be understood as the measured
resistance in the near absence of bias current. More specifically,
due to the small scale of the TMR sensor, it is not possible to
measure R with a truly neutral bias current because electric fields
and currents are inherent at that scale. Thus, R.sub.COLD is
measured when the effective bias current in the desired direction
across the TMR sensor is brought as close to zero or neutral as
possible. To achieve the change in R.sub.COLD, the TMR diagnostic
program 101 compares currently measured R.sub.COLD with a
previously measured or expected R.sub.COLD for the TMR sensor prior
to the operational anomaly. In cases where the previously measured
value for the same device is not known, for example in a newly
operational TMR sensor, R.sub.COLD may be compared with a measured
value for a similar TMR sensor of similar geometry and composition.
In the contemplated embodiment, the TMR diagnostic program 101
computes |.DELTA.R.sub.COLD| by taking the magnitude or absolute
value (i.e., regardless of whether absolute resistance has
increased or decreased) of R.sub.COLD. The inventors have observed
and/or recognized that a change in absolute resistance measured in
the effective absence of bias current without a corresponding
.DELTA.R.sub.+ or .DELTA.R.sub.- (i.e.,
|.DELTA.R.sub.COLD|>.DELTA.R.sub.COLD.sub.LIMIT, where
.DELTA.R.sub.COLD.sub.LIMIT is a near-zero positive predetermined
limit value sufficient to distinguish effectively zero resistance
change), is diagnostic as to a change in magnetic state of the TMR
sensor as a whole device (step 360), which the inventors have
observed to occur with some probability in TMR sensors. As a cause
of a resistance drop, a change in magnetic state is understood by
the inventors to be distinct from a short, however caused, and thus
not necessarily requiring corrective action.
[0098] Referring now to FIG. 4, FIG. 4 is a flowchart diagram for a
TMR diagnostic program 101, in accordance with at least one
embodiment of the present invention. For the depicted embodiment,
the method is understood to be performed by one or more processors
in electronic communication with a TMR sensor, wherein said
tunneling magnetoresistive sensor is a component of a magnetic tape
drive or hard disk configured to read magnetic data from a magnetic
tape or hard disk. Diagnostically, the properties and inferences
applied in a TMR diagnostic program 101 according to FIG. 4 rely
upon the assumption of a physical model of a magnetic medium,
specifically the tape or disk, running across the sensor in
conjunction with detecting the below-described electrical
properties. In the depicted embodiment, at step 400, the TMR
diagnostic program 101 detects a short contemporaneously with the
tape (or hard disk or other magnetic storage medium) running across
the TMR sensor. It should be noted that contemporaneous operation
permits a measurement of voltage amplitude across the TMR sensor,
because the movement of the tape or other medium is what causes
resistance, and thus measured voltage, to vary with time. By
contrast, only stopping the motion of the tape or other medium
permits static resistance of the TMR to be measured, because motion
of the medium causes resistance to vary. As described above, the
short is detected by observing a drop in resistance across the TMR
sensor. At step 410, the TMR diagnostic program measures a voltage
amplitude across the TMR sensor as a function of a fractional
current through said tunneling magnetoresistive sensor to yield a
voltage amplitude data set. At step 420, the TMR diagnostic program
101 fits the voltage amplitude data set to a power law using any
function matching and/or regression technique.
[0099] If the resistance of a TMR sensor drops while running
against tape, the inventors have observed and/or recognized that
the short can be ascribed to a magnetic shunt shorting across the
tunnel junction if the amplitude, Amp, versus fractional current
through the TMR sensor, given by Equation 8, above, can be fit with
the power law equation:
Amp 0 = ( I TMR I 0 ) .beta. Equation 19 ##EQU00033##
[0100] Where .beta. is greater than 2.5, and where Io is a fixed
bias current. R is the resistance with the short, and R.sub.TMRo is
the TMR resistance prior to the short and Amp.sub.o is the
amplitude measured with the bias current I.sub.o prior to the
short. Also, thus, fractional current may be defined as a bias
current multiplied by a measured resistance after the short and
divided by an initial resistance prior to the short. Also, thus,
the power law includes an initial voltage amplitude prior to the
short multiplied by a ratio raised to the exponent. The ratio
includes the fractional current divided by said bias current.
