NL1044072B1 - Method of operation for reduction of drift in a differential pressure sensor - Google Patents

Abstract

Title: Method of operation for reduction of drift in a differential pressure sensor Abstract Method of operation for reduction of drift in a differential pressure sensor (1), wherein each single measurement of a physical quantity S in a given medium (for example flow of a fluid flowing through a duct) is defined as the difference of the results of two partial measurements, said two partial measurements being made in succession by said sensor (1) with different relative orientations between sensor (1) and medium (M), said relative orientations corresponding to distinct sensitivities of said sensor to S. The invention also provides a system configured for sensing pressure or differential pressure in a fluid flowing through a fluid duct (W). 1044072

Description

Title: Method of operation for reduction of drift in a differential pressure sensor The invention relates to a method of operation for reduction of drift in a differential pressure sensor. An aspect of the invention also relates to a system configured for sensing pressure or differential pressure in a fluid flowing through a fluid duct.

SUMMARY In the real-world application of sensors drift often forms the limiting factor in the performance of the sensor. In particular, additive drift tends to constitute the limiting factor in sensitivity and dynamic range, rather than noise.

The invention offers a means of operation which can eliminate or at least significantly reduce additive drift in vector sensors. The invention is exemplified in its application to differential pressure sensors, in which case a dynamic range of 1:100 is obtained, well iv excess of what is feasible by standard differential pressure measurement techniques (1:10).

BACKGROUND OF THE INVENTION The invention provides a method for the reduction of drift in vector sensors. To set notation, a vector sensor is understood to be measuring a physical vectorial quantity S in an input energy domain Di by converting S into a signal in an output energy domain Do. In particular, we think here of electric or electronic sensors for the measurement of gas flow or pressure. Thus, sensors are seen to simultaneously possess a state A in the input energy domain Di and a state B in the output energy domain Do.

The new method of operation concerns the real-world application of sensors in the presence of imperfections. Imperfections are generally present in both the sensor itself and, in the case of active sensors, in the biasing of the sensor and its output amplifiers. While ideal sensors as abstract devices possess zero sensitivity for physical quantities other than S, sensors in the real-world possess imperfections due to technological limitations and, consequently, possess finite sensitivity for quantities other than S. Errors in the measurement of S due to finite sensitivity to quantities other than S are manifest in the form of offset with unpredictable behavior. The resulting drift in offset persists additively as drift in the output signal, when the input signal, S, is kept constant. Typically, the resulting offset is predominant over noise, 1.e., S/offset<<S/noise, so that offset defines the lowest limit at which S can be reliably measured. In other words, offset typically determines the dynamic range of the sensor at hand, rather than noise. We remark that drift in the sensitivity of a sensor to its measurand S does not influence the dynamic range of the sensor. For this reason, only the for mentioned drift in offset, i.e, additive drift is considered here. The current approach towards reduction of drift is taken by a continuous effort towards ever more precise manufacturing processes. In particular, modern manufacturing processes involve special correction techniques such as laser trimming, with which imperfections resulting from the basic production process are greatly reduced. In such approach, the final performance of the sensor is determined by the quality of the overall production process. With our new method and system, proposed below, technological limitations can be of secondary importance.

SUMMARY OF THE INVENTION The present invention is defined by the features of each of independent conclusies 1, 6 and 7.

The new method is proposed for reduction of drift in differential pressure flow sensors. The method applies to sensors with directional sensitivity, i.e, anisotropic sensitivity in two or more directions. Sensors of this type are also known as vector sensors. Sensors with this property define an orientation with respect to which S is measured.

An example of a vector sensor is given by the flow sensor by A.F.P. van Putten (U.S. Pat. Nos. 4,548,077 and 3,996,799), where the gas flow velocity, S, induces a thermal gradient (a vectoral quantity) in a silicon integrated chip. This silicon flow sensor is heated electrically and obtains temperature gradients in the presence of gas flow. An integrated Wheatstone bridge reads off the temperature gradient induced by such flow.

Another example is given by a MEMS differential pressure sensor, where differential pressure DP induces a deformation in a silicon integrated chip. The deformation causes stress in the on-board resistors, which results in change of resistance, which can be measured by an on-board or external electronic circuit.

