Absolute position detection apparatus and error compensation methods therefor

• US 5,677,686 A
• Filed: 01/27/1995
• Issued: 10/14/1997
• Est. Priority Date: 01/28/1994
• Status: Expired due to Term

• Current Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Inventors: Mizutani, Takao, Sakuma, Hirokazu, Aoki, Yukio, Kazama, Tsutomu, Kachi, Mitsuyasu
• Primary Examiner(s): Williams, Howard L.
• Application Number:
  US08/379,739
• Time in Patent Office: 991 Days
• Field of Search: 341/8, 341/11, 341/115
• US Class Current: 341/8
• CPC Class Codes: G01D 5/2451 Incremental encoders G01D5/2454 takes precedence
• CPC Class Codes: H03M 1/645 for position encoding, e.g. using resolvers or synchros H03M1/485 takes precedence

Abstract

An absolute position detection apparatus which comprises sine and cosine wave generators for generating one or more sets of sine and cosine waves within a cycle, analog-to-digital converters for converting the incoming sine and cosine waves generated by the sine and cosine wave generators into digital values. An arithmetic unit is used for calculating a compensation for errors including offset, amplitude and phase errors on the basis of the digital values from the analog-to-digital converters. In the apparatus, the arithmetic unit operates on a phase angle from phase angles found by the arithmetic performed prior to or during said error compensations and the digital values from the analog-to-digital converters, whereby a low-priced, highly reliable absolute position detection apparatus can be achieved without the addition of compensation circuits to the hardware.

Claims

1. In an absolute position detection method wherein one or more sets of sine and cosine waves are generated within a cycle of a rotating device, the generated sine and cosine waves are analog-to-digital converted into digital values, and error compensation is performed using an arithmetic process to compensate for errors including offset, amplitude and phase errors on the digital values, the error compensation method comprises the steps of:

determining maximum and minimum values of the digital values, said maximum and minimum values representing peak values of the sine and cosine waves;

calculating an amplitude value for each of said sine and cosine waves using a difference between said maximum and minimum values;

determining an amplitude damping factor which varies as a function of an angular speed of said rotating device;

calculating an amplitude compensation factor for each of said sine and cosine waves as a function of said amplitude value, said amplitude damping factor and a reference amplitude value;

amplitude compensating each of said sine and cosine waves by multiplying each of said digital values by said amplitude compensation factor, thereby adjusting the amplitude of each of said sine and cosine waves to a reference amplitude.

2. The error compensation method as defined in claim 1, further comprising the steps of:

using a sum of said minimum and maximum values to determine an offset voltage for each of said sine and cosine waves;

for each of said sine and cosine waves, adjusting said digital values by said offset voltage to effect offset compensation.

3. The error compensation method as defined in claim 2, wherein higher-frequency sine and cosine waves are generated consecutively by arithmetic operation using the sine and cosine waves compensated for.

4. An error compensation method as defined in claim 1, further comprising the steps of:

when an angle of said digital values falls within predetermined ranges of 0°

, 90°

, 180° and

270°

, calculating respective amplitudes R0, R90, R180 and R270 from a sum of the square of the digital value representing the sine wave and the square of the digital value representing the cosine wave;

determining an offset voltage for said cosine wave by comparing amplitude R0 at 0°

with amplitude R180 at 180°

which is 180°

out of phase therewith and determining an offset voltage for said sine wave by comparing amplitude R90 at 90°

with amplitude R270 at 270°

; and

for each of said sine and cosine waves, adjusting said digital values by said offset voltage to effect offset compensation.

5. The error compensation method as defined in claim 4, further comprising the steps of:

storing in a memory phase compensation values as a function of phase compensation angle α

;

when an angle of said digital values falls within predetermined ranges of 45°

, 135°

, 225° and

315°

, calculating respective amplitudes R45, R135, R225 and R315 from a sum of the square of the digital value representing the sine wave and the square of the digital value representing the cosine wave;

determining the phase compensation angle α

as a function of R45, R135, R225 and R315; and

phase compensating the digital values representing said sine and cosine waves using the phase compensation values corresponding to the determined phase compensation angle α

.

6. The error compensation method as defined in claim 5, wherein higher-frequency sine and cosine waves are generated consecutively by arithmetic operation using the sine and cosine waves compensated for.

7. The error compensation method as defined in claim 5, wherein the phase compensation values are sin(α

) and cos(α

), and the phase compensation value α

is incremented from a previous value when the quantity {R45+R225-R135-R315} is greater than 0, and decremented from a previous value when the quantity {R45+R225-R135-R315} is less than 0.

8. The error compensation method as defined in claim 4, wherein higher-frequency sine and cosine waves are generated consecutively by arithmetic operation using the sine and cosine waves compensated for.

9. The error compensation method as set forth in claim 1, further comprising:

storing in a memory a preset sine wave sin(α

) and a preset cosine wave cos(α

) as phase compensation values;

sequentially determining a waveform 90°

out of phase with one of the generated sine and cosine waves from the phase compensation values and the digital values representing the generated sine and cosine waves.

