U.S. patent application number 17/575100 was filed with the patent office on 2022-07-14 for encoderless motor with improved quantization and methods of use and calibration.
The applicant listed for this patent is Cepheid. Invention is credited to Richard J. Casler, JR., Jeffrey Davis, Rajesh Nerkar.
Application Number | 20220224259 17/575100 |
Document ID | / |
Family ID | 1000006135586 |
Filed Date | 2022-07-14 |
United States Patent
Application |
20220224259 |
Kind Code |
A1 |
Davis; Jeffrey ; et
al. |
July 14, 2022 |
ENCODERLESS MOTOR WITH IMPROVED QUANTIZATION AND METHODS OF USE AND
CALIBRATION
Abstract
A DC electric motor having a stator mounted to a substrate, the
stator having a coil assembly having a magnetic core, a rotor
mounted to the stator with permanent magnets distributed radially
about the rotor, the permanent magnets extending beyond the
magnetic core, and sensors mounted to the substrate adjacent the
permanent magnets. During operation of the motor passage of the
permanent magnets over the sensors produces a substantially
sinusoidal signal of varying voltage substantially without noise
and/or saturation, allowing an angular position of the rotor to be
determined from the sinusoidal signals by utilizing a
transformation matrix or piece-wise algorithm applied in
substantially linear portions of the sinusoidal signals without
requiring use of additional hardware encoder or position sensors
and without requiring noise-reduction or filtering of the
signal.
Inventors: |
Davis; Jeffrey; (Sunnyvale,
CA) ; Nerkar; Rajesh; (Sunnyvale, CA) ;
Casler, JR.; Richard J.; (Sunnyvale, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Cepheid |
Sunnyvale |
CA |
US |
|
|
Family ID: |
1000006135586 |
Appl. No.: |
17/575100 |
Filed: |
January 13, 2022 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
63136766 |
Jan 13, 2021 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H02P 6/16 20130101 |
International
Class: |
H02P 6/16 20060101
H02P006/16 |
Goverment Interests
STATEMENT AS TO RIGHTS TO INVENTIONS MADE UNDER FEDERALLY SPONSORED
RESEARCH AND DEVELOPMENT
[0002] This invention was made with U.S. Government support under
Agreement No. W15QKN-16-9-1002 awarded by the ACC-NJ to the MCDC.
The Government has certain rights in the invention.
Claims
1. An n-phase encoder for use in a mechatronic system, the encoder
comprising: a movable element that applies a magnetic field with
period, S, the period representing a total displacement; a
stationary support with n magnetic field sensors mounted thereon
and positioned on the stationary support so as to measure the
magnetic field imparted by the movable element, wherein n is
greater than 1; and a processor communicatively coupled to the n
magnetic field sensors and configured to determine displacement of
the movable element based on n signals from the n magnetic field
sensors by processing the n signals utilizing a transformation
matrix.
2. The encoder of claim 1, wherein the total displacement is 2.pi.
radians of a field angle .PHI..
3. The encoder of claim 2, wherein the processor is configured to
process the n signals from the n sensors by: computing a sine and
cosine of the field angle, .PHI., by pre-multiplying an n-by-1
vector by a 2-by-n mathematical transformation matrix, M; and
computing the field angle, .PHI., as
.PHI.=tan-1(sin(.PHI.),cos(.PHI.)).
4. The encoder in claim 1, where S is a rotary displacement.
5. The encoder in claim 1, where S is a linear displacement.
6. The encoder in claim 3, wherein the mathematical transformation,
M, is configured such that calculation of the field angle, .PHI.,
is independent of an amplitude and bias of the magnetic field
sensors.
7. The encoder in claim 1, where the magnetic field sensors are
uniformly distributed within the period, S.
8. The encoder claim 1, where the system is configured such that
the applied magnetic field is represented by a sum of first and at
least one of higher-order harmonics.
9. The encoder of claim 1, wherein the processor is configured to:
store a runout represented by a spatially-varying signal
representing a difference between a true field angle and a sensed
field angle and utilizes the runout to compensate for the
difference thereby removing any runout error.
10. The encoder in claim 1, where the encoder is utilized in a BLDC
motor configured for operation of a mechatronic system within a
diagnostic assay system.
11. The encoder in claim 10, wherein the mechatronic system of the
diagnostic assay system comprises any of: a syringe, valve,
cartridge loading or door mechanism.
12. A processing method comprising: providing an n phase encoder of
a mechatronic system that includes a movable element that applies a
magnetic field with period, S, the period representing a total
displacement, and a stationary support with n magnetic field
sensors mounted thereon and positioned on the stationary support so
as to measure the magnetic field imparted by the movable element,
wherein n is greater than 1; obtaining signals from the n magnetic
field sensors corresponding to the measurements of the magnetic
field imparted by the movable element; and processing n signals
from the n magnetic field sensors by utilizing a transformation
matrix to determine a displacement of the movable element.
13. The processing method of claim 12, wherein the total
displacement is 2.pi. radians of a field angle .PHI..
14. The processing method of claim 13, wherein processing the
signals from the n sensors comprises: computing a sine and cosine
of the field angle, .PHI., by pre-multiplying an n-by-1 vector by a
2-by-n mathematical transformation matrix, M; and computing the
field angle, .PHI., as .PHI.=tan-1(sin(.PHI.),cos(.PHI.)).
15. The processing method of claim 12, wherein the processing
comprises normalizing an amplitude of at least one of the signals
from the n magnetic field sensors to an arbitrary value.
16. The processing method of claim 15, wherein processing of the
signals further comprises subtracting a signal bias of at least one
of the signals from the n magnetic field sensors before a
normalization operation.
17. The processing method of claim 16, wherein processing further
comprises storing in memory or outputting to the mechatronic
system, one or more signal bias coefficients
18. The processing method of claim 12, where S is a rotary
displacement.
19. The processing method in claim 12, where S is a linear
displacement.
20. The processing method in claim 12, wherein processing n signals
comprises processing only a substantially linear portion(s) of the
signals.
21. A calibration method for an n-phase encoder as in claim 1 in
which the amplitude, bias and phase-shift of the signal of the n
sensors is computed and stored in a memory of the encoder or the
mechatronic system in which it is employed.
22. The calibration method of claim 21, wherein a transformation
matrix, M(.PHI.)), specific to the mechatronic system accounts for
irregularities or otherwise actual phase-to-phase angle offsets.
