Title of Invention

"A SPEED SENSING SYSTEM FOR MEASURING THE SPEED OF A TARGET OBJECT"

Abstract A method and apparatus for measuring the speed of a target object passing a pair of sensor units (12) displaced apart by a predetermined distant L in the direction of motion of the target object (16). Passage of one or more features of the target object (16) past the first sensor unit (12A) results in the generation of a signal (x1), and passage of the same feature of the target object (16) past the second sensor unit (12B) results in the generate of a second signal, (x2). A signal processor (18) is configured to determine a mathematical correlation between signals (x1) and (x2), and an associated time delay (τ0). The speed (v) of the target object (16) is calculated by the signal processor (18) as the ratio of the predetermined distance (L) to the time delay (10).
Full Text The present application is related to, and claims priority from,
U.S, Provisional Patent Application No. 60/417,839 filed on October 11,
5 2002.
Technical Field
The present invention relates generally to speed sensors
configured to monitor the speed of a moving body such as a shaft or
axle, and in particular, to an improved speed sensing system utilizing a
10. pair of sensors each configured to detect random targets on a moving
body, and a signal processor configured to measuring a phase shift
between each target detection signals, the phase shift proportional to a
speed of the target.
Background Art
15 Speed sensing plays an important role in monitoring, and thus
controlling, machine operations. An accurate and reliable speed sensor
is critical. Over the years numerous speed-sensing techniques and
devices have been developed. Mechanical speedometers, electro-
mechanical speed sensors, magnetic speed sensors, and optical speed
20 sensors are just a few examples. Most popular speed sensing systems
often include a single sensor, an electronic control unit, and a target
whose speed relative to the single sensor is measured.
Depending upon the type of speed being measured, i.e., linear or
angular speed, and on the sensor technology that is employed, a target
25 may be constructed in a variety of ways and may take many different
forms. Conventionally, speed sensing targets have been made from
marked bars and toothed wheels, from multi-polar magnetic-strips and
magnetic-rings, and from linear and angular bar-encoders. As the target
moves relative to the sensor, a conventional sensor output signal takes
30 the form of a series of pulses, with the pulse frequency being
proportional to the target wheel speed.

WO 2004/034065 PCT/US2003/031601
-2-
The resolution or accuracy of these conventional speed sensing
systems depend heavily, among other factors, on the accuracy of the
spacing between the teeth in a toothed target, the spacing of the
magnetic poles in a magnetic target, and the spacing of the bars in a bar
5 encoder. Thus, or a precision system, a target with high spacing
accuracy is preferred.
However, the target manufacturing cost is proportional to the
target spacing accuracy requirements, and it is not always economical to
construct a large outer diameter angular target wheel or a long linear
10 target with high spacing accuracy. Accordingly, it would be
advantageous to introduce a speed sensing system which maintains a
high degree of speed measurement accuracy without requiring the
production and application of a precision speed sensing target.
Summary of the Invention
15 Briefly stated, the present invention sets forth a speed sensor
system comprising a pair of sensing elements disposed in a directionally
spaced relationship adjacent a surface of a moving object from which a
speed measurement will be acquired. A target, having substantially
random features is disposed on or beneath the surface, and is moved
20 directionally past the pair of sensing elements by the movement of the
object from which a speed measurement will be acquired. The pair of
sensing elements are directionally spaced apart by a predetermined
distance in the direction of the object's movement. Signals from each of
the pair of sensing elements, generated by the passage of the target,
25 are conveyed to a signal processor. The signal processor is configured
to determine a phase shift between the generated signals which is
inversely proportional to the speed at which the target passed the pair of
sensor.
As a method for measuring a target speed, the present invention
30 includes the steps of observing at a first point, a plurality of random
features of said target, generating a first signal representative of said
observations at said first point, observing at a second point displaced

