Abstrict An optimized Coriolis mass flow meter is disclosed which has improved
stability to excitations caused by external influences. A primary
source of improvement involves determining by modal analysis of
the flow conduit a location for the sensor means that minimizes
the influence of external excitation of one or more of the first
in phase bending mode, the first out of phase bending mode, the
first out of phase twist mode, the second out of phase twist mode,
the second out of phase bending mode and the third out of phase
bending mode.
Claims We claim:
1. A mass flow meter for flowable materials wherein mass flow rates
for flowable materials are determined based on at least one measured
effect of Coriolis forces, said flow meter comprising: a support
means; at least one continuous flow conduit which is free of pressure
sensitive joints or sections, each of said conduits being solidly
mounted to said support means at inlet and outlet ends for said
conduits; driver means for oscillating each of said conduits about
bending axes adjacent each of said solid mountings; a pair of sensor
means mounted on each of said conduits for monitoring motion of
said conduits while flowable materials are flowing therethrough
and said conduits being oscillated by said driver means about said
bending axes, monitored motion including motion caused by Coriolis
forces about twist axes for each of said conduits, said sensor means
generating signals related to all motions of said conduits; and
signal processing means to detect and convert said signals to mass
flow rate values; in which the improvement comprises:
fixed mounting of each of said pair of sensor means on each of
said conduits to monitor motions of said conduits including motions
about said twist axes, where each sensor means is mounted between
nodes of a pair of vibration modes for said conduit, said pair of
vibration modes being selected from a pairing of the first in phase
bending mode, first out of phase bending mode, first out of phase
twist mode, second out of phase twist mode, second out of phase
bending mode, or third out of phase bending mode.
2. A mass flow meter according to claim 1 in which said continuous
flow conduit has an essentially straight inlet leg and an essentially
straight outlet leg which converge toward one another at said support
and are interconnected opposite said support by the remainder of
said continuous conduit.
3. A mass flow meter according to claim 2 in which said inlet and
outlet legs are interconnected opposite said support by an essentially
straight portion of said conduit which curves at either end to meet
each of the essentially straight leg portions.
4. A mass flow meter according to either of claim 2 or claim 3
in which members of said sensor pairs are placed between nodes of
the second out of phase twist mode and the second out of phase bending
mode on the respective inlet and outlet sides of each flow conduit.
5. A flow meter according to claim 1 having at least two flow conduits
clamped together by brace bars at points along the inlet and outlet
legs which are spaced from the support, said points having been
determined by modal analysis to be those points which provide optimum
separation between the frequencies of the first in phase bending
mode and the first out of phase bending mode so as to minimize effects
of the first in phase bending mode on meter sensitivity.
6. A flow meter according to claim 5 in which said continuous flow
conduit has an essentially straight inlet leg and an essentially
straight outlet leg which converge toward one another at said support
and are interconnected opposite said support by the remainder of
said continuous conduit.
7. A flow meter according to claim 5 in which the brace bars include
nipple shaped mounting sleeve means.
8. A flow meter according to either of claim 2 or claim 3 having
at least two flow conduits, wherein the support comprises a flangeless
inlet and outlet plenum which further comprises two separated flow
chambers, one for flow separation on the inlet side and one for
flow recombination on the outlet side.
9. A flow meter according to claim 1 further comprising a pressure
tight case of essentially the same geometric configuration as said
flow conduit which encases all of said conduit, said sensor means
and said driver, and is welded to said support.
10. A flow meter according to either of claim 2 or claim 3 further
comprising a pressure tight case of essentially the same geometric
configuration as said flow conduit which encases all of said conduit,
said sensor means and said driver, and is welded to said support.
Description In the art of measuring mass flow rates of flowing substances it
is known that flowing a fluid through an oscillating flow conduit
induces Coriolis forces which tend to twist the conduit in a direction
essentially transverse to the direction of fluid flow and also to
the axis about which oscillation occurs. It is also known that the
magnitude of such Coriolis forces is related to both the mass flow
rate of the fluid passing through the conduit and the angular velocity
at which the conduit is oscillated.
One of the major technical problems historically associated with
efforts to design and make Coriolis mass flow rate instruments was
the necessity either to measure accurately or control precisely
the angular velocity of an oscillated flow conduit so that the mass
flow rate of the fluid flowing through the flow conduits could be
calculated using measurements of effects caused by Coriolis forces.
