Measurement Techniques, Vol. 37, No. 9, 1994

MECHANICAL MEASUREMENTS

STANDARDIZATION OF FLOWMETERS

WITH CONSTRICTORS

P . P. Kremlevskii, N. K. Vasil'ev, and V. A. Kuz'min

UDC 681.121..8.089.6

Shortcomings of Regulations RD 50-213-80 are noted, recommendations are made for eliminating them, and a suitable structure and content for new regulations are considered.

Flowmeters with constrictors are the principal instruments for measuring the flow rate of a liquid, gas, or steam, in in particular for technological process control. They are also used extensively to measure amounts (mass and volume) of substances in pipelines, especially gas and steam, for commercial calculations. In view of this, it is very important to have correctly and clearly formulated standards documentation, which regulates the application and rules of flowmeter operation for the maximum possible accuracy of measurement. The current Regulations RD 50-213-80 have long needed a considerable reworking. The first version of new regulations with the emphasis on measurement of the flow rate and amount of natural gas was drafted by Gazpriboravtomatika (Moscow). These draft regulations were reworked at the VINIIR (All-Union Scientific-Research, Design-Construction, and Technological Institute of Relays, Kazan) in 1993. The final editing of the draft was done at the D. I. Mendeleev VNIIM (All-Union Scientific-Research Institute of Metrology, St. Petersburg). Early in 1994 the draft was sent to a number of organizations for discussion. In this article we note the principal shortcomings of Regulations RD 50-213-80, make recommendations for eliminating those shortcomings, and consider a suitable structure and content of the new regulations. The major shortcomings of Regulations RD 50-213-80 are: 1. The text proper of the Regulations accounts for only 70 pages of the very large document (320 pp.). Three-quarters of the document consists of appendices in the form of tables, some of which have been revised while some others are not sufficiently accurate or are already outdated. 2. The overall composition of Regulations RD-213-80 and the order of presentation of the material, in our opinion, are substantially inferior to all preceding editions. As an example, let us refer to the f'LrStchapter, "Fundamental Flow Rate Equations." It considers various subjects that are not logically related. It does not give the fundamental flow rate equation with a coefficient of 0.01252, but does give particular cases of that equation with coefficients of 0.2109 and 3.553, which are valid only for gases. Then, in contradiction to the chapter heading, the conditions for cavity-free flow of a liquid are considered and a table of values is given for the relative solubility of a gas in a liquid. 3. The method of calculation that is given is exceedingly complicated, especially for processing daily diagrams of the pressure drop and for determining the flow-rate coefficient that corresponds to the true Reynolds number. 4. The treatment of the allowable length of straight portions before the constrictor, which is a very important matter in practice, raises objections. Some of the requirements lead to a paradoxical result. For example, if a gate valve is installed ahead of the diaphragm at a distance 10D (where D is the diameter of the pipeline) and a group of rings is installed in various planes ahead of the gate valve at a distance of 25D, there is no such length. But if another gate valve is installed between the gate valve and the group of rings at a distance of 10D from the first gate valve, the length meets the requirements of the regulations. Furthermore, the requirements for lengths of straight portions before Venturi tubes are much more stringent than those in the international standard ISO 5167, thus greatly hindering the application of these modern constrictors. 5. In the Regulations the root-mean-square error of Venturi tubes is erroneously doubled relative to the international standard ISO 5167.

Translated from Izmeritel'naya Tekhnika, No. 9, pp. 40-41, September, 1994.