WO 2004/034065 PCT/US2003/031601
-3-
from said first point in a direction of motion of said target, said plurality
of random features of said target, generating a second signal
representative of said observations at said second point, and calculating
a phase shift between said first signal and said second signal, said
5 phase shift inversely proportional to a speed of said target.
The foregoing and other objects, features, and advantages of the
invention as well as presently preferred embodiments thereof will
become more apparent from the reading of the following description in
connection with the accompanying drawings.
10 Brief Description of Drawings
In the accompanying drawings which form part of the
specification:
Figure 1A is a simplified diagrammatic view of one embodiment
of a speed sensor system of the present invention in relation to a linearly
15 moving object;
Figure 1B is a perspective view of the speed sensor system of
Fig. 1A in relation to a rotationally moving object;
Figure 2A graphically illustrates sample eddy-current sensor
signals received from a pair of adjacent speed sensors units of the
20 present invention;
Figure 2B graphically represents a cross correlation function
between the two signals shown in Figure 2A;
Figure 3A illustrates a first speed sensor configuration relative to
the direction of motion of a target object;
25 Figure 3B illustrates a second speed sensor configuration relative
to the direction of motion of a target object;
Figure 3C illustrates a third speed sensor configuration relative to
the direction of motion of a target object; and
Figure 4 compares results in measuring angular speed from a
30 speed sensor system of the present invention using a pair of eddy
current sensors together with a toothless target wheel against the results
from a conventional variable reluctance (VR) speed sensor system.

WO 2004/034065 PCT/US2003/031601
-4-
Corresponding reference numerals indicate corresponding parts
throughout the several figures of the drawings.
Best Mode for Carrying Out the Invention
The following detailed description illustrates the invention by way
5 of example and not by way of limitation. The description clearly enables
one skilled in the art to make and use the invention, describes several
embodiments, adaptations, variations, alternatives, and uses of the
invention, including what is presently believed to be the best mode of
carrying out the invention.
10 Turning to Figures 1A and 1B, the basic components of a speed
sensor system 10 of the present invention are shown. A pair of speed
sensor units 12A and 12B are disposed in a spaced relationship
adjacent a toothless target surface 14 of a moving object 16, for
example, the circumferential surface of a bearing race as shown in
15 Figure 1B. Preferably, the speed sensor units 12A and 12B are spaced
apart by a predetermined distance L between the speed sensor unit
centers, aligned with the direction of motion of the target surface 14,
shown by the arrow in Figure 1A. The sensor units 12A and 12B are
operatively couples to a signal processing unit 18. Preferably, as shown
20 in Figure 1A, the sensor units 12A and 12B are directly coupled to the
signal processing unit 18 by electrically conductive wires 20 configured
to communicate respective signals x1 and x2 from the sensor units 12A
and 12B to the signal processing unit 18. However, those of ordinary
skill in the art will recognize that a variety of components may be utilized
25 to couple the sensor units 12A and 12B to the signal processing unit 18,
including wireless transmission components.
The operating principles of the current invention are based on
generating and analyzing two mathematically correlated signals x1 and
x2 from the pair of sensor units 12A and 12B in the speed sensor system
30 10. By detecting the phase shift between corresponding points in each
of the two signals x1 and x2, a time delay can be determined. The phase
variation in signals is solely related to the speed of motion. The speed of

WO 2004/034065 PCT/US2003/031601
-5-
motion can be calculated as the ratio of the distance L between the
sensor units 12A and 12B, and the determined time delay.
For accuracy considerations, it is highly desirable to use random
signals, each preferably having a high frequency content. To this end, a
5 target surface 14 with random or near random topographical features
(roughness) is employed as target to produce randomly variable signals.
With certain types of sensor units 12A and 12B, such as eddy current
sensors, the randomness of the signals x1 and x2 can be enriched by
subsurface material property variations.
10 The first sensing unit 12A and the second sensing unit 12B are
configured to be substantially sensitive to surface and/or subsurface
features of the target surface 14, and are functionally similar in that each
sensor unit 12A, 12B produces identical signals or substantially similar
signals when passing over the same surface or subsurface features on
15 the target surface 14. Alternative sensor units 12 may include optical
sensors sensitive to optical variations on the target surface 14.
As the target surface 14 moves relative to the speed sensor
system 10, the first sensing unit 12A and the second sensing unit 12B
each generate signals, such as exemplified in Figure 2A, at an identical
20 sampling rate f that is substantially higher than the signal variation rate,
allowing the speed sensing system 10 to resolve high frequency surface
features even at the highest target speeds.
In general, a correlation exists between the first signal x1=[x11,
x12, x13,... x1j, ..., 25 second signal x2=[x21, x22, x23,... x2j, ..., x2n] generated by the second
sensing unit 12B in response to the passage of surface or subsurface
features on the target surface 14, where n represents the sample size
(number of data points in a sample). There is, however, a time delay of