Description
CROSS-REFERENCES TO RELATED APPLICATIONS
[0001] This application is a Non-Provisional of and claims the
benefit of priority of U.S. Provisional Application No. 63/136,766
filed on Jan. 13, 2021, the entire contents of which are
incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0003] The present invention relates to the field of electric
motors, in particular brushless DC electric motors, and pertains to
commutation and encoding for such motors.
[0004] This application is generally related to: U.S. Pat. No.
10,348,225 entitled "Encoderless Motor with Improved Granularity
and Methods of Use" issued Jul. 9, 2019; U.S. application Ser. No.
15/217,893 entitled "Molecular Diagnostic Assay System" filed Jul.
22, 2016; U.S. patent application Ser. No. 13/843,739 entitled
"Honeycomb tube," filed on Mar. 15, 2013; U.S. Pat. No. 8,048,386
entitled "Fluid Processing and Control," filed Feb. 25, 2002; U.S.
Pat. No. 6,374,684 entitled "Fluid Control and Processing System,"
filed Aug. 25, 2000; each of which is incorporated herein by
reference in its entirety for all purposes.
BRIEF SUMMARY OF THE INVENTION
[0005] In one aspect, the invention provides system and methods for
encoding a brushless DC electric motor in a manner that produces
extraordinarily high resolution and positional accuracy without
requiring use of any additional encoder hardware as taught herein.
The same system and methods can also provide for commutation of the
motor. In some embodiments, the system allows for encoding of a
brushless DC motor without use of a hardware encoder or additional
positional sensors and without requiring any noise-filtering of a
measured voltage signal.
[0006] In some embodiments, the invention provides a motor system
that includes a stator comprising a magnetic core, a movable
element (e.g. linear stage, rotor that is rotatably mounted
relative to the stator) having a plurality of permanent magnets
distributed therein), and multiple voltage sensors at fixed
positions relative the stator and disposed adjacent a path of the
plurality of magnets during movement of the movable element. The
system further includes a processor module communicatively coupled
with the multiple sensors and configured to determine a
displacement of the motor from the voltage signals from the sensors
without requiring use of a hardware encoder or additional
position-based sensor and/or without error correction of the
signal. In some embodiments having a rotatable rotor, the plurality
of magnets extend a distance (e.g. about 1 mm or greater) beyond
the magnetic core of the stator such that the signal from the
sensors is substantially without noise.
[0007] In some embodiments, the system includes a processing module
configured to: receive a measured voltage signal from each of the
sensors during rotation of the rotor, the signal being a
substantially sinusoidal signal of varying voltage during rotation
of the rotor; and determine a displacement of the motor from the
sinusoidal signals from the sensors. In some embodiments, the
processing module is configured to process the signals from at
least two sensors by utilizing a matrix transformation from which
the motor displacement (e.g. angular displacement of rotor) is
determined. Advantageously, utilizing matrix transformation from
the signals of at least two sensors that eliminates electrical
cycle harmonics that may arise from piece-wise applied algorithms
that may adversely affect the accuracy of a signal from an
individual sensor so as to provide increased resolution and
quantization and reduced runout--the difference between the actual
and the determined displacement--for determination of motor
displacement.
[0008] In some embodiments, the system includes a DC electric motor
having a stator mounted to a substrate and a rotor mounted to the
stator. The stator includes a coil assembly having a core and
electrical windings, the coil assembly having an outside diameter,
a proximal extremity, and a distal extremity. The rotor includes
permanent magnets disposed along an outer edge (e.g. mounted to a
cylindrical skirt), the rotor having an outside diameter, an inside
diameter, and a distal edge. In some embodiments, the permanent
magnets extend beyond the distal extremity of the magnetic core of
the stator (i.e. the coil assembly). The system further includes
multiple sensors mounted to the substrate adjacent the permanent
magnets. While the sensors are mounted to the substrate in this
embodiment, it is appreciated that the sensors could be mounted to
any support or element that is adjacent to the magnets of the
movable element (e.g. rotor, translator) and that remains
stationary with the stator. In some embodiments, the rotor is
fabricated using a series of separate permanent magnets arranged in
a pattern of alternating opposite polarity of the adjacent magnets
at the distal edge of the skirt. In some embodiments, the rotor is
defined as a single piece of magnetic material (e.g. a
ferromagnetic or ferrimagnetic material) in the shape of a strip, a
ring or a disk, that is then magnetized to create the pattern of
alternating opposite magnetic polarity at the distal edge of the
skirt. Both fabrication methods are suitable for use with the
invention. In some embodiments, the core is a core of magnetic
material, typically a metal or other paramagnetic material.
Non-limiting exemplary materials suitable for use in the core of
the instant invention include iron, especially soft iron, cobalt,
nickel, silicon, laminated silicon steel, silicon alloys, special
alloys (e.g. mu-metal, permalloy, supermalloy, sendust), and
amorphous metals (e.g. metglas). The core may also include air, and
in some embodiments, the core is an air core. During operation of
the motor, passage of the permanent magnets over the sensors
produces a substantially sinusoidal signal of varying voltage
substantially without noise and/or saturation, thereby allowing an
angular position of the rotor relative the substrate to be
determined from the sinusoidal signal without requiring use of
encoder hardware or positional sensors. Thus, displacement of the
motor can be determined and controlled with a high degree of
accuracy and resolution. For example, a motor as described herein
comprising 12 permanent magnets and 9 poles and using 3 Halls
sensors and an 11-bit analog to digital converter as a processing
module can deliver a resolution of about 0.01 degrees mechanical
rotation, without use of any encoder hardware or positional sensors
or noise filtering. The resolution and accuracy of the system can
be increased or decreased by changing the number of poles, the
number of permanent magnets, or using a higher or lower bit
ADC.
[0009] In some embodiments, the multiple sensors mounted are
positioned relative to the extended edge of the permanent magnets.
The multiple sensors can be mounted on the substrate that the
movable element is mounted to or can be mounted to a stationary
support element that remains stationary with the stator while the
movable element (e.g. translator, rotor) moves. The position is
defined such that a clearance from the extended edge of the
permanent magnets to the multiple sensors is sufficient to provide
a DC voltage signal substantially without noise and/or saturation.