1042

0543-1972/94/3709-1042512.50

9

Plenum Publishing Corporation

6. The Regulations wrongly suggest that the flow-rate coefficient of Venmri nozzles can be assumed to be equal to that of a nozzle. They are not equal because the exit pressure is sampled at different places. The standard ISO 5167 gives a special formula for determining the flow-rate coefficient of Venturi nozzles and at the same time their range of application is limited to Reynolds from 1.5-l0 s to 2.106. 7. The correction factor Kb for the leading edge blunting of the diaphragm does not correspond to the actual blnntin,g~ of the edge. If, for example, I~ = 1.015, i.e., then in the first few months of operation when the diaphragm is still ~harp, we introduce a systematic positive error of 1.5% as a result of flow-rate measurement. 8. No recommendations are made for a rational selection of the final value of the relative area m of the diaphragm or nozzle. As a result, computation of the diameter of the diaphragm or nozzle orifice takes much longer. 9. The reliability and validity of the recommendations and requirements in RD 50o213-80 that do not appear in the standard ISO 5167 and Regulations 28.64 cannot be evaluated because there are no references to the sources of those recommendations and requirements. Besides those mentioned above there are other, secondary shortcomings and inaccuracies. Becoming acquainted with Regulations RD 50-213-80, and all the more mastering them, requires much more time than did previous editions. The first version of the new Regulations, by decision of a meeting held at VNIIM in 1992 with the participation of representatives of VNIIR and the gas industry, was based on the international standard ISO 5167. The first ten sections of the new Regulations, which were drawn up at Gazpriboravtomatika, therefore, are almost a word-for-word translation of ISO 5167. Slight revisions and supplements were introduced into those sections at VNIIM. The principle of measurement was considered and an initial formula was obtained for the flow rate. Particularly formalized definitions of the flow-rate coefficient c~, the discharge coefficient C, the entry velocity coefficient E, and the expansion coefficient t, have been replaced by definitions that reveal their physical meaning. Moreover, large-radius nozzles, which are not used in our country, have been excluded. Unfortunately, ISO 5167 gives exhaustive information about constrictors but the other parts of flowmeters are discussed very little. Some new sections of the Regulations deal with those parts. The first section considers the instruments and conditions for measuring pressure and pressure drop, temperature, and density, as well as the requirements for connecting lines and auxiliary devices on them. The version from Gazpriboravtomatika has been retained virtually unchanged. The next section, dealing with calculation, was worked out at VNIIM. First the main computational formulas are given with the necessary interpretation of their numerical coefficients. Redundant formulas that would only encumber the Regulations have been omitted. In agreement with VNIIR, all the computational formulas retain the flow-rate coefficient ~ generally used in our country and, moreover, the relative area m and the product m~ instead of the complicated product 82, especially since the dependence of ~ = EC on B = d/D is less linear than the dependence on m = (d/D) 2. Further on the same section gives recommendations for the selection of type of constrictor, its relative area m, and the maximum pressure drop; it is proposed that in calculations for differential manometers with percentage scales the choice of the maximum pressure drop not be associated with the standard series of digital scales. This makes it possible to increase the accuracy of measurement. The recommended computational procedure, both direct (determination of the constrictor orifice diameters) and reverse (determination of the flow rate), contains much that is new. A formula that simplifies the iteration process of selecting the product m~ has been obtained for direct calculation while for calculation of diaphragms with angular tapping in tubes with D > 300 mm a table gives the dependence of flow-rate coefficients on the product m,~ for a number of values of the Reynolds number Re and m 2. This table greatly simplifies diaphragm calculations. A formula obtained for the reverse calculation of diaphragms also makes it substantially simpler to determine the actual values of the Re number and the coefficient ~,. In the section "Errors of Measurement ~ inaccuracies due to inclusion of the errors of measurement of d and D have been removed and the subsection dealing with errors of measurement of the mass and volume of the material have been enlarged. The VNIIR version of the section concerned with checking of constrictors and metrological operation has been retained and supplemented with recommended schedules for checking constrictors. Ten appendices are an integral part of the first version of the Regulations. The first appendix contains 13 tables: of these seven give values of the discharge coefficients; two tables give expansion coefficients; one reflects the dependence of ~ on m , for diaphragms with D > 300 mm, one reflects the dependence of m on mo~ for wear-resistance diaphragms, and two reflect the properties of the materials. Appendix 2 contains instructions for determining the correction factor k~ for the roughness pipeline and Appendix 3, for determining the correction factor kb for the leading-edge blunting and, moreover, it gives a procedure for calculations for a wear-resistant diaphragm with a forward-blunted edge. 1043

Appendix 4 contains formulas derived at VNIIR for determining the additional error for the length of straight portions of pipes while Appendix 5 gives the characteristics of local resistances. Appendix 6 contains an algorithm for calculating the flow rate of a dry gas and Appendix 7, a procedure for machining diaphragms with planimeters developed at VNIIR. Appendix 8 discusses the possible application of constrictors to measure the flow rates of some two-phase substances. Appendix 9 contains the form for the technical specifications of a flowmeter and Appendix 10, the list of references cited in the text. In conclusion, we point out that the first version of the new Regulations has been drawn up for measurement of the flow rate of natural gas, but is also fully applicable to the measurement of the flow rate of steam, liquid, and other gases when supplemented with tables or formulas for the densities of those substances, as well as some of their thermophysical properties (viscosity, adiabatic exponent, and compressibility). In our opinion, the final version of the Regulations should contain that information so that the Regulations for Flow-Rate Measurement would be universal. It is unacceptable and wasteful to produce separate regulations for measuring specific substances. The flow rates of gas, steam, and liquid must be measured in most enterprises. Please send comments and proposals for the new Regulations to: 198005, St. Petersburg, Moscow Prospekt 19, NPO VNIIMM imeni D. I. Mendeleeva (All-Union Scientific-Research Institute of Metrology). Attention: Director V. S. Aleksandrovich.

1044