Even if the angular velocity of a flow conduit could be accurately
determined or controlled, precise measurement of the magnitude of
effects caused by Coriolis forces raised another severe technical
problem. This problem arose in part because the magnitude of generated
Coriolis forces is very small in comparison to other forces such
as inertia and damping, therefore resulting Coriolis force-induced
effects are minute. Further, because of the small magnitude of the
Coriolis forces, effects resulting from external sources such as
vibrations induced, for example, by neighboring machinery or pressure
surges in fluid lines, may cause erroneous determinations of mass
flow rates. Such error sources as discontinuities in the flow tubes,
unstable mounting of the tubes, use of tubes lacking mechanically
reproducible bending behavior, etc., often completely masked the
effects caused by generated Coriolis forces, greatly diminishing
the practical use of a mass flow meter.
A mechanical structure and measurement technique which, among other
advantages: (a) avoided the need to measure or control the magnitude
of the angular velocity of a Coriolis mass flow rate instrument's
oscillating flow conduit; (b) concurrently provided requisite sensitivity
and accuracy for the measurement of effects caused by Coriolis forces;
and (c) minimized susceptibility to many of the errors experienced
in earlier experimental mass flow meters, is taught in U.S. Pat.
Nos. Re 31450 entitled "Method and Structure for Flow Measurement"
and issued Nov. 29 1983; 4422338 entitled "Method and Apparatus
for Mass Flow Measurement" and issued Dec. 27 1983; and 4491025
entitled "Parallel Path Coriolis Mass Flow Rate Meter"
and issued Jan. 1 1985. The mechanical arrangements disclosed in
these patents incorporate curved continuous flow conduits that are
free of pressure sensitive joints or sections, such as bellows,
rubber connectors or other pressure deformable portions. These flow
conduits are solidly mounted at their inlet and outlet ends, with
their curved portions cantilevered from the support. For example,
in flow meters made according-to any of the aforementioned patents,
the flow conduits are welded or brazed to the support, so that they
are oscillated in spring-like fashion about axes which are located
essentially contiguous with the solid mounting points of the-flow
conduits or, as disclosed in U.S. Pat. No. 4491025 essentially
at the locations of solidly attached brace bar devices designed
to clamp two or more conduits rigidly at points located forward
of the mounting points.
By so fashioning the flow conduits, a mechanical situation arises
whereby, under flow conditions, the forces opposing generated Coriolis
forces in the oscillating flow conduits are essentially linear spring
forces. The Coriolis forces, opposed by essentially linear spring
forces, deflect or twist the oscillating flow conduits containing
flowing fluid about axes located between and essentially equidistant
from the portions of those flow conduits in which the Coriolis forces
manifest themselves. T-he magnitude of the deflections is a function
of the magnitude of the generated Coriolis forces and the linear
spring forces opposing the generated Coriolis forces. Additionally
these solidly mounted, continuous flow conduits are designed so
that they have resonant frequencies about the oscillation axes (located
essentially at the locations of the mountings or brace bars) that
are different from, and preferably lower than, the resonant frequencies
about the axes relative to which Coriolis forces act.
Various specific shapes of solidly mounted curved flow conduits
are disclosed in the prior art. Included among these are generally
U-shaped conduits "which have legs which converge, diverge
or are skewed substantially" (Re 31450 col. 5 lines 10-11).
Also disclosed in the art are straight, solidly mounted flow conduits
which work on the same general principles as the curved conduits.
As stated above, the Coriolis forces are generated when fluid is
flowed through the flow conduits while they are driven to oscillate.
Accordingly, under flow conditions, one portion of each flow conduit
on which the Coriolis forces act will be deflected (i.e. will twist)
so as to move ahead, in the direction in which the flow conduit
is moving, of the other portion of the flow conduit on which Coriolis
forces are acting. The time or phase relationship between when the
first portion of the oscillating flow conduit deflected by Coriolis
forces has passed a preselected point in the oscillation pathway
of the flow conduit to the instant when the second portion of that
conduit passes a corresponding preselected point in that pathway
is a function of the mass flow rate of the fluid passing through
the flow conduit. This time difference measurement may be made by
various kinds of sensors, including optical sensors as specifically
exemplified in U.S. Pat. No. Re 31450 electromagnetic velocity
sensors as specifically exemplified in U.S. Pat. Nos. 4422338
and 4491025 or position or acceleration sensors as also disclosed
in U.S. Pat. No. 4422338. A parallel path double flow conduit
embodiment with sensors for making the preferred time difference
measurements is described in U.S. Pat. No. 4491025. This embodiment
provides a Coriolis mass flow rate meter structure which is operated
in the tuning fork-like manner earlier described in U.S. Pat. No.