In some embodiments, the edge of the permanent magnets extends
beyond the distal extremity of the coil assembly by about 100
microns. In some embodiments, the permanent magnets extend beyond
the distal extremity of the coil assembly by less than 100 microns,
e.g., 90, 80, 70, 60, 50, 40, 30, 20, 10 microns or less, depending
on the particular embodiments of the motor. In some embodiments,
the permanent magnets extend beyond the distal extremity of the
coil assembly by more than 100 microns, for example, 200, 300, 400,
500, 600, 700, 800, 900, 1000 microns, including all values between
about 100 microns and 1000 microns, or more depending on the
particular embodiments of the motor. In some embodiments, the
permanent magnets extend beyond the distal extremity of the coil
assembly by about 1 mm or more, including but not limited to about
1 mm, 2 mm, 3 mm, 4 mm, 5 mm, 6 mm, 7 mm, 8 mm, 9 mm, 10 mm, or
more. The exact distance that the permanent magnet extends beyond
the distal extremity of the coil assembly depends on the particular
characteristics and embodiments of the motor, and is well within
the skill of an ordinary artisan to determine based on the guidance
provided herein. In some embodiments, the one or more sensors are
linear Hall-effect sensors, spaced apart by a common arc length
along an arcuate path of the rotor.
[0010] In one aspect, the invention pertains to an n-phase encoder
for use in a mechatronic system. Such encoders can be utilized in a
BLDC motor configured for operation of a mechatronic system or
sub-assembly, such as within a diagnostic assay system. The
mechatronic system for a diagnostic assay system can include any
of: a syringe, valve, cartridge loading or door mechanism. In some
embodiments, the encoder system can include: a movable element that
applies a magnetic field with period, S, the period representing a
total displacement in the coordinate frame of the movable element;
and a stationary support with n magnetic field sensors mounted
thereon and positioned on the stationary support so as to measure
the magnetic field imparted by the movable element, wherein n is
greater than 1. In some embodiments, the encoder includes a
processor communicatively coupled to the n magnetic field sensors
and configured to determine displacement of the movable element
based on n signals from the n magnetic field sensors by processing
the n signals utilizing a transformation matrix. Typically, the
total displacement is a radians of a field angle .PHI.. In some
embodiments, the processor is configured to process the n signals
from the n sensors by: computing the sine and cosine of the field
angle, .PHI., by pre-multiplying the n-by-1 vector by a 2-by-n
mathematical transformation matrix, M; and computing the field
angle, .PHI., as .PHI.=tan-1(sin(.PHI.),cos(.PHI.)). The period S
can be a rotary displacement or a linear displacement.
[0011] In some embodiments, the mathematical transformation, M, is
configured such that calculation of the field angle, .phi., is
independent of an amplitude and bias of the magnetic field sensors.
One of the advantages of the implementations described herein is
that the encoder is insensitive to the common-mode amplitude and
bias in the n signals, the term "common-mode" referring to the
"mean" of the signal attributes. Preferably, the magnetic field
sensors are uniformly distributed within the period, S. In some
embodiments, the system is configured such that the applied
magnetic field is represented by a sum of first and at least one of
higher-order harmonics. This provides that the
position/displacement can be derived even if there are higher-order
harmonics in the signals. In some embodiments, the processor is
configured to store a runout represented by a spatially-varying
signal representing the difference between the true field angle and
the sensed field angle and utilizes the runout to compensate for
the difference so as to remove a runout error. It can be
advantageous to calibrate the runout function and subtract the
known error by storing this from a prior calibration or in-situ
self-test prior to operating the system. In some embodiments, the
runout can be specified as a linear sum of sine and cosine of
harmonics of S. This may improve accuracy in operation of the
mechatronic system (e.g. syringe mechanism in aspirating and
dispensing).
[0012] In another aspect, the invention pertains to methods of
determining displacement utilizing an encoder as described above.
Such methods can includes: providing an n phase encoder of
mechatronic system that includes a movable element that applies a
magnetic field with period, S, the period representing a total
displacement within the coordinate frame of the movable element,
and a stationary support with n magnetic field sensors mounted
thereon and positioned on the stationary support so as to measure
the magnetic field imparted by the movable element, wherein n is
greater than 1; obtaining signals from the n magnetic field sensors
corresponding to the measurements of the magnetic field imparted by
the movable element; and processing n signals from the n magnetic
field sensors by utilizing a transformation matrix to determine a
displacement of the movable element. The displacement S can be
rotary or linear. Typically, the total displacement is a radians of
a field angle .PHI.. In some embodiments, processing the signals
comprises: computing the sine and cosine of the field angle, .PHI.,
by pre-multiplying the n-by-1 vector by a 2-by-n mathematical
transformation matrix, M; and computing the field angle, .PHI., as
.PHI.=tan-1(sin(.PHI.),cos(.PHI.)).
[0013] In some embodiments, processing of the multiple signals can
include normalizing an amplitude of at least one of the signals
from the n magnetic field sensors to an arbitrary value. In some
embodiments, processing can further include subtracting the signal
bias of at least one of the signals from the n magnetic field
sensors before the normalization operation. In some embodiments,
the system and methods include determining one or more signal bias
coefficients. In some embodiments, the processing further includes
storing in memory or outputting to the mechatronic system, the
signal bias coefficients. These additional aspects of signal
processing can be advantageous in the case where the n sensors have
uneven amplitude and bias, so as to compensate for any
phase-to-phase differences in amplitude and bias.
[0014] In another aspect, the invention pertains to a method of
calibrating the encoder described above. Such calibration methods
can include storing any or all of: a signal amplitude, a bias, and
a phase-shift of the sensors. In some embodiments, the calibration
method includes computing the signal amplitude, bias and
phase-shift of the n sensors and storing in a memory of the
software encoder or the mechatronic system in which it is employed.
In some embodiments, the calibration method includes computing the
signal amplitude, bias and a matrix transformation that
incorporates the 2-by-n mathematical transformation matrix,
M(.PHI.) specific to the calibrated phase-to-phase angles .PHI., in
the mechatronic system. These additional aspects of signal
processing can be advantageous in the case where the n sensors are
irregularly spaced in relation to a nominal phase relationship as
might arise in component placement and soldering during mechatronic
system fabrication.
