Sensor conditioning method and apparatus

• US 4,873,655 A
• Filed: 08/21/1987
• Issued: 10/10/1989
• Est. Priority Date: 08/21/1987
• Status: Expired due to Term

• Current Assignee: Board of Regents of the University of Texas System (University of Texas System)
• Original Assignee: Board of Regents of the University of Texas System (University of Texas System)
• Inventors: Kondraske, George V.
• Primary Examiner(s): Lall, Parshotam S.
• Assistant Examiner(s): Ramirez, Ellis B.
• Application Number:
  US07/088,087
• Time in Patent Office: 781 Days
• Field of Search: 364/553, 364/571, 364/573, 364/582, 364/577, 364/723, 364/571.02, 364/571.07, 73/1 R, 324/207
• US Class Current: 702/86
• CPC Class Codes: G01D 18/008 with calibration coefficients stored in memory
• CPC Class Codes: G01R 1/38 Arrangements for altering the indicating characteristic, e.g. by modifying the air gap

Abstract

An adaptable method and apparatus for calibrating and conditioning sensor output to conform to a desired transfer function is provided. In one form, the invention allows use of a sensor system having nonlinear output characteristics when measuring a physical quantity and provides a response which varies linearly to changes in the physical quantity. In the calibration phase, a look-up table is built based on inputs from the sensor system and a reference transducer which behaves in the desired fashion to changes in the physical quantity. In the operation phase, the sensor system measures the unknown physical quantity and enters the look-up table to determine a corresponding reference value. Thus, the present invention adapts a sensor system to an application without restrictive considerations to individualized operating performance of the sensor system or environment of the application.

Claims

1. A method for conditioning the output characteristics of a sensor to conform to a desired transfer function, where the normal output of the sensor in measuring a physical quantity of interest deviates from the desired transfer function, comprising the steps of:

(a) measuring the physical quantity in a first condition with a reference transducer having the desired transfer function;

(b) measuring the physical quantity in the first condition with the sensor;

(c) repeating steps (a) and (b) with the physical quantity in a plurality of different conditions;

(d) building a changeable look-up table with the plurality of measurements of steps (a)-(c), where for each condition a sensor measurement corresponds to a reference transducer measurement conforming to the desired transfer function;

(e) measuring the physical quantity in an unknown condition using the sensor;

(f) conditioning the sensor measurement of step (e) to obtain an output conforming to the desired transfer function by adjusting the sensor output by providing real time feedback signals to obtain an adjusted sensor output and including the substeps

determining a corresponding reference transducer measurement using the adjusted sensor output from the look-up table, and

outputting the reference transducer measurement from the look-up table.

2. The method of claim 1, wherein the desired transfer function is linear and the sensor performs in a nonlinear fashion.

3. The method of claim 1, wherein in step (f) the reference transducer measurement is determined by interpolating between two reference transducer values in the look-up table.

4. The method of claim 3, wherein the desired transfer function is linear and the interpolation is linear.

5. The method of claim 1, wherein in step (b), including the step of measuring the physical quantity in the first condition with a second sensor.

6. The method of claim 5, wherein in step (d) for each condition a single reference transducer measurement corresponds to a unique pair of measurements from the first and second sensors.

7. A method for obtaining a linearized output from a sensor, where the sensor normally has an output that is a nonlinear function of the physical quantity being measured, comprising the steps of:

calibrating the sensor by building a changeable look-up table for a plurality of conditions of the physical quantity in which the sensor signals correspond to linearized reference values, including the substeps of

measuring the physical quantity in a plurality of conditions with both the sensor and a reference transducer that outputs a reference value that is a linear function of the physical quantity being measured,

outputting for each condition a sensor signal and a reference value to establish a changeable look-up table with a unique reference value for each condition measured;

operating the sensor to obtain a linearized output corresponding to the physical quantity in an unknown condition including the substeps of

measuring the physical quantity in the unknown condition with the sensor to produce a sensor signal, adjusting the sensor signal for gain and offset in real time,

outputting the adjusted sensor signal to the look-up table,

entering the look-up table with the adjusted sensor signal,

determining a reference value that corresponds to the adjusted sensor signal and is a linear function of the physical quantity being measured, and

outputting the reference value from the look-up table as the linearized output corresponding to the physical quantity in the unknown condition.

