Measuring and testing – Volume or rate of flow – By measuring vibrations or acoustic energy
Reexamination Certificate
1999-10-18
2001-12-18
Kwok, Helen (Department: 2856)
Measuring and testing
Volume or rate of flow
By measuring vibrations or acoustic energy
Reexamination Certificate
active
06330831
ABSTRACT:
BACKGROUND
This invention relates in general to ultrasonic measurement systems for ascertaining density or other characteristics of a fluid. More particularly, it relates to determining density of fluids flowing in pipes, including fluids like liquids, gases, and some two-phase combinations or mixtures.
It has been recognized in the ultrasonic literature since the 1960s, and perhaps earlier, that by measuring the acoustic impedance and sound speed of a fluid, one can determine the fluid density, dividing the impedance by the sound speed. In the prior art it is also known that impedance can be obtained from reports of the reflection coefficient at a fluid/solid reflecting interface. The 1982 U.S. Pat. No. 4,320,659 for Ultrasonic System for Measuring Fluid Impedance or Liquid Level of inventors Lynnworth, Seger and Bradshaw, teaches, in connection with
FIGS. 4-9
thereof, that the reflecting surface should be substantially flush with the interior surface of the vessel or conduit in which the reflection coefficient is measured. The clamp-on systems of those figures used the existing internal wall of the vessel as the surface at which R is sensed. This sensor was minimally intrusive, and thereby minimally disturbed the process operations or the flow, and introduced little pressure drop. Generally, since a protrusion could be a site for buildup, an intrusive reflector is viewed as undesirable. This idea of flush mounting a reflecting surface is still prevalent, and is found in several recent technical papers on the subject, examples being Van Deventer and Delsing, 1997, An Ultrasonic Density Probe, pp. 871-875 in
Proc.
1997
IEEE Ultrasonics Symposium
, and the peer-reviewed paper by Adamowski, Buiochi and Sigelmann, Ultrasonic Measurement of Density of Liquids Flowing in Tubes,
IEEE Trans UFFC
45 (1), pp. 48-56 (January 1998). These two papers may be taken to represent the current teaching in this field regarding reflective determinations of fluid density. They also teach that, to achieve sensitivity to fluid density, the characteristic acoustic impedance (Z
0
) of the reflector should be comparable to that of the fluid. The Z
0
of many liquids is on the order of one to two mks rayls, water having a value of 1.5 in these units. If one scans a handbook of physics table of characteristic acoustic impedances of solids, one finds Z
0
of about 2 to 3 for plastics, and about 17 to 50 for commonly-used metals. This leads one to select a plastic as the reflector. If a metal were to be used, the reflection coefficient R at a metal/fluid interface is typically so close to unity, that the small changes in fluid density that are of interest would be very difficult to detect, let alone measure with high accuracy. This would lead one to conclude that using a metal reflector would not be of much practical use in a reflectometer designed for sensing fluid density. This would appear particularly true for a metal such as stainless steel (SS), having Z
0
about 45 mks rayls, or nickel, with Z
0
of 54 mks rayls. In the 1982 '659 patent mentioned above, a high-Z
0
sensor is shown.
Nonetheless, there remains a need for sensing or determining fluid properties, such as density, or properties such as mass flow rate, that depend on density.
OVERVIEW OF THE INVENTION IN RELATION TO THE ART
The present invention provides improved density or related fluid measurements through reflection of one or more signals from specially configured or positioned reflectors. Applicant surprisingly finds high-Z metal reflectors to be particularly useful as key parts of a density sensing system, when used in certain novel configurations. In exemplary embodiments: (a) the high-Z reflector provides, even at normal incidence, a useful reference echo amplitude nearly independent of the fluid impedance and the reflector temperature; (b) an even better reference echo is obtained at sufficiently oblique angles of incidence, e.g. 45°, which may for example be achieved in a vee block internal right-angled corner reflector. Applicant utilizes the fact that metal generally has a sound speed so much greater than that of the fluid that total internal reflection within the fluid occurs at angles of incidence between about 30 and 75° to provide a reference echo from the metal that is substantially independent of fluid impedance and reflector temperature. As a third example: (c) through impedance transformation, a high-Z material is made to behave like a low-Z plastic, yet retain better stability than a plastic over a temperature range, and, unlike some plastics, not absorb or change characteristics when immersed in fluid for extended periods.
In some instances an exceptionally robust system of the invention is made by combining the reflection principles with transmission principles. To illustrate this idea we may consider dry steam at high temperature and high pressure. In one such combination, sound speed is measured accurately in the dry steam, along with pipe wall temperature, without requiring penetration of the pipe. (The pipe wall temperature can be determined from the speed of sound in the wall, which is determinable from the arrival time of the pipe-borne noise, sometimes called short circuit noise or crosstalk.) From such measurements of speed of sound in steam at a known (measured) temperature, the pressure in dry steam is calculable. We shall take the steam temperature to be equal to the pipe wall temperature. From the steam pressure and temperature, steam density is obtained, by reference to steam tables. This provides a useful cross-check on the density obtained by reflection, and so yields a better, more robust mass flowmeter for dry steam. In the example just cited, transmission principles are used not merely as an algebraic equivalent related to the basic energy coefficient equation T+R=1.
Other embodiments of a system provide robustness through redundancy. One can, for example, install in the fluid, by means of a spoolpiece or other known means, a wave guide in which a torsional or flexural or breathing-mode guided wave has a phase velocity that slows down as the density of the surrounding fluid increases. This provides an average density over the sensing length of that wave guide in the fluid, typically across the diameter. Sometimes just the speed of sound across such a diameter path provides a reasonably accurate measure of average fluid density over that path. Sometimes the temperature of the fluid provides the density information, if the fluid composition is known well enough.
A problem with prior-art flush mounted reflection coefficient sensors is that they are easily fouled, when placed in a residue-bearing fluid, because the low flow at the walls (theoretically zero), allows settling of sediment or debris. Fouling can also take the form of microbubbles attaching to the wall, yielding a reflection coefficient that is not representative of the fluid average density. However, if the fluid is clean and does not deposit residue or microbubbles, then it may be acceptable for the reflecting surface to be at the inner wall of the pipe, or very near that wall. The debris problem may be addressed by directing a high-intensity beam of ultrasound towards the reflector, cleaning it by creating acoustic streaming over the reflector; if the beam is focused it may cause cavitation at or near the reflector. Recognizing that steam is often used for cleaning, one can surmise that a pipe conveying dry steam may indeed remain clean and a flush-mounted reflectometer could be expected to not become fouled. Even with presumably clean fluids, however, to avoid bubble or residue problems at the pipe wall, it is advisable to not utilize the very top or very bottom regions, respectively. This is often expressed in a guideline for installations that one should avoid the six and twelve o'clock positions. Instead, sensors are preferably installed between 1 and 5 o'clock or between 7 and 11 o'clock positions. Applicant notes that if the reflection coefficient is to be utilized to determine fluid density,
Liu Yi
Lynnworth Lawrence C.
Kwok Helen
Nutter & McClennen & Fish LLP
Panametrics, Inc.
LandOfFree
Stream-cleaned differential reflection coefficient 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 Stream-cleaned differential reflection coefficient sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Stream-cleaned differential reflection coefficient sensor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2569600