between the first signal x1 and second signal x2 where m
30 represents the number of shifted data points. The direction of the signal

WO 2004/034065 PCT/US2003/031601
-6-
shifting corresponds the direction of the motion of the target surface 14
relative to the sensor units 12A and 12B.
Thus, a cross correlation function y(τ) between the signals x1 and
x2 may be defined by the equation:
5
Equation (1)

which reaches a maximum value when t τ= Τ0.
The time delay τ0 can be determined by finding the maximum
value of the cross correlation function of the signals x1 and x2. That is:
Equation (2)

10 where ψ is an inverse function to the cross correlation function
y(τ) defined in Equation (1).
During operation, signal processor 18 receives and processes
signals x1 and x2. The incoming signals x1 and x2 are initially processed
to remove any direct current (DC) components, resulting in a pair of
15 signals each having zero-mean such as shown in Figure 2A. The signal
processor 18 further performs the cross correlation analysis of the two
signals, preferably using a Fast Fourier Transform (FFT) based
algorithm for fast computation. Next, the signal processor 18 determines
the time delay to between the two signals by calculating the maximum
20 value for the cross correlation function y(τ) defined in Equation (1).
Finally, the speed of motion v for the target surface 14 past the sensor
units 12A and 12B is computed by the signal processor 18 as:
Equation (3)
Optionally, the signal processor 18 may be configured to compute
25 a relative position of the target surface 14 by integrating the computed
speed v with respect to time.
Returning to Figure 2A, the signals x1 and x2 illustrated
graphically are representative of signals from a pair of independent eddy

WO 2004/034065 PCT/US2003/031601
-7-
current speed sensor units 12A and 12B positioned 0.788 inches apart
along the circumferential direction of motion for a rotating target surface
14, such as shown in Figure 1B. For a sampling rate of approximately
48 kHz, the resulting sample size is 3700 data points for a single
5 revolution of the target surface 14. In the graph shown in Figure 2A, the
horizontal axis represents sample sequencing and vertical axis is the
strength of the signals in volts. The cross correlation function of signals
x1 and x2, illustrated in Figure 2B, identifies a maximum value at data
point 4036 that corresponds to a shifting of 336 data points (4036-
10 3700=336). The corresponding time delay between the first signal x1
and second signal x2 is τ0 =336/48000 = 0.007 sec. The surface speed
of the target 14 is then v = 0.788/0.007 = 112.5 in/sec. The direction of
motion is determined by the direction of signal shifting.
Based on the selected sensor technology, the signal processor
15 18 and sensor units 12A, 12B could be integrated into a single unit 20,
such as shown in Figure 1B using modern ASIC fabrication techniques
with Digital Signal Processing (DSP) computation ability.
To ensure a good correlation between the two signals x1 and x2
under less than ideal installation and/or application conditions,
20 differential sensing combinations of speed sensor units 12 may
optionally be used. In this case one sensor combination may contain
more than two speed sensing units 12. A comparison of signals from
each of the speed sensor units 12 comprising the differential sensing
combinations permits removal or cancellation of signal components
25 common to all sensing units 12. such as noise or interference, which are
present at each speed sensor unit location. These common signal
components usual y carry no information with respect to signal phase
shifting.
Figures 3A through 3C illustrate three differential sensing
30 combinations and the positioning of the associated speed sensing units
12 in relationship to the direction of motion of the target surface 14. In
Figure 3A. a first sensor combination 100 contains four speed sensor

WO 2004/034065 PCT/US2003/031601
-8-
units 12A - 12 D positioned at the corners of a rectangle to form two
differential sensing pairs. The first differential sensing pair is formed by
speed sensor units 12A and 12C, and the second differential sensing
pair is formed by speed sensor units 12B and 12D. In the first differential
5 sensing pair, speed sensor units 12A and 12C are separated by a
center distance W perpendicular to the direction of motion. In the
second differentia sensing pair, speed sensor units 12B and 12D are
similarly separated by the center distance W, perpendicular to the
direction of motion. Each differential sensing pair is spaced apart by a
10 distance L substantially in the direction of motion.
Figure 3B illustrates an alternate arrangement for a sensing
system 200 wherein the differential sensing pairs 12A, 12C and 12B,
12D are disposed at the corners of a parallelogram, i.e., where the
center lines between speed sensor units 12A and 12C and between
15 speed sensor units 12B and 12D are not perpendicular to the center line
defined by the position of speed sensor units 12A and 12B. The
included angle α between speed sensor units 12A, 12C and speed
sensor units 12A, 12B is set equal to the included angle β between
speed sensor unite 12B, 12D and speed sensor units 12A, 12B. That is
20 α = β, such that the placement of the speed sensor units 12A - 12D
defines a parallelogram having two sides parallel to the direction of
motion of the target surface 14. In general, α and β could each vary
from 0 to 360 degrees.
Figure 3C shows an alternate arrangement for a sensing system
25 300 similar to that shown in Figure 3A, but where the first pair of
differential sensing units 12 A, 12C and the second pair of differential
sensing units 12B, 12D are disposed in two different sensor housings,
and hence are spaced apart by a distance L' > L. As is shown in Figure
3A. the centerline between the centers of sensing elements 12A and
30 12B substantially aligns with the direction of motion of the target surface
14. Corresponding y, the center line that connects the centers of the

WO 2004/034065 PCT/US2003/031601
-9-
sensing elements 12C and 12D is also substantially aligned with the
direction of motion. The center lines connecting the centers of the first
pair of differential sensing units 12A, 12C is parallel with the center line
that connects the centers of the second pair of differential sensing units
5 12B, 12D, and substantially perpendicular to the direction of motion for
the target surface 14.
The current invention is not confined to any specific type of speed
sensors units. However, the speed sensor units 12 are preferably eddy
current sensors capable of generating signal variations induced both by
10 topographical features on the target surface 14 and by subsurface
material property changes in the target object 16. This allows the
sensing system to be used not only for rough target surfaces 14 but also
for smooth target surfaces 14 where the signal variation is induced
primarily by subsurface material property changes rather than by surface
15 topographical features.
Figure 4 graphically illustrates the validity of the sensing system
10 and techniques of the present invention in measuring angular speed
using a pair of eddy current speed sensor units 12A, 12B and a
toothless target object 16. The graph of Figure 4 plots the angular speed
20 of the target object 16 as measured by the sensing system 10 of the
present invention versus the angular speed of the target object 16 as
measured by a conventional variable reluctance (VR) speed sensor
system, illustrating a close correlation between the two sensor systems.
It should be understood that the sensing system 10 and
25 techniques of the present invention are applicable to a host of
applications such as for use in bearing application, and particularly in
bearing applications wherein the target surface 14 is a bearing seal.
In view of the above, it will be seen that the several objects of the
invention are achieved and other advantageous results are obtained. As
30 various changes could be made in the above constructions without
departing from the scope of the invention, it is intended that all matter

WO 2004/034065 PCT/US2003/031601
-10-
contained in the above description or shown in the accompanying
drawings shall be interpreted as illustrative and not in a limiting sense.