According to an aspect the present new method defines a single measurement through two partial measurements obtained from the same differential pressure sensor in succession with different sensor states Ain the input energy domain Di. In the presence of directional sensitivity the two different sensor states A can be realized through at least two different relative orientations between sensor and medium. Thus, according to an embodiment, in the new method the aforementioned two partial measurements are obtained by successively changing the relative orientation of the sensor with respect to the medium, which relative orientations can be precisely those that possess different sensitivities to the physical quantity S. In the new method, which we shall refer to as the Alternating Direction Method (ADM), the final measurement result can be defined as the difference between the two results from the two partial measurements. We shall refer to the output signal thus obtained as the

ADM signal. Because in ADM each output signal is calculated from (at least) two sensor measurements, we refer to the physical, sensor measurements as partial measurements. We conclude that a sensor which discriminates between S and quantities other than S which may interfere with the measurement of S, in the sense that the sensor possesses aforementioned directional sensitivity to S, while its sensitivity is independent of direction to quantities other than S which may interfere with measurement of S, offers an ADM signal which contains no or substantially no drift.

It follows that according to an aspect there is provided an innovative method of operation for reduction of drift in a differential pressure sensor, wherein each single measurement of a physical quantity S in a given medium (for example flow of a fluid flowing through a duct) is defined as the difference of the results of two partial measurements, said two partial measurements being made in succession by said sensor with different relative orientations between sensor and medium, said relative orientations corresponding to distinct sensitivities of said sensor to S.

Furthermore, advantageously, according to an aspect there is provided a method for sensing pressure or differential pressure in a fluid flowing through a fluid duct, for example for measuring fluid flow, the method for example including a method according to the above- mentioned aspect, wherein a differential pressure sensor is used that has a first fluid input port and a second fluid input port, the differential pressure sensor in particular being configured for sensing a difference between pressure of fluid present at the first input port and pressure of fluid present at the second input port, wherein during a first measurement step: - a first fluid communication is provided between the first sensor input port and a first fluid duct measuring location, and a second fluid communication is provided between the second sensor input port and a second fluid duct measuring location, wherein the sensor provides a first partial measurement result; wherein during a second measurement step: 5 - a third fluid communication is provided between the second sensor input port and the first fluid duct measuring location, and a fourth fluid communication is provided between the first sensor input port and the second fluid duct measuring location, wherein the sensor provides a second partial measurement result; wherein the first and second partial measurement result are processed to provide an overall measurement result, the processing in particular including subtracting the first partial measurement result and second partial measurement result.

In this way, the above-mentioned advantages can be achieved.

Moreover, according to an aspect of the invention there is provided a system configured for sensing pressure or differential pressure in a fluid flowing through a fluid duct, for example a system for carrying out a method according to the invention, wherein the system includes: -a differential pressure sensor having a first fluid input port and a second fluid input port, the differential pressure sensor in particular being configured for sensing a difference between pressure of fluid present at the first input port and pressure of fluid present at the second input port; -a valve structure, switchable between a first valve state and a second valve state, wherein the valve structure in its first valve state provides a first fluid communication between the first sensor input port and a first fluid duct measuring location, and a second fluid communication between the second sensor input port and a second fluid duct measuring location, wherein the valve structure in its second valve state provides a third fluid communication between the second sensor input port and the first fluid duct measuring location, and a fourth fluid communication between the first sensor input port and a second fluid duct measuring location, wherein the sensor is configured to provide a first partial measurement result when the valve structure is in its first state and a second partial measurement result when the valve structure is in its second state, wherein the system includes processing means configured for receiving the partial measurement results from the sensor and for processing those results to provide an overall measurement result, the processing in particular including subtracting the first partial measurement result and second partial measurement result.

ADM APPLIED TO DIFFERANTIAL PRESSURE

MEASUREMENT The power of ADM can be exemplified in differential pressure measurement using a low-cost differential pressure MEMS sensor, which is widely available on the market. In the manufacturing process of these sensors asymmetries occur in the order of a few percent. The sensors are also prone to thermal stress when powered up. A realistic measurement, therefore, represents not only differential pressure signal, but contains a component as function of ambient temperature as well. Drift in the offset as a function of temperature, humidity, pollution, age etc., deteriorates the measurement signal, and, consequently, is independent of relative orientation between sensor and flow medium (a common mode signal in the flow medium). The said differential pressure sensor therefore, offers the unique possibility for reduction or elimination of drift in offset through application of ADM. This may be made more precise in gas volume measurement as follows:

In gas volume measurement with ADM we integrate the electric output signal of the sensor, DP, as induced by the flow velocity S, over consecutive time intervals 0<t1<t2 <.,.<tk <tk+1 <. with tk+1 -tk ident..DELTA.t. Introducing the quantity qk the measured volume over the time-interval tk -tk-1 with flow over the sensor alternating with 180 DEG (90 DEG) relative orientation, then we have = [FT vd = (CDE Vand + [TF Varigedt (1).

