Measuring and testing – Volume or rate of flow – By measuring thrust or drag forces
Reexamination Certificate
2000-05-08
2003-11-11
Lefkowitz, Edward (Department: 2855)
Measuring and testing
Volume or rate of flow
By measuring thrust or drag forces
C073S861740
Reexamination Certificate
active
06644132
ABSTRACT:
BACKGROUND
1. Field of the Invention
This disclosure relates to flow conditioning in pipes and, more particularly, to an apparatus for conditioning flow characteristics in pipe flow systems.
2. Description of the Related Art
Measuring flow in pipes may be performed in many ways. On such way includes sonic energy propagated into a fluid stream and measuring transit time to determine flow characteristics. There are primarily two types of transit-time ultrasonic flowmeters, the clamp-on type and the wetted transducer type. The clamp-On type is limited to either diametric direct or reflect transducer mounting for each sonic path. The wetted transducer is capable of either of these or chordal mounting.
It is well known that the shape o the flow profile as a liquid flows down a pipe is a function of the Reynolds number, a function of flow rate and liquid viscosity. This shape is “pointier” for lower Reynolds numbers and flatter for higher Reynolds numbers, transitioning to “Turbulent” or flatter shapes to “Laminar” or “pointer” shapes occurs when Reynolds number increases to about 2000 to 4000 as it passes through the “Transition” region.
The calibration of transit-time flowmeters is affected by this shape or profile since the transit time difference between upstream and downstream travel is a linear function of the distance that the beam travels through each different velocity. The volumetric flow is obviously not linear with flow profile, since the area of high flow near the center of the pipe is much lower than the low velocities which apply near the pipe wall.
Since the shape of the flow profile usually has a known “mathematical” relationship with Reynolds number, it is possible to automatically correct for this relationship as the flow profile changes with flow rate if the viscosity of the liquid is known. The actual shape of the flow profile is known accurately only if the flow profile is “fully developed”. Fully developed flow refers to a flow profile which has reached a steady state shape. This typically occurs at a distance of a number of pipe diameters of “straight run” pipe past any change in pipe diameter, change of pipe direction or obstructive disturbance.
Another flow profile characteristic is called “crossflow” which is a tendency for flow not to be primarily axial to the pipe, as occurs just past a pipe bend or obstructive disturbance. This is usually more disturbing to ultrasonic flowmeter accuracy since the basic flowmeter calibration assumes a known angle between the ultrasonic beam(s) and the pipe axis.
Only if the shape of the flow profile is known and properly compensated for mathematically, or conditioned by flow profile conditioners, and if crossflow is avoided or compensated, can an ultrasonic flowmeter reach its full potential for accuracy.
The “shape” of the flow profile is different from that predicted by the current Reynolds number prior to it being fully developed. If the upstream condition is, say, entry into a pipe from a large tank, the profile gradually transitions from a “flat” or “plug” flow shape, to that demanded by the Reynolds number. This also applies, in general, to the effect of a reduction or substantial increase in pipe diameter, both of which have the tendency to flatten the profile. However, if the upstream condition is a bend, either an in-plane or out-of-plane single or double elbow, the flow profile can change radically, displacing the center of flow from the center of the pipe, as well as flattening its shape and causing crossflow. Each such condition affects the relationship between the profile shape and the volume flow rate integration of the effect of the profile on the sonic beam's transit-time for sonic beam flowmeters. The flattening of the flow profile, in some cases, may be so profound as to cause a lower flow velocity in the center of the pipe, as compared to regions closer to the pipe wall. Therefore, flow profile effects, if not controlled, or compensated, can seriously affect accuracy of ultrasonic flowmeters.
Some of the flow profile conditions that need accommodation to achieve ultimate clamp-On transit-time flowmeter accuracy include:
a) Correction for the effect of the shape of the “standard” Reynolds number predicted flow profile, for flow volume integration;
b) Correction for the difference in the actual flow profile shape relative to the predicted Reynolds number related shape;
c) Correction for the displacement of the center of flow profile from the central axis of the pipe;
d) Correction for crossflow; and
e) Correction for the effects of liquid swirl, usually induced as flow changes direction, as in an elbow, due to Coriolis forces.
Correcting or mitigating the effects noted above may respectively include:
a) Correction for the shape of the standard flow profile.
Since the shape of the “standard” flow profile shape is mathematically predictable, transit-time flowmeters may include a compensating algorithm. This results in changing the basic computed flow rate, which is predicated at “plug” flow, by an amount directly related to the actual effect of the predicted shape on flow calibration, resulting in elimination of any resultant flow profile error.
Since this compensation algorithm needs knowledge of Reynolds number, a site setup of the meter requires identification of the liquid viscosity and the pipe diameter. Together with the computed flow rate, this is sufficient information to compute Reynolds number, which is then fed into the flow profile calibration correction algorithm.
Note that in this case, where the flow profile is that which was predicted, and the computation of viscosity is accurate, only one sonic path is needed, since all diametric clamp on paths are symmetric, all being affected by a symmetric flow profile in an identical way.
b) Correction for aberration of the standard flow profile shape.
Since local pipe configuration sometimes aberrate the expected shape of the flow profile from that expected relative to the current Reynolds number, it is possible that the standard correction algorithm does not fully correct the flow profile error. In other cases, the actual Reynolds number is not known precisely, due either to the liquid's viscosity being unknown, or being highly variable.
c) Correction for the displacement of the center of flow profile from the central axis of the pipe.
In cases where the “axis” of the flow profile is displaced from the axis of the pipe, it is desirable to assure that the diametric clamp-on sonic beam is oriented so as to pass through the flow profile axis. This can generally be accomplished by orienting the clamp-on transducers in the plane of the bend, within two diameters of the bend itself. This may be difficult to do and may require several iterations.
Alternatively, one or more additional sonic paths may be employed, so as to assure that the higher flow rate which occurs on the flow profile axis is scanned by a sonic beam. This is more expensive than orienting a single path appropriately, as noted above. This is generally only recommended for applications involving a wide variety of different liquids, and in cases where extremely high accuracy is required under these difficult conditions.
d) Correction for crossflow.
As noted above, crossflow is a condition usually associated with the existence of a bend in the pipe, or some non-symmetric disturbance. The error crossflow causes is due to the axis of flow being divergent from the axis of the pipe.
By mounting the clamp-on transducers in “reflect” mount, crossflow error may be corrected, as the angle between the beam and the stream in the direct and reflected paths is equally increased and decreased respectively, thus substantially canceling out any error.
To assure in accomplishing this result it is desirable that the longer reflect mount path not result in excessive sonic signal attenuation. In cases where liquid is especially viscous, and capable of absorbing or scattering ultrasound, wide beam transducers are needed, as they are more efficient at injecting high levels of ultrasound into the liquid.
F. Chau & Associates LLP
Lefkowitz Edward
Mack Corey D.
LandOfFree
Flow profile conditioner for pipe flow systems does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Flow profile conditioner for pipe flow systems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Flow profile conditioner for pipe flow systems will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3124564