[0015] In yet another aspect, the encoder approach can be modified
for a multi-speed mechatronic system. Such an encoder system can
include: a movable element that applies at least two of a
spatially-varying field, a first applying a field of period, S1 and
a second, S2 where S2 is an integer multiple of the period, S1; and
a stationary support having n1 sensors arranged within the period
S1 and n2 sensors arranged within the period S2, where n1 and n2
are each greater than or equal to two, the sensors being configured
to measure the magnetic field of the movable element. The system
can further include a processor that obtains the signals from the
sensors and applies a transformation matrix to determine field
angles, .PHI. 1 and .PHI. 2. In some embodiments, the processor is
configured to: process n1 and n2 sensor signals from the n1 and n2
sensors; apply a mathematical transformation to compute sine and
cosine of the field angles, .PHI. 1 and .PHI. 2, respectively; and
compute .PHI. 2 with substantially equivalent resolution to .PHI.
1. The above described approaches can also be utilized with various
other encoding approaches as well.
BRIEF DESCRIPTION OF THE DRAWINGS
[0016] FIG. 1 is a plan view diagram illustrating elements of a
brushless DC electric motor in an exemplary embodiment of the
present invention.
[0017] FIG. 2A is a side elevation view, partly in section, of the
motor depicted in FIG. 1, mounted on a substrate.
[0018] FIG. 2B is a magnified elevation view of area 2b of FIG. 2A,
illustrating spacing between magnets and sensors in an exemplary
embodiment of the invention.
[0019] FIG. 3A is a plan view of the substrate of FIG. 2B with the
motor removed, illustrating placement of sensors in an exemplary
embodiment of the invention.
[0020] FIG. 3B illustrates the placement of the permanent magnets
in the rotor in a pattern of alternating polarity at the distal
edge, and showing the fringing fields of the adjacent permanent
magnets according to an exemplary embodiment of the invention.
[0021] FIG. 4 illustrates an essentially sinusoidal variable
voltage pattern produced by passage of permanent magnets of a motor
rotor over a first Hall-effect sensor in an exemplary embodiment of
the invention.
[0022] FIG. 5 illustrates a sinusoidal variable voltage pattern
produced by passage of permanent magnets of a rotor over a second
Hall-effect sensor, the pattern superimposed over the pattern of
FIG. 4.
[0023] FIG. 6 illustrates a sinusoidal variable voltage pattern
produced by passage of permanent magnets of a rotor over a third
Hall-effect sensor, with the pattern superimposed over the patterns
of FIG. 5 from which displacement can be determined in accordance
with some embodiments of the invention.
[0024] FIG. 7 is a diagram depicting circuitry in an exemplary
embodiment of the invention for controlling a DC motor using output
of the Hall-effect sensors.
[0025] FIG. 8 is a control schematic using PID control to control
PWM and drive direction of a motor mechanism.
[0026] FIG. 9 illustrates a method of determining displacement of a
motor during operation in accordance with some embodiments of the
invention.
[0027] FIG. 10 illustrates a linear motor with integrated sensor
encoder in accordance with some embodiments of the invention.
DETAILED DESCRIPTION OF THE INVENTION
[0028] FIG. 1 is a plan view diagram illustrating elements of a
brushless DC (BLDC) electric motor 100 in a non-limiting exemplary
prototype. It is appreciated that such a motor could be used in a
wide variety of applications, and is of particular use for
operation of a small-scale mechanical mechanism requiring a high
level of accuracy and quantization. Some embodiments include a
motor system having improved resolution in the determination of
motor displacement without use of hardware encoders and/or
noise-filtering, for example, a resolution of about 0.1 degrees of
mechanical rotation, or preferably about 0.01 degrees mechanical
rotation, or even about 0.001 degrees of mechanical rotation or
less. One such application is operation of a syringe drive to
effect highly precise fluid metering, or operation of a valve
assembly of a diagnostic assay system that interfaces with a sample
cartridge in order to facilitate a complex sample processing and/or
analysis procedure upon fine-tuned movement of the valve assembly.
Examples of such applications can be found in U.S. Pat. No.
10,562,030 entitled "Molecular Diagnostic Assay System" and U.S.
Pat. No. 8,048,386 entitled "Fluid Processing and Control," filed
Feb. 25, 2002; U.S. Pat. No. 6,374,684 entitled "Fluid Control and
Processing System," filed Aug. 25, 2000, the entire contents of
which are incorporated herein by reference.
[0029] In one aspect, the BLDC motor includes a rotor, a stator,
and multiple analog voltage sensors configured to produce a
smoothly varying Hall-effect voltage without any need for filtering
or noise reduction of the individual signals. In some embodiments,
this feature is provided by use of permanent magnets within the
rotor that extend a distance beyond the magnetic core of the
stator. In some embodiments, the BLDC motor includes as many analog
voltage sensors as phases of the motor, which are positioned such
that the motor can be controlled based on the measured voltage
patterns received from the sensors. In one aspect, the voltage
signals from the sensors are processed utilizing a transformation
matrix, thereby avoiding any inaccuracies that might otherwise
arise from algorithms applied in a piecewise method. In some
embodiments, this includes spacing the sensors radially uniformly
about the stator such that the measured voltage waveforms are
offset uniformly. For example, a three-phase BLDC can include three
Hall-effect sensors spaced 40 degrees radially from each other,
thereby allowing the system to control a position of the sensor
within an increment of 40 degrees. The signals from the sensors can
be used to determine the displacement of the motor with a high
degree of accuracy. When utilized individually, although each
signal is substantially free from noise and/or saturation, a
velocity ripple or other motor harmonic arising from the piecewise
method may adversely affect displacement determinations based on a
single sensor signal. A velocity ripple describes cyclical
variations or oscillations of rotational speed over time in
comparison to the actual speed or displacement. Therefore, to
overcome any adverse effects attributable to a velocity ripple on
an individual signal, the control methods can obtain and process
multiple signals from the distributed sensors to determine a given
displacement of the motor. In some embodiments, a processor
communicatively coupled to the motor is configured to obtain and
process the signals from the multiple sensors by application of a
mathematical transformation matrix (e.g. for three sensors a
2.times.3 transformation matrix) so as to determine the motor
displacement with a high level of accuracy despite the presence of
any velocity ripple affecting an individual signal.