-11-
SUBSTITUTE SHEET
Claims
1. A speed sensing system for measuring the speed of a
target object, comprising:
a first speed sensor unit operatively disposed adjacent a surface
5 of said target object, said first speed sensor unit configured to generate
a first signal responsive to the passage of at least one random feature of
said target object;
a second speed sensor unit operatively disposed adjacent a
surface or said target object and displaced at a predetermined distance
10 from said first speed sensor unit substantially in a direction of motion of
the target object said second speed sensor unit configured to generate
a second signal responsive to the passage of said at least one random
feature of said target object; and
a signal processor configured to receive said first and second
15 signals, said signal processor further configured to apply a cross
correlation analysis with a Fast Fourier Transform-based algorithm to
determine a phase shift between said first and second generated
signals, said phase shift inversely proportional to a speed of said target
object.
20 2. The speed sensing system of Claim 1 further including:
a third speed sensor unit operatively disposed adjacent a surface
of said target object, said third speed sensor unit configured to generate
a third signal responsive to the passage of at least one feature of said
target object:
26 a fourth speed sensor unit operatively disposed adjacent a
surface of said target object and displaced at a predetermined distance
from said third speed sensor unit substantially in a direction of motion of
the target object, said fourth speed sensor unit configured to generate a
fourth signal responsive to the passage of said at least one feature of
30 said target object; and
wherein said signal processor is further configured to receive said
third and fourth signals, and to utilize said third and fourth signals to
AMENDED S HEET

30-07-2004 4 13:33 FAX 3142382401
EPO NETHERLANDS US0331601
006 30.07.2004
-12-
SUBSTITUTE SHEET
cancel signal components common to said first, second, third, and
fourth signals.
3. The speed sensing system of Claim 2 wherein said signal
processor is further configured to provide differential signals between
5 the first and third signals, and between the second and fourth signals,
and said signal processor is further configured to determine a phase
shift between said third and fourth generated signals, said phase shift
inversely proportional to a speed of said target object.
4. The speed sensing system of Claim 2 wherein said first
10 and third speed sensing units define a first differential sensing pair;
wherein said second and fourth speed sensing units define a
second differential sensing pair; and
wherein said first and second differential sensing pairs are
spaced apart by a predetermined distance parallel to said direction of
15 motion of the target object.
5. The speed sensing system of Claim 1 wherein said first
and second speed sensing units are eddy current sensors; and
wherein said at least one feature is a random subsurface target
feature.
20 6. The speed sensing system of Claim 1 wherein said first
and second speed sensing units are optical sensors.
7. The speed sensing system of Claim 1 wherein said signal
processor is configured to filter direct-current components from said first
and second generated signals such that said generated signals have a
25 zero signal mean.
8. The speed sensing system of Claim 1 wherein said signal
processor is configured utilize a Fast Fourier Transform-based algorithm
to determine a cross correlation function between said generated
signals, said cross correlation function defined by:
30

where x1 is said first generated signal;
AMENDED SHEET

30-07-2004 4 13:33 FAX 3142382401
EPO NETHERLANDS US0331601
006 30.07.2004
-13-
SUBSTITUTE SHEET
x2 is said second generated signal;
t Is a signal time; and
τ is a time delay between said generated signals.
9. The speed sensing system of Claim 8 wherein said phase
5 shift is associated with a maximum value for said cross correlation
function: and wherein said signal processor is further configured to
determine a maximum value for said cross correlation function;
wherein a speed v of said target object is determined from:

10 where L is said predetermined distance; and
τ0 is a time delay corresponding to said determined maximum
value for said cross correlation function.
10. The speed sensing system of Claim 1 wherein said first
speed tensor unit and said second speed sensor unit are disposed
15 within a common housing.
11. The speed sensing system of Claim 1 wherein said at least
one target leature is a random surface feature of the target object.
12. The speed sensing system of Claim 1 wherein said at least
one target feature is a random subsurface feature of the target object.
20 13. The speed sensing system of Claim 1 where each of said
first and second speed sensing units has an identical sampling rate; and
wherein said identical sampling rate is substantially greater than a
signal variation rate for said first and second speed sensing units.
14. A method for speed measurement of a target object,
25 comprising the steps of:
obseiving at a first point, a passage of at least one random
feature of the target object.
generating a first signal responsive to said passage of said at
least one random feature at said first point.

