Acoustic resonance anemometer

Abstract

An acoustic resonance fluid motion sensor, comprising a housing comprising a first reflective surface, a second reflective surface and four pillars, wherein the second reflective surface is parallel to the first reflective surface so as to define a resonance cavity between the reflective surfaces, wherein the resonance cavity is open in a direction of fluid motion, and wherein the first reflective surface and the second reflective surface are spaced apart by the four pillars. Three electro-acoustic transducers are arranged within the first reflective surface.

Description

[0001] Acoustic resonance anemometer

[0002] Field

[0003] This disclosure relates to a configuration of an acoustic resonance fluid sensor.

[0004] Background

[0005] Acoustic resonance fluid sensors, comprising a resonance cavity between two parallel reflector surfaces, are capable of measuring the speed and direction of a fluid flow. During operation, a transducer present in the resonance cavity of the sensor emits an acoustic wave at a resonance frequency of the resonance cavity, generating a standing wave in the cavity. Transducers are switched so that, in one switching state a first transducer emits an ultrasonic signal whilst the second transducer acts as a receiver. In a second switching state the second transducer emits an ultrasonic signal whilst the first transducer acts as a receiver. The signals measured in the respective switching states are processed and a phase difference between them is determined. Measurements of this phase difference, facilitate the determination of the of the fluid through the cavity. The direction of the fluid flow within the cavity can be determined by using more than one transducer pairing. The transducers that form these pairings are arranged in a 2D plane, to allow measurement of fluid flow speeds in more than one direction.

[0006] Also known in the art are time-of-f light sensors, comprising only one reflector surface, which measure the time taken for an ultrasonic pulse emitted by one transducer to be received at another transducer. The transducers are angled at substantially 45 degrees to the normal, such that the emitted ultrasonic pulse is incident first on the reflector surface, and then reflected for receipt at the second transducer. Time-of-flight sensors differ substantially from acoustic resonance sensors, not least because they do not make use of a standing wave during the measurement process.

[0007] Summary

[0008] An acoustic resonance fluid motion sensor, comprising a housing comprising a first reflective surface, a second reflective surface and four pillars, wherein the second reflective surface is parallel to the first reflective surface so as to define a resonance cavity between the reflective surfaces, wherein the resonance cavity is open in a direction of fluid motion, and wherein the first reflective surface and the second reflective surface are spaced apart by the four pillars. Three electro-acoustic transducers are arranged within the first reflective surface. In an embodiment the housing is at least partially or fully formed of plastic, metal or a composite material.

[0009] In an embodiment the housing is formed by injection molding.

[0010] In an embodiment the four pillars are located equiangularly around the perimeter of the resonance cavity.

[0011] In another embodiment there is provided a housing for an acoustic resonance fluid motion sensor, comprising a first reflective surface, a second reflective surface and four pillars. The second reflective surface is parallel to the first reflective surface so as to define a resonance cavity between the reflective surfaces. The resonance cavity is open in a direction of fluid motion. The first reflective surface and the second reflective surface are spaced apart by the four pillars. Mounting locations for three electro-acoustic transducers are arranged within the first reflective surface.

[0012] In another embodiment there are provided computer program instructions for execution by an additive manufacturing apparatus, wherein the computer program instructions, when executed by the additive manufacturing apparatus cause the additive manufacturing apparatus to manufacture a housing as described above.

[0013] Brief Description of the Drawings

[0014] Figure 1 shows a high level diagram of the components of an acoustic resonance wind sensor.

[0015] Figure 2 shows an external view of a conventional acoustic resonance wind sensor.

[0016] Figure 3 shows a view of the first reflective surface of the conventional acoustic resonance wind sensor.

[0017] Figure 4 shows a set of graphs pertaining to directional error results for an acoustic resonance wind sensor with three pillars.

[0018] Figure 5 shows an external view of an acoustic resonance wind sensor with four pillars, according to embodiments.

[0019] Figure 6 shows a view of the first reflective surface of the acoustic resonance wind sensor with four pillars, according to embodiments.

[0020] Figure 7 shows a set of graphs pertaining to direction error results for an acoustic resonance wind sensor according to embodiments.