[0030] As described herein, the sensor and processor configuration
provides for highly accurate determination of motor displacement
from multiple sensor signals that detect the magnetic field of the
permanent magnets that also effect movement of the movable element
(e.g. translator, rotor). Although this configuration is referred
to herein as an "encoder", this concept is distinguishable from
conventional hardware encoders or additional position-based sensors
that require additional encoder hardware on the motor (e.g.
additional optical, electrical, or magnetic features on the shaft
or rotor dedicated to encoding). Thus, the "encoder" described in
further detail below is greatly simplified as compared to a
conventional hardware encoder since it only requires sensing of
existing magnets in the system. This provides for improved
integration within the motor design and associated control unit as
it does not require any additional encoder hardware or additional
position-based detection components beyond the sensors noted below.
It is further noted that the encoder described herein, allows for
accurate determination from the sensor signals without filtering,
noise filtering, or error correction of individual signals and
without iteration or recursive solutions.
[0031] In one aspect, the encoder in accordance with the present
invention is used to effect movement of a mechatronic sub-assembly.
In a preferred embodiment, the encoder is utilized in a small-scale
motor or mechatronic system within a diagnostic assay system, for
example, a syringe, valve, cartridge loading or door mechanism, or
other mechatronic sub-assembly. In the embodiments described below,
the encoder is a multi-phase encoder corresponding to the number of
phases of the motor that it encodes. Although a three-phase motor
is described in the following examples, it is appreciated that the
concepts described herein could be applied to any multi-phase motor
configuration.
[0032] For a three-phase encoder, the system can include: a movable
element (e.g. rotor, linear stage) that applies a magnetic field
with period, S, the period representing a displacement of 2.pi.
radians of a field angle, .PHI.; a stationary element with 3
magnetic field sensors (e.g. Hall-effect sensors) configured to
measure the magnetic field imparted by permanent magnets of the
movable element; and a processor (e.g. microprocessor, control
unit) that is configured to obtain the signals from the three
sensors and to process the signals from the three by: (a) computing
the sine and cosine of the field angle, .PHI., by pre-multiplying
the 3-by-1 vector by a 2-by-3 mathematical transformation matrix,
M; and (b) computing the field angle .PHI., as .PHI.=tan-1 (sin
(.PHI.), cos (.PHI.)). The following equations can be used for
implementation of this integrated sensor encoder approach for three
sensors placed 2703 radians apart.
[ hs .times. .times. 1 hs .times. .times. 2 hs .times. .times. 3 ]
= V A * [ sin .function. ( .phi. ) sin .function. ( .phi. + 2
.times. .pi. 3 ) sin .function. ( .phi. + 4 .times. .pi. 3 ) ]
.times. .times. where , .times. hs .times. .times. 1 = hall .times.
.times. sensor .times. .times. 1 .times. .times. hs .times. .times.
2 = hall .times. .times. sensor .times. .times. 2 .times. .times.
hs .times. .times. 3 = hall .times. .times. sensor .times. .times.
3 eqn .function. ( 1 ) [ hs .times. .times. 1 hs .times. .times. 2
hs .times. .times. 3 ] = V A * [ sin .function. ( .phi. ) sin
.function. ( .phi. ) .times. cos .function. ( 2 .times. .pi. 3 ) +
cos .function. ( .phi. ) .times. sin .function. ( 2 .times. .pi. 3
) sin .function. ( .phi. ) .times. cos .function. ( 4 .times. .pi.
3 ) + cos .function. ( .phi. ) .times. sin .function. ( 4 .times.
.pi. 3 ) ] eqn .function. ( 2 ) [ hs .times. .times. 1 hs .times.
.times. 2 hs .times. .times. 3 ] = V A * [ 1 0 cos .function. ( 2
.times. .pi. 3 ) sin .function. ( 2 .times. .pi. 3 ) cos .function.
( 4 .times. .pi. 3 ) sin .function. ( 4 .times. .pi. 3 ) ] * [ sin
.function. ( .phi. ) cos .function. ( .phi. ) ] eqn .function. ( 4
) [ hs .times. .times. 1 hs .times. .times. 2 hs .times. .times. 3
] = V A * [ 1 0 ( - 1 2 ) ( 3 2 ) ( - 1 2 ) ( - 3 2 ) ] * [ sin
.function. ( .phi. ) cos .function. ( .phi. ) ] eqn .function. ( 5
) [ hs .times. .times. 1 hs .times. .times. 2 hs .times. .times. 3
] = V A * A * [ sin .function. ( .phi. ) cos .function. ( .phi. ) ]
eqn .function. ( 6 ) [ sin .function. ( .phi. ) cos .function. (
.phi. ) ] = 1 V A * ( A T .times. A ) - 1 .times. A T * [ hs
.times. .times. 1 hs .times. .times. 2 hs .times. .times. 3 ] eqn
.function. ( 7 ) [ sin .function. ( .phi. ) cos .function. ( .phi.
) ] = 1 V A * [ 2 3 - 1 3 - 1 3 0 3 3 - 3 3 ] * [ hs .times.
.times. 1 hs .times. .times. 2 hs .times. .times. 3 ] eqn
.function. ( 8 ) ##EQU00001##
[0033] The matrix M in equation (8) can then be computed using the
phase difference of 2.pi./3 radians between sensor signals.
[ sin .function. ( .phi. ) cos .function. ( .phi. ) ] = 1 V A * M *
[ hs .times. .times. 1 hs .times. .times. 2 hs .times. .times. 3 ]
eqn .function. ( 9 ) .phi. = tan - 1 .function. ( sin .function. (
.phi. ) , cos .function. ( .phi. ) ) eqn .function. ( 10 )
##EQU00002##
In another aspect, the sensors can be calibrated for magnitude,
offset and phase, by modifying equation (1). This phase calibration
will compute a new M matrix in equation (8).
[ m 1 * hs .times. .times. 1 + off 1 m 2 * hs .times. .times. 2 +
off 2 m 3 * hs .times. .times. 3 + off 3 ] = V A * [ sin .function.
( .phi. ) sin .function. ( .phi. + ( 2 .times. .pi. 3 + .0. 1 ) )
sin .function. ( .phi. + ( 4 .times. .pi. 3 + .0. 2 ) ) ]
##EQU00003##
[0034] It is appreciated that the above is an example
implementation of this approach applied to a three-phase encoder,
as described herein, and that this same approach could be adjusted
and utilized for encoding of any multi-phase system or motor or
even a multi-speed motor.