10. The error compensation method as defined in claim 9, wherein said waveform 90°

out of phase is calculated as;

space="preserve" listing-type="equation">sin(APH)={sin(AAM)+cos(AAM)sin(α

)}/cos(α

)

where sin(AAM) and cos(AAM) are the generated sine and cosine waves, sin(AAM) is shifted 90+α

from cos(AAM), and sin(APM) is 90°

out of phase with cos(AAM).

11. The error compensation method as defined in claim 1, wherein the amplitude damping factor decreases as function of the angular speed of said rotating device, and wherein the amplitude compensation factor equals the reference amplitude divided by the product of the amplitude value and the amplitude damping factor.

12. The error compensation method as defined in claim 1, wherein the amplitude damping factor is stored in a look-up table addressable by the angular speed of said rotating device.

13. The error compensation method as defined in claim 1, wherein the step of determining the maximum and minimum values comprises:

when an angle of a digital value is within a first predetermined range, comparing the digital value to a previous maximum value and storing the digital value as the maximum value if the digital value exceeds the previous maximum value;

when an angle of a digital value is within a second predetermined range, comparing the digital value to a previous minimum value and storing the digital value as the minimum value if the digital value is less than the previous minimum value.

14. The error compensation method as defined in claim 1, wherein the maximum and minimum values are determined by averaging maximum and minimum values detected in a plurality of cycles.

15. In an absolute position detection method wherein one or more sets of sine and cosine waves are generated within a cycle of a rotating device, the generated sine and cosine waves are analog-to-digital converted into digital values, and error compensation is performed using an arithmetic process to compensate for errors including offset, amplitude and phase errors on the digital values, the error compensation method comprising the steps of:

when an angle of said digital values falls within predetermined ranges of 0°

, 90°

, 180° and

270°

, calculating respective amplitudes R0, R90, R180 and R270 from a sum of the square of the digital value representing the sine wave and the square of the digital value representing the cosine wave;

calculating an amplitude compensation factor for each of said sine and cosine waves as a function of R0, R90, R180 and R270;

determining an amplitude damping factor which varies as a function of an angular speed of said rotating device;

amplitude compensating each of said sine and cosine waves by multiplying each of said digital values by said amplitude compensation factor, thereby adjusting the amplitude of each of said sine and cosine waves to a reference amplitude.

16. The error compensation method as defined in claim 15, further comprising the steps of:

storing in a memory phase compensation values as a function of phase compensation angle α

;

when an angle of said digital values falls within predetermined ranges of 45°

, 135°

, 225° and

315°

, calculating respective amplitudes R45, R135, R225 and R315 from a sum of the square of the digital value representing the sine wave and the square of the digital value representing the cosine wave;

determining the phase compensation angle α

as a function of R45, R135, R225 and R315; and

phase compensating the digital values representing said sine and cosine waves using the phase compensation values corresponding to the determined phase compensation angle α

.

17. The error compensation method as defined in claim 16, wherein higher-frequency sine and cosine waves are generated consecutively by arithmetic operation using the sine and cosine waves compensated for.

18. The error compensation method as defined in claim 15, wherein higher-frequency sine and cosine waves are generated consecutively by arithmetic operation using the sine and cosine waves compensated for.

19. An absolute position detection apparatus, comprising:

sine and cosine wave generators (1a, 1b, 2a, 2b) for generating one or more sets of sine and cosine waves within a cycle;

analog-to-digital converters (5a, 5b) for converting the incoming sine and cosine waves generated by said sine and cosine wave generators (1a, 1b, 2a, 2b) into digital values; and

an arithmetic unit (17) for performing arithmetic to compensate for errors including offset, amplitude, and phase errors on the basis of the digital values from said analog-to-digital converters (5a, 5b), said arithmetic unit determining maximum and minimum values of the digital values, said maximum and minimum values representing peak values of the sine and cosine waves, calculating an amplitude value for each of said sine and cosine waves using a difference between said maximum and minimum values, determining an amplitude damping factor which varies as a function of an angular speed of said rotating device, calculating an amplitude compensation factor for each of said sine and cosine waves as a function of said amplitude value, said amplitude damping factor and a reference amplitude value, and amplitude compensating each of said sine and cosine waves by multiplying each of said digital values by said amplitude compensation factor, thereby adjusting the amplitude of each of said sine and cosine waves to a reference amplitude.

20. The absolute position detection apparatus as defined in claim 19, wherein said arithmetic unit uses a sum of said minimum and maximum values to determine an offset voltage for each of said sine and cosine waves, and

for each of said sine and cosine waves, adjusts said digital values by said offset voltage to effect offset compensation.

21. The absolute position detection apparatus as defined in claim 19, further comprising:

a memory storing phase compensation values,

wherein said arithmetic unit sequentially determines a waveform 90°

out of phase with one of the generated sine and cosine waves from the phase compensation values and the digital values representing the generated sine and cosine waves.