There are numerous design criteria available for gravel-packs based on the well-known gravel-to-sand grain size ratio. These rules are mainly derived from the interpretation of compatibility experiments. The favourable gravel-to-sand ratios are deter-mined within the limits set by successful and unsuccessful tests. Because of empirical approach, various influences have to be considered due to different test setup and evaluation of experi-ments. In this paper, the types of design criteria, the similarities and differences between the various methods, the uncertainties in the use of these design criteria, and the gaps in current knowl-edge are described. Applicability of these purely geometrical methods under fluctuating flow, methods relating to sand reten-tion, permeability impairment, clogging and acceptable quantity of sand within a certain depth of penetration into the gravel are discussed. IntroductionSince gravel-packs are recognized as the most popular sand prevention method. numerous gravel-pack design criteria have been developed roughly a quarter of a century a@,o. These methods have ranged from using simple geometrical approach on easily measurable sand and gravel properties to performing filtration tests with representative reservoir sand and the proposed gravel-pack. The latter have generally been regarded as being better indi-cators of future gravel-pack performance. However, as sand parti-cles of reservoirs are structured at random, it is almost impossible to describe interaction between sand and -ravel theoretically in a rigorous manner. All existing gravel-pack design criteria are more or less based on laboratory tests with defined sLL[id-gravel combi-nations and can therefore onlv be applied in connection with the respective combinations and test conditions. However, despite careful application of these rutes, hiah pressure drop and reduc-tion in production rate are still associated with this exclusion tech-niclue. Consequently, a gravel-pack seems to change its structure, permeability and pressure differential over the course of a filtra-tion period. Thus, there is a certain unpredictability about the process ot' sand exclusion due to the inability of t'ully describing the mechanisms and the properties of ,t-avel-sand/fluids system and of relating properties and mechanistils to the performance of materials. So far, there is not much known @lbout the reliability of gr@ivel-packs that are designed on the basis ot' different geometri-cal criteria, in particular in cases where the @rain size distribution of the produced formations is not well known. lii addition, perfor-mance tests only qualitatively evaluate gr@ivel-packs, and therefore discourage improvement of gravel-pack design by not providing January 1995, Volume 34, No. 1 quantitatiN e perf4:)rmance measures.Gravel-pack clesign methods based on gravel-to-sand size ratio results have been more readily accepted because they are easy to use and e@isily implemented into practice. However, there is con-siderable variability between these methods, ...
