Abstract
The object of this investigation was to compare the relative merits of eccentric and segmental orifice plates as a flow measurement device, and to collect sufficient experimental data to enable a national standard to be prepared. In view of the statistical aspect of the work the experimental programme had to be extensive. Some 600 flow tests involving over 1500 differential pressure readings were carried out using incompressible fluids. A number of pitot traverses and friction factor tests were made to augment the discharge coefficient data on secondary influences. Two computer programmes were devised to facilitate the processing and interpolation of the experimental data obtained. Basic discharge coefficient values were obtained for pipes of 6", 8" and 12" in diameter. In the case of eccentric orifices these values are in good agreement with the corresponding data obtained by other researchers (Beitler, West). Segmental orifices data, on the other hand, though close to Witte's results showed on average a 3% discrepancy from Lohman's and nearly 8% from Beitler's results. The effect of pipe roughness and asymmetrical velocity distribution on the value of the discharge coefficients were investigated, and found to be of the same order as in the concentric orifice plates. The major part of the investigation (over 50 pages) deals with the influence of pipe roughness on the discharge coefficients of segmental orifice plates. This aspect of flow measurement, although recognised now as of considerable importance, has not been studied extensively partly because of the complex experimental technique required to produce reliable results on naturally roughened pipes. This technique is discussed in detail, and a computer programme devised to prepare comprehensive tables of correction factors. A comparison of the two types of devices tested shows that the eccentric orifice plate:- 1. is easier to manufacture (and consequently less likely to have manufacturing errors); 2. has a more constant discharge coefficient as a function of area ratio (only 5% variation compared with over 10% for segmental); 3. has a more constant discharge coefficient as a function of pipe size (less than 1/2% variation compared with 3-4% for segmental), particularly at very low and very high area ratios. 4. provides less effective passage for solids in suspension, where errors are likely due to accumulation of material. For these reasons, eccentric orifice plates give more accurate results in most industrial applications, except in those cases where the size of particles in suspension, or extremely low velocity of flow, are likely to cause build up of material on the front face of the plate. With the addition of comprehensive data from this investigation it will now be possible to draft both a national and an international standard on the use of eccentric orifice plates for flow measurement. However, there is as yet no comparable confidence in the validity of the data on segmental orifice plates. More work will have to be completed before a reliable standard could be drafted. It is hoped, above all, that this work, in addition to providing useful data, has thrown new light on this neglected aspect of flow measurement.