Here, we have written the output signal as it is composed of the pure Vdp and the undesired drift, Vdrift. The ADM quantity, QnADM, is now given by tn QnirM = > (—1)* lg, — dui] = | Vapdt + At k 0 (2) With ADM remainder: tx nse DI | Varigedt. k Ees (3) The ADM remainder has the property that tn tl Ears | Vrije dt | Varigedt. tres 0 (4) in the limit as A,— 0 with t, = const. ADM substantially eliminates, therefore, undesired contributions due to drift in offset in gas volume measurement up to a remainder in the order of ZAtV ie: In the above formulas 1-4 the following applies: qx = velocity signal or flow signal in m/sec or m3/hr k = time interval (number) t = time (seconds); Vap= differential pressure signal in Volt or as binary number

Varit= differential pressure drift component, in Volt or as binary digital value QnAPM = The resulting flow value e= The ADM remainder Further advantageous embodiments of the invention are described in the dependent conclusies. The invention will now be explained in more detail with reference to the drawings. Therein shows: Figures 1A, 1B schematically a first example of a system and method for reduction of drift in a differential pressor sensor; Figure 1C schematically show a further state of the first example of Figures 1A, 1B, showing a valve purging state providing purging valve and sensor ducts; Figures 2A, 2B schematically a second example of a system and method for reduction of drift in a differential pressor sensor; and Figures 3A, 3B schematically a third example of a system and method for reduction of drift in a differential pressor sensor; and In the drawings, similar or corresponding features are denoted by similar or corresponding reference signs.

SURVEY OF DRAWINGS Application of ADM in differential pressure measurement is illustrated by two following two implementations. First possible implementation. Figures 1A, 1B schematically show a system configured for sensing pressure or differential pressure in a fluid (of a medium M) flowing through a fluid duct W. Arrows F indicate the flow of medium M in the duct W. The system includes a differential pressure sensor having a first fluid input port S1 and a second fluid input port S2. The present differential pressure sensor

1 is configured for sensing a difference between pressure of fluid present at the first input port S1 and pressure of fluid present at the second input port S2. Further, the system includes a valve structure SR, SL, switchable between a first valve state (shown in in Fig. 1A) and a second valve state (shown in Fig. 1B).

As follows from the drawings (Fig. 1A), the valve structure SR, SL in its first valve state provides a first fluid communication between the first sensor input port S1 and a first fluid duct measuring location F1. Also, in that first valve state the valve structure SL, SR provides a second fluid communication between the second sensor input port S2 and a second fluid duct measuring location F2.

As 1s shown in Fig. 1B, the valve structure SR, SL in its second valve state provides a third fluid communication between the second sensor input port S2 and the first fluid duct measuring location F1. In its second state the valve structure SL, SR provides a fourth fluid communication between the first sensor input port S1 and a second fluid duct measuring location F2.

In particular, respective ‘fluid pressure sensing communications’ (allowing pressure sensing at the sensor ports) provided by said fluid communications of the valve structure SR, SL are schematically indicated by double-arrows in the drawings 1A, 1B, 2A, 2B, 3A, 3B (generally, during pressure sensing, fluid will be substantially stationary in respective valve ducts of the valve structure SR, SL, as will be clear to the skilled person).

Moreover, preferably, the sensor 1 is configured to provide a first partial measurement result when the valve structure SR, SL 1s in its first state (see Fig. 1A) and a second partial measurement result when the valve structure SR, SL is in its second state (see Fig 1B).

Preferably, the system includes processing means C configured for receiving the two (or more) partial measurement results (i.e. a first and a second partial measurement result) from the sensor 1 and for processing those results to provide an overall measurement result, the processing in particular including subtracting the first partial measurement result and second partial measurement result.

Such processing means C (or a system controller) can communicate with the sensor 1 in various ways, for example for receiving sensor data (in particular for receiving each said partial measurement result) from the sensor 1, for example via a wired and/or wireless communication link (not shown). Also, such processing means (or system controller) C can e.g. be integrated with the sensor 1 but that is not required, The sensor data processing means C can be configured in various ways, e.g. including controller/processor software and/or hardware, micro- electronics, a control unit and/or the-like, as will be appreciated by the skilled person.

The switching of the valve structure SR, SL between respective states can e.g. be controlled by a valve control means or system controller, which can be configured in various ways (e.g. including controller/processor software and/or hardware, micro-electronics, a control unit and/or the-like as will be appreciated by the skilled person). Such valve control means (or system controller) can e.g. be part of the sensor processing means and/or can e.g. be integrated with the valve structure SR, SL or be separate therefrom. Also, such processing means C (or system controller) can communicate with the valve structure SL, SR in various ways, for example for sending valve control signals to the valve structure, for example via a wired and/or wireless communication link (not shown).

In particular, FIG. 1A, 1B show an implementation of a gas flow meter 1, in which the two different relative orientations between sensor S and medium are obtained through successively changing the differential pressure signal by means of two valves SR, SL (i.e. the valve structure).

As follows from the two valve states shown in the drawings, ADM can now be applied to the differential pressure sensor as described in formentioned U.S. patents, with 180 DEG bidirectional sensitivity, by obtaining two partial measurements from successively clockwise and counter clockwise differential pressure directions.

Tests have shown that with ADM the formentioned differential pressure sensor yields a dynamic range up 1:1000, from 0.1 mbar to 100 mbar, resulting in a flow range of ~ 1:30 (3 times better than any regular differential pressure based flow meter). With this new method of operation, measurements are made possible which go beyond those feasible with current differential pressure techniques.

DESCRIPTION OF PREFERRED EMBODIMENTS Referring more specifically to the drawings, for illustrative purposes a first non-limiting example of the present invention is embodied in each of the apparatus generally shown in FIG. 1A and 1B. It will be appreciated that the embodiment of the invention may vary as to the particular sensor and as to the details of the parts without departing from the basic concepts as disclosed herein.

Referring to FIG. 1A, 1B, the valves SL, SR of the valve structure can possess three ports (e.g. opening pipes) for pressure: ports P2 and R1 connect to sensor port S1, ports P1 and R2 connect to sensor port S2. In this example, valve port Al connects to fluid duct port F1 and valve port A2 connects to fluid duct port F2.

Alternatively, the valve structure arrangement can be reversed, resulting in the configuration shown in FIGURES 2A, 2B, showing two valve structure states respectively.

According to a further embodiment, the system shown in the drawings (see FIG. 1A, 1B) can also provide the option for purging sensor and valve ducts, channels and/or ports by switching the SL and SR valves in mode A1-P1 and A2-R2 (providing a channel purging state of the valve structure), the result is shown in FIG. 1C. Another purging state is switching the SL and SR valves in mode A1-R1 and A2-P2 (providing a second channel purging state of the valve structure). In each of the purging states, the sensor 1 1s being bypassed, wherein one of the sensor ports is not in fluid communication with fluid flowing through the duct W, whereas the other sensor port is in fluid communication with fluid ducts that can be purged with fluid flowing between the two duct ports F1, F2 and respective valve sections of the valve structure SR, SL. In Fig. 1C a respective fluid flow is indicated by arrows.

In other words, in this way, a ‘positive and negative side’ of the DPsensor channels can be cut short, and any fluid in the channels can flow out. For example, a valve controller of the system can be configured for providing a purging step, wherein the valve structure is switched to a said purging state.

The three ports (e.g. pipe openings) in each of the valves SL and SR, the pipes BL and BR, and the T-elements TL and TR all preferably are sized equally, with a low “dead volume” and identical lengths of connecting pipes, to ensure symmetry and proper drainage of any liquid content in the gas. Preferably, respective connecting pipes and valves are made of durable material and designed to be able to withstand system pressure. The sensor 1 may be any differential pressure sensor, or the silicon flow sensor as described in U.S. Pat. No. 4,548,077, or any other sensor with allows for directional sensitivity.

Figures 2A, 2B are similar to Figures 1A, 2B, showing an alternative arrangement of the valve structure SR, SL with respect to the sensor 1 and the fluid duct W.

Referring to FIG. 1, 2 the measuring section provided at the fluid duct W can be a pitot style device.

Referring to the alternative system embodiment shown in FIG. 3A, 3B (showing two respective valve structure states, similar to those of Figures 1A, 1B) the measuring section Q at the duct W’ can be an orifice or similar venturi, wedge, cone differential pressure generating shape, wherein the measuring section is located between two respective duct ports P1, P2 that are in fluid communication with the valve/sensor structure SL, SR, 1.

In each of the embodiments as shown in FIG. 1, 2 and 3, the entire system may be under a common mode pressure as found in typical industrial applications.

It is self-evident that the invention is not limited to the above- described exemplary embodiments. Various modifications are possible within the framework of the invention as set forth in the appended conclusies.

Although the description above is in the context of differential pressure measurement, this presentation should not be construed as limiting the applicability of ADM but merely as providing illustration of a real-world application of ADM. Thus, the scope of ADM should be determined by the appended conclusies and their legal equivalents.

For example, each of the valves SL, SR can be a bi-stable valve, or a different type of valve, as will be appreciated by the skilled person.

Claims (7)

Conclusions A method for reducing drift in a differential pressure sensor (1), wherein each measurement of a physical quantity S in a particular medium (e.g. flow of a liquid or gaseous medium flowing through a channel) is defined as the difference in the results of two partial measurements, wherein these two partial measurements are successively performed by this sensor (1) with different relative orientations between sensor (1) and medium (M), and these relative orientations correspond to distinguishable sensitivities of this sensor to S. A method as claimed in claim 1, wherein the two relative orientations between sensor (1) and medium (M) are obtained by alternating the orientation of this medium (with physical quantity S), the orientation of this sensor remaining unchanged . The method as claimed in claim 1, wherein the two relative orientations between sensor (1) and medium (M) are obtained by alternating the position of the sensor in the medium (with physical quantity S), the orientation of this medium remains unchanged. A method as claimed in any one of the preceding claims, wherein the two relative orientations between sensor (1) and medium (M) are obtained by using bistable valves or valves (SR, SLD). Method according to any one of the preceding claims, wherein the method is a method for measuring pressure or differential pressure in a liquid or gas flowing through a channel (W), for example for measuring the flow rate, wherein the differential pressure sensor (1) having a first liquid input port (S1) and a second liquid input port (S2), the differential pressure sensor (1) configured to detect a difference between the pressure of liquid (M) present at the first liquid input port (S1 ) and the pressure of liquid (M) present on the second liquid inlet port (S2); wherein during a first measuring step: - a first liquid communication is possible between the first liquid input port (S1) and a first flow channel measuring location (F1; P1) and a second liquid communication is provided between the second liquid input port (S2) and a second flow channel measuring location (F2; P2), wherein the sensor (1) provides a first partial measurement result; wherein during a second measuring step: - a third fluid communication is provided between the second fluid input port (S2) and the first flow channel measurement location (F1; P1) and a fourth fluid communication is provided between the first fluid input port (S1) and the second flow channel measurement location (F2; P2) wherein the sensor (1) provides a second partial measurement result; wherein the first and the second partial measurement results are processed to provide the single measurement result of the physical quantity S, the processing subtracting at least the first partial measurement result and the second partial measurement result. A system configured for measuring pressure or differential pressure in a liquid flow through a flow channel (W), for example for measuring liquid flow according to any of the preceding claims, wherein a differential pressure sensor is used having a first liquid input port (S1) and a second liquid input port (S2), in particular the differential pressure sensor (1) is configured to detect a difference between the liquid pressure present at the first input port (S1) and the liquid pressure present at the second input port (S2), wherein during a first measuring step: - a first liquid communication provided between the first liquid input port (S1) and a first measuring location for the flow channel (F1; P1) and a second liquid communication is provided between the second liquid input port (S2) and a second measuring location for the flow channel (F2 P2), the sensor (1) yielding a first partial measurement result; wherein during a second measuring step: - a third liquid communication is provided between the second liquid input port (S2) and the first measuring location of the flow channel (F1; P1) and a fourth liquid communication is provided between the first liquid input port (S1) and the second measuring location of the flow channel (F2; P2), the sensor (1) yielding a second partial measurement result; wherein the first and second partial measurement result are processed into an overall measurement result, in particular the processing including subtraction of the first partial measurement result and the second partial measurement result. | A system configured for measuring pressure or differential pressure in a fluid flow through a flow channel (W), e.g. a system for performing a method according to any preceding claim, wherein the system includes a differential pressure sensor having a first fluid input port ( S1) and a second liquid input port (S2), in particular the differential pressure sensor (1) being configured to detect a difference between the pressure of the liquid present at the first liquid input port (S1) and the pressure of liquid present on the second liquid input port (S2); - a valve structure switchable between a first valve state and a second valve state; - wherein the valve structure in the first valve state provides a first fluid communication between the first sensor input port (S1) and a first fluid channel measurement location (F1; P1), and a second fluid communication between the second sensor input port (S2) and a second fluid channel measurement location (F2; P2), - wherein the valve structure in the second valve state provides a third fluid communication between the second sensor input port (S2) and the first fluid channel measurement location (F1; P1) and a fourth fluid communication between the first sensor input port (S1) and a second fluid channel measurement location (F2; P2), wherein the sensor (1) is configured to provide a first partial measurement result when the valve structure is in the first state and a second partial measurement result when the valve structure is in the second state, wherein the system comprises processing means (C) which are configured to receive the partial measurement results from the sensor (1) and for processing these results into a general measurement result, the processing being subtracted in particular the first partial measurement result and the second partial measurement result.