Geometrical instruments – Distance measuring – Scale reading position sensor
Reexamination Certificate
2001-08-03
2002-12-03
Bennett, G. Bradley (Department: 2859)
Geometrical instruments
Distance measuring
Scale reading position sensor
C033S702000
Reexamination Certificate
active
06487787
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to position transducers, and more specifically, to a system and method for determination of error parameters for performing self-calibration and other functions in a position transducer system.
BACKGROUND OF THE INVENTION
Various movement or position transducers for sensing linear, rotary or angular movement are currently available. These transducers are generally based on either optical systems, magnetic scales, inductive transducers, or capacitive transducers.
In general, a transducer may comprise a read head and a scale. In an example of a 2-phase system, the transducer outputs two signals S
1
and S
2
that vary sinusoidally as a function of the position of the read head relative to the scale along a measuring axis. In one common concept for transducers, the signals S
1
and S
2
are intended to be identical except for a quarter-wavelength phase difference between them. The transducer electronics use these two signals to derive the instantaneous position of the read head relative to the scale along the measuring axis.
Ideally, the signals S
1
and S
2
are perfect sinusoids with no DC offsets, have equal amplitudes, and are in exact quadrature (i.e., a quarter-wavelength out of phase relative to each other, also referred to as “orthogonal” herein). In practice, the signals S
1
and S
2
have small DC offsets, their amplitudes are not equal, and they have some orthogonality error. In addition S
1
and S
2
may have distorting spatial harmonic components. In addition, the transducer electronics can introduce additional errors such as offset, gain, and non-linearity errors.
In many practical transducers, the dominant sources of errors are offsets, and amplitude mismatches and phase errors between the phases. These errors are shown in the following equations for the case of a 2-phase system.
S
1
=
C
1
+
V
1
sin
2
π
λ
(
x
-
φ
1
)
(
1
)
S
2
=
C
2
+
V
2
cos
2
π
λ
(
x
-
φ
2
)
(
2
)
In the above equations, x is a position, and &lgr; is a wavelength to the transducer output. The terms C
1
and C
2
produce offset errors. The terms V
1
and V
2
(if V
1
≠V
2
) produce amplitude mismatch errors. The terms &phgr;
1
and &phgr;
2
(if &phgr;
1
-&phgr;
2
≠0°) produce phase mismatch errors, that is, phase relationship errors.
Alternatively, some transducers utilize a 3-phase system. The equations for a 3-phase system are shown below.
U
R
=
C
0
+
A
0
sin
2
π
λ
(
x
-
φ
0
)
(
3
)
U
S
=
C
0
′
+
A
0
′
sin
2
π
λ
(
x
+
λ
3
-
φ
0
′
)
(
4
)
U
T
=
C
0
″
+
A
0
″
sin
2
π
λ
(
x
-
λ
3
-
φ
0
″
)
(
5
)
In the above equations, the terms C
0
, C
0
′, and C
0
″ produce offset errors. The terms A
0
, A
0
′, and A
0
″ (if they are not identical) produce amplitude mismatch errors. The terms &phgr;
0
, &phgr;
0
′, and &phgr;
0
″ (if not all equal) produce phase mismatch errors.
One method for addressing errors such as those shown above is to calibrate the transducer. Calibrating the transducer and compensating for these errors requires determining or comparing the DC signal offsets, the amplitudes of the fundamental signals, the phase error between the fundamental signals, and insuring that they are adjusted or compensated to be equal. For further error compensation, the amplitudes of the harmonic components must also be considered.
One commonly used prior transducer calibration method is the “Lissajous” method. The Lissajous method typically comprises inputting two nominally orthogonal read head signals to an oscilloscope, to drive the vertical and horizontal axis of the oscilloscope. The read head is continually scanned relative to the scale to generate changing signals. The oscilloscope display is observed, and the read head is physically and electronically adjusted until the display indicates a “perfect” circle, centered at zero on both axes. Under this condition, the amplitude, orthogonality, and offset of the two signals are properly adjusted.
The Lissajous method assumes that the two signals are both perfect sinusoids. Typically, there is no adjustment for harmonic errors which can distort the circle, as it is hoped that these are made insignificant by fixed features of the transducer design and assembly. The Lissajous method is well known to those skilled in the art, and has been performed by sampling the two signals with computer-based data acquisition equipment. However, to use this method in the case of a 3-phase system, the signals must generally to be converted to orthogonal signals before processing. Many three-phase systems either lack such signals, or access to such signals is either inconvenient or costly, making the Lissajous method inappropriate for many 3-phase transducer systems and products.
Alternatively, it has also been common to accept any transducer errors due to imperfect amplitudes, orthogonality, harmonics, and offsets, and to use an external reference, such as a laser interferometer, to accurately correct position errors from the read head at the system level, at predetermined calibration positions relative to the scale.
Position transducers typically require initial factory calibration, and periodic calibration or certification thereafter. In both cases, there is a cost for the associated equipment and labor. When the transducer is located in a remote location, it is difficult to set up the external data acquisition equipment and/or accurate external reference required for calibration. As a result, the transducer often has to be transported to another site or shipped back to the factory for calibration. This results in long downtime and increased costs.
Even in cases where the transducer does not have to be transported for calibration, the special tools and increased time required to set up the external display and/or reference result in increased costs and downtime. Thus, calibration and recalibration is often minimized or avoided, in practice. Since most practical position transducers are sensitive to variations during production, installation, and use, measurement errors normally increase in the absence of calibration, and in the periods between recalibration.
One example of a self-calibration method for an inductive position sensor is illustrated in U.S. Pat. No. 5,742,921. However, the method of the '921 patent is closely tied to the operation of an associated motor, teaches positioning the transducer in order to acquire the desired data, and emphasizes calibration based on a signal range which is of little or no importance in certain implementations.
Another exemplary self-calibrating position transducer system that better-addresses some of the issues outlined above is illustrated in U.S. Pat. No. 6,029,363, which is commonly assigned and hereby incorporated by reference in its entirety. The method described in the '363 patent samples two orthogonal output signals of the transducer at a plurality of evenly spaced positions within the one scale period, using the transducer itself as a position reference. The method then determines calibration values for the DC signal offsets, the amplitudes and non-orthogonality of the fundamental signals, and the amplitudes of the signal harmonic components using Fourier analysis techniques. Finally, the method corrects the signals using the determined calibration values. In summary, the method of the '363 patent determines the error parameters by: (1) measuring the output voltages of the transducer as a function of a known position, and (2) analyzing the data using Fourier series techniques to derive the offset, amplitude, and phase mismatch parameters.
The method of the '363 patent is effective for addressing many of the calibration issues described above, including the ability to compensate spat
Mawet Patrick H.
Nahum Michael
Bennett G. Bradley
Christensen O'Connor Johnson & Kindness PLLC
Mitutoyo Corporation
LandOfFree
System and method for determination of error parameters for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for determination of error parameters for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for determination of error parameters for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2958907