[0021] Detailed Description Fig.1 shows a high level diagram of an acoustic resonance fluid sensor 100. Acoustic resonance fluid sensors typically comprise a first and second reflective surface, 110 and 120 respectively, separated by a vertical distance and substantially parallel to each other. The first and second reflective surfaces are generally circular, and are substantially the same size. A resonance cavity 130 is defined by the space in between the first and second reflective surfaces. The resonance cavity 130 is open around its perimeter for the movement of fluid through the cavity, in a plane parallel to the first and second reflective surfaces.

[0022] Retained in the first reflective surface 110 are a number of electro-acoustic transducers, connected to an electronics unit 150. Each electro-acoustic transducer is configured to convert an electrical signal from the electronics unit 150 into a corresponding acoustic signal. Each transducer is also configured to convert a received acoustic signal into a corresponding electrical signal and send it to the electronics unit 150.

[0023] In operation, a first electro-acoustic transducer 141 receives a transmitted electronic signal from the electronics unit 150, and converts it into an emitted acoustic signal. The electronic signal driving the transducer 141 has a bandwidth that includes the expected resonance frequency of the resonance cavity 130. The emitted acoustic signal is incident on the second reflective surface 120, and is reflected back towards the first reflective surface 110, generating a standing wave inside the cavity 130.

[0024] At least one other electro-acoustic transducer 142 in the cavity is configured to receive an acoustic signal and convert it into an electronic signal, for transmission to the electronics unit 150. This second electro-acoustic transducer 142 therefore receives the acoustic standing wave formed by the first electro-acoustic transducer 141 , and converts it into a corresponding received electronic signal. This received electronic signal can be compared with the transmitted electronic signal at the electronics unit 150. The received electronic signal will possess a phase shift relative to the transmitted electronic signal, dependent on the time taken for the acoustic wave to propagate between the emitting transducer 141 and the receiving transducer 142. When the roles of the transducers are reversed (i.e. a signal is emitted by the second transducer 142 and received by the first transducer 141), another phase difference may be determined. The speed of fluid flow along an axis defined by the locations of the emitting and receiving transducers in the first reflective surface 110 can be determined using the difference between the first and second phase differences.

[0025] A third electro-acoustic transducer 143 (not shown in Fig. 1) is retained by the first reflective surface 110. The third electro-acoustic transducer 143 is positioned externally to the axis defined by the positions of the first and second electro-acoustic transducers. The third electro-acoustic transducer 143 may also receive the acoustic standing wave present in the resonance cavity 130, with a difference in phase to the transmitted signal. By repeating the above process between different pairs of the transducers present in the resonance cavity 130, the overall velocity of fluid flow through the cavity 130 may be determined using the components of the velocity along each axis.

[0026] Embodiments discussed herein use air as the example medium through which the signal propagates, to determine the speed and direction of wind. However, the invention is not limited to wind sensors only.

[0027] Figure 2 shows a known implementation of the above sensor 100, with an external view. The sensor is housed by a first housing portion 101 and a second housing portion 102. The first housing portion 101 holds the electro-acoustic transducers 141 , 142, 143 and the electronics unit 150, as well as the first reflective surface 110 (not shown in Fig. 2). The second housing portion 102 holds the second reflective surface (not shown in Fig. 2), and is mounted onto a support member 105. The resonance cavity 130 is bounded around its perimeter by pillars 160. In the embodiments shown in Figure 2, the acoustic resonance wind sensor 100 possesses six pillars 160, positioned equiangularly around the perimeter of the first and second reflective surfaces. However, the resonance cavity 130 is still substantially open to fluid flow parallel to the first and second reflective surfaces. The electronics unit 150 is shown in the first housing portion in Fig. 2, but could also be located in the second housing portion.

[0028] Figure 3 illustrates the layout of components as retained by the first reflective surface 110. The first reflective surface 110 retains the electro-acoustic transducers 141 , 142, 143. The transducers are positioned symmetrically around the center of the reflective surface in a triangular configuration. As discussed above, by positioning the third transducer 143 outside of the axis defined by the positions of the first two transducers 141 and 142, it is possible to determine the overall speed and direction of the wind through the cavity 130. Additionally, six geometrically identical pillars 160 are distributed equiangularly around the perimeter of the first reflective surface 110. Known acoustic resonance wind sensors 100 use six pillars 160 to provide symmetry to the system (the six pillars 160 being a multiple of the three transducers).