[0101] Referring still to the embodiment of FIG. 4, the TMR
diagnostic program 101, at decision block 430, tests whether the
exponent exceeds a threshold. As above, an exemplary range of the
exponent threshold is between 2.0 and 2.5. At step 440, in the
depicted embodiment, the TMR diagnostic program, responsive to the
voltage amplitude data set fitting a power law, wherein the power
law comprises an exponent, and wherein the exponent is greater than
the exponent threshold, determines that the short is caused by a
magnetic shunt. At step 450, the TMR diagnostic program set the
bias current to a normal value, the normal value being effective in
the absence of the short. As above, exemplary values of a normal
value are between.
[0102] By contrast, where the exponent does not exceed the
threshold, or where there is no power law fit, then, at step 445
(decision block 430, NO branch), the TMR diagnostic program 101 may
conclude that the short is due to dielectric breakdown or a
nonmagnetic shunt. Accordingly, at step 445, the TMR diagnostic
program 101 may limit the bias voltage on the TMR sensor to a safe
level, as described more particularly above.
[0103] In another embodiment of the invention, for TMR sensors
installed in a read head of a tape drive, hard disk drive, or
similar device, a TMR diagnostic program 101 measures the read-back
amplitude, Amp.sub.m (i.e., voltage amplitude) and the resistance,
R.sub.m, for track m (wherein the read had includes multiple
parallel tracks for reading from the tape, disk, or other magnetic
storage medium). In such an embodiment, a TMR diagnostic program
101 measures the average or median voltage amplitude, (Amp), and
resistance, (R) for neighbor or all tracks in the read head. If for
a given track, m, (Amp.sub.m-(Amp))<dAmp.sub.Error. That is,
where dAmp.sub.Error is a predetermined value, and the ratio
Amp m - Amp R m - R ##EQU00034##
is greater than a predetermined value, dAmpdR.sub.Limit, then the
TMR diagnostic program determines that track m has suffered a short
from a magnetic shunt.
[0104] FIG. 18 is a block diagram depicting components of a
computer 1800 suitable for executing the TMR diagnostic program
101. FIG. 18 displays the computer 1800, the one or more
processor(s) 1804 (including one or more computer processors), the
communications fabric 1802, the memory 1806, the RAM, the cache
1816, the persistent storage 1808, the communications unit 1810,
the I/O interfaces 1812, the display 1820, and the external devices
1818. It should be appreciated that FIG. 18 provides only an
illustration of one embodiment and does not imply any limitations
with regard to the environments in which different embodiments may
be implemented. Many modifications to the depicted environment may
be made.
[0105] As depicted, the computer 1800 operates over a
communications fabric 1802, which provides communications between
the cache 1816, the computer processor(s) 1804, the memory 1806,
the persistent storage 1808, the communications unit 1810, and the
input/output (I/O) interface(s) 1812. The communications fabric
1802 may be implemented with any architecture suitable for passing
data and/or control information between the processors 1804 (e.g.,
microprocessors, communications processors, and network processors,
etc.), the memory 1806, the external devices 1818, and any other
hardware components within a system. For example, the
communications fabric 1802 may be implemented with one or more
buses or a crossbar switch.
[0106] The memory 1806 d persistent storage 1808 are computer
readable storage media. In the depicted embodiment, the memory 1806
includes a random access memory (RAM). In general, the memory 1806
may include any suitable volatile or non-volatile implementations
of one or more computer readable storage media. The cache 1816 is a
fast memory that enhances the performance of computer processor(s)
1804 by holding recently accessed data, and data near accessed
data, from memory 1806.
