Measuring and testing – Instrument proving or calibrating – Span or set point adjustment
Reexamination Certificate
1999-02-25
2001-05-29
Noland, Thomas P. (Department: 2856)
Measuring and testing
Instrument proving or calibrating
Span or set point adjustment
C073S708000, C073S001620, C702S087000, C702S098000, C702S104000
Reexamination Certificate
active
06237394
ABSTRACT:
BRIEF DESCRIPTION OF THE INVENTION
This invention relates generally to sensing devices. More particularly, this invention relates to a technique of correcting drift in a sensor.
BACKGROUND OF THE INVENTION
A piezoresistive or capacitive sensor generates a change in electrical resistance or capacitance proportional to a physical force applied to it. Such sensors are commonly positioned on a silicon membrane to measure the deflection of the silicon membrane.
A number of problems arise in connection with the packaging of such sensors. For example, a silicon-based piezoresistive pressure sensor is attached to a package or substrate by means of a very soft, compliant adhesive, such that there is no residual stress or force transmitted from the package to the sensor. Should stress be transmitted to the sensor, the membrane of the sensor is deflected, creating a measured resistance value at the sensing element, which erroneously indicates a change in measured pressure. Stresses from the attachment process may add or subtract to this resistance change, depending upon the direction and magnitude of the pressure induced membrane deflection. In general, most commonly accepted calibration techniques adequately compensate for the non-ideal behavior of the pressure sensor caused by manufacturing variations and the necessity of attaching the sensor to a rigid base.
A problem with existing calibration techniques arises when the stresses from the attachment process change over time due to stress relief of the attachment material. This stress relief can be caused by exposure of the sensor, changes in the sensor surroundings, changes in the attachment material, and/or changes in the base to varying pressures and/or temperatures. Should the base, sensor, and attachment material not be closely matched in thermal expansion, then any temperature excursion produces a stress differential in the materials proportional to the difference in thermal expansion and temperature differential. In the case where the die attach material is a viscoelastic substance, it will flow over time in a fashion so as to decrease the stress. This will clearly have an impact on the piezoresistive membrane, changing the resistance value over time as well.
Typical calibration methods cannot account for the changing nature of the resistance values described above. The resistance of most sensors is a function of both temperature and pressure. One approach to model this physical activity is to use a polynomial equation of the form:
P=M
0
+M
1
*T+M
2
*T
2
+M
3
*T
3
+(M
4
+M
5
*T+M
6
*T
2
+M
7
*T
3
)*V+(M
8
+M
9
*T+M
10
*T
2
+M
11
*T
3
)*V
2
+(M
12
+M
13
*T+M
14
*T
2
+M
15
*T
3
)*V
3
(Equation I)
P is the differential pressure applied across the membrane. Each M term is a derived coefficient. T is the temperature of the sensor membrane. V is the measure of the resistance of the sensor structure. For example, V may be the voltage produced by a constant current across the piezoresistive element.
The order of the polynomial required to fit a particular pressure sensor depends upon the individual manufacturing process, the attachment technique and the degree of accuracy required. A second order fit with respect to pressure and temperature commonly suffices, although a second order fit with respect to pressure and a third order fit with respect to temperature is sometimes required.
When a non-rigid die attach material is used, the M0 term of Equation I typically changes with time as well as temperature. This term is usually referred to as the “offset”, referring to the fact that in most sensors when the pressure is “zero”, there is no deflection on the membrane, but there is still a non-zero signal. This is true even for a sensor using a Wheatstone bridge resistor configuration which, ideally, may be designed to give a zero signal, but with manufacturing tolerances, will produce a small signal.
This aging phenomenon can create signal level shifts of approximately 0.5 psi. This shift or drift is relatively constant across the entire pressure range of interest. At a pressure level of 40 psi, this amounts to a 1.25% error, while at 5 psi, this amounts to a 10% error. Such errors are unacceptable in most applications.
It is not practical or acceptable for a customer to re-calibrate a pressure sensor which has drifted. Thus, it would be highly desirable to provide an improved sensing device that has the capability to correct drift. Ideally, such a sensor would automatically self-correct for drift in response to a relatively simple set of conditions invoked by a customer in the field. A thorough technique would compensate for drift factors, such as sensor aging, as a function of voltage and temperature conditions.
SUMMARY OF THE INVENTION
A method of correcting drift in a sensor includes the step of identifying a calibration command. A nominal zero pressure sensor drift correction factor for the sensor is identified by relying upon a calibration voltage value and a calibration temperature value secured at a nominal zero pressure condition. Sensor output is subsequently adjusted according to the nominal zero pressure sensor drift correction factor.
The apparatus of the invention includes a micro-controller and associated electronics configured to establish a nominal zero pressure sensor drift correction factor for a sensor in response to a calibration voltage value and a calibration temperature value secured at a nominal zero pressure condition. The sensor output is subsequently modified by the micro-controller according to the nominal zero pressure sensor drift correction factor.
The invention provides an improved technique to correct drift in piezoresistive and capacitive sensors. The technique automatically self-corrects for drift in response to a relatively simple set of external conditions. Advantageously, the technique compensates for drift factors, such as sensor aging, as a function of voltage and temperature conditions as measured at a nominal zero pressure condition. The zero pressure sensor drift correction factor is then applied to other pressure conditions to achieve a correction in sensor drift.
REFERENCES:
patent: 4051712 (1977-10-01), Zias et al.
patent: 4079625 (1978-03-01), Mann
patent: 4927170 (1990-05-01), Wada
patent: 5170549 (1992-12-01), Julien et al.
patent: 5247171 (1993-09-01), Wlodarczyk et al.
patent: 5287254 (1994-02-01), Baert et al.
patent: 5422478 (1995-06-01), Wlodarczyk et al.
patent: 5469750 (1995-11-01), Lloyd et al.
patent: 5551301 (1996-09-01), Cowan
patent: 5668320 (1997-09-01), Cowan
patent: 3143061 A1 (1983-05-01), None
patent: 2034592 (1980-06-01), None
patent: 62-191259 (1987-08-01), None
Basov et al Abstract of RU 2062993C1 by Derwent Information LTD, ACC-No 1997-106932, Derwent Week 199710 “Correction of Characteristic of Pressure Differential Flowmeter—In Which Pressure Differential at 0.09 of Nominal Value is Set and 0.3 Nominal Reading of Indictor is Set by Corrector, Before Setting Zero on Instrument After Release of Pressure” Jun. 27, 1996.
Arkilic Errol
Harris James M.
Taheri Babak
Noland Thomas P.
Pennie & Edmonds LLP
Redwood Microsystems, Inc.
LandOfFree
Apparatus and method for correcting drift in a sensor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for correcting drift in a sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for correcting drift in a sensor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2555267