8. The method of claim 7, wherein said entering step includes:

determining the two sensor signal values in the look-up table nearest the entered sensor signal,

linearly interpolating between the entered sensor signal and the two look-up table sensor signal values to determine an interpolation factor.

9. The method of claim 8, using the reference values corresponding to the two sensor signal values and applying the interpolation factor to the reference values to determine a reference value related to the sensor signal.

10. The method of claim 7, said calibrating step including enlarging said look-up table by interpolation based on said measured sensor signals and measured reference values to obtain interpolated sensor signals and corresponding interpolated reference values.

13. The method of claim 7, said calibrating step including enlarging said look-up table by extrapolation based on said measured sensor signals and measured reference values to obtain extrapolated sensor signals and corresponding extrapolated reference signals.

11. An apparatus to adjust the output of a sensor which produces an undesirable output signal in relation to changes in condition of a physical quantity being measured by the sensor, comprising:

reference means coupled to the physical quantity for producing an output which varies in a desirable fashion with changes in the condition of the physical quantity;

microprocessor means having a changeable look-up table for receiving the values of the sensor signals and reference means outputs, and for matching the values of the signals and outputs by adjusting the sensor output by providing realtime feedback signals to obtain an adjusted sensor output for a given physical quantity condition, including

means for receiving an adjusted sensor signal indicative of the physical quantity in an unknown condition for determining a reference value from the look-up table using only the adjusted sensor signal of the unknown condition as the input; and

means coupled to the microprocessor means for indicating the reference value.

12. An apparatus according to claim 11, said microprocessor means including means for determining portions of the changeable look-up table by interpolating to derive reference values.

18. The apparatus according to claim 11, said microprocessor means including means for determining portions of the changeable look-up table by extrapolating to derive reference values.

19. The apparatus according to claim 11, said receiving and determining means being operable to use said sensor signal of the unknown condition as a pointer to the look-up table to determine said reference value.

20. The apparatus according to claim 11, said sensor signals being analog and said receiving and determining means including an analog-to-digital converter.

21. The apparatus according to claim 11, said reference means being decoupleable from said physical quantity and from said microprocessor means.

14. A method of providing a measurement output which varies linearly with changes in a physical quantity being measured using a sensor having an output which varies nonlinearly with changes in the physical quantity being measured, comprising the steps of:

measuring the physical quantity in a plurality of conditions with the nonlinear sensor and with a linear transducer having an output which varies linearly with changes in the physical quantity being measured;

outputting a nonlinear sensor signal adjusted in real time for gain and offset and a linear transducer value for each condition measured to build a look-up table matching a linear transducer value with a nonlinear sensor signal for each condition measured;

measuring the physical quantity in an unknown condition with the nonlinear sensor to produce an object sensor signal;

comparing the object sensor signal with the look-up table sensor signals to identify a look-up table location;

retrieving the linear transducer value from the identified look-up table location; and

outputting a measurement of the physical quantity in the unknown condition using said identified linear transducer value.

15. The method according to claim 14, including the step of expanding the look-up table beyond the measured nonlinear sensor signals and linear transducer values using interpolation.

16. The method according to claim 14, the comparing step including the steps of:

determining the two sensor signal values in the look-up table bounding the object sensor signal;

the retrieving step including the steps of;

retrieving the two bounding transducer values matching the two bounding sensor signal values;

approximating an object linear transducer value using the two bounding transducer values.

17. The method according to claim 15, the approximating step including the steps of:

interpolating between the two bounding signal values and the object sensor signal to determine an interpolation factor;

applying the interpolation factor to the two bounding transducer values to identify said object linear transducer value.