[0107] Program instructions for the TMR diagnostic program 101 may
be stored in the persistent storage 1808 or in memory 1806, or more
generally, any computer readable storage media, for execution by
one or more of the respective computer processors 1804 via the
cache 1816. The persistent storage 1808 may include a magnetic hard
disk drive. Alternatively, or in addition to a magnetic hard disk
drive, the persistent storage 1808 may include, a solid state hard
disk drive, a semiconductor storage device, read-only memory (ROM),
electronically erasable programmable read-only memory (EEPROM),
flash memory, or any other computer readable storage media that is
capable of storing program instructions or digital information.
[0108] The media used by the persistent storage 1808 may also be
removable. For example, a removable hard drive may be used for
persistent storage 1808. Other examples include optical and
magnetic disks, thumb drives, and smart cards that are inserted
into a drive for transfer onto another computer readable storage
medium that is also part of the persistent storage 1808.
[0109] The communications unit 1810, in these examples, provides
for communications with other data processing systems or devices.
In these examples, the communications unit 1810 may include one or
more network interface cards. The communications unit 1810 may
provide communications through the use of either or both physical
and wireless communications links. TMR diagnostic program 101 may
be downloaded to the persistent storage 1808 through the
communications unit 1810. In the context of some embodiments of the
present invention, the source of the various input data may be
physically remote to the computer 1800 such that the input data may
be received and the output similarly transmitted via the
communications unit 1810.
[0110] The I/O interface(s) 1812 allows for input and output of
data with other devices that may operate in conjunction with the
computer 1800. For example, the I/O interface 1812 may provide a
connection to the external devices 1818, which may include a
keyboard, keypad, a touch screen, and/or some other suitable input
devices. External devices 1818 may also include portable computer
readable storage media, for example, thumb drives, portable optical
or magnetic disks, and memory cards. Software and data used to
practice embodiments of the present invention may be stored on such
portable computer readable storage media and may be loaded onto the
persistent storage 1808 via the I/O interface(s) 1812. The I/O
interface(s) 1812 may similarly connect to a display 1820. The
display 1820 provides a mechanism to display data to a user and may
be, for example, a computer monitor.
[0111] The programs described herein are identified based upon the
application for which they are implemented in a specific embodiment
of the invention. However, it should be appreciated that any
particular program nomenclature herein is used merely for
convenience, and thus the invention should not be limited to use
solely in any specific application identified and/or implied by
such nomenclature.
[0112] The present invention may be a system, a method, and/or a
computer program product at any possible technical detail level of
integration. The computer program product may include a computer
readable storage medium (or media) having computer readable program
instructions thereon for causing a processor to carry out aspects
of the present invention.
[0113] The computer readable storage medium can be a tangible
device that can retain and store instructions for use by an
instruction execution device. The computer readable storage medium
may be, for example, but is not limited to, an electronic storage
device, a magnetic storage device, an optical storage device, an
electromagnetic storage device, a semiconductor storage device, or
any suitable combination of the foregoing. A non-exhaustive list of
more specific examples of the computer readable storage medium
includes the following: a portable computer diskette, a hard disk,
a random access memory (RAM), a read-only memory (ROM), an erasable
programmable read-only memory (EPROM or Flash memory), a static
random access memory (SRAM), a portable compact disc read-only
memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a
floppy disk, a mechanically encoded device such as punch-cards or
raised structures in a groove having instructions recorded thereon,
and any suitable combination of the foregoing. A computer readable
storage medium, as used herein, is not to be construed as being
transitory signals per se, such as radio waves or other freely
propagating electromagnetic waves, electromagnetic waves
propagating through a waveguide or other transmission media (e.g.,
light pulses passing through a fiber-optic cable), or electrical
signals transmitted through a wire.