[0035] In the foregoing specification, the invention is described
with reference to specific embodiments thereof, but those skilled
in the art will recognize that the invention is not limited
thereto. As used throughout, the term "about" can refer to the
.+-.10% of the recited value. Various features and aspects of the
above-described invention can be used individually or jointly. It
is appreciated that any of the aspects or features of the
embodiments described herein could be modified, combined or
incorporated into any of the embodiments described herein, as well
as in various other types and configurations. Further, the
invention can be utilized in any number of environments and
applications beyond those described herein without departing from
the broader spirit and scope of the specification. The
specification and drawings are, accordingly, to be regarded as
illustrative rather than restrictive.
[0036] In some embodiments, such as that shown in FIG. 1, the motor
includes an internal stator assembly 101 having nine pole teeth
extending radially from center, each pole tooth ending in a pole
shoe 103, and each pole tooth having a winding providing an
electromagnetic coil 102. The motor further includes an external
rotor 104 having an external cylindrical skirt 105 and twelve
permanent magnets 106 arranged with alternating polarity around the
inner periphery of the skirt 105. The permanent magnets are shaped
to provide a cylindrical inner surface for the rotor with close
proximity to outer curved surfaces of the pole shoes. The BLDC
motor in this example is a three-phase, twelve pole motor. Controls
provided, but not shown in FIG. 1, switch current in the coils 102
providing electromagnetic interaction with permanent magnets 106 to
drive the rotor, as would be known to one of skill in the art.
While an internal stator and external rotor are described here, it
is appreciated that this approach can also be used in a motor
having an internal rotor and external stator, or a linear motor as
well.
[0037] It should be noted that the number of pole teeth and poles,
and indeed the disclosure of an internal stator and an external
rotor are exemplary, and not limiting in the invention, which is
operable with motors of a variety of different designs.
[0038] FIG. 2A is a side elevation view, partly in section, of the
motor of FIG. 1, cut away to show one pole tooth and coil of the
nine, ending in pole shoe 103 in close proximity to one of the
twelve permanent magnets 106 arranged around the inner periphery of
cylindrical skirt 105 of external rotor 104. The pole teeth and
pole shoes of stator assembly 101 are a part of the core, and
define a distal extremity of the core at the height of line 204.
Stator assembly 101 is supported in this implementation on a
substrate 201, which in some embodiments is a printed circuit board
(PCB), which can include a control unit and traces configured for
managing switching of electrical current to coils 102 so as to
provide electromagnetic fields interacting with the fields of
permanent magnets 106 to drive the rotor. The PCB substrate can
also include control circuitry for encoding and commutation. Rotor
104 engages physically with stator 101 by drive shaft 107, which
engages a bearing assembly in the stator to guide the rotor with
precision in rotation. Details of bearings are not shown in FIG.
2A, although it is appreciated that there are many conventional
ways such bearings can be implemented. Drive shaft 107 in this
implementation passes through an opening for the purpose in PCB 201
and can be engaged to drive mechanical devices.
[0039] Three linear Hall-effect sensors 202a, 202b and 202c are
illustrated in FIG. 2A, supported by PCB 201, and positioned
strategically according to some embodiments so as to produce
variable voltage signals that may be used in a process to encode
and provide commutation for motor 100. In FIG. 2A the overall
height of skirt 105 of rotor 104 is represented by dimension D.
Dimension d1 represents extension of the distal extremity of the
rotor magnets below the distal extremity of the magnetic core at
line 204. In some embodiments, the direction of this extension is
transverse, typically perpendicular, to the plane along which the
rotor rotates.
[0040] FIG. 2B is a magnified elevation view of area 2b of FIG. 2A
illustrating clearance d2 between the distal edge of the permanent
magnets of rotor 104 and the structural bodies of the Hall-effect
sensors 202a, 202b and 202c on PCB 201.
[0041] FIG. 3A is a plan diagram of a portion of PCB 201 taken in
the direction of arrow 3 of FIG. 2A, showing placement of
Hall-effect sensors 202a, 202b and 202c relative to the distal edge
of rotor 104, which may be seen in FIG. 2 to extend below the
distal edge of the core by dimension d. In FIG. 3 the rotation
track of rotor 104 including the twelve permanent magnets 106 is
shown in dotted outline 212. The rotor rotates in either direction
213 about the rotational center 211 depending on details of
commutation. It is appreciated that the approaches described herein
can be used regardless of the direction of rotation of the
rotor.
[0042] As illustrated in this non-limiting exemplary prototype,
each of Hall-effect sensors 202a, 202b and 202c is positioned
beneath the distal edge of the rotor magnets, just toward the
inside, radially, of the central track of the rotating magnets.
Hall-effect sensor 202b is located to be forty degrees arc from
Hall-effect sensor 202a along the rotating track of the magnets of
the rotor. Similarly, Hall-effect sensor 202c is located a further
forty degrees around the rotor track from Hall-effect sensor
202b.
[0043] FIG. 3B is a perspective view of three of the permanent
magnets 106 in relation to two of the Hall-effect sensors 202a and
202b in this non-limiting exemplary prototype. The permanent
magnets in the rotor are placed with alternating polarity as shown
in FIG. 3B, and at the distal edge of the rotor fringing fields 214
are exhibited between adjacent permanent magnets. It is these
fringing fields between adjacent permanent magnets that the analog
sensors (e.g. Hall-effect sensors) are placed and spaced to sense
the fringing fields, and the curved aspect of the inner surfaces of
the permanent magnets dictate the placement of the Hall-effect
sensors somewhat to the inside, radially, of the central track of
the magnets in rotation. The Hall-effect sensors can be placed
inside by a small distance, such as 1 mm or more (e.g. 2, 3, 4 mm)
depending on the overall size of the device and the strength of the
magnetic fields being dictated. It is advantageous for the space
between the sensors and the permanent magnets (i.e. d2) to be
minimized in order to substantially eliminiate noise in the
detection signal. Similarly, in a linear motor the sensors can be
mounted on an support element that extends between the magnetic
core of the stator and the permanent magnets along the movable
element (e.g. linear stage) such that a space between the sensors
and the magnets is less than the space between the magnetic core of
the stator and the magnets, thereby substantially eliminating noise
from the signal. Alternatively, in a linear motor the sensors can
be mounted on a movable stage that extends between the magnetic
core of the stage and the permanent magnets along the stationary
element such that a space between the sensors and the magnets is
less than the space between the magnetic core of the stage and the
magnets, thereby substantially eliminating noise from the
signal.
