Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
NON-CONTACT TORQUE MEASUREMENT APPARATUS AND METHOD
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation-in-part of United States
Application No.
12,869,447, filed on August 26, 2010.
BACKGROUND
Technical Field
[0002] This invention relates, generally, to apparatus and methods used in
measuring
torque in a shaft. In particular, this invention relates to non-contact
torsion measurement on a
rotating shaft assembly, comprising two rigid shafts segments connected by a
flexible torsion
member.
[0003] This invention relates to torque measurement in the following
conditions while
not compromising on performance and accuracy:
1) Harsh environments, High Pressure, High Temperature;
2) Presence of Corrosive, Particle Laden and Non-Clear (dirty) fluids;
3) High Rotating Speeds;
4) Wide Span of Torque Ranges over several orders of magnitude;
5) End of Shaft not accessible;
6) Large Offset Distances between Shaft and Sensing elements; and
7) Vibration and Noise.
[0004] In one embodiment, this invention relates to testing apparatus and
methods for
monitoring mixing torque on liquids, gels, slurries or pastes enclosed under
specific pressure and
temperature conditions and, in particular, apparatus and methods for testing
fluid mixtures and
1
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
slurries for use in subterranean wellbores under simulated wellbore
conditions.
Background Art
[0005] When drilling, completing, and treating subterranean hydrocarbon wells,
it is
common to inject materials in fluid form with complex structures, such as
suspensions,
dispersions, emulsions and slurries. These injected materials are present in
the wellbore with
materials such as water, hydrocarbons, and other materials originating in the
subterranean
formations. The materials present in the wellbore will be referred to herein
as "wellbore fluids"
or "wellbore liquids." The flow of these fluids and mixtures cannot be
characterized by a single
viscosity value, instead the apparent viscosity and shear stresses changes due
to other factors
such as temperature and pressure and the presence of other materials. Two
fluids are
incompatible if undesirable physical or chemical interactions occur when the
fluids are mixed.
Incompatibility is characterized by undesirable changes in apparent viscosity
and shear stresses.
When apparent viscosity of the mixed fluids is greater than apparent viscosity
of each individual
fluid, they are said to be incompatible at the tested shear rate. An example
of when compatibility
of fluids is important may include the scenario below.
[0006] It is common to determine optimum wellbore liquids and incompatibility
of
those liquids in a laboratory by running a series of tests of different liquid
mixtures under
wellbore conditions. Testing various ratios of mixtures of wellbore liquids is
done to replicate
the changes in the wellbore concentrations of the fluids. Testing a series of
samples of actual
wellbore mixtures during well treatment is also common. Viscosity, visco-
elasticity, shear
stress, and consistency are rheological characteristics that need to be
measured for a given fluid
or mixture.
[0007] Known prior art devices used to test fluids for these characteristics
include
viscometers, rheometers and consistometer. Examples include those illustrated
and described in
2
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
United States Patent Numbers 3,435,666, 4,668,911, 5,535,619 and 6,951,127,
which are
incorporated herein by reference for all purposes. Testing comprises filling a
test chamber with a
first mixture, bringing the chamber to pressure and temperature test
conditions, and then
conducting tests of the fluids characteristics. In some prior art testing
devices, apparent viscosity
is tested by measuring the torque required to rotate a paddle in a closed or
sealed housing
containing the test fluid. In these prior art devices, the tests are conducted
at elevated
temperatures and pressures. For instance rheology measurement applications in
harsh
conditions, requires an accurate but non-invasive measurement technique.
Typically, these
devices use a paddle rotated in a test fluid. The torque required to rotate
the paddle in the fluid
corresponds to the apparent viscosity of the fluid.
[0008] Other applications include monitoring torque during mixing, and
measuring
torque on a rotating engine shaft and the like. Non-contact measurements have
been previously
carried out using optical encoders and non-offset magnetic systems. Optical
encoders require
clear fluids, and non-offset systems need the end of the shaft to be
accessible to mount the
sensor. Prior art systems have been deployed to measure angular displacement
only, such as:
Gear Tooth Detection Devices; Non-Offset Systems; and Systems with High
Proximity between
magnet and sensor.