[0114] Computer readable program instructions described herein can
be downloaded to respective computing/processing devices from a
computer readable storage medium or to an external computer or
external storage device via a network, for example, the Internet, a
local area network, a wide area network and/or a wireless network.
The network may comprise copper transmission cables, optical
transmission fibers, wireless transmission, routers, firewalls,
switches, gateway computers and/or edge servers. A network adapter
card or network interface in each computing/processing device
receives computer readable program instructions from the network
and forwards the computer readable program instructions for storage
in a computer readable storage medium within the respective
computing/processing device.
[0115] Computer readable program instructions for carrying out
operations of the present invention may be assembler instructions,
instruction-set-architecture (ISA) instructions, machine
instructions, machine dependent instructions, microcode, firmware
instructions, state-setting data, configuration data for integrated
circuitry, or either source code or object code written in any
combination of one or more programming languages, including an
object oriented programming language such as Smalltalk, C++, or the
like, and procedural programming languages, such as the "C"
programming language or similar programming languages. The computer
readable program instructions may execute entirely on the user's
computer, partly on the user's computer, as a stand-alone software
package, partly on the user's computer and partly on a remote
computer or entirely on the remote computer or server. In the
latter scenario, the remote computer may be connected to the user's
computer through any type of network, including a local area
network (LAN) or a wide area network (WAN), or the connection may
be made to an external computer (for example, through the Internet
using an Internet Service Provider). In some embodiments,
electronic circuitry including, for example, programmable logic
circuitry, field-programmable gate arrays (FPGA), or programmable
logic arrays (PLA) may execute the computer readable program
instructions by utilizing state information of the computer
readable program instructions to personalize the electronic
circuitry, in order to perform aspects of the present
invention.
[0116] Aspects of the present invention are described herein with
reference to flowchart illustrations and/or block diagrams of
methods, apparatus (systems), and computer program products
according to embodiments of the invention. It will be understood
that each block of the flowchart illustrations and/or block
diagrams, and combinations of blocks in the flowchart illustrations
and/or block diagrams, can be implemented by computer readable
program instructions.
[0117] These computer readable program instructions may be provided
to a processor of a general purpose computer, special purpose
computer, or other programmable data processing apparatus to
produce a machine, such that the instructions, which execute via
the processor of the computer or other programmable data processing
apparatus, create means for implementing the functions/acts
specified in the flowchart and/or block diagram block or blocks.
These computer readable program instructions may also be stored in
a computer readable storage medium that can direct a computer, a
programmable data processing apparatus, and/or other devices to
function in a particular manner, such that the computer readable
storage medium having instructions stored therein comprises an
article of manufacture including instructions which implement
aspects of the function/act specified in the flowchart and/or block
diagram block or blocks.
[0118] The computer readable program instructions may also be
loaded onto a computer, other programmable data processing
apparatus, or other device to cause a series of operational steps
to be performed on the computer, other programmable apparatus or
other device to produce a computer implemented process, such that
the instructions which execute on the computer, other programmable
apparatus, or other device implement the functions/acts specified
in the flowchart and/or block diagram block or blocks.
[0119] The flowchart and block diagrams in the Figures illustrate
the architecture, functionality, and operation of possible
implementations of systems, methods, and computer program products
according to various embodiments of the present invention. In this
regard, each block in the flowchart or block diagrams may represent
a module, segment, or portion of instructions, which comprises one
or more executable instructions for implementing the specified
logical function(s). In some alternative implementations, the
functions noted in the blocks may occur out of the order noted in
the Figures. For example, two blocks shown in succession may, in
fact, be executed substantially concurrently, or the blocks may
sometimes be executed in the reverse order, depending upon the
functionality involved. It will also be noted that each block of
the block diagrams and/or flowchart illustration, and combinations
of blocks in the block diagrams and/or flowchart illustration, can
be implemented by special purpose hardware-based systems that
perform the specified functions or acts or carry out combinations
of special purpose hardware and computer instructions.
* * * * *