[0044] Referring back to FIG. 2A, dimension d1 refers to an
extension distance of the distal extremity of the rotor magnets
below the distal extremity of the core at line 204. In conventional
motors, there is no reason or motivation to extend this edge below
the extremity of the core, particularly since this can increase the
height of the motor and require increased clearance between the
rotor and substrate. In fact, the skilled artisan would limit
dimension D so there is no such extension, as the added dimension
would only add unnecessary cost and bulk to a conventional motor.
Furthermore, in conventional motors at the distal extremity of the
rotor, at the height of or above the distal extremity of the core,
switching of current in coils 102 creates a considerable field
effect, and a signal from a Hall-effect sensor placed to sense
permanent magnets at that position would not produce a smoothly
varying Hall-effect voltage. Rather, the effect in a conventional
motor is substantially noise corrupted. The conventional approach
to this dilemma is to introduce noise-filtering, or more commonly
to utilize a hardware encoder.
[0045] Advantageously, extending the rotor magnets below the distal
extremity of the iron core avoids the corrupting effect of the
switching fields from the coils of the stator on the signal from
the Hall-effect sensors. The particular extension d1 will depend on
several factors specific to the particular motor arrangement, and
in some embodiments will be 1 mm or more (e.g. 2 mm, 3 mm, 4 mm, 5
mm, 6 mm, or greater), while in some other embodiments the
extension will be less than 1 mm. In some embodiments, the distance
is a function of the size of the permanent magnets and/or the
strength of the magnetic field. In some embodiments of the
exemplary prototype as detailed herein, 1 mm of extension is
sufficient to produce a sinusoidal signal of varying voltage
without noise or saturation. Placement of the Hall-effect sensors
at a separation d2 to produce a Hall-effect voltage produces a
smoothly variable voltage, devoid of noise. In some embodiments,
the Hall-effect sensors produce a smoothly variable DC voltage in
the range from about 2 volts to about 5 volts devoid of noise or
saturation. The dimension d2 may vary depending on choice of
sensor, design of a rotor, strength of permanent magnets in the
rotor, and other factors that are well known to persons of skill in
the art. A workable separation is readily discovered for any
particular circumstance, to avoid saturation of the sensor and to
produce a smoothly variable DC voltage substantially devoid of
noise.
[0046] FIG. 4 illustrates a sinusoidal variable voltage pattern 40
produced by passage of permanent magnets 106 of rotor 104 over
Hall-effect sensor 202a in a three-phase BLDC motor. The 0 degree
starting point is arbitrarily set to be at a maximum voltage point.
Three complete sine waveforms are produced in one full 360 degree
revolution of the rotor.
[0047] FIG. 5 illustrates a substantially noise free sinusoidal
variable voltage pattern 50 produced by passage of permanent
magnets 106 of rotor 104 over Hall-effect sensor 202b, with the 50
pattern superimposed over the 40 pattern of FIG. 4. As Hall-effect
sensor 202b is positioned at an arc length of 40 degrees from the
position of Hall-effect sensor 202a, sinusoidal pattern 50 is
phase-shifted by 20 degrees from that of sinusoidal pattern 40.
[0048] FIG. 6 illustrates a substantially noise free sinusoidal
variable voltage pattern 60 produced by passage of permanent
magnets 106 of rotor 104 over Hall-effect sensor 202c, with the 60
pattern superimposed over the 40 and 50 patterns of FIG. 5. As
Hall-effect sensor 202c is positioned at an arc length of 40
degrees from the position of Hall-effect sensor 202b, sinusoidal
pattern 50 is phase-shifted by 120 degrees from that of sinusoidal
pattern 40. The patterns repeat for each of the electrical cycles
comprising the 360 degree rotation of the rotor. As can be seen,
each of the signals is offset such that the signals can be combined
by utilizing a transformation matrix for determination of
displacement despite any inaccuracy associated with an individual
signal due to electrical cycle harmonics or other cyclical
factor.
[0049] In some embodiments, the approaches described herein provide
for a high degree of accuracy and precision for mechanisms driven
by motor 100. In the non-limiting example described above using an
11-bit ADC, the motor position can be controlled to 0.0005 degree
mechanical. Coupled with gear reduction extremely fine control of
translation and rotation of mechanisms can be attained. In some
embodiments, motor 100 is coupled to a translation drive for a
syringe-pump unit to take in and expel fluid in an analytical
chemical processes.
[0050] FIG. 7 is a diagram depicting circuitry in one embodiment of
the invention for controlling motor 100 using the output of the
Hall-effect sensors and the unique method of processing the signals
from the phase-separated curves produced by the sensors by
utilizing a transformation matrix as described above. Output of the
Hall-effect sensors 202a, 202b is provided to a
proportional-integral-derivative (PID) motion control circuitry for
commutation purpose, and the waveforms produced by interaction of
the rotor magnets with the Hall-effect sensors is provided to
multiplexer circuitry as shown in FIG. 9. In some embodiments,
displacement of the motor can be determined by processing two or
more sinusoidal curves of the voltage signals measured by two or
more sensors of the motor.
[0051] FIG. 8 is a control schematic depicting control over
adjusting a Pulse-width-modulation (PWM) and drive direction of a
motor by use of a PID controller. The PID controller incorporates
the difference between a desired commanded position of the motor
displacement and the measured position. The difference in these
inputs is commonly referred to as the following error. While in
conventional devices, the measured encoder position is provided by
a hardware encoder or position-based sensor, in some embodiments,
this input can be provided by the measured voltages from the analog
sensors without requiring use of any hardware sensor or
position-based sensor. Thus, the approach described herein allows
for determination of an input that is conventionally provided by a
hardware encoder, without otherwise altering the control
configuration. It is appreciated, however, that a processing unit
would be adapted to determine the encoder position input.