SUMMARY OF THE INVENTIONS
[0009] According to the apparatus and methods of one embodiment of the present
invention, torque in a shaft can be measured accurately under difficult
conditions using a magnetic-
detector configuration. Permanent magnets placed along the shaft induce
current or a voltage output
in the detectors and, from the phase shift in the detector outputs, the torque
can be calculated,
knowing the properties of the shaft. The detectors can be used with Hall
Effect sensors or other
magnetic field sensors to detect magnet polarity.
3
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
BRIEF DESCRIPTION OF THE FIGURES
[0010] The advantages and features of the present invention can be understood
and
appreciated by referring to the drawings of examples attached hereto, in
which:
[0011] Figure 1 is a schematic diagram of the testing apparatus of the present
inventions;
[0012] Figures 2 ¨ 4, and 6 illustrate a magnet configuration and mounting for
use in
the present inventions; and
[0013] Figure 5 illustrates the non-contact torque measuring device of the
present
invention on a motor shaft.
DETAILED DESCRIPTION OF THE INVENTIONS
[0014] Referring now to the drawings, wherein like or corresponding parts are
designated
by like or corresponding reference numbers throughout the several views, there
is schematically
illustrated in Figure 1, a fluid testing apparatus 10 embodying the method and
apparatus of the
present inventions. The apparatus 10 comprises a housing 12, enclosing a test
chamber 14
containing the fluid 16 to be tested. The space in the housing 12 above the
test chamber can be
filled with an inert fluid. In the preferred embodiment, the housing and the
test chamber are
sealed enclosures that can be raised in pressure to perform tests on the fluid
in the chamber 14.
A shaft formed by two shaft segments SA and SB extends into housing 12.
According to the
present inventions, a paddle 18 mounted on shaft SB is rotated in the test
chamber 14 while in
contact with the test fluid 16 to measure the apparent viscosity of the test
fluid.
[0015] In one embodiment, a resilient member embodied as a torsion spring 22
with the
spring constant k, couples or connects shaft segment SA to shaft segment SB.
As used herein,
the term "resilient member" refers to a member that has the ability to absorb
energy when it is
4
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
deformed elastically and release that energy upon unloading. Resilient members
include springs
and elastic items that are capable of returning to an original shape or
position after having been
deformed. The spring should be selected with a constant k that is linear (or
within Hookean
range) for the operating range of the apparatus. Alternatively, the spring
embodiment could be a
Flexural Pivot Bearing, such as the Cantilevered Single Ended Pivot Bearings
or the Double
Ended Pivot Bearings supplied in various sizes by Riverhawk Flexural Pivots
Company of
Hartford, New York.
[0016] The shaft segment SA is, in turn, mechanically coupled at 22 to a
driver 24.
Typically, the driver 24 is an electrical motor which is coupled to the shaft
segment SA through
the wall of the housing 12, using a conventional a magnetic coupling 22. The
magnet coupling
has a driver magnet outside the housing coupled by magnetic forces to drive a
follower magnet
located inside the housing. Suitable bearings (not shown) can be used to
maintain the shaft in
position in the housing. Alternative to using an externally mounted driver 24
with a through wall
coupling 22, the driver 24 could be mounted in whole or part inside the
housing.
[0017] Magnets MA and MB are mounted to rotate with shaft segments SA and SB
with their magnetic field substantially perpendicular to the shaft,
respectively. Detector DA is
located outside the housing 12 in the proximity of (distance dl) magnet MA.
Detector DB is
located outside the housing 12 in the proximity of (distance d2) the magnet
MB. Detectors DA
and DB are connected to data processing unit 30 which, in the present
embodiment, determines
the phase shift between detectors DA and DB when torque is applied to the
shaft segments. The
Processing unit also includes a counter and clock to provide output data
regarding the shaft speed
and time. A display-data storage unit 40 is connected to the output of the
unit 30 for recording
data from detectors DA and DB and processed data from unit 30. Data
acquisition can be carried
out using the PX14330 card supplied by National Instruments. Waveform analysis
can be
carried out to extract information on phase shift, using standard Digital
Signal Processing (DPS)
techniques. In addition, phase detection can be carried out on the waveforms
generated by the
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
AMR sensors, using Model 7270 DPS Lock in amplifier and the SR810 or SR830
from
Signalrecovery and Stanford Research Systems, respectively.
