Electromagnetic flowmeter evaluation in real facilities: Velocity profiles and error analysis

The reliability and accuracy of a flowmeter depends on its installation conditions, as an improper installation may distort the end results. The objective of this work was to determine the errors obtained in an electromagnetic flowmeter of 100 mm diameter, installed in common real conditions, that is, not obeying the straight pipelines length recommended in flowmeter standards. To achieve this objective of the work, the electromagnetic flowmeter was installed in four different conditions: installed downstream of a 90° curve and downstream of two curves of 90°, both of short radius; downstream of a gate valve, openings of 100% and 50%. During the research, flow velocity profiles were obtained, with the purpose of evaluating the influence of profile distortion on the performance of the meter. The tests were carried out in the Laboratory of Hydraulics and Fluid Mechanics of FEC-Unicamp, Brazil, and the results are consubstantiated in comparative graphs, which demonstrate the influence of the velocity profile on the flowmeter response and accuracy for each setup bench.     

Influence of cone angle on the performance of a wafer cone flowmeter

The present study explores the effect of upstream disturbances like a single 90°bend, double 90° bends (in plane and out of plane) on the performance of wafer cone flowmeters with same beta ratio (β) of 0.77 but different half cone angles (α) of 30° and 45°. The influence of these disturbances on the upstream and downstream axial velocity (u) profiles are studied experimentally. The orientation effects, if any, are also studied experimentally. The minimum upstream distances required to get a fully developed flow for these disturbances vary with type of upstream disturbance, beta ratio (β) and half cone angle (α) of the wafer cone flowmeter. The study is carried out for a single phase flow with air as working medium at high Reynolds number (Re_D = 144000). From the results obtained from this study, it may be concluded that the wafer cone flowmeter with a beta ratio (β) of 0.77 and a cone angle of 30° requires less upstream distance compared to the wafer cone flowmeter with a beta ratio (β) of 0.77 with a cone angle of 45° for all the disturbances under consideration.     

Pressure measurement technique and installation effects on the performance of wafer cone design

The present study explores novel pressure averaging technique for wafer cone flowmeter design and its robustness in the presence of double 90° bend (out-of-plane) and gate valve as a source of upstream flow disturbance. The wafer cone flowmeter is tested in a circular pipe (inside diameter of 101 mm) with water as the working medium for the flow Reynolds number ranging from 1.19×10⁵ to 5.82×10⁵. Influence of the half cone angle (α) on the coefficient of discharge (C_d) of wafer cone flowmeter is studied with this new pressure averaging technique. Half cone angles considered in this study are 30° and 45° with a constant constriction ratio (β) of 0.75. The upstream static pressure tap is located at 1D upstream of the wafer cone. The downstream pressure averaging technique comprises eight circumferential holes of diameter 2 mm on the maximum diameter step of the wafer cone. The pressure taps are communicated through the support strut which serves as a downstream static pressure tap. The disturbance causing elements are individually placed at 1.5D, 5.5D, 9.5D and 13.5D upstream to the wafer cone flowmeter. The wafer cone flowmeter is also tested with gate valve opening of 25%, 50% and 75% for all the arrangements considered. The 30° cone is found to be better than 45° cone for the range of Reynolds number covered in the present study. The results show that the 30° wafer cone flowmeter with novel downstream pressure averaging technique is insensitive to the swirl flow created by a double 90° bend (out-of-plane) and requires an upstream length of 9.5D with a gate valve as a source of flow disturbance.     

Velocity and turbulence measurements downstream of flow conditioners

The flow downstream of three conditioners, a tube bundle and two perforated plates, is investigated experimentally by means of particle image velocimetry and hot wire anemometry for Reynolds numbers of the order 10⁵. The conditioners are exposed to the flow disturbed by two different installations: a 90° single bend and a 2×90° out-of-plane double bend. Velocity profiles, turbulent fluctuations and Reynolds‘ stress are measured. The jets issuing from the holes and tubes of the conditioners are visualised in the near field which extends up to approximately four pipe diameters downstream of the conditioners. The disturbance imposed on the flow by the conditioners disappears at this position, while the decay of the disturbance caused by the installations takes place in the far field. The decay rate in the far field depends on the specific installation. It is found that this decay is more rapid for the double bend. While the velocity profiles match the profile for fully developed flow approximately at a position of 25 diameters downstream of the conditioners, the turbulent equilibrium state is not even reached at 50 diameters. The results also show that the perforated plates have a higher efficiency than the tube bundle in conditioning the disturbed flow.     

Numerical analysis of installation effects in Coriolis flowmeters: Single and twin tube configurations

A fully coupled, partitioned, numerical model that accounts for fluid–structure interaction is applied for a study of the installation effects of Coriolis flowmeters. The modeled configurations include a single straight-tube full-bore flowmeter and two different twin tube flowmeters with straight and U-shaped measuring tubes. Three different flow disturbance elements positioned upstream of the flowmeter are considered in the study, as well as two different types of flow splitters in the case of the twin tube configurations. The installation effects are estimated by comparing the mass-flow sensitivities under the disturbed and fully developed flow conditions at the inlet of the flowmeter. For the modeled twin tube flowmeters they are found to be of the order of one-tenth of a per cent. These relatively small values of the installation effects are related to the presence of flow splitters and to the averaging of the motion of both measuring tubes in the twin tube configurations. Similarly, averaging the response from two sensor pairs instead of only a single sensor pair reduces the circumferential variations and the peak values of the installation effects for asymmetric flows in the single straight-tube flowmeter.     

Study on the design and performance of a bi-directional cone flowmeter

The present study explores a novel design of cone flowmeter for bi-directional flow metering application. Two identical cone shapes are machined with their base circle surfaces joined together with a small step in between them and differential pressure measurement is done across the apex of the cones. The bi-directional cone flowmeter is tested under fully developed flow conditions and its performance under double 90° bend (out-of-plane) is also evaluated. The bi-directional cone flowmeter is tested in a circular pipe (inside diameter of 101 mm) with water as the working medium for the flow Reynolds number ranging from 1.18×10⁵ to 5.48×10⁵. Influence of the half cone angle (α) and the location of static pressure taps on the coefficient of discharge (C_d) of a cone flowmeter are studied. Two cones with half cone angles α=30° and α=45° with a constant constriction ratio (β) of 0.75 are studied. Static pressure taps are located on both sides of the bi-directional cone. Two sets of locations of static pressure taps are studied. First set includes two static pressure taps on the pipe wall in the planes of apexes of the bi-directional cone—called apex taps. Second set includes pressure taps on the pipe wall in the planes at a distance D/4 away from the apexes of the bi-directional cone—called D/4 taps. Double 90° bend (out-of-plane) is placed at 1.5D, 5.5D, 9.5D and 13.5D upstream to the bi-directional cone flowmeter. It is observed that the apex static pressure taps located in the plane of apexes of the bi-directional cone result in statistically consistent coefficient of discharge for all Reynolds numbers covered in this study. The results suggest that the bi-directional cone flowmeter is insensitive to the swirl created by double 90° bend (out-of-plane) placed at the upstream of cone flowmeter, if placed at a distance of 9.5D or more.     

Performance analysis of the two-stage axial compressor with 90-degree bend inlet conditions

The 90-degree bend is one of the basic connection components applied in industrial flowing systems. This bend influences the aerodynamic performance of the downstream connecting equipment. In this study, the performance of a two-stage axial compressor with a 90-degree bend inlet was numerically and experimentally analyzed under low rotating speed. The testing results show that the outflow of the bend was numerically non-uniform in the circumferential and radial directions. To analyze the performance influence of the bend on the downstream compressor, a full passage compressor model with and without the inlet bend was simulated. The size of the distortion region gradually spread to the entire cascade, and the intensity of the distortion obviously dropped after the first stage as the compressor ran with the bend. The deterioration of the compressor performance, especially in the first stage, was verified numerically and experimentally. The total pressure coefficient and isentropic efficiency decreased by 2.6 % and 1.13 %, respectively. To save simulation cost, a model with a downstream single blade passage and cylindrical inlet was proposed, and the distorted flow was set as the inlet boundary condition. In addition, the unsteady numerical simulation was performed with the rotating inlet distortion region. Compared with the full blade passage model, the unsteady single blade passage model obtained the downstream spread characteristic of the distortion and performance deterioration. The latter can therefore be suggested as a compromised approach for obtaining the propagation characteristics of the inlet distortion with acceptable accuracy and low computational cost.

     

Development of side wall type permanent magnet flowmeter for sodium flow measurement in large pipes of SFRs

Sodium cooled Fast Reactors (SFR) require measurement of liquid sodium flow in its primary and secondary circuits. For the primary system of the pool type concept of SFR design, flowmeters have to be immersed in sodium pool and require flow sensors which can withstand high temperatures up to 550 °C, nuclear radiation and chemically reactive sodium environment. Secondary circuits and safety grade decay heat removal (SGDHR) circuits of SFR need flow measurement in stainless steel (SS) pipes of diameter varying from 15 mm to 800 mm. For small pipes, flowmeters with permanent magnet flowmeter with ALNICO-V magnet assembly is the unanimous choice. Conventional permanent magnet flowmeters (PMFM) for large pipelines become bulky, heavy and have installation problems. For sodium flow measurement in large pipelines a few other alternate methods are considered. In the case of Prototype Fast Breeder Reactor (PFBR), which is at an advanced stage of construction at Kalpakkam, flow in the 800 mm diameter secondary main circuit is measured by means of a bypass flowmeter. Other sensors that could be deployed include eddy current flowmeters (ECFM), which are introduced into the pipe to measure flow velocity in the pipe, ultrasonic flowmeters and permanent magnet based side wall flowmeters. In permanent magnet based side wall flowmeter (SWFM), a permanent magnet block is mounted on one side of the large pipe and the magnetic field produced by the magnet penetrates through the pipe and interacts with the flowing sodium and induces an electro motive force (emf) proportional to the flow. This is a compact, cost effective and fairly accurate method for flow measurement in large pipelines of SFR circuits. SWFM is suitable for pipelines of 100 mm and above. In the present work a side wall flowmeter for 100 mm pipe is designed, manufactured, calibrated and tested in an existing sodium facility. Voltage signal developed in SWFM for different flowrates was simulated with three dimensional Finite Element Model (FEM) and validated with experimental results. Effect of asymmetric magnetic field on flowmeter voltage signal and dependence of flowmeter voltage signal on position of electrodes was also analyzed with model. The feasibility of use of this type of flowmeter for large pipelines of SFRs is demonstrated.     

A method based on a novel flow pattern model for the flow adaptability study of ultrasonic flowmeter

Ultrasonic flowmeters are widely used in industry for accurate measurement. Flow adaptabilities of meters in non-ideal flow fields are usually concerned about by researchers. This paper presents a theoretical analysis method to study the measurement performance of ultrasonic flowmeter. For the specific water flow in single elbow pipe, a novel three-dimensional flow pattern model is invented by the trust region Newton algorithm based on computational fluid dynamics simulation results. In order to verify the correctness of the model, a typical ultrasonic flowmeter with single diametric acoustic path is mainly analyzed. By comparing flow adaptabilities of the meter downstream of the single elbow with both the novel theoretic model analysis approach and simulation method, good agreement is achieved. It is indicated that both the three-dimensional model and its invention method are valid for this study, which is not only helpful to get knowledge of characteristics of disturbed flows, but also provides a practical method to study the flow adaptability of ultrasonic flowmeter in non-ideal flow fields.     

Measurement of steam flow rates using a clamp-on ultrasonic flowmeter with various wetness fractions

Most of the heat in industrial plants is supplied by steam. To minimize energy waste, measuring the steam flow rates in existing pipes is important. Clamp-on ultrasonic flowmeters are used for this purpose, for which the sensors are attached to the pipe wall. However, flow conditions that can be used are limited because the signal-to-noise ratio of the ultrasonic signal in a steam flow is low. Furthermore, the steam wetness increases with heat losses, which may affect measurement results. Therefore, flow rate measurements in wet steam flows using clamp-on ultrasonic flowmeters have not been fully established. In this study, steam flow rates with various wetness fractions and system pressures were measured using a laboratory-made clamp-on ultrasonic flowmeter. The results show that flow rates in wet steam could be determined within a 10% error under general conditions in a steam piping system, although the conversion factor from line-average to area-average velocities was calibrated in superheated conditions, and the speed of sound in saturated conditions at each pressure was used. However, the error of the flow rates tended to increase with the wetness fraction and was biased toward positive values. The speed of sound and liquid volume fraction were evaluated at different wetness fractions. The flow rate error due to the change in sound speed was less than 1%, and 1.2% of the flow rates were overestimated owing to the liquid volume fraction. The velocity distribution in wet steam was considered different from that in the superheated steam owing to the existence of the liquid phase, and the change in velocity profile may lead to an overestimation of the steam flow rates in the wet steam condition.     

An investigation of heat transfer characteristics of swirling flow in a 180° circular section bend with uniform heat flux

An experiment was performed to obtain the local heat transfer coefficient and Nusselt number in a circular duct with a 180° bend for Re=6X 10⁴, 8X 10⁴ and 1 X 10⁵ under swirling flow and non-swirling flow conditions. The test tube with a circular section was made from stainless steel having a curvature ration of 9.4. Current heat flux of 5.11 kW/m² was applied to the test tube by electrical power and the swirling motion of air was produced by a tangential inlet to the pipe axis at 180°. Measurements of local wall temperatures and the bulk mean temperatures of air were made at four circumferential positions at 16 stations. The wall temperatures showed a reduced distribution curve at the bend for the non-swirling flow, but this effect did not appear for the swirling flow. The Nusselt number distributions for the swirling flow, which was calculated from the measured wall and the bulk temperatures, were higher than that of the non-swirling flow. The average Nusselt number of the swirling flow increased by about 90-100%, compared to that of the non-swirling flow. The Nu/Nu_DB values at the 90° station for non-swirling flow and swirling flow were approximately 2.5 and 4.8 at Re=6X 10⁴ respectively. The values agree well with Said’s results for non-swirling flow.     

A study of the flow characteristics of developing turbulent pulsating flows in a curved duct

Flow characteristics of turbulent pulsating flows in a square-sectional curved duct were experimentally investigated. Experimental studies for air flow were conducted to measure axial velocity profiles, secondary flow and pressure distributions in a square-sectional 180° curved duct by using an LDV system with a data acquisition and processing system which includes a Rotating Machinery Resolve (RMR) and PHASE software. Measurements were made at the seven cross-sections from the inlet (ø=0°) to the outlet (ø=180°) of the duct with 30° intervals. Pressure was measured by using a magnetic differential pressure gage. The experiment was conducted in nineteen sections from the inlet to the outlet of the duct at 10° intervals.Velocity profiles for turbulent pulsating flows were large at the outer wall for a bend angle of ø=30° because of the centrifugal force. The velocity profiles were similar to those of turbulent steady flows. The secondary flow of the turbulent pulsating flow had a positive value at a bend angle of 150° without regarding the phase. The dimensionless value of the secondary flow became gradually weak and approached to zero in the region of a bend angle of 180° regardless of the ratio of velocity amplitude. The pressure difference of turbulent pulsating flows was the largest near the region of a bend of angle of 90° in the case of the middle region and became small beyond 90.     

Effects of gas leaks in oil flow on single-phase flowmeters

The effect of gas entrainment in oil on the performance of a range of single-phase flowmeters has been investigated experimentally using the National Standard Multiphase Flow facilities at NEL. The flowmeters tested were 4-inch and 2-inch positive displacement, venturi, helicoidal and flat-bladed turbine meters and 2-inch U-tube and 1.5-inch straight tube Coriolis meters. The flowmeters were tested in oil flow with gas fractions up to 15% by volume. The aim of the project was to quantify the effect of second-phase fluid components on the basic uncertainty of a range of single-phase flowmeters and, as a consequence, identify which generic types of single-phase flowmeter were most suitable in applications where such components may be present. These tests have provided evidence of the suitability of particular flowmeters for two-component flow applications. Comparisons have been made between generic type and size of flowmeter. At low gas fractions, the positive displacement and venturi flowmeters were more accurate than the other meters and estimated the total flowrate to within ±2%. Over 9% gas fraction, there was an improvement to the response from some of the flowmeters with increasing gas fractions. This was considered to be indicative of improved mixing in the flow.     

The performance characteristics of a novel multi-electrode electromagnetic flowmeter

The design of a novel multi-electrode electromagnetic flowmeter, which provides nine conventional electromagnetic flowmeters housed in a single body, has been used in adverse flow conditions immediately downstream from a header tank. The results show that careful initial calibration, in accord with the pipeline conditions recommended in international standards, enables errors of under 1% to be achieved. The results show that the two electrode planes closest to the header provide results with the greatest errors. The performance with the electrodes mounted in the vertical plane is generally better than that achieved with the electrodes mounted in the horizontal plane. Pressure measurements together with laser Doppler anemometry (LDA) and particle image velocimetry (PIV) techniques have been used to provide details of the magnitude of the disturbance caused to the flow field by the header and outflow combination.     

Installation effects on ultrasonic and electromagnetic flowmeters: a model-based approach

The paper describes the basic ideas of a model-based installation-effects analysis method and presents results from a collaborative research programme being conducted at the National Institute of Standards and Technology (NIST) in Gaithersburg, MD, USA, and at Tampere University of Technology in Finland. The analysis method is based on a combination of flow modelling and flowmeter modelling. In this paper, installation effects on electromagnetic and travel-time-difference ultrasonic flowmeters are discussed. The installation cases are single elbow and double elbow out-of-plane piping configurations. The results show that significant shifts from ‘ideal’ meter performance can occur in such disturbed flow conditions. The flowmeter modelling results also show that these significant shifts can be reduced by altering the operational arrangements and performance characteristics of the meters. It is concluded that such flowmeter modelling can be very useful in: (1) redesigning flowmeters to make them less susceptible to installation effects, (2) constructing new meters, or (3) arranging special meter configurations to deal with specific disturbed flows.     

Influence of blockage and upstream disturbances on the performance of a vortex flowmeter with a trapezoidal bluff body

Experimental investigations are conducted on vortex flowmeter with the differential wall pressure measurement method. The bluff body employed is trapezoidal in shape and water is used as the working fluid. Three different blockages (0.14, 0.24 and 0.3) are considered in this study. The performance of the vortex flowmeter is studied both under fully developed condition and in the presence of flow disturbances. The flow disturbance is created using 45° swirl generator and gate valve placed at different upstream distances. The performance of the flowmeter is also evaluated in the presence of a Laws Vanes flow conditioner placed downstream of the swirl generator. The blockage ratio of 0.3 is found to be the best among all the blockages studied under both disturbed and undisturbed conditions.     

Water flow measurement in large bore pipes: an experimental comparison between two different types of insertion flowmeters

In this paper the metrological behavior of two different insertion flowmeters (magnetic and turbine types) in large water pipes is described. A master-slave calibration was carried out in order to estimate the overall uncertainty of the tested meters. The experimental results show that (i) the magnetic insertion tested flowmeter performs the claimed accuracy (+/- 2%) within all the flow range (20:1); (ii) the insertion turbine tested meter, instead, reaches the claimed accuracy just in the upper zone of the flow range.     

Numerical investigation on the installation effects of electromagnetic flowmeter downstream of a 90° elbow–laminar flow case

An electromagnetic flowmeter installed downstream of a 90° elbow to measure the flowrate of laminar flow is numerically simulated to investigate installation effects by varying the location of the electromagnetic flowmeter at a distance up to 22D from the elbow, and the angle between the electrodes plane and the symmetry plane of the elbow at ϕ=0, 45 and 90°. Effects of the curvature radius (R_c) and the Reynolds number (Re) based on a diameter D are also scrutinized in the range of 400≤Re≤1500 and R_c=1.5D and 3.0D.For the simulation of an electromagnetic flowmeter, a commercial code FLUENT(ver. 4.4) is applied for flow field analyses and a three-dimensional numerical code is developed for analyzing the magnetic field. The developed code adopts a finite volume method to solve a Poisson-type voltage equation for the magnetic field.It is found that the deviations of the flow signal due to the disturbance from the elbow is strongly dependent on the pattern of axial velocity contours. Cases for ϕ=45° are found to permit significantly better measurement accuracy in comparison with ϕ=0° and ϕ=90°, and the effect of the curvature on the optimum installation distance depends on the Reynolds number. The present numerical simulation method is found to be a useful tool for the performance analysis of the electromagnetic flowmeter.     

Flow characteristics of back supported V-cone flowmeter (wafer cone) using PIV

The present experimental study has been carried out to evaluate the performance and flow characteristics of the Wafer cone flowmeter using Particle Image Velocimetry (PIV). Two equivalent diameters (β) of 0.62 and 0.72 with combination of two vertex angles (ϕ) namely 30°and 45°are used for the evaluation of the performance of the flowmeter in the range of Reynolds number of 3 × 10³ to 8.19 × 10⁴. The investigation shows that the coefficient of discharge seems to be independent of β-value with the increase in vertex angle. Further, the appropriate location of the downstream pressure tap is also estimated for the cone configuration of β = 0.62 and ϕ = 30°. It is observed that the downstream pressure tap location of 0.8D distance gives a higher value of discharge coefficient compared to 0.0D distance with the error being also lower marginally. PIV data has been analysed for the cone configuration of β = 0.62 and ϕ = 30°at four Reynolds numbers of 3028, 6057, 52755 and 74488 in terms of axial velocity and turbulent intensity. The measurements reveal an interesting phenomenon in terms of the rapid decay of turbulent kinetic energy on the downstream of the cone. This may be due to the interference of the cone wake with the support wake resulting in fast decay. This unique phenomenon leads to the reduction in the requirement of the downstream straight length for the Wafer cone flow meter, unlike other obstruction type flowmeters.     

Evaluation of flowmeters for heat metering

Heat flowmeters are expected to be reasonably priced, be very reliable, and have high measurement accuracy. Various types of heat flowmeters have been developed and they are widely used in large residential and industrial buildings. In this study, three types of heat flowmeters (turbine, electromagnetic and ultrasonic) were tested for accuracy, effect of installation position and vibration, durability and performance in the field for several installation positions and in the presence of vibration. We used a liquid flow standard system and a customized durability test system in accordance with the International Organization of Legal Metrology (OIML) R 75-2 heat meter testing method. The field test was conducted in eight different locations from winter to summer. All flowmeters were calibrated before and after the field test, and the measurement deviation and the relative expanded uncertainty were calculated. The mean deviations obtained were–0.21%,–0.07%, and 0.11%, with the relative expanded uncertainties 0.48%, 0.17%, and 0.40% for turbine, electromagnetic, and ultrasonic flowmeters, respectively. The results of position and rotation tests, mean deviations by rotation angles at 90°, 180°and 270°relative to 0°(horizontal position) were–1.24%,–1.07% and–0.80%, respectively. For the vibration tests at 1 m/s² and 5 m/s² vibration acceleration, the turbine flowmeter, the electromagnetic flowmeter and the ultrasonic flowmeter showed deviations that ranged from −0.2% to −0.5%, −0.6% (2.6 m³/h), and 0.0% (negligible), respectively. In the durability tests, the accuracy of all three types of heat flowmeters remained at ±1% or less, showing sufficient durability. In the field test, the deviation of the turbine flowmeter and the ultrasonic flowmeter showed ±2.5% or less deviation. However, the electromagnetic flowmeter seems to be inaccurate below 6.9% of the maximum flow rate.