-14-
SUBSTITUTE SHEET
observing at a second point, displaced at a predetermined
distance from said first point in a direction of motion of said target
object, said passage of said at least one random feature of the target
object;
5 generating a second signal responsive to said passage of said at
least one random feature at said second point;
filtering direct-current components from said first and second
generates signals;
applying a cross correlation analysis with a Fast Fourier
10 Transform based algorithm to calculate a phase shift between said
filtered first signal and said filtered second signal, said phase shift
inversely proportional to a speed of said target object.
15. The method of Claim 14 for speed measurement of an
object wherein said phase shift is associated with a maximum value of a
15 cross correlation function between said filtered first and second
generated signals, and wherein said step of applying further includes
calculating said maximum value of said cross correlation function
between said filtered first and second generated signals, said cross
correlation function defined by:
20

where x1 is said first generated signal;
x2 is said second generated signal:
f is a signal time; and
Τ IS a time delay between said generated signals.
25 16. The method of Claim 16 for speed measurement cf an
object, further including the step of determining a speed v of said target
object from:

where L is said predetermined distance;
AMENDED SHEET

-15-
SUBSTITUTE SHEET
and τ0 is a time delay corresponding to said determined maximum
value for said cross correlation function.
17. The method of Claim 14 for speed measurement of an
object further including the steps of:
5 observing at a third point a passage of an additional feature of
the target object;
generating at least one additional signal responsive to said
passage of said additional feature at said third point;
utilizing said at least one additional signal to cancel common
10 elements present in each of said first and second generated signals.
18. A method of Claim 14 for speed measurement of a target
object further including the step of:
determining a relative position of the target object from said
calculated phase shift.
16 19. The method of Claim 18 for determining a relative position
of a target object wherein said determining step includes the step of
integrating a calculated speed of said the target object with respect to
time.


AMENDED SHEET

A method and apparatus for measuring the
speed of a target object passing a pair of sensor units (12)
displaced apart by a predetermined distant L in the direction
of motion of the target object (16). Passage of one or more
features of the target object (16) past the first sensor unit
(12A) results in the generation of a signal (x1), and passage
of the same feature of the target object (16) past the second
sensor unit (12B) results in the generate of a second signal,
(x2). A signal processor (18) is configured to determine a
mathematical correlation between signals (x1) and (x2), and
an associated time delay (τ0). The speed (v) of the target
object (16) is calculated by the signal processor (18) as the
ratio of the predetermined distance (L) to the time delay (10).

Documents:


Patent Number 218582
Indian Patent Application Number 00615/KOLNP/2005
PG Journal Number 14/2008
Publication Date 04-Apr-2008
Grant Date 02-Apr-2008
Date of Filing 11-Apr-2005
Name of Patentee THE TIMKEN COMPANY,
Applicant Address 1835 DUEBER AVENUE, S.W. CANTON, OH 44706 UNITED STATES OF AMERICA, A CORPORATION OF THE STATE OF OHIO,
Inventors:
# Inventor's Name Inventor's Address
1 HWANG, WEN-RUEY, 8609 DEACON AVENUE NW, NORTH CANTON, OH 44720 CHINESE CITIZEN, U.S.A.
2 AI, XIAOLAN 4480 NOBLE LOON STREET NW, MASSILLON, OH 44646 U S CITIZEN
3 VARONIS, ORESTES, J, 1340 IRONDALE CIRCLE NE, NORTH CANTON, OH-44720 U S CITIZEN
PCT International Classification Number G01P 3/80
PCT International Application Number PCT/US2003/031601
PCT International Filing date 2003-10-06
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 60/417,839 2002-10-11 U.S.A.