[0018] In operation, test fluid 16 is added to test chamber 14, and the
pressure and
temperature test conditions are applied. Motor driver 24 is activated to
rotate shaft A at a speed
Omega as illustrated by arrow 26. Torsion spring 20 couples shaft segments SA
and SB and
transfers the rotation of shaft segment SA to shaft segment SB. As shaft
segment SB rotates, the
paddle 18 is rotated in the test fluid 16. Contact between the paddle 18 and
the test fluid 16 retards
the rotation shaft B which, in turn, causes twisting or relative rotation
(angular deflection) between
shaft segments SA and shaft SB due to the deflection in torsion spring 20. The
term "torque"
(measured in force multiplied by distance) is used herein to indicate applying
a twisting force to an
object to tend to cause rotation. The term "torsion" is used herein to
describe the shearing stress in a
shaft or other object when torque is applied. Torsion, of course, varies from
zero at the axis to a
maximum at the outside surface of a shaft. By calibrating the device and
measuring the relative
rotation between shafts A and B, the apparent viscosity of the test fluid 16
can be determined.
[0019] The detectors used in the apparatus of the present invention are
Wheatstone
bridge-type elements. Magnetic field detectors can comprise a coil wound with
insulated
conducting material. These detectors comprise resistive elements whose
resistance changes
with the orientation of the magnetic field and preferably are Anisotropic
Magneto Resistance
(AMR) effect sensors supplied by Honeywell Inc. Sensors using this technology
are classified as
saturation mode or liner mode sensors. For example, position sensors (HMC1512
supplied by
Honeywell Inc) are classified as saturation mode sensors. The output of these
sensors is an
electrical signal and, in some sensors, is in the form of a sinusoidal wave,
having twice the
frequency of the rotation of the shaft. These sensors can be used with Hall
Effect Sensors that
act as polarity detectors as to which pole of the magnet is rotating. The
phase shift between the
detector outputs is measured to determine applied torque.
6
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
[0020] The second kind of AMR detectors that can be used are supplied by
Honeywell
Inc. For example, HMC 1512, 102X, 104X and 105X sensors offered by Honeywell,
Inc could
be used to infer magnetic field by measuring the voltage response. These
sensors are available in
single-axis, 2-axis and 3-axis configurations to measure the magnetic fields
in space. These
sensors work on the same principle as saturated mode sensors but provide a
full 360-degree
detection and exhibit a linear relationship between the output voltage and the
magnetic field. In
addition, Giant Magneto Resistance sensors from NVE Corporation may be used in
some
applications.
[0021] Additionally, in lieu of Hall effect sensors or AMR, GMR sensors,
one may also
use detectors made of multiple turns of coil wound using insulated metal wires
such as insulated
copper wire which generates induced voltage in the presence of the rotating
magnets MA and
MB.
[0022] According to a particular feature of the present invention, the
preferred
generally rectangular shape of the magnets A and B is illustrated in Figures 2
¨ 4. This shape is
particularly suited to measuring the relative angular position of shafts
rotating at high speeds. In
these figures, the magnet is identified, generally by reference numeral 30.
The magnet body can
best be described as having a generally rectangular cross section with two
opposed, flat faces 32
formed by parallel straight lines and two opposed curved faces 34 formed by
arcs. The arcs are
preferably semicircular. The magnet is designed with bore 38 positioned to
receive a shaft to be
rotated about the geometric center of the generally rectangular cross section.
As used herein, the
term "generally rectangular" refers to a shape that has sides and is elongated
in the direction
between its magnetic poles. The end faces 36 of the magnet are planar. The
magnetic field
orientation M is diametrical. A central bore 38 extends through the magnet
between the ends 36.
In Figure 6 the magnet 32 is illustrated clamped onto the shaft SA by a
mounting bracket M. As
illustrated in Figure 6, the bore 38 is of a size to receive shafts A or B for
mounting. The flat
faces 32 are used to fix the magnets in an angular position on the shaft in
bracket M. The
7
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
bracket M is fixed to rotate with the shaft SA. Bracket M has a bifurcated
portion forming a
straight sided slot in which the magnet 32 is nested. The sides of the slot
fit snugly against the
faces 32 to prevent rotation of the magnet 32 with respect to the bracket and
shaft. A set screw
or the like is used to releasable hold the magnet in axial position in the
bracket. By mounting the
magnets in the manner the magnets can be easily changed out as required. In an
alternative
embodiment, the flat faces are replaced with planar faces, giving the magnet a
rectangular, cross-
sectional shape.