Re 31450. Detailed discussion of methods and means for combining
motion sensor signals to determine mass flow rate appears in U.S.
Pat. Nos. Re 31 450 and 4422338 and in application PCT/US88/02360
published as W089/00679.
In the aforementioned meter designs, the sensors are-typically
placed at symmetrically located positions along the inlet and outlet
portions of the flow conduit which provide acceptable sensitivity
to enable the selected sensors to make measurements yielding a mass
flow rate that is accurate within +/- 0.2 percent.
On the order of about 100000 Coriolis mass flow meters have been
built using the inventions of one or more of U.S. Pat. Nos. Re 31450
4422338 and 4491025 and these meters have had extensive commercial
use. More than ten years' experience in the commercial application
of these meters to mass flow rate measurement with a variety of
diverse fluid products has shown that in general, the end users
are satisfied with the sensitivity and accuracy of their performance
but desire that the meters be improved in overall stability, including
zero stability, thus reducing plant maintenance related to these
meters, including meter recalibration. Meter instability, in general,
results from susceptibility of the meters to the unwanted transfer
of mechanical energy from sources external to such meters. Such
forces can also affect the zero (i.e., measured value at no flow)
stability of the flow meters.
While commercial experience as described above has shown essentially
no problem in practical use with fatigue failure of the flow conduits,
it is recognized that potential improvements in conduit life span
by reducing possible sources of fatigue failure represent a forward
step. Similarly, providing a sealed pressure-tight case increases
the suitability of the meters for hazardous materials applications
at significant pressures which may range up to 1000 psi and even
higher. Even when achievable pressure rating is balanced against
cost considerations involved in fabricating the case, the use of
a case as herein described affords a pressure rating for the meter
at least as high as 300 psi for flow tube outside diameter sizes
up to about 21/2 inches and as high as 150 psi for larger sized
flow tubes.
SUMMARY OF THE INVENTION
The present invention provides an improved mass flow meter with
considerably increased overall stability, including reduced susceptibility
to external forces and increased zero stability, reduced pressure
drop characteristics and better resistance to fluid pressures. A
number of design changes to the Coriolis mass flow meters manufactured
in accordance with one or more of the previously cited patents have
resulted in optimizing their already successful features and operating
characteristics.
The present invention relates to Coriolis mass flow rate meters
that include one or more flow conduits which are driven to oscillate
at the resonant frequency of the flow conduit containing fluid flowing
therethrough. The drive frequency is maintained at this resonance
by a feedback system, heretofore described, which detects a change
in the resonant behavior of the fluid-filled conduit as a result
of the fluid mass change due to changes in fluid density. The flow
conduits of these Coriolis mass flow rate meters are mounted to
oscillate about an oscillation axis located substantially at the
mounting points or at the location of the brace bars. The resonant
frequency of oscillation is that associated with the oscillation
axis. The flow conduit also deforms (twists) about a second axis
which is that axis about which the flow conduit deflects or twists
in response to Coriolis forces generated as a result of the flow
of fluid through the oscillating flow conduit. This latter axis
associated with Coriolis-caused deflections is substantially transverse
to the oscillation axis. The present invention provides an improved
flow meter with enhanced stability having reduced susceptibility
to the influence of outside forces, primarily because of optimized
sensor placement as explained more fully hereinafter. Other improvements
which contribute to overall stability of the improved meter include
reducing by at least fourfold the mass of the sensors and driver.
In a preferred embodiment, a modified U-shaped flow conduit design
is provided, having two essentially straight inlet and outlet legs
which converge towards each other at the process line manifold,
and bends, at two symmetrical locations along the length of the
conduit, separated by an essentially straight middle portion. It
is also contemplated that some modified U-shape flow conduits will
have convergent inlet and outlet legs which are separated by a continuously
curved middle portion, rather than a straight middle portion and
that others will have substantially parallel inlet and outlet legs
in accordance with current commercial embodiments. Attached to each
flow conduit at symmetrical locations are two motion sensors, so
located that the susceptibility to external forces of the signals
which they detect and transmit to the meter electronics is dramatically
reduced over that of previously known commercial mass flow meters.
