IMEKO Event Proceedings Search

   

Page 411 of 977 Results 4101 - 4110 of 9762

M. Anklin, G. Eckert, S. Sorokin, A. Wenger
EFFECT OF FINITE MEDIUM SPEED OF SOUND ON CORIOLIS MASSFLOWMETERS

A theory is presented for the interaction between a vibrating tube and the enclosed flowing fluid with finite speed of sound. From this the influence of the speed of sound on the massflow measurements is derived. This model is valid for compressible mediums and for flow velocities which may be quite high (~ 100 m/s) but still subsonic. Such velocities can easily be reached in a Coriolis mass flowmeter (CMF) for gas application. The exact theoretical calculation is complex and the results can not be written in a simple one line equation. Nevertheless, we approximate the exact formula to estimate the deviation between true mass flow and the CMF reading. This deviation can be described with mainly two terms:
The first term is proportional to (kd)² = (2 π f d / c)², with f being the resonant frequency, c being speed of sound of the medium, and d being the diameter of tube. This term contributes most to the observed deviation. The second term depends on density and velocity of fluid. A comparison of this model with experimental data shows a good agreement.

P. Kesic, V. Damic
THE EFFECTS OF ASYMMETRIC FLOWS IN THE CORIOLIS MASS FLOWMETER WITH STRAIGHT PARALLEL TUBES: NUMERICAL SIMULATION

The effects of flow disturbances on the performance of the Coriolis mass flowmeter(CMF) with two straight parallel tubes were invetigated numerically. Two types of flow disturbances were studied numerically using the bond-graph presentation. The first one represents a local restriction in one of the pipes that causes asymmeric flow in pipes. In the second case of operating conditions, uniformly obliterated tubes were considered. The effect of the obliterated layer thickness with different densities was analysed. It was found that the local restriction at the tube entrance does not effect the sensitivity coefficient, while the flowrate in the tube changes from zero to a rate equal to that in the parallel tube. The obliteration does not effect the sensitivity coefficient either, but changes the first natural mode of tube vibration. If the CMF is used for density measurement, the obliteration may cause significant error.

H. Gehlhaar, B. Nath
LDA MEASUREMENTS IN HIGH-PRESSURE GAS PIPELINES

In industrial gas volume measurement, the reliable, precise, and economically efficient determination of the gas quantities by measurement is of utmost importance. Standby meter sections, Z-connection, and permanent series connection of two gas meters of different designs represent the state of the art. The permanent or temporary series connection for the comparison and test sections require expenses for a the establishment of a second complete measuring device. An alternative at present is the dismounting of the gas meter and checking it in an high-pressure test facility. A test according to requirements, using a non-interfering measuring instrument which should be portable, if possible, and usable with other gas meters, can be a bridge between these two options. To prove this, measurements with an LDA system were carried out in a natural gas high-pressure pipeline up-stream of a turbine gas meter. After extensive profile and single-point measurements have been made, it can be said that this test using the single-point method is feasible with an uncertainty of ±0.5%, providing that certain prerequisites are met.

H. Jeanneau, M. Piguet
PIPE FLOW MODELING FOR ULTRASONIC FLOW MEASUREMENT

In many industrial applications, the precise measurement of volumetric flow- rates is of crucial importance for economic or safety reasons. Flowmeters are generally installed in perturbed flow conditions. The measured value is then affected by a systematic error which can reach several %. EDF has developed a numerical tool to estimate the error introduced by the flow disturbances on the flow-rate measurement with ultrasonic flowmeters or differential pressure devices.
The tool is based on a 3D CFD program which determines the flow profile. A module allows the simulation of ultrasonic flowmeters and differential pressure devices.
This paper deals with ultrasonic flowmeters which are very sensitive to the flow profile through the parameter K (profile factor). Two configurations are presented in this paper:
- a double bend out of plane,
- a pipe representative of the primary loop in nuclear power plant.
In each case, both modelisation and experimental tests have been carried out to determine the profile factor K representative of the flow conditions, which should be used for the measurement with the ultrasonic flowmeter to reduce the error introduced by the disturbances. The influence of different parameters has been studied. Experimental and calculated profile factors are compared. In both configurations, the simulation results are in good agreements with the experimental data. The tendency is well predicted by the tool and the typical difference between experimental and numerical K is around 2% or 3%.
The profile factors are also compared to the classical K (Prandtl Model) used and proposed by manufacturers, which is calculated from an analytic model of fully developed profile. The gain realized by using simulation is significant.

Th. Hüwener, E. von Lavante, Reinhard Ernst, W. M. Schieber
NUMERICAL INVESTIGATION OF FLOW EFFECTS ON ACCURACY IN TURBINE FLOW METERS

The present paper describes 2-D numerical simulations of a new class of turbine flow meters with two counterrotating rotors. The main goal was to investigate stator/rotor interactions and to improve the accuracy of the meter. The impact of stator/rotor interaction on the variation of coefficient of tangential force is shown. An improved design with a modified guide vane is proposed. Experimental results of the accuracy curves for the original and the modified guide vane are shown.

S. Perpéet, E. von Lavante, H. Windorfer, V. Hans
EFFECTS OF DISTURBED INFLOW ON ACCURACY OF VORTEX-SHEDDING FLOWMETERS

Vortical flows are mostly considered as disturbances, however, in a vortex-shedding flowmeter controlled vortex shedding from a bluff body is used for throughflow metering. In this case, any additional vortices, swirl, or pulsations introduced upstream will interfere with the meter’s function. In the present work, the effect of inflow vortices on the accuracy and function of a vortex-shedding flowmeter are investigated numerically. The inflow vortices are induced by a single and double bend positioned upstream of the bluff body. The resulting flow is analyzed using visualization and DFT for various types of bluff bodies.

C. J. Bates
THE PERFORMANCE OF A MODIFIED ELECTROMAGNETIC FLOWMETER WHEN ABUTTED TO A SMALLER MISALIGNED UPSTREAM DIAMETER PIPE

Electromagnetic flowmeters are today widely used for industrial flow measurement. For industrial users, due to physical constraints, the installation recommendations contained in national and international standards may not be achievable. The actual installation then results in the flowmeter being exposed to velocity profiles which are not fully developed, and do not accord with the situation for which the meter was designed and calibrated. The influence of a smaller upstream diameter pipe together with pipeline misalignment, when abutted to an electromagnetic flowmeter, is discussed.

J. Escande, P. Gajan, A. Strzelecki
3D VELOCITY PROFILE RECONSTRUCTION OF GAS FLOW IN A PIPE WITH ULTRASONIC TOMOGRAPHY

Beyond possibilities of simple flowmeters, ultrasonic tomography allows to reconstitute the three-dimensional velocity field over a circular pipe cross section. The differences of transit times between transducers distributed around the pipe are used in an iterative algebraic reconstruction algorithm in order to reconstruct the three velocity components. This experimental method has been tested successfully to quantify the 3D velocity field of various flows (fully developed, asymmetric and swirling air flows).

P. Brassier, B. Hosten, F. Vulovic
AN ULTRASONIC GAS FLOWMETER USING HIGH FREQUENCY TRANSDUCERS AND CORRELATION TECHNIQUE

The research activities carried out in the past years showed that in gas networks some constraints exist on the ultrasonic working frequency of flowmeters. The study of the transmission and the attenuation of the signal, and noise mechanisms gives us a defined suitable frequency range. Thereby, a working frequency of 500 kHz proves to be the most suitable to avoid noise effects. This frequency is above the noise level detected in gas pipelines. Moreover, the signal lost due to attenuation of ultrasound in gas is still negligible. In parallel, the use of this frequency allows the application of efficient numerical techniques of signal processing such as the correlation method. An initialisation process based on this method is developed for gas flowmeters. It provides low uncertainty on parameters involved in the flow measurement process.
A single path flowmeter equipped with this system has been tested on the Gaz de France test facilities. Its accuracy is better than 1.5% from 250 to 1000 m³ / h without initial adjustment. The auto-calibration process also forms part of the system, using a systematic comparison between ultrasonic measurement of sound velocity and a theoretical approach. The signal to noise ratio remains large enough to perform correctly the measurements even with a control valve installed close to the flowmeter.

V. Skwarek, V. Hans
THE ULTRASONIC CROSS-CORRELATION FLOW METER – NEW INSIGHTS INTO THE PHYSICAL BACKGROUND

During the last years of research, flow measurement by the cross correlation method made much progress concerning accuracy and reliability. Especially for gaseous fluid the signal processing was far improved by new demodulation techniques such as undersampling and complex demodulation. This allows new insights into the physical principle of interaction between ultrasound and fluid introduced in this paper.

Page 411 of 977 Results 4101 - 4110 of 9762