[0023] Preferably, the magnet is made from materials that can operate at
high
temperatures, for example, Sm5Co 17 or Alnico. The size of the magnets are
selected such that
the minimal measurable magnetic field is at least in the order of the
measuring range of the
sensor.
[0024] An alternative application of the present invention is
illustrated in Figure 5.
Motor M drives shaft S which is connected to a load L. As illustrated, the
motor is located
outside a sealed enclosure or housing H (depicted as dotted lines), however,
the system is useful
in applications where no housing is present, such as where access to the shaft
is limited. In
operation, shaft S is twisted by the force applied by the motor to the shaft.
The torque required
to drive the load at a given speed can be measured by installing axially
spaced magnets M1 and
M2 to rotate with shaft S. Detectors-magnet pairs D1-M1 and D2-M2 sense the
position of the
shaft at the detectors as shaft rotates. Torque in the shaft can be determined
by measuring the
twist in the shaft between the detectors. In this application, the two shaft
segments are formed
into a unitary shaft without the torsion spring used in the Figure 1
embodiment. The actual twist
or distortion in the rotating shaft is used to determine the torque in the
shaft.
[0025] The housing itself can have an effect on the performance of the
measurements.
For the magnets MA and MB to "transmit" their fields to sensor A and sensor B,
respectively,
the housing which typically holds the pressure and temperature should be
formed, at least in part,
8
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
from non-magnetic materials. The term "non magnetic" is used to refer to
materials that do not
stick to magnets, such as materials like SS-316L, inconel 718, MP35N, etc. Non-
magnetic
materials will have a magnetic relative permeability value of about 1.
Preferably, the housing
will be formed from non-metallic materials so that the magnetic field will be
transmitted through
the housing wall. In some embodiments, it is preferable that a portion of the
housing structure
between the magnet and its sensor will be made out of a non-magnetic material.
[0026] According to the present inventions, the position of magnet MA on shaft
segment
SA acts as a reference point against which the position of magnet MB on shaft
segment SB is
measured. In applications where shaft segment SA and shaft segment SB are
connected by a
flexible member, such as a torsion spring, flexural pivot, or the like, the
reference point may
determined alternatively. In applications where the shaft segments are
relatively rigid, the
reference point may be determined at different locations in the system.
[0027] When driver 24 is a motor, magnet MA may be mounted on the motor shaft
outside the housing, provided there is no material slippage between the motor
and the shaft
segment SA. If the driver is coupled to shaft segment SA by a belt, magnet MA
may be mounted
on the driven pulley shaft. In another embodiment, the field lines of the
magnetic coupling 22
can be sensed as a reference. The waveform from the coupling is of the
type:
Asin(wt+a)+Bsin(2wt+b). The primary frequency signal can still be processed on
the fly by
extracting the multitone information and digital signal processing to filter
out the required
reference signal [Asin(wt+a)]. Therefore, in one of the embodiments, more than
one magnet
may be disposed in the mounting location to provide measurements.
[0028] Quality of the data is very important to get meaningful torque
measurements.
Before applying a phase measurement algorithm like convolution, cross
correlation or using a
lock-in-amplifier, it has to be ensured that both the signals (reference
magnet/Magdrive as well
as bottom magnet) are perfectly sinusoidal with minimal distortion. Both
signals are of the same
9
CA 02868847 2014-09-26
WO 2013/177332 PCT/US2013/042306
frequency before applying the phase shift algorithms which can be done both in
the time domain
as well as frequency domain. Application of filters and noise reduction should
be done,
ensuring, however, meaningful data is not lost. The calculated phase
difference will not
"stabilize" if the quality of data is bad.
[0029] While the preceding description contains many specificities,
however, it is to be
understood that same are presented only to describe some of the presently
preferred
embodiments of the invention, and not by way of limitation. Changes can be
made to various
aspects of the invention, without departing from the scope thereof.
[0030] Therefore, the scope of the invention is not to be limited to the
illustrative
examples set forth above, but encompasses modifications which may become
apparent to those
of ordinary skill in the relevant art.