[0052] FIG. 9 illustrates a method in accordance with some
embodiments. The method includes: Operating a DC motor having a
rotor with permanent magnets distributed about an outer periphery
and a stator having a magnetic core. During operation of the motor,
the system receives an analog sinusoidal signal from each of at
least two analog sensors in a fixed position relative the stator,
the sensors being spaced apart from each other such that the
sinusoidal signals are offset from each other. The system can
include multiple analog sensors, such as Hall-effect sensors,
distributed uniformly along at least a portion of the motor's
rotational path. In some embodiments, such a configuration includes
at least three such sensors separated by about 40 degrees from each
other, as shown in FIG. 3A. The system then determines a
displacement of the motor based on the sinusoidal signals by
processing the multiple signals utilizing a transformation matrix,
in accordance with the approach described above. The displacement
of the motor can be used by the system to inform various other
processes or functions of the system, or can be used to control the
motor by use of a control loop which includes the motor
displacement as an input to a controller, such as a PID controller.
In some aspects, this approach can be used to facilitate or
fine-tune operation of a small scale valve mechanism or a syringe
drive mechanism in a diagnostic assay system or other such fluid
processing system.
[0053] In some cases, the above algorithms may not be suited for
operation on a simple microprocessor. In some embodiments, the
algorithms utilize a look-up table combined with a Newton-Raphson
or equivalent iterative numerical solution. This can be implemented
as a simple subroutine call in a PSoC processor having a floating
point implementation. An alternative approach of utilizing only
substantially the linear portions of the sinusoidal signals, such
as the centroid approach described in U.S. Pat. No. 10,348,225
entitled "Encoderless Motor with Improved Granularity and Methods
of Use" issued Jul. 9, 2019, utilizes only involves multiplication
and division such that it can be implemented in simpler
microprocessors where more complex facilities are not possible.
[0054] As described above in the non-limiting exemplary
embodiments, an ADC is used to produce the division of the straight
portions of the phase-separated waveforms and motor 100, which can
be driven by, for example, a DRV83 13 Texas Instruments motor
driver circuit. It is understood that there are other arrangements
of circuitry that might be used while still falling within the
scope of this approach. In some embodiments the circuitry and coded
instructions for sensing the Hall-effect sensors and providing
motor encoding may be implemented in a programmable system on a
chip (PSOC) on the PCB.
[0055] FIG. 10 illustrates the concepts described herein applied to
a linear motor. In this example, the linear motor 300 includes an
elongated cylindrical stator 301 and a movable translator 304 that
moves back and forth through the stator in a linear manner. The
stator has a series of electromagnetic coils 302 that are
distributed linearly within. The cylindrical translator 304 has a
series of alternating permanent magnets 305 that effect
back-and-forth movement of the translator upon selective activation
of the electromagnetic coils of the stator by the control unit 312
having a processor therein. Movement of the translator is
faciltiated by one or more bearings 306. The spacing of the
permanent magnets is referred to as the pole pitch 307, which
corresponds to the spacing of the electromagnetic coils referred to
as the slot pitch 303. The linear motor 300 further includes
multiple magnetic field sensors 312 (e.g. Hall Effect sensors) that
are disposed on a stationary support element 310 that remains
stationary with the stator to detect the magnetic field of the
permanent magnets of the translator. In this embodiment, there are
three sensors 312 spaced apart and positioned so that a gap between
the sensors and the permanent magnets is less than a gap between
the coils of the stator and the permanent magnets. This allows the
signal (e.g. voltage measurement) from each of the sensors to be
substantially without noise and/or without saturation so that
multiple signals can be processed to determine displacement by
utilizing a transformation matrix, as described above, to allow for
accurate determination of the displacement of the translator.
[0056] It is appreciated that a variety of alterations can be made
in the embodiments described herein without departing from the
scope of the invention. For example, electric motors of different
designs might be incorporated and controlled in alternative
embodiments of the invention by placement of sensors to generate
substantially sinusoidal phase-separated waveforms in a manner that
the circuitry takes into account only the substantially straight
portions of the resulting, intersecting curves, with additional
resolution provided by dividing the straight portions into equal
length segments, effectively dividing the voltage increments into
equal known segments to be associated with fractions of rotor or
stator rotation, depending on mechanical design of the motor.
[0057] Some non-limiting exemplary uses and applications for a DC
electric motor according to the invention include the
following:
[0058] Diagnostic applications: With increasing use of robotics for
use in high-throughput processing of fluid samples and performing
of diagnostic assays, high resolution control of mechanical
mechanisms has become extremely useful. Particularly, as diagnostic
devices have trended toward small-scale and microdevices, which are
more efficient and require smaller sample sizes, control over
small-scale movements is of particular interest.
[0059] Medical applications: With increasing use of robotics for
remote surgery techniques, extremely well controlled movement of
remotely controlled implements have become essential. For example,
in ophthalmology or neurology procedures where manipulation of
retinal cells or nerve endings require movements with microscopic
resolution. In order to effect these movements, which are far finer
than is possible with a human hand with eye coordination, computers
are used to move actuators in concert with feedback from suitable
sensors. A motor with high resolution positional encoding
capabilities as disclosed herein can assist the computer, and
therefore the surgeon, in performing these delicate procedures.
[0060] Semiconductor fabrication: Systems for fabrication of
semiconductor devices rely on fine movement of the silicon wafer
and manipulator arms. These movements are regulated by means of
positional feedback. A motor with high resolution positional
encoding capabilities as disclosed herein suitable in these
applications.
[0061] Aerospace and satellite telemetry: High resolution angular
position feedback can be used for precise targeting and for antenna
positioning. In particular, satellite communication antenna dishes
need to precisely track orbiting satellites. Satellite trajectory
combined with precise angle feedback from a motor as described
herein mounted to the antenna and power spectrum from the antenna
can assist precise tracking. In addition, because the motor as
described herein is small, inexpensive and robust, it is an ideal
choice for use on satellites and in other extra-terrestrial
applications that will be well known to persons of skill in the
art.
[0062] Remote controlled vehicles: the small size and reduced cost
of the motor disclosed herein makes it desirable for use in remote
controlled vehicle applications, including drones. In particular
the high resolution positional encoding features of the motor make
it ideal for steering (directional control) and acceleration (power
control) in both commercial and recreational uses of remote
controlled vehicles. Additional uses will be apparent to persons of
ordinary skill in the art.
[0063] Human Augmentation: The small size and reduced cost of the
motor disclosed herein makes it desirable for use in prosthetic,
orthotic or humanoid applications as these might be applied to
augment or substitute for leg, arm or hand mechanics.
[0064] Further to the above, the skilled person will be aware that
there are a variety of ways that circuitry may be arranged to
provide granular control for a motor thusly equipped and sensed.
The invention is limited only by the claims that follow.
* * * * *