[0029] For some applications of the acoustic resonance wind sensor 100, it is advantageous to manufacture the first and second housing portions 101 , 102 and the pillars 106 out of plastic. Manufacturing the sensor using plastic reduces the overall cost of the sensor 100, as the plastic used is significantly cheaper than the metal conventionally used to produce the sensor housing. However, to produce the sensor 100 using plastic, adjustments must be made to the design of the parts, to ensure that the sensor is sufficiently robust. For instance, the diameter of the sensor and size of the pillars must increase, to ensure structural stability in the sensor. In some embodiments, the sensor may be manufactured using injection molding, in which case adjustments must be made to facilitate this manufacturing method. These adjustments include a constant thickness of parts to allow for even cooling, drafted walls and allowance for the movement of the injection mold tool section. In another embodiment, the sensor is manufactured using additive manufacturing.

[0030] The design for the wind sensor intended to be manufactured from plastic has a larger diameter and uses larger pillars in the housing than were present in the conventional sensor. It was determined, through testing, that the number of pillars separating the first housing portion 101 and the second housing portion 102 should be reduced, to prevent the larger pillars from impacting the airflow into the resonance cavity 130. It was believed that alignment between the layout of the pillars and transducers was required. For anemometers comprising three transducers this necessarily means the use of three pillars to maintain this alignment with the positions of the transducers.

[0031] Typically, a wind sensor may be tested by mounting it, via the support member 105, to a robotic arm and placing it in a wind tunnel. The wind sensor may be tested by rotating it, via the robotic arm, from 0-360° throughout a testing interval of time, during a constant wind speed and direction through the wind tunnel. The measured direction of the wind from the wind sensor can be compared to the known rotation of the robotic arm throughout the testing interval, and from this comparison a record of the error in the direction measurement across the rotation can be determined.

[0032] Fig 4A shows such a record, for a constant wind speed of 30 m / s. The error in the direction measurement in degrees is displayed on the Y axis, and the rotation undergone by the wind sensor is displayed on the X axis. The shaded columns on the graph represent angles of the rotation in which a sensor is blocked by pillars 160.

[0033] After the test, the data is processed to calibrate the measurement error. This is done by comparing the speed and direction data during testing at various wind speeds (e.g. 15m / s, 30m / s and 65m / s), and comparing the measurements to the known wind speed and rotation angle of the arm. The measurement results can be averaged every few degrees of rotation to reduce noise, resulting in a smoothed error curve at each tested wind speed. The error curves are then interpolated and extrapolated to produce a continuous three-dimensional error curve across a wider range of speeds. Discrete points are found on the direction error for each speed, and a function is applied to “zero” the direction at these points, generating calibration tables. This method is called Advanced Yaw Calibration for Direction (“AYC-D”). These calibration tables are written to the wind sensor after the wind tunnel process to correct the wind direction output of the wind sensor. Fig. 4B shows the same measurements recorded in Fig. 4A, after undergoing AYC-D calibration.

[0034] Figs. 4C and 4D correspond to Figs. 4A and 4B respectively, but with the X axis wrapped around a circle. In this way, the results shown in Figs. 4A and 4B can be intuitively visualized with respect to the layout of the components in the first reflective surface 110, which are shown on the graphs. The empty circles shown on Figs. 4C and 4D represent the transducers, while the shaded circles shown on Figs. 4C and 4D represent the pillars. Ideally, for a wind sensor which is behaving correctly, the direction data is able to be calibrated so that the final result is a flat line on the Cartesian plot and a circle on the polar plot.

[0035] Figs. 4A and 4B show the results of the testing process described above, acquired for a wind sensor design with three pillars, intended for production out of plastic. It can be seen that instead of the desired behaviour described above, the direction error instead behaves erratically, immediately switching from a high positive error to a high negative error at points throughout rotation. When this data is calibrated, these jumps are too large and abrupt to be removed by calibration. Consequently, they appear as spikes in the calibrated results. Figures 4C and 4D show the graphs of 4A and 4B with the X axis wrapped around a circle, for intuitive visualization.

[0036] Therefore, a correction to the geometry of the sensor is required to facilitate its production from plastic without introducing significant error into the direction measurement. After extensive testing of many different geometries, a new configuration of the sensor was determined to minimise this calibration issue.

[0037] Fig. 5 shows an implementation of this new configuration of the sensor 600. It can be seen that this configuration possesses four pillars 660 separating the first and second housing portions 601 and 602, rather than the six pillars shown in Fig. 2, and the three pillars tested in Fig. 4. As a result of using four pillars, the rotational symmetry in the layout between the transducers and pillars used in the known six pillar configuration is broken. By reducing the number of pillars 660 from six, the impact the larger pillars have on the airflow through the sensor 600 is reduced. Surprisingly, by using four pillars instead of three the error present in the direction measurement is reduced, as will be discussed further below.

[0038] Fig. 6 illustrates the layout of components on / in the first reflective surface 610, according to an embodiment. The first reflective surface 610 still retains three electro-acoustic transducers 641, 642, 643 in a substantially triangular arrangement, but four pillars 660 are now located around the perimeter of the surface. The pillars 660 are distributed substantially equiangularly around the perimeter of the surface 610 in Fig. 7. Although not essential, in the embodiment one of the transducers 643 lies on an axis defined between two of the pillars, whereas the other transducers 641 and 642 do not.

[0039] Testing of a wind sensor according to embodiments of the invention according to the testing method disclosed above produced the results shown in Fig. 7. It can be seen that instead of the drastic shifts in the directional error displayed in Figs. 4A and 4B, the behavior of the direction error is markedly smoother throughout the rotation (as shown in Fig. 7A), allowing for a more successful calibration of the sensor (as shown in Fig. 7B). Figs. 7C and 7D again show polar plots corresponding to Figs. 7A and 7B.

[0040] Therefore, it can be seen that the configuration according to the invention provides an improvement to the direction error of the sensor, which is an unexpected and surprising result. Prior to making the present invention, it was expected in the field that acoustic resonance anemometers require symmetry in the configuration of the transducers. Consequently, it would be expected that the number of pillars chosen in an attempt to reduce the pillar number is three. Surprisingly the measurement error behaved in the manner discussed above with reference to Figs. 4A to 4D. Even more surprising was the finding, after a large number of experiments, that, by abandoning the expectation that the pillars and transducers should be symmetrically arranged and choosing four pillars instead of three, these measurement errors are markedly reduced.

[0041] Whilst the present invention was made in the context of the development of an acoustic resonance anemometer that may be manufactured from plastic, it will be appreciated that the fact that four pillars can be used in combination with three transducers is not limited to sensors made from plastic. Alternatively, sensors having this geometry may be manufactured from other materials, such as from metals, including aluminium, ceramics or composite materials.

[0042] While certain embodiments have been described, these embodiments have been presented by way of example only and are not intended to limit the scope of the invention. Indeed, the novel methods, devices and systems described herein may be embodied in a variety of forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the invention. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the invention.

Claims

Claims:

1. An acoustic resonance fluid motion sensor, comprising: a housing comprising a first reflective surface, a second reflective surface and four pillars, wherein the second reflective surface is parallel to the first reflective surface so as to define a resonance cavity between the reflective surfaces, wherein the resonance cavity is open in a direction of fluid motion, and wherein the first reflective surface and the second reflective surface are spaced apart by the four pillars; three electro-acoustic transducers arranged within the first reflective surface.

2. The sensor of claim 1 , wherein the housing is at least partially or fully formed of plastic, metal or a composite material.

3. The sensor of claim 2, wherein the housing is formed by injection molding.

4. The sensor of claim 1 , wherein the four pillars are located equiangularly around the perimeter of the resonance cavity.

5. A housing for an acoustic resonance fluid motion sensor, comprising: a first reflective surface, a second reflective surface and four pillars, wherein the second reflective surface is parallel to the first reflective surface so as to define a resonance cavity between the reflective surfaces, wherein the resonance cavity is open in a direction of fluid motion, and wherein the first reflective surface and the second reflective surface are spaced apart by the four pillars; mounting locations for three electro-acoustic transducers arranged within the first reflective surface.

6. Computer program instructions for execution by an additive manufacturing apparatus, wherein the computer program instructions, when executed by the additive manufacturing apparatus cause the additive manufacturing apparatus to manufacture a housing as claimed in claim 5.

Images