This is accomplished in one preferred embodiment by locating the
motion sensors between but as close as possible to the nodes on
each side of the conduit of the second out of phase twist mode and
the third out of phase bending mode of the flow tube and placing
the driver equidistant between these sensors. The masses-of the
motion sensors plus their mountings and of the driver plus its mounting
are substantially reduced in relation to the corresponding parts
of the mass flow meters heretofore in commercial use. The susceptibility
of the flow conduit to fatigue failure may optionally be reduced
by providing novel brace bars having a novel nipple shaped sleeve,
which serve to define the axis about which each flow conduit oscillates,
but conventional brace bars may alternatively be used and in some
embodiments, brace bars are omitted. In one embodiment, advantage
may be taken of the convergent U-shape to provide a wafer configuration
manifold structure, without flanges, for connecting to the process
line to be monitored. A special sealed pressure tight case is provided
which encloses the flow conduit, motion sensors, driver and associated
electrical connectors. Several embodiments of such a case are disclosed
herein, among which the embodiment of FIGS. 8 and 9 is preferred
.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates an optimized Coriolis mass flow meter of this
invention partially within a case
FIG. 1A illustrates the location of the motion sensors as a result
of modal analysis for the FIG. 1 embodiment.
FIGS. 2A and 2B illustrates an option novel brace bar configuration.
FIG. 3 illustrates an optimized Coriolis mass flow meter of this
invention with a wafer manifold structure partially within a case
as shown by FIG. 4;
FIG. 4 illustrates an optional high pressure case design; and
FIGS. 5A-7I illustrate shaker table stability test results.
FIG. 8 illustrates another optional high pressure case design which
is preferred based on cost and ease of fabrication.
FIG. 9 gives further detail regarding the FIG. 8 case design.
FIGS. 10-10F and 11A-11L illustrate further test results as hereinafter
described.
DETAILED DESCRIPTION OF THE INVENTION INCLUDING THE DRAWINGS
The major feature of the present invention the minimization of
the influence of external forces upon meter stability--is achieved
through optimal placement of sensors and, to a limited extent, of
brace bars. It has been found that there are essentially six modes
of vibration, within the frequency range of 0 to 2000 Hz, excitation
of which is likely to result in loss of meter stability. They are
identified as (1) the first in phase bending mode (of lower frequency
than the drive frequency), (2) the first out of phase bending mode,
which corresponds to the fundamental drive frequency, except that
the drive frequency is the natural frequency of the fluid--filled
tube (whereas modal analysis is conducted on the empty tube), (3)
the first out of phase twist (also called torsion or deflection)
mode, (4) the second out of phase twist mode, (5) the second out
of phase bending mode and (6) the third out of phase bending mode.
Optimal placement of sensors is achieved by conducting a modal analysis
of the flow tube to locate the two nodes for each of the six modes
on that tube and to determine, based upon tube geometry, size and
material, those node points to which the sensors should be placed
most closely proximate. For a flow tube of the geometric configuration
shown in FIG. 1 for example, modal analysis has shown the sensors
should optimally be placed intermediate the second out of phase
twist node and the third out of phase bending node on either side
of the flow tube, as close to each of these nodes as possible. Those
skilled in the art will appreciate that, depending upon the geometry
and other characteristics of the flow tube, node points for the
six enumerated modes may be located differently in relation to one
another. In some flow tube shapes, two or more node points of different
modes may actually coincide, making it possible to locate sensors
at the location of coincidence and thereby enhancing flow tube sensitivity.
The present invention embraces the discovery, as a rule of thumb,
that meter stability is enhanced by locating the sensors as close
as possible on each side of the tube to at least two node points,
each of which is a node point for a different one of the modes above,
and especially of those designated (3) and (6) above. The invention
further embraces the discovery that the influence of mode 1 the
first in phase bending mode, may be minimized or eliminated by a
placement of the brace bars which separates harmonics of mode 1
from those of mode 2.
FIG. 1 illustrates a preferred embodiment of a Coriolis mass flow
meter of optimized modified U-shape and motion sensor positioning
and mounting. As in the current commercial meters, the flow conduits
112 are solidly mounted to the manifold 130 at points 131. Brace
bars 122 are solidly mounted to the flow conduits 112 thereby
defining oscillation axes B' when the flow conduits 112 are driven
in tuning fork fashion by driver 114. When flowable material flows
through conduits 112 the Coriolis forces cause the conduits to
deflect about deflection axes A'. Electrical connectors 125 from
the driver, 126 from motion sensors 118 and 127 from motion sensors
116 may be connected and supported in stable stress minimizing fashion
to bracket 128 as shown, or alternatively the bracket may be dispensed
with as discussed hereinafter, in favor of printed circuit boards
mounted to the case. The connectors shown in FIG. 1 are individual
wires or ribbon-shaped flexible connectors with embedded wires,
which are mounted in a stable, stress minimizing half-loop shape.
