Sand retention testing is often required when selecting screen media and media size. It has also been used to define the limitations of retention for different media . However, the results from such tests are very dependent on experimental conditions and prone to artefacts. Small and apparently innocuous changes to the test conditions can cause wide variations in results. Some of the factors affecting test results are discussed. The problems of artefacts can be overcome, at least in part, by considering trends in data rather than absolute results.
A previous paper described the performance of Dutch twill weaves (DTW) and identified a selection criteria based on particle size distributions measured by laser light scattering1. This paper extends that work to include sintered media from other premium screens and wire wrap screens. The general trend in the data from a number of tests shows good correlations between some aspects of the sand distribution and retention performance of the individual media. From these data, an aperture sizing criteria for sand screens based on laser particle size measurements is proposed, and screen selection parameters considering both retention and "plugging" performance are identified.
Laboratory sand retention tests are routinely used to assist in sand screen selection ,3. Generally identifiying the best performing screen from a set of results is fairly straightforward, but trying to identify minimum requirements in terms of retention from laboratory tests is much more problematic. There is often a temptation to over-interpret results from retention tests in relating them to the field. It is hoped that by pointing out some of the problems and artefacts in such tests that the results will be viewed with more caution. Sand retention tests are particularly prone to artefacts because the tests cannot accurately replicate what happens in the field. This is partly due to limitations in test equipment but also to the lack of understanding/knowledge of what happens downhole. