Measurement Library

International Symposium on Fluid Flow Measurement Publications (2006)

North American Fluid Flow Measurement Council

Calibration Of Rotating-Element Current-Meters Using An Annular Tow Tank
Author(s): R. lvarez-Breton, E. Dominguez-lvarez P. Cortez-Mejia, J. Martinez-Navarro, A. Aguilar-Chvez, S. Tamari
Abstract/Introduction:
Water-filled straight tow tanks are commonly used to calibrate current-meters. However, the use of these facilities is tedious. As an alternative, an annular tank with two rotating arms was built at the Mexican Institute of Water Technology (IMTA). Such a facility can a priori simulate flow velocities during an infinite time, provided that the water in the tank remains reasonably still. Therefore the tank has several pairs of buffers to set water at rest soon after the passing of an arm with a meter under calibration. Some tests were performed to evaluate the ability of the annular tank to calibrate rotating-element current-meters. Two Price AA and two Pigmy current meters, which had been previously calibrated in a straight tow-tank, were used for this purpose. Results show that the annular tank can be used to calibrate rotating-element current meters with reasonable accuracy in the range from 0.15 to 3 m/s, provided that: (1) at least eleven pairs of buffers are used, (2) only one Price AA current meter (or a smaller device) is under calibration, (3) water is initially set at rest during 300 s, (4) simulated flow velocities are gradually increased and test duration is not more than 1200 s. Under these conditions, residual velocity produced by the passing of the meter under calibration remains typically smaller than 2 % of the test velocity.
Go to Download Page
Email Reference
Document ID: 7FFEBE01

Determination Of Liquid Flowmeter Characteristics For Precision Measurement Purposes By Utilizing Special Capabilities Of Ptbs Hydrodynamic Test Field
Author(s): Rainer Engel, Hans-Joachim Baade
Abstract/Introduction:
In flow measurement applications with high-accuracy requirements, it is necessary to have an exact knowledge of the meter characteristics under different process conditions. Liquid flowmeters reveal both temperature and pressure dependence of their steady-state metering properties. So it requires the flowmeter error curves to be measured with the fluid temperature and pressure being varied as process parameters over their ranges in which the meters are to be applied. PTBs new national water flow standard and high-accuracy calibration facility provides unique capabilities to perform meter characterization measurements which meet those requirements. That has been proven in a series of calibrations.
Go to Download Page
Email Reference
Document ID: F3566E6F

Performance Testing Of Liquid Petroleum Meters
Author(s): Thomas Kegel, Raymond Kalivoda
Abstract/Introduction:
As the price petroleum products increase in the world market, the need to accurately measure escalates. Even seemingly small measurement errors can add up to millions of dollars over just a few months. Dynamic measurement with flow meters has played an important part in the accurate measurement of petroleum products since the mid 1930s. Initially Positive Displacement meters were the only devices approved by the American Petroleum Institute (API) for the custody transfer of petroleum products. Later, in the early 1970s, the API issued the Turbine Meters Standard, which was used for refined products and light crude oils. More recently, the helical turbine meters were introduced to measure more viscous products. Most recently API Standard 5.6 Measurement of Liquid Hydrocarbons by Coriolis Meters (2002) and API Standard 5.8 Measurement of Liquid Hydrocarbons by Ultrasonic Flowmeters Using Transit Time Technology ( 2005) were issued. Choosing the right meter and verifying the meters performance over the flow and product range is important for an accurate, trouble-free custody meter application. Verification of performance on hydrocarbon products of like characteristics by the manufacturer is an important step in this process. The details of this critical step in the custody meter manufacturing process are the subject of this paper.
Go to Download Page
Email Reference
Document ID: 004DF152

Investigations On Installation Disturbances Of Flow Sensors Of Heat Meters
Author(s): Franz Adunka, Thomas Lederer
Abstract/Introduction:
This article contributes shortly to the results of the previous studies concerning installation disturbances. The experiments are not finished yet, so that the results are preliminary. The initial measurements on this topic had been carried out on the test rig of the Fern-warme Wien company under the direction of Karlheinz Lechtermann and also on the old test rig at the BEV by F. Adunka, starting in 1996. By the end of 1998 the BEV had built up its new test rig for measurements with a very low uncertainty 1,2. Additionally a Laser-Doppler-Velocimeter (LDV) was purchased by which axial flow profiles after disturbances could be studied. Moreover, the flow profile development in the lines of test rigs could be examined thus giving valuable data on the influence of disturbances on meters under test. The presented results had been published in extracts in 3 and 4, recent measurements on the BEV test rig have been published partly, too (see 4 - 9). The measurements of this article are a sum of the most recent findings, but they are - as mentioned above - preliminary. The investigations were carried out on velocity meters for winged wheel type water me-ters, Woltman meters, ultrasonic meters (US meters), electromagnetic meters (MIDs), with water as the heat carrier at a temperature of 35 C 1. The measurements at the BEV were per-formed with inlet pipes up to 60 D (D inner diameter of the connection pipe of the meter). Because the BEV does not provide of longer test sections, the whole LDV-system was trans-ferred to the PTB-IB. The outcome there was that for distances approx. 92 D significant influences of disturbances on the flow profile could not be discovered any more. For the tests only pipes made of stainless steel had been used, with flanges and with di-ameters matching exactly the inner diameter of the connection flanges of the meters. Some-times problem arose because some of the meters fitted exactly, whereas others had significant dimensional differences. It had taken some time that these discrepancies were discovered. For the recent investigations adapters had been designed as diffusers/convectors with a very small opening angle (approx. 1) for a smooth transition from the standardized inner diameter of the pipes to the flange diameter of the meters. For the recent part of the project, where we also used the LDV-system, only meters with flanges and with a nominal diameter of their connections DN 40 and DN 50 had been investi-gated. The first nominal diameter DN 40 was representative for the large number of small-sized flow sensors (e.g. winged wheel type water meters), the second diameter DN 50 for the group of big-sized flow sensors (e.g. Woltman meters). This restriction was also necessary in order to realise on the test rig the inlet and outlet pipes as long as possible.
Go to Download Page
Email Reference
Document ID: F5B1A417

Computational Fluid Dynamics (CFD) And Its Application In Fluid Measurement Systems
Author(s): Jose De Jess Casillas Maldonado, Diego Nelson Moncada Benavides
Abstract/Introduction:
Some types of configurations (headers, simple, multiple or serial elbow pipes etc), flow control elements (valves, regulators, filters etc) or flow measurement equipment (orifice plates, turbines etc) may distort gas flow in its path, generating asymmetric velocity profiles, swirls, a combination of both or some kind of flow separation phenomenon. Separation occurs when flow momentum in the border layer is insufficient, and the flow of the layer adjacent to the solid surface separates from the surface (example, reduction in a valve port, flow regulators or orifice plates).
Go to Download Page
Email Reference
Document ID: 92670A15

Paper Summarizing API Investigative Testing For Existing Square Edge Flange Tap Orifice Expansion Factor Equations
Author(s): Casey Hodges
Abstract/Introduction:
When measuring compressible fluid flow through a differential producing flowmeter, an expansion factor or coefficient must be used to correct for the changes in density and pressure as the fluid is accelerated through the flowmeter. In the 1930s, Buckingham and Bean developed the expansion correction that is currently being used in API 14.3 and ANSI 2530, which is applied in the United States and other countries. In the mid 1980s, work by Kinghorn showed that the Buckingham equation was in error. Further work by Seidl in the mid 1990s confirmed the errors seen by Kinghorn. In the late 1990s, Reader-Harris collected data from several facilities, and formed a new equation for the expansion factor. This equation has been implemented into ISO 5167-2. This paper describes the results of testing to determine the validity of existing expansion factor equations for flange-tapped sharp-edged orifices. Tests were performed at two independent facilities and on four different line sizes 2, 4, 6, and 10 inch, with diameter ratios ranging from 0.2 to 0.66. After initial testing, it was decided to also test a 3 inch line. At the Colorado Engineering Experiment Station Inc. (CEESI) facility dry air was used as the compressible flowing fluid, while at the Southwest Research Institute (SwRI) facility natural gas was used. The ratio of differential pressure across the orifice to the static inlet pressure was varied from near zero, to approximately 0.25 at a constant Reynolds number. The expansion factor was determined from the change in discharge coefficient with relation to the change in the differential pressure to inlet static pressure ratio at a constant Reynolds number. This paper shows that new equations fit the API expansion factor data set more effectively than the existing Buckingham and ISO 5167.2 equations.
Go to Download Page
Email Reference
Document ID: A699B6BE

The Revision Of ISO/TR 15377 Orifice( Plates, Nozzles And Venturi Tubes Beyond The Scope Of ISO 5167)
Author(s): Dr Michael Reader-Harris, Dr Jeff Gibson, Dr David Hodges, And Mr Ronnie Rushworth
Abstract/Introduction:
This is a necessary consequence of the revision of ISO 5167, but the opportunity is being taken to make other changes to ISO/TR 15377 either to respond to work carried out since the previous revision or to give clarity. The particular areas in which significant change will be made are: orifice plates or nozzles with drain holes, orifice plates with small pipe diameter, orifice plates with no upstream or no downstream pipeline, and Venturi tubes with convergent angle 10.5.
Go to Download Page
Email Reference
Document ID: A29A37E6

EUROMET Project 852: A Comparison Between Italian And Swiss Gas Flow Standards In The Range 0.3 To 25 m3/h
Author(s): Henri Baumann, Giorgio Cignolo, Mario Clausen, Roberto Goria
Abstract/Introduction:
An interlaboratory comparison (EUROMET Project 852) of four primary standard gas provers was carried out between the Italian and Swiss NMIs (IMGC/INRIM and METAS) from 0.3 m3/h to 25 m3/h using ambient air. The transfer standard was a G25 rotary piston meter. The comparison showed a full compatibility of all four provers above 4 m3/h, whereas at lower flows some discrepancies occurred, whose causes are identified and discussed. The results confirm the improved accuracy of the large piston provers pioneered by both laboratories, if compared to bell provers, a couple of which were also tested.
Go to Download Page
Email Reference
Document ID: E0EED798

Water Flow Comparison Measurements Between Centro Nacional De Metrologia Mexico() And Physikalisch - Technische Bundesanstalt Germany() By Using A Cenam 100 mm Double-Turbine Meter Transfer Standard
Author(s): Dario Alejandro, Loza Guerrero, Rainer Engel
Abstract/Introduction:
This comparison test program 2004 CENAM - PTB was in order to assess the confidence and performance of the Mexican primary standard for liquid flow measurement. This program was an initiative from Centro Nacional de Metrologia (CENAM) in Mexico in collaboration with the Physikalisch-Technische Bundesanstalt (PTB) Liquid flow laboratory in Germany. We expected that the results differences between PTB and CENAM were inside CENAMs uncertainty limits, nevertheless test results from both laboratories, confirm that the measuring capabilities of meter 2 (MTR 2) are insufficient to accomplish the objective brought up above. Moreover, MTR 2 had a negative influence over the fluid velocity profile in configuration 1, causing an awkward response on the downstream turbine meter (MTR 1).
Go to Download Page
Email Reference
Document ID: D52D0BE8

Internal Comparison Between Mexicos Primary Standard For Water Flow Measurements And A Compact Prover Using A Turbine Meter Transfer Standard
Author(s): Jose Lara Manriquez, Emmanuel Rios Carrizalez And Dario A. Loza Guerrero
Abstract/Introduction:
This paper describes the results of an internal comparison between two calibration techniques - Mexicos primary standard for liquid flow measurements, based on the static weighing principle and a compact piston prover, volumetric method - using a turbine meter of 100 mm as transfer standard. The transfer standard was calibrated at several flow rates simultaneously for both the gravimetric and the volumetric methods. It is concluded from the results that the maximum difference between these two techniques was 0,05 % at 900 L/min.
Go to Download Page
Email Reference
Document ID: 950C028A

Intercomparison Of Bell Provers Using Sonic Nozzles
Author(s): Kyung-Am Park, Hae Man Choi, Jiunn-Haur Shaw Denis Boam Jiuan C. Gervacio
Abstract/Introduction:
An intercomparison exercise was conducted to confirm the uncertainty of gas flow measurement using bell provers. Four national metrology institutes (KRISS, NEL, CENAM and CMS) participated in the exercise, which was organized by KRISS. A sonic nozzle package was developed as a transfer standard to do the intercomparison. Three sonic nozzles were tested at stagnation pressures of 0.3, 0.4 and 0.5 MPa. The sonic nozzles were installed upstream of the bell prover using inlet and outlet tube sections provided with the nozzle holder. The discharge coefficients of the sonic nozzles were compared as a function of Reynolds number. At low flow rates the differences between results from individual laboratories and the average was small. However, at the highest flow rates, the differences were, in some cases, larger than the expected value from uncertainties claimed for the measurements.
Go to Download Page
Email Reference
Document ID: DB5DAB52

Models For Evaluating Flow Meter Inter-Comparisons
Author(s): Peter Lau
Abstract/Introduction:
Calibration inter-comparisons have long time been used to detect possible systematic differences between laboratories. In the flow area with a large variety of reference equipments and methods this is especially important. Traditionally inter-comparison results have predominantly been compared in a graphical manner. With the MRA focus is put on the Key Comparison Reference Value and the Degree of Equivalence as means to demonstrate a laboratories calibration capability. The paper describes critically suggested approaches how to achieve these measures, stresses on the need of corrections for deviating test conditions and ads alternatives like representative k-factors or using meter transfer functions.
Go to Download Page
Email Reference
Document ID: 19E0C7C5

The International BIPM/CIPM Key Comparison Reference Values For Compressed Air And Nitrogen.
Author(s): D. Dopheide, B. Mickan, R. Kramer, H-J Hotze
Abstract/Introduction:
Under auspices of the CIPM (International Conference for Weight and Measures), which is the highest metrological authority worldwide, so called Key Comparisons (KC) have been conducted to get international reference values for all quantities of interest. Among these KCs, the flow area is of economic importance and Key Comparisons for natural gas flow at high pressure and larger flow rates as well as for compressed air have been conducted successfully among all interested National Metrology Institutes.
Go to Download Page
Email Reference
Document ID: 23475967

The International BIPM/CIPM Key Comparison Reference Values For High Pressure Natural Gas
Author(s): D. Dopheide, B. Mickan, R. Kramer, H.-J. Hotze
Abstract/Introduction:
Under auspices of the CIPM (International Conference for Weight and Measures), which is the highest metrological authority worldwide, so called Key Comparisons (KC) have been conducted to get international reference values for all quantities of interest. Among these KCs, the flow area is of economic importance and Key Comparisons for natural gas flow at high pressure and larger flow rates as well as for compressed air have been conducted successfully among all interested National Metrology Institutes.
Go to Download Page
Email Reference
Document ID: AA43BAB1

Comparisons By Ptb, NIST, And Lne-Ladg In Air And Natural Gas With Critical Venturi Nozzles Agree Within 0.05 %
Author(s): B. Mickan, R. Kramer, D. Dopheide, A. Johnson, J. Wright
Abstract/Introduction:
The PTB, NIST, and LNE-LADG compared gas flow measurement standards in air and in natural gas in the pressure range from 0.9 to 42 bar using four critical nozzles with ISO standard toroidal shape 1. The four facilities generated calibrated flows through the four nozzles to determine curves of the discharge coefficient (Cd) versus Reynolds number (Re) spanning the range from 2 105 to 2 107 in Re. Three of the facilities used air the fourth facility used natural gas. The data show the laminar-to-turbulent transition at throat Reynolds numbers in the interval 1 106 to 2 106. All of the laboratories discharge coefficients agreed within 0.15 % or less, well within the uncertainty of the comparison. Comparing the average Cds for the four nozzles, all of the laboratories agree within 0.05 %.
Go to Download Page
Email Reference
Document ID: 6DE121C5

Ultrasonic Measurement Of Natural Gas To 10,000 PSI
Author(s): Mike Hopkins P.E.
Abstract/Introduction:
In 1996, much of the natural gas industry was just beginning to consider accepting ultrasonic meters for custody transfer measurement. At that time, the American Gas Association, Transmission Measurement Committee was writing what would eventually become AGA Report No. 9 (AGA-9), titled Measurement of Gas by Multipath Ultrasonic Meters, and published in June 1998. Technical information relating to the use of ultrasonic meter performance in 1996 consisted primarily of a limited number of magazine articles and published papers, some of which would later become part of AGA-9, primarily in Appendix C. Therefore, at that time, orifice meters or turbine meters appeared to be the only industryaccepted options since both meter types were covered by well-accepted, and welldocumented, gas industry measurement standards.
Go to Download Page
Email Reference
Document ID: 94399873

Performance Of 4-INCH Ultrasonic Meters In High Pressure Natural Gas Flow
Author(s): Robert Mcbrien, John Geerligs
Abstract/Introduction:
The results of testing performed on three different 4-inch multi-path ultrasonic flow meters (USMs) are reported. Tests were performed in high pressure natural gas flows over a pipe Reynolds number range of 8.5x105 to 1.0x107. Results were obtained with meters in fully developed flow and downstream of single elbow and two-elbows-out-of-plane installations, with and without flow conditioning. It was concluded that, where there is potential for other than fully developed flow conditions, flow conditioning should be used with 4-inch USMs to minimize the amount of metering error. Detailed description and discussion of the tests and results are presented in this paper.
Go to Download Page
Email Reference
Document ID: E43FD9A5

Laser 2 Focus Flow Optical Flow Meters For The Measurement Of Natural Gas
Author(s): Derek Montgomery, Casey Hodges
Abstract/Introduction:
Photon Control Inc. has developed a new class of optical flow meters based on the laser two focus method, for measuring natural gas in pipelines. Two laser beams, focused into the gas stream, detect the passage of microscopic particulate entrained in the flow. The optical signals are coupled by optical fibers to an electronic processing unit, which computes the velocity of the particulate using correlation analysis. The flow rate is calculated using calibration tables obtained from empirical testing against traceable reference standards.
Go to Download Page
Email Reference
Document ID: B028CBF0

Error Estimation In A Turbine Flowmeter Under Insufficient Straight Sections Of Pipe
Author(s): Josefina Ortiz m., Francisco Ramirez N., Edmundo Pedroza G.
Abstract/Introduction:
In agricultural pumping plants flowmeters can be installed with insufficient straight sections of pipe, both upstream and downstream from the meter. It has been found that flowmeter manufacturers are not aware of the error rate even though they know the correct position. In 1998, Hanson and Schwankl found that a propeller flowmeter had an error of 3.5% when it was installed a distance of two pipe diameters downstream from an elbow. The present study investigated the performance of a turbine flowmeter under non-optimal conditions (not considered by Hanson and Schwankl). Measurements were made in different flowmeter positions at various flow rates. Results from this project show that in any case the flowmeter doesnt have a large error rate. However, the lowest error rate is obtained when it is installed in the position recommended by the manufacturer.
Go to Download Page
Email Reference
Document ID: F1707856

Impulse Force Flow Meter - Functional Principle And First Results
Author(s): Helmut Tbben, Jrg Riedel
Abstract/Introduction:
A novel flow meter, for which a patent has been applied, will be presented at the symposium. By means of force measuring techniques, the mass flow rate can be determined. In the present contribution, the functional principle, first results having been achieved with a prototype as well as proposals for improvements are demonstrated.
Go to Download Page
Email Reference
Document ID: E45AA5F3

Testing The Wafer V-Cone Flowmeters In Accordance With API 22 Testing Protocol Section 2 - Differential Pressure Flow Measurement Devices In The CEESI Colorado Test Facility
Author(s): Dr R.J.W.Peters, Casey Hodges, Steve Caldwell
Abstract/Introduction:
The paper describes the testing of the Wafer V-Cone Meters in accordance with the new API 22.2 Testing Protocol for Differential Pressure Flow Measurement Devices in the Colorado Test Facility. The use of this new API standard and some of the points which had to be addressed in order to implement it are recorded. The results of testing 2 and 4 Wafer V-Cone meters in gas will be presented. The new nonstandard testing requirements in the API 22.2 will be discussed. The conclusions reached were: the results verify the conditioning effect of the V-Cone as it meters the fluid API 22.2 tests the claims of the manufacturer of the meter in a more demanding manner than API 5.7. The results of the Wafer V-Cone Testing Uncertainty are discussed.
Go to Download Page
Email Reference
Document ID: BF5C94B8

Liquid Property And Diameter Effects On Venturi Meters Used With Wet Gas Flows
Author(s): Richard Steven, Josh Kinney, Charlie Britton
Abstract/Introduction:
Wet gas flow metering technologies for the natural gas production industry are very important and their use is predicted to grow over the following decade. The state of the art of wet gas metering is steadily changing but it is generally true to say that the majority of the wet gas meter designs utilize Differential Pressure (DP) meter technology. Wet gas flows affect the gas flow predictions of all single phase gas flow meters but it has been found by experiment that DP meter designs are relatively reliable meters for wet gas flow applications. Whereas many gas meter designs fail to give any output or give erroneous unrepeatable gas flow rate predictions when a flow is a wet gas the DP meter type continues to operate and gives incorrect but repeatable gas flow rate predictions for wet gas flows. These incorrect (or apparent) gas flow rates are reported to always have a positive error and hence the liquid induced error on the gas flow rate prediction is commonly known as the DP meter wet gas flow over-reading.
Go to Download Page
Email Reference
Document ID: 6F696FC5

Wet Gas Flow Measurement With Ultrasonic And Differential Pressure Metering Technology
Author(s): Martin Van Werven, Jan Drenthen, Geeuwke De Boer, Martin Kurth
Abstract/Introduction:
Wet gas meters can offer significant cost savings by eliminating the need for separators and by allowing several fields to share common processing facilities. In allocation situations the main issue is one of measurement accuracy, either solely in the gas, or in both the gas and any condensate present. The development of wet gas meters that can satisfy the allocation accuracy requirements of the oil and gas companies is a key requirement for the industry. In well management applications wet gas meters offer continuous data output giving valuable information of the performance of wells.
Go to Download Page
Email Reference
Document ID: FA578880

Velocity Profile Effects On Multipath Ultrasonic Meters
Author(s): Gregor Brown, Don Augenstein, Terry Cousins
Abstract/Introduction:
Ultrasonic meters based on the transit time principle have a number of recognised benefits relative to more traditional methods of flow measurement. In recent years a new generation of ultrasonic meters have emerged and the technology is now being widely accepted for custody transfer applications in the oil and gas industry.
Go to Download Page
Email Reference
Document ID: ECE42402

Effect Of Surface Roughness Of Pipe Wall For Transit Time Ultrasonic Flowmeter
Author(s): Noriyuki Furuichi, Hiroshi Sato, Yoshiya Terao
Abstract/Introduction:
To investigate the effect of wall roughness for a transit-time ultrasonic flowmeter, velocity profile in a pipe was measured by the pulse Doppler method. Flowrate as measured by the transit-time ultrasonic flowmeter is affected by variation in roughness of the pipe wall, due to changes in the velocity profile.
Go to Download Page
Email Reference
Document ID: 27B9E888

The Performance Of Ultrasonic Flow Meters In The Presence Of Water-In-Oil
Author(s): T. Cousins, D. Augenstein, G. Brown, S. Eagle, D. Thearle
Abstract/Introduction:
Initial flow tests were carried out mainly with high proportions of water in order to verify meter operation. These tests showed that at higher velocities, the meter would still operate, although the accuracy could not be judged. At lower flow rates the water separated and formed a river along the pipe bottom.
Go to Download Page
Email Reference
Document ID: 2325847A

The Calibration, Proving And Validation Of Ultrasonic Flow Meters
Author(s): Klaus J. Zanker
Abstract/Introduction:
The gas industry has been spoiled by the Orifice Plate, which does not require flow calibration and the Dual Chamber Fitting, which allows field proving and validation. The USM construction and operation varies considerably, discouraging standardization and making flow calibration necessary. Liquid meters are commonly checked against volume provers, but they expect real time pulses, which are sometimes not available from the USM. Volume provers are difficult to use in gas flow, master metering offering a more practical approach. The USM produces an abundance of diagnostic data that could possibly be used to validate performance. These ideas are explored and developed further.
Go to Download Page
Email Reference
Document ID: 006A3D1C

Uncertainty In The Measurement Of Natural Gas Composition With On-Line Chromatography And Its Contribution To Total Uncertainty In The Measurement Of Energy Flow
Author(s): Ruben Moldes, Marcelo Moscoso
Abstract/Introduction:
Evaluation of the uncertainty of the composition of natural gas when on-line chromatography is used. Theoretical analysis to identify different uncertainty sources, estimation of sensibility of fluid compressibility with pressure and temperature at typical operation condition. Evaluation of minimum and maximum practical limits in heating value, specific gravity and fluid compressibility. Field round robin to estimate reproducibility of on-line chromatographs. Monte Carlo simulation to estimate uncertainty contribution in the total uncertainty of energy flow: it is almost double when Ultrasonic flowmeter is the primary meter compared with an orifice plate as primary meter.
Go to Download Page
Email Reference
Document ID: 3D62BDB0

How Todays Usm Diagnostics Solve Metering Problems
Author(s): Daniel Measurement
Abstract/Introduction:
This paper discusses both basic and advanced diagnostic features of gas ultrasonic meters (USM), and how capabilities built into todays electronics can identify problems that often may not have been identified in the past. It primarily discusses fiscal-quality, multi-path USMs and does not cover issues that may be different with non-fiscal meters. Although USMs basically work the same, the diagnostics for each manufacturer does vary. All brands provide basic features as discussed in AGA 9 Ref 1. However, some provide advanced features that can be used to help identify issues such as blocked flow conditioners and gas compositional errors. This paper is based upon the Daniel USM design and the information presented here may or may not be applicable to other manufacturers.
Go to Download Page
Email Reference
Document ID: DA5AD7F2

Flow Range Extension And Uncertainty Estimation Of Light Oil Flow Standard System With A Build-Up Technique
Author(s): Ki Won Lim
Abstract/Introduction:
The Light Oil Flow Standard System (LOFSS) in KRISS was designed as a national hydrocarbon flow standard system. In order to extend the flow capacity from 120 to 200, a build-up technique was applied with two positive displacement(PD) flowmeters. For an uncertainty analysis, the repeatability of the master flowmeters, the variation of the fluid density, and the pipe volume due to temperature change were scrutinized. A turbine flowmeter as a transfer package was tested with LOFSS and two master flowmeters for the feasibility of the uncertainty analysis. Both of the results showed good agreement, which means the build-up technique was carried out successfully.
Go to Download Page
Email Reference
Document ID: D54697FA

Inside Diameter Measurements For The Cylinder Of A 20 L Piston Prover
Author(s): Jesus J. Aguilera, T. T. Yeh, John D. Wright
Abstract/Introduction:
We made inside diameter measurements of a stainless steel cylinder used in a piston prover for hydrocarbon liquid flow measurements. We used a 3-point micrometer and two traceable setting rings to measure diameter at 12 radial positions at each of 21 lengthwise positions, on three different occasions. The diameter measurements showed a change in diameter (taper) of about 20 m near the middle of the cylinder and non-circularity at one end. These shape imperfections lead to a diameter uncertainty of 6 m (k2) for the cylinder in its present form. This amounts to 75 parts in 106 uncertainty in the cross sectional area or flow measurements. The cylinder will be re-honed to improve the uniformity of diameter along the length and chrome plated to increase the surface hardness (to prevent scratches). Based on the measurements presented herein, diameter uncertainty of 2 m or less appears attainable by this measurement approach (26 parts in 106 for flow).
Go to Download Page
Email Reference
Document ID: CFD094A5

Calibration Performance Of Small Volume Prover For Several Types Of Hydrocarbon Flowmeters
Author(s): Takashi Shimada, Ryouji Doihara, Yoshiya Terao, Masaki Takamoto
Abstract/Introduction:
Several kinds of flowmeters, namely, Coriolis flowmeters, turbine meters, ultrasonic flowmeters, vortex-shedding flowmeters and positive displacement flowmeters, have been calibrated using the primary standard in Japan, which is based on static and gravimetric methods with a flying start and finish, and a small volume prover (SVP) in order to investigate the effect of the sudden pressure change caused by the proving action of the SVP. Furthermore, the relationship between the type of flowmeter and the additional uncertainty due to the SVP has been discussed in detailed on the basis of calibration results.
Go to Download Page
Email Reference
Document ID: 4046BCBC

Design Of A New Calibration Facility For Hydrocarbon Flow At 0.1 To 15 m3/h
Author(s): Ryouji Doihara, Takashi Shimada, Yoshiya Terao, Masaki Takamoto
Abstract/Introduction:
The NMIJ has established a national primary standard facility for large oil flow rates (3 - 300 m3/h). Currently, another calibration line is being constructed in order to extend the calibration range to smaller flow rates, below 0.1 m3/h. This middle oil flow primary standard facility applies a static and gravimetric method with flying start and finish. A rotating double wing diverter (RDWD) and improved weighing tank system have been developed for this new facility. Experiments have been carried out on a small prototype rig (water) in advance of constructing the new oil facility. After adjustment of the start and stop signal timings, the diverter timing error was certain to fall within 1 ms.
Go to Download Page
Email Reference
Document ID: 2E058955

Determination Of Discharge Coefficient Of Critical Nozzles Based On Their Geometry And The Theory Of Laminar And Turbulent Boundary Layers
Author(s): B. Mickan, R. Kramer, D. Dophaido
Abstract/Introduction:
THE EXISTING THEORIES FOR LAMINAR BOUNDARY LAYERS AND THE INCLINATION OF ISOTCHE LINES INSIDE OF CRITICAL NOZZLES NOWADAYS ALLOW A RELIABLE CALCULATION OF THE DISCHARGE CO-EFFICIENTS OF TORIODAL SHAPED NOZZLES BASED ON THEIR MEASURED GEOMETRY FOR THE OPERATION WITH LAMINAR BOUNDARY LAYERS
Go to Download Page
Email Reference
Document ID: 7A092F00

Evaluation Of Theoretical Cfv Flow Models In The Laminar, Turbulent, And Transition Flow Regimes
Author(s): Aaron Johnson, John Wright
Abstract/Introduction:
State of the art dimensional metrology was used to measure the throat diameter and throat curvature of nine critical flow venturis (CFVs) with nominal throat diameters ranging from 5 mm to 25 mm. The throat curvature was used calculate the theoretical discharge coefficients while the throat diameter was used to compute the experimental discharge coefficients. The nine CFVs were calibrated in dry air using two NIST primary flow standards with expanded uncertainties of 0.05 % and 0.09 %, respectively. The calibration data spans a Reynolds number range from 7.2 104 to 2.5 106, including laminar, turbulent, and transition flow regimes. By correcting for both the throat diameter and curvature, the agreement between predicted and measured discharge coefficients was less than 0.17 % in the turbulent regime and less than 0.07 % in the laminar regime.
Go to Download Page
Email Reference
Document ID: BA6BE8CA

The Use Of Micro-Nozzles Under Sonic And Subsonic Conditions With Various Gases
Author(s): B. Mickan, R. Kramer, D. Dophaido
Abstract/Introduction:
THE USE OF CRITICAL NOZZLES AS SECONDARY FLOW STANDARDS AND TRANSFER STANDARDS BECAME MORE AND MORE IMPORTANT IN THE RECENT YEARS.
Go to Download Page
Email Reference
Document ID: 0EF5535B

A Translating Rig For Low-Speed Anemometer Testing
Author(s): Aline Piccato, Riccardo Malvano, Pier Giorgio Spazzini
Abstract/Introduction:
The low air-speed primary standard of INRIM called Towing Tank is described and its measurement uncertainty is discussed, giving details about its measurement chain. The Towing Tank is a 30 m long facility, whose useful measurement range is 0.05 m/s U 1.20 m/s, with a best measurement uncertainty equal to 2% at the maximum speed. This facility is complementary to the companion test device (the Rotating Arm 1), whose motion is rotational and continuous, while in the Towing Tank motion is linear, but limited in time. In both facilities speed measurement is traceable to length and time primary standards
Go to Download Page
Email Reference
Document ID: C48DF0AB

A Rotating Rig For Low-Speed Anemometer Testing
Author(s): Riccardo Malvano, Pier Giorgio Spazzini, Aline Piccato
Abstract/Introduction:
The present paper is aimed at describing the main low-speed anemometric test device of INRIM. This device consists of a 3.5 m long rotating arm which carries the instrument to be tested or calibrated. Details about the test rig device are provided together with a discussion about uncertainty sources and uncertainty evaluation.
Go to Download Page
Email Reference
Document ID: 2D9E6846

Diverter Uncertainty Less Than 0.01% For Water Flow Calibrations
Author(s): Iryna Marfenko, T.T. Yeh, John Wright
Abstract/Introduction:
We describe the primary water flow standard at NIST and tests that determined the uncertainty of a new diverter design called a uni-diverter. The tests compared the flow measurement results for the uni-diverter (operated with the collection / bypass unit) and the diverter operated in the traditional manner with a time correction (as described in ISO 4185). The uni-diverter introduces less than 0.01% into the total flow uncertainty.
Go to Download Page
Email Reference
Document ID: CF7E1790

Triple Point: A Unique Calibration Facility For Turbine Gas Meters
Author(s): Larry Fraser
Abstract/Introduction:
Terasen Gas Inc. has constructed a facility for calibrating turbine gas meters that uses carbon dioxide as a test medium. Carbon dioxide is heavier than most fluids used to calibrate gas meters and so permits operation over a wide range of Reynolds numbers and densities at relatively low pressures. Moreover, the phase behaviour of carbon dioxide permits the temperature in the test loop to be controlled by direct injection of liquid carbon dioxide. In addition, it has been demonstrated that the facility can operate on air, thus enabling direct comparison of meter calibrations obtained on two fluids.
Go to Download Page
Email Reference
Document ID: CAA83FEA

Comparisons Of Viscous-Sealed Provers With Lne And Studies Of Piston - Cylinder Leakage
Author(s): Harvey Padden
Abstract/Introduction:
Dry piston provers (DryCals) are used for high-precision primary calibration of transfer standards and MFCs. In the course of our NVLAP ISO 17025 accreditation1, a large amount of validation had been done at flows above 1000 sccm. However, we desired additional validation at low flows.
Go to Download Page
Email Reference
Document ID: D7CB1CDF

A Primary Calibration System For The Support Of High Performance Gas Flow Transfer Standards
Author(s): P. Delajoud, m. Bair, C. Rombouts, m. Girard
Abstract/Introduction:
The practical application of high performance gas flow transfer standards requires means to efficiently and reliably calibrate them with very low measurement uncertainty. A unique primary gas flow calibration system has been developed to support gas flow transfer standards based on laminar flow and sonic nozzle based elements in the range of 2 x 10-5 gs-1 (1 Ncc min-1) to 50 gs-1 (2500 Nlmin-1). The primary calibration system is made up a gravimetric standard up to 0.2 gs-1 (10 Nl min-1) and a group of sonic nozzle based flow elements to extend the gravimetric reference flow measurements to higher values. Any non-corrosive gas can be studied. The gravimetric system measures depleted gas mass real time using a force balanced load cell and an automated taring system to eliminate force measurement drift over time. The group of nozzles uses the extensive nature of flow and a successive addition technique to build up the gravimetric reference measurements to higher flows. The primary calibration system supports a calibration chain made up of a group of laminar flow and sonic nozzle based elements characterized with a variety of gases. The structure of the group allows verification of the coherence of flow measurements up and down the chain and the precision of the chain to be quantified and maintained over time. The calibration chain is used for day to day calibration of gas flow transfer standards. A complete uncertainty analysis for the primary gravimetric system and the calibration chain has been performed.
Go to Download Page
Email Reference
Document ID: 808EB3AE

Construction Of A Re-Circulation Type High-Pressure Air Flow Measurement Facility And Uncertainty Evaluation
Author(s): Jiunn-Haur Shaw, Fong-Ruey Yang, Yi-Lin Ho
Abstract/Introduction:
Increasing conscience on environmental protection has prompted the use of cleaner energy, such as natural gas in stead of fuel oil or coal, for electric power generation. Taiwan has over 15 years history in importing natural gas from Indonesia and distributing it for household gas burner as well as two major electrical power plants. The sizes of the gas flow measurement device range from small diaphragm gas meters to large capacity, 600-750mm orifices or ultrasound meters. To promote fair trade and safeguard Taiwans legal metrology infrastructure in gas flow measurement, the National Measurement Institute constructed a gravimetric type air-flow calibration system with sonic nozzle array as the working standard in mid-90s. The flow measurement capacity at 1 bar and 23 ? is from 15 m3/h to 20000 m3/h with operating pressure range from 1 bar to 60 bar. Apparently, the flow capacity does not meet the calibration need from large capacity flow meters. Due to such limitation at NMI, and to disseminate measurement traceability to Taiwan Power Company for custody transfer, a larger flow calibration station was constructed at Chinese Petroleum Corp in southern part of Taiwan. The pressure range is also from 1 bar to 60 bars with flow range from 20 m3/h to 4000 m3/h under actual pressure. Based on a re-circulating principle and flow driven by three sets of Howden high pressure blower, the facility set up the traceability with NMI through two sets of 100mm Instromet Rotary Piston Prover(IRPP). Various sets of turbine meters ranging from 100 mm to 200 mm are used in parallel as working standards and to cascade up the flow rate to 4000 m3/h. The largest meter size to be calibrated is 300 mm at present and 400 mm is expected. The calibration system consists of three flow loops with the expanded uncertainties at 95% confidence level at 0.24%, 0.34% and 0.42% respectively. This paper describes the design features, operation equipments, control logic and instrumentation, and uncertainty evaluation of this flow facility. An international proficiency testing with Korea Gas Company is planned for year 2006, and optical based flow measurement technology is under development with assistance from CMS.
Go to Download Page
Email Reference
Document ID: 36736603

A New Primary Gas Flow Standard For Flow Rate Measurement From 0.001 To 1000 Nano mol/S
Author(s): Wu Jian, Chua Hock Ann
Abstract/Introduction:
A primary gas flow standard was designed and developed in SPRING for the calibration of gas flow meters or leak rate standard at nano-flow rate range of 10-6 to 10-12 mol/s. It is also used to generate very low gas flow for SPRINGs continuous expansion vacuum primary standard. The primary standard consists of two piston-cylinder flow units with diameters of 10mm and 25mm, which provide wider measurement ranges. A stepping motor is used to drive the piston moving at precise speed via a digital micrometer. Two methods, constant pressure method and constant volume method12, are used for the measurement of flow rate. Flow meters can be calibrated automatically using a software program developed at SPRING. The expanded measurement uncertainty of the primary gas flow standard is estimated to be 1% of the readings.
Go to Download Page
Email Reference
Document ID: 3C59C87B

Simplifying Industry Standard Flow Measurement Equations, Fluid Property Calculations And Uncertainty Estimates
Author(s): Michael Boyd
Abstract/Introduction:
In recent years, the number and complexity of industry standards relating to flow measurement has continued to increase relentlessly - for example the AGA Report Number 10 contains some 120 pages of C++ code required to calculate the speed of sound of natural gas mixtures. These standards are typically intended for fiscal quality applications, and as such impose calculation overheads that mean they are not always suitable for PLCs / DCSs or petroleum field allocation models. Such standards require an investment of time to understand and implement that is not available to many Engineers
Go to Download Page
Email Reference
Document ID: D4314A3D

Energy Metering At The Flow Measurement Station
Author(s): Reed Mitchell, Barry Balzer, Paul Kizer
Abstract/Introduction:
A grass-roots gas Chromatograph that can combine gas flow measurement with analysis so that it is possible to output total energy from the single box is described. This design provides an on-line GC in a small, field worthy explosion proof package that provides lab quality analysis. The built-in GC controller has a powerful 32-bit microprocessor running a Windows CE based operating system. Data from the unmatched repeatability will be presented. The software allows simultaneous gas flow measurement while the unit is processing Chromatographic analysis. The register based data storage and calculations allow the flow and heating value data to be combined for use as a direct energy meter.
Go to Download Page
Email Reference
Document ID: C01610D9

Theoretical And Experimental Investigations Of Rotary Piston Flow Meters
Author(s): E. Von Lavante, U. Banaszak, V. Ltz-Dauer, K. Enste, J. Bergervoet, H. Dietrich
Abstract/Introduction:
Rotary piston meters have been demonstrated as stable and reliable flow meters, having a large operating range. Their favourable properties render them suitable for usage as reference meters. Their main working principles have been known for some time. However, irregular behaviour of the deviation between the indicated and true flow rate has been occasionally observed during their operation. The goal of the present study was to present an unified theory explaining the generation of leakage flow, responsible for a major portion of the deviation. To this end, an equation considering most of the leakage effects has been developed. Several experiments aimed at supporting the theory have been carried out. It has been demonstrated that the theory is qualitatively correct and incorporates most of the important characteristics of the rotary piston flow meters.
Go to Download Page
Email Reference
Document ID: E30869CE

The Estimation Of Measurement Uncertainty Of New Calibration Facilities For Waterflow And The Results Of Internal Comparison In Nmij
Author(s): Hiroshi Sato, Yoshiya Terao, Masaki Takamoto
Abstract/Introduction:
NMIJ has been providing calibration service for water flowmeter at flow rate from 50 up to 3000 m3/h by using 50 t weighing tank system for many years. This service has already been registered at BIPM MRA Appendix C.
Go to Download Page
Email Reference
Document ID: 519572D3

A New Low Pressure Calibration Facility Using 8-PATH Ultrasonic Meters As Working Standards
Author(s): Dr. Volker Herrmann, Dr. Matthias Wehmeier, Toralf Dietz, Dr. Rainer Kramer, Dr. Bodo Mickan
Abstract/Introduction:
This report describes a completely new designed low-pressure test facility for flow rates from 10 to 10,000 cubic metres per hour m/h. In order to ensure the lowest possible measuring uncertainty, both thermal and flow properties of the test stand were optimised. Special 8-path ultrasonic meters with a very large measuring range and integrated diagnostic capabilities are used as measuring (working) standards.
Go to Download Page
Email Reference
Document ID: 0F685D83

A Novel Concept Of A Gas Flow Calibration Rig
Author(s): A. Pfau, J. Honegger, H. Hafelfinger, m. Carr
Abstract/Introduction:
Within a project period of one and a half years a new calibration facility at Endress+Hauser Flowtecs production centre in Reinach, Switzerland, was developed, built and commissioned. The first run with a meter under test happened in May 2005. In January 2006, the facility was accredited according to ISO/IEC 17025 to within 0.3% best uncertainty covering a range of air mass flow from 0.05 to 10000kg/h. The stated objective was to provide a facility capable of operating to very high levels of accuracy over a wide range of flows. To achieve this, a combination of rotary piston gas counters, turbines and critical nozzles were chosen for the reference meters, all operating at near atmospheric conditions.
Go to Download Page
Email Reference
Document ID: 78466664

Process Variable Stability, Data Processing And Installation And Environmental Influences During Ultrasonic Meter Calibration
Author(s): Jairo Mantilla, Wayne Haner
Abstract/Introduction:
This paper is a synopsis of TCCs data acquisition methodology, processing protocols and supervisory system means, reference values and integrity procedures, presented together with a detailed description of the stability requirement of pressure, temperature and flows at the laboratory during ultrasonic meter calibration to minimise Type A contributions. TCC calibration data, including repeatability and shortterm reproducibility values, taken from a meter calibration will be presented and used as baseline to illustrate the sensitivity of USM response to environmental and calibration variables. Monte Carlo simulation modeling will be used as replication tool. To provide a real-time illustration of TCCs operational capability, at the end of the presentation of this paper a calibrated meter will be verified live remotely by running the calibration laboratory from the conference room.
Go to Download Page
Email Reference
Document ID: 33975E57

High Accuracy Measurements Of Natural Gas-Like Densities Using A Single-Sinker Magnetic Suspension Densitometer And A Validity Check Of AGA8-92DC
Author(s): Mert Atilhan, Saquib Ejaz, Prashant Patil, Kenneth R. Hall
Abstract/Introduction:
Accurate knowledge of the PVT behavior of natural gas at reservoir and pipeline conditions is necessary for several practical applications. The most important economic concerns relate to the custody transfer of natural gas. Furthermore, high-accuracy density and phase change data can validate equations of state (EOS). Development of AGA8-DC92, currently the industry standard for natural gas custody transfer, used a round robin project (RR) and a databank covering gas mixtures with compositions containing up to 0.2 mole percent of the C6+ fraction. Hopefully, AGA8-DC92 could predict densities of natural gas mixtures containing higher mole percentages of C6+ with the same accuracy. A need exists for high-accuracy density data to check the validity of AGA8-DC92 over the entire range of pressure, temperature and composition encountered in custody transfer. It is also important to compare the RR and earlier databank for AGA8-DC92 EOS to the latest natural gas data measurements. For these reasons, we have used data on two different natural gas samples, RR-NIST2 and M91C1, to discover the differences that heavy components make on both predicted and measured phase envelopes. We have used a state-of-the-art, high pressure, high temperature, compact single-sinker magnetic suspension densitometer to collect densities of a simulated natural gas mixture over the range (270 to 340) K (26.33 to152.33) oF, (-3.15 to 66.85) oC and (3.447 to 34.474) MPa (500 to 5,000) psia. These data can serve as a test for the validity of AGA8-DC92 or any other EOS.
Go to Download Page
Email Reference
Document ID: 35A5C9A1

Fluid Composition Analysis By Multiple Gamma-Ray Beam And Modality Measurements
Author(s): Geir Anton Johansen, Stein-Arild Tjugum
Abstract/Introduction:
The feasibility of using multiple gamma-ray beams to identify the cross sectional flow pattern has been demonstrated for three phase gas/ oil/ water pipe flow. A fan beam collimated 241Am source with principal emission at 59.5 keV is used because this relatively low energy enables efficient collimation and thereby shaping of the beams, as well as compact detectors. Results from a flow loop experiment using a system two transmission detectors and two scatter detectors are presented. The scatter response from the fluid is used to compensate for changes in salinity of the water component.
Go to Download Page
Email Reference
Document ID: 53271218

Development Of A Comprehensive And User-Friendly Fluid Meter Uncertainty Calculator
Author(s): Casey Hodges
Abstract/Introduction:
While the primary objective of any flow metering system is to obtain accurate flow measurement, the quantification of measurement uncertainty has historically required the use of theoretical, complex, and time consuming calculations. This has been a major impediment to the widespread application of uncertainty determination in the field.
Go to Download Page
Email Reference
Document ID: D05BE832

Traceability Chain And Uncertainty Budget Calculation For The French High Pressure Gas Flow Measurement Laboratory Lne-Ladg And The Impact Of The Harmonization Process With Ptb-Pigsar And Nmi-Vsl On This Calculation
Author(s): Jean-Pierre Vallet, Christophe Windenberger
Abstract/Introduction:
This paper describes the uncertainty budget calculation for the LNE-LADG which is the French MNI for the high pressure gas flow measurement. The LNE-LADG uses critical flow Venturi flow nozzles (sonic nozzles) as transfer standards for more than 25 years. These nozzles used to generate and measure the reference flow on calibration and testing operating as well as in air and in natural gas. The full tractability chain from mass and time up to any calibration of gas meters are explain as well as the uncertainty analysis. All the input variables and their standard uncertainty are listed and given. There are evaluated to either type A or B method presented in the GUM. All the calculation is conform to EA-4/02 requirements. This paper presents the different calculation models which can occur when using sonic nozzles (primary or secondary (comparison) calibration, different nature of gas (air or natural gas))
Go to Download Page
Email Reference
Document ID: 558FD8AC

Uncertainty In The Measurement Of Natural Gas Composition With On-Line Chromatography And Its Contribution To Total Uncertainty In The Measurement Of Energy Flow
Author(s): Ruben Moldes, Marcelo Moscoso
Abstract/Introduction:
Evaluation of the uncertainty of the composition of natural gas when on-line chromatography is used. Theoretical analysis to identify different uncertainty sources, estimation of sensibility of fluid compressibility with pressure and temperature at typical operation condition. Evaluation of minimum and maximum practical limits in heating value, specific gravity and fluid compressibility. Field round robin to estimate reproducibility of on-line chromatographs. Monte Carlo simulation to estimate uncertainty contribution in the total uncertainty of energy flow: it is almost double when Ultrasonic flowmeter is the primary meter compared with an orifice plate as primary meter
Go to Download Page
Email Reference
Document ID: E560731B

Revised Uncertainty Analysis Of NIST 26 m3 Pvtt Flow Standard
Author(s): Aaron Johnson, John Wright
Abstract/Introduction:
We re-analyze the uncertainty of NISTs 26 m3 pressure, volume, temperature, and time (PVTt) primary flow standard used to calibrate gas flow meters in the range 200 L/min to 77,000 L/min.1 During the past few years, this standard has been upgraded by improving measurements of gas temperatures and vacuum conditions within the collection tank, implementing a symmetric diversion process, and fully automating the calibration process. The upgrades reduced the expanded uncertainty of this standard, based on the GUM procedure, from 0.2 % to 0.09 % at the ninety-five percent confidence level (i.e., k 2) over the full flow range.
Go to Download Page
Email Reference
Document ID: 2AC5D91D

Significance Of Gas Ultrasonic Meter Pressure Effects
Author(s): Charles W. Derr
Abstract/Introduction:
Pressure changes the ultrasonic meter bodys longitudinal and radial dimensions and the geometric parameters that go into the calculation of flow. These can be calculated or experimentally determined and corrections applied. The degree of pressure related uncertainty varies as a function of the meter body mass, alloy choice and transducer mounting technique. In general, Ultrasonic Meters have reasonably low Pressure Coefficients.
Go to Download Page
Email Reference
Document ID: 171E67BC


Copyright © 2017