It is contemplated that flexible connectors, as described in U.S.
patent application Ser. No. 865715 filed May 22 1986 now abandoned;
its continuation U.S. patent application Ser. No. 272209 filed
Nov. 17 1988 now abandoned, and its continuation U.S. patent application
Ser. No. 337324 filed Jul. 10 1989 may be used. Such flexible
connectors also provide a stable, stress minimizing half-loop shape.
In FIG. 1 experimental motion sensors 116 are located at the ends
of the inlet or outlet leg 117 just before the bend. Motion sensors
118 are located at a position determined by modal analysis according
to this invention, which effectively minimizes the influence of
external forces. In a commercial meter embodiment, only motion sensors
118 would be present and motion sensors 116 used experimentally
for comparison purposes, would be eliminated. The meter of FIG.
1 includes case 140 which encloses the flow conduit and associated
attachments and is fixed to the manifold 130 but it is contemplated
that a meter case as shown in FIG. 4 or FIG. 8 and described below
might preferably be used.
Another feature of the flow meter design for the embodiment of
FIG. 1 which minimizes the effects of external forces involves balancing
of the flow conduits and their attachments and employing reduced
mass sensor and driver components.
In Coriolis mass flow meters, the flow conduits serve as springs,
with the spring forces acting predominantly on the inlet and outlet
legs. The flow conduits of the FIG. 1 embodiment are of modified
U-shape having no permanent deformation from bending in the inlet
and outlet legs. No bending during the fabrication process results
in the absence of permanent deformation in the four regions in which
spring forces act in the double conduit meters. As a result, all
four regions display essentially the same response to spring forces
if similar materials of essentially the same dimensions are used.
This improves the ability of the flow meter manufacturer to balance
the flow conduits. Balance of the flow conduits is further enhanced
by decreasing the masses of the motion sensors and driver by using
lighter mass magnets and coils and reducing the size of their mountings.
The motion sensors and drivers are comprised, as in current commercial
meters, of a magnet and a coil. The sensors are of the type disclosed
in U.S. Pat. No. 4422338 which linearly track the entire movement
of the conduit throughout its oscillation pathway. In the FIG. 1
embodiment and in other embodiments of this invention, the total
mass of these sensors and of the driver are reduced over their total
mass in current commercial meter embodiments by a factor of at least
about 4 and preferably by a factor of 5 to 6 or more. This reduction
in mass is accomplished by use of a lightweight bobbin having molded
pin connections and by winding the coil with 50 gauge wire while
continuing to use the same mass magnets. The lightweight sensors
and driver are mounted directly on the flow conduits, thus eliminating
mounting brackets. The lightweight coil used, e.g., in a meter embodiment
of the size and shape of that in FIG. 1 with a flow tube of approximately
0.25 inch outside diameter, has a mass of approximately 300 milligrams.
Prior coils used with comparable sized mass flow meters made pursuant
to U.S. Pat. Nos. Re. 31450 4422338 and 4491025 and having
approximately the same flow tube outside diameter size had a mass
of approximately 963 milligrams. In the same sized meter embodiment
according to FIG. 1 the new assemblies add a total of approximately
3.9 grams to the mass flow meter (1 driver coil and 2 sensors coils
at 300 grams each, and 3 magnets at 1 gram each). By contrast, in
comparable earlier commercial embodiments, the corresponding assemblies
added 22.2 grams (1 driver coil, 2 sensor coils, 1 coil bracket,
1 magnet bracket and 3 magnets) to the mass flow meter.
This resulting sensor location is not only between, but in the
closest possible proximity to each of the two node points, on each
side of the conduit. As those skilled in the art will readily appreciate,
by performing modal analyses on flow conduits of other precise shapes,
dimensions and materials, each of the node points for all of the
modes enumerated above can be located and resulting sensor locations
can readily be optimized.
In some embodiments of the improved meters of this invention, the
fundamental driving frequency (the first out of phase bending mode)
is increased relative to current commercially available flow meters
made by applicants' assignee, thereby increasing the values of its
harmonics. This results in better separation of the individual harmonics
for the drive mode from that of other modes. In the FIG. 1 embodiment
of the size stated, for example, the harmonics of the other five
modes of interest are each separated from harmonics of the driving
frequency by at least 20 Hz, for all frequencies below 2000 Hz.
In the FIG. 1 embodiment, the placement of the brace bars has the
effect of separating the first in phase bending frequency from the
fundamental driving frequency and thereby eliminating possible effects
of excitation of the first in phase bending frequency. The effects
of external forces operating at frequencies corresponding to the
remaining four modes of interest are in part minimized by the balanced
flow conduit design. In addition, the effects of the second out
of phase twist mode and the third out of phase bending mode are
also minimized in this embodiment by locating the motion sensors
between, but in close proximity to the nodes of both these two modes,
which nodes happen to be located close together. It is contemplated
that other location selections can be made to minimize the effects
of those modes that most affect stability of any particular conduit,
taking into account through modal analysis its size, shape and material.
Testing of the FIG. 1 embodiment of the current commercial Model
D meters manufactured by applicants' assignee and of current commercial
Coriolis mass flow meters manufactured by others at varying fluid
pressures ranging from less than 10 psi up to about 1000 psi established
that at fluid pressures approaching 1000 psi, variations in drive
frequency and twist frequency are induced which adversely affect
the accuracy of mass flow measurements. To date, it has been determined
that these effects of high fluid pressure are minimized by increasing
the flow tube wall thickness by approximately 20% and by enclosing
the flow tube assembly in a specially designed fluid-pressure-insensitive
case, as discussed below.
Applying the wall thickness increase to the meter embodiment of
FIG. 1 for example, for a tube of outside diameter 0.230 inches,
the wall thickness is increased from about 0.010 inches to about
0.012 inches in order to minimize instabilities caused by high fluid
pressure.
FIG. 2 shows an optional brace bar design according to this invention.
Each brace bar 122 is formed, as by punching a piece of metal (e.g.,
316L or 304L stainless steel) or other suitable material, to provide
two sleeves with nippled transitions 121 from holes having the outer
diameter of flow conduit (hole 124), to larger holes 120. These
brace bars are contemplated to be brazed or welded to the flow conduits
in order to reduce stress concentrations at the point of attachment
123 the primary locus about which the conduit is oscillated. It
is within the scope of the present invention, however, to utilize
conventional brace bars as earlier disclosed in the art.
FIG. 3 shows an optimized Coriolis meter as in FIG. 1 with an
exploded view of the process line attachment. Instead of the typical
prior art flanged manifold, a novel wafer flangeless structure 230
is provided for which the ends 232 can be bolted between the existing
flanges in a manufacturing or other commercial process line, by
means of threaded connectors 234 passing through flange holes 238
and held in place by nuts 235.
FIG. 4 illustrates one form of case which minimizes pressure effects.
This form can be used to enclose the entire flow conduit and sensor
attachment assembly. It comprises a pipe 350 of sufficient diameter
to enclose the flow conduits 312 driver, motion sensors and associated
wire attachments (not shown). The pipe is bent in the shape of the
flow conduit. It is then cut longitudinally into two essentially
equal halves. The flow conduits 312 are fitted into it along with
the associated driver, motion sensors and wiring. The other half
is fitted over this combined assembly and welded along the two longitudinal
seams and at the connections to the manifold. Thus, a pressure tight
case is provided which is suitable for applications involving hazardous
fluid containment and able to withstand significant pressures on
the order of at least 300 psi and up to 500 pounds per square inch
or more. For some embodiments, a printed circuit board may be attached
to the inside of the case, with flexible connections running from
the driver and motion sensors to the circuit board. A junction box
may then be attached to the case and connected to the circuit board
by wires which can be run through pressure tight fittings at the
top of the case. The junction box is in turn connected to means
for processing electronically the signals from the sensors to give
mass flow readout values and, optionally fluid density readout values.
An alternative case, preferred for ease of fabrication, is shown
in FIG. 8 and a section thereof is shown in FIG. 9. This form of
case is made from stamped steel pieces of half-circular cross section
as shown in FIG. 9 (which depicts a piece 2 or 4 from FIG. 8) welded
together to form the case. As specifically applied to the embodiment
of FIG. 1 this case is formed of ten pieces labelled 1-10 on FIG.
8 which are assembled in the following manner:
Five pieces (1 2 3 4 and 5 as labelled on FIG. 8) comprising
one half of the case--i.e., when assembled covering one half the
outer circumference of the flow tube--are welded to the support
comprising the inlet-outlet manifold of the flow tube. Printed circuit
boards, not shown in any of the figures, are affixed to the case
at locations as near as possible to the placement of the pick-off
coil portions of the sensors on the flow tube and the pick-off coil
terminals are connected to these printed circuit boards by flexures
containing wires of the type referred to hereinabove or by individual
half-loop shape wires. Wires are then run along the case to the
center straight section of the case (i.e. section 3 which encloses
the straight flow conduit section 112 of FIG. 1) where the wiring
feed-through to the meter electronics is located. This feed-through,
which is not shown in FIGS. 8 and 9 may comprise posts to which
the wires are directly connected or may comprise a third printed
circuit board to which the wires are connected and which is, in
turn, connected to feed-through posts and then to a junction box,
not shown, positioned on section 3 of the case at its midsection.
After the wiring is completed, the remaining five pieces (not shown
in FIG. 8 which is a plan view of the case) are welded in place
to one another, to the support and to the previously assembled and
welded portion, preferably by means of automated welding. These
latter five pieces comprise one half of the case. The welds between
pieces are as shown by the lines on FIG. 8. In addition, welds are
made along the inner and outer periphery of the case at seam lines
which are not shown, but which connect the top and bottom halves
of the case both inside the enclosure formed by the meter tube and
support and outside that enclosure.
The case embodiments of FIGS. 4 and 8 are illustrative only. Those
skilled in the art will readily recognize that similar cases can
readily be fashioned to any size and shape of curved or straight
tube Coriolis mass flow meter and that, depending upon the precise
shape involved, the embodiment of FIGS. 8 and 9 may advantageously
be made with other numbers of stamped steel half-circumferential
pieces. As is also readily apparent, other wiring arrangements may
be readily devised by those skilled in the art without departing
from the essential principles of this invention.
Shaker table tests were performed to test the influence of external
vibration forces and process line noise in exciting the flow conduit
with its associated attachments. Such external forces are frequently
present during plant operations. FIGS. 5A through 5F show experimental
shaker table test results for a current commercial Micro Motion,
Inc. Model D25 Coriolis mass flow meter. FIGS. 6A through 6F show
results for similar tests for a current commercial Micro Motion,
Inc. Model D40 meter. The Model D25 has a 0.172 inch inner diameter
flow conduit; the D40 has a 0.230 inch inner diameter. FIGS. 7A
through 7I show experimental test results for similar tests for
a Coriolis mass flow meter similar to that shown in FIG. 1.
In each of FIGS. 5A through 7I, the x-axis is the axis through
the meter flanges (i.e., parallel to the oscillation axis B'--B'),
the y-axis is parallel to the plane of the flow conduits (i.e.,
parallel to the deformation axis A'--A'). The z-axis is perpendicular
to the plane of the flow conduits.
These shaker table experiments were performed on complete mass
flow meter assemblies without cases. The output indicated on the
strip chart recordings of FIGS. 5A through 7I are of motion sensor
readings in response to the corresponding external vibration. The
sequence shown in each series of charts is the meter's response
to a linear frequency ramp ranging from 15 Hz to 2 KHz and then
vibration inputs at random frequencies (FIGS. 5C, 5F, 6C, 7C, 7F,
7I). The frequency ramp occurs over a ten minute period and random
vibrations occur over an approximately five minute period. FIGS.
5A, 5B, 5D, 5E, 6A and 6B each indicate the influence of external
vibrations of various frequencies in exciting harmonics of the six
modes of motion of the D25 and D40 meters that are discussed above.
FIGS. 7A through 7F show susceptibility to excitation due to external
vibrations of a meter of this invention of the FIG. 1 embodiment
about the x and z axes (the same axes as in FIGS. 5A through 5F)
and are to the same respective scale. FIGS. 7G through 7I are taken
about the y-axis. (The random vibrations were conducted first in
FIGS. 7D-7F). It is noted that the optimized meter shows dramatically
reduced influence of external vibrations in exciting harmonics of
the six modes of motion. Thus, the optimized design is shown effectively
to isolate the meters from effects of external forces.
In addition, tests were performed to test the influence of external
vibrations on zero stability. Such tests provided results for the
stability of the time difference (.DELTA.t) measurement at no flow,
the so-called jitter test. A frequency counter was used to directly
measure the pulse width of an up/down counter prior to any averaging
or filtering of the electronics. The test was performed on a shaker
table using random frequency input over a range of accelerations
in the x, y, and z directions. The results showed that, as the accelerations
were increased, the influence of external vibrations resulted in
pulse width divergence of one or more multiples of the average value
from the average value for both the D25 and D40 meters. Such divergence,
for each axis, is markedly reduced for the optimized meter. Thus,
as in the case of the vibration tests, the jitter tests showed that
the optimized meter design effectively isolates the meters from
the effects of external forces.
FIGS. 10A-10G inclusive are plots of further data obtained with
a flow meter embodiment constructed as in FIG. 1 having conventional
brace bars and a 20% thicker flow tube than comparably sized current
commercial meters, as herein disclosed, with sensors placed in accordance
with the teachings of this invention between, but as near as possible
to the node points of the modes labelled as 4 and 5 herein above.
As tested, the flow tube of this flow meter embodiment was half
covered (i.e., one half of the circumference of the pipe) by a case
of the type shown in FIGS. 8 and 9 and a junction box (not shown
in the drawings) was appended to the outside of the case where the
wires feed through the case. The junction box was conventionally
connected to another box (called the "remote flow transmitter"
or "RFT") containing the meter electronics and having
readout panels for mass flow rate and density values, from which
data was collected for FIGS. 10A-10D inclusive. The inner diameter
of the flow tube wall on this meter embodiment was approximately
0.206 inches.
FIGS. 10A and 10B each represent calibration plots of accuracy
versus flow rate using water at mass flow rates from 0 to 45 pounds
per minute. FIG. 10A differs from FIG. 10B in that FIG. 10A represents
a "22 point" calibration curve with the first measured
points taken at mass flow rates of about 3 to 4 pounds per minute.
FIG. 10B covers a "45 point" calibration curve in which
more data points, especially for mass flow rates below 5 pounds
per minute, (commencing at about 0.5 pound per minute) were collected.
FIG. 10B also shows fluid line pressure drop data as measured for
mass flow rates from 0 to about 45 pounds per minute. FIGS. 10A
and B combine to show that the meter embodiment of this invention
performs well within the published accuracy values of .+-.0.2% which
characterize the current commercial meters of applicants' assignee,
Micro Motion, Inc. FIG. 10B also illustrates the very acceptable
fluid line pressure drop performance of this meter embodiment.
FIGS. 10C and D are, respectively, plots of measured mass flow
rate and density analog drift values against fluid pressure of water
at values from 0 to approximately 2000 psi. In each instance, a
comparison appears on the plot of average historical standard deviation
measurements for commercial mass flow meters of the D-series sold
by Micro Motion, Inc. In FIG. 10C, flow rate analog drift and flow
rate standard deviation data points are shown in units of seconds.
In FIG. 10D, density analog drift and density standard deviation
are shown in units of grams per cubic centimeter. In both cases,
the data show the meter built in accordance with this invention
to perform well within the measured standard deviation data.
FIGS. 10E, F and G are plots of shaker table test data obtained
similarly to the data depicted in FIGS. 5A to 7I, but presented
as plots of shaker table frequency (in Hertz) against analog output
(i.e. mass flow rate) disturbance in seconds, whereas FIGS. 5A to
7I are reproductions of strip charts plotting shaker table frequency
in Hertz against motion sensor readings per se. In addition, the
data in FIGS. 10E, F and G were collected on the same meter assembly
with half case attachment as that to which FIGS. 10A-D inclusive
apply. For comparison purposes, similar plots for two different
D25 flow conduit units, each without case attachment, are presented
in FIGS. 11A-11L inclusive. In each instance, the x, y and z - axes
are as defined above, the shaker table vertical scale frequency
sweep is from 15 to 2000 Hertz, the remote frequency transmitter
from which analog output data were obtained had a span of 5 grams
per second and a calibration factor of 1.
A linear sweep of the vibration table expends the same amount of
time in moving through each 100 Hertz vibration interval so that,
e.g. a sweep of 15 to 115 Hertz occurs in the same time interval
as e.g. 1000 to 1100 Hertz. In log sweep, an amplified time interval
is consumed at low frequencies, e.g., from 15 to 400 Hertz and a
shortened (or speeded up) time interval is consumed at the higher
frequency end. In both instances, the total sweep time is the same.
As can be seen, log sweep clearly points out the frequencies at
which external excitations have given rise to harmonic disturbances
in current commercial Model D meters. FIGS. 10E, F and G show the
meters of this invention to be markedly less susceptible to such
influences than the D25 flow conduits.
Although the preferred embodiment is illustrated for a dual flow
conduit mass flow meter, it is contemplated that the invention described
herein can be embodied in a Coriolis mass flow meter having only
one flow conduit, either in conjunction with a member such as leaf
spring, or a dummy conduit, that forms a tuning fork with the flow
conduit or under circumstances where the single flow conduit is
of very small mass and is mounted to a base of relatively very large
mass.
While the foregoing detailed discussion focuses, for exemplary
purposes, upon one size and shape of flow tube, numerous changes
and modifications in the actual implementation of the invention
described herein will be readily apparent to those of ordinary skill
in the art, and it is contemplated that such changes and modifications
may be made without departing from the scope of the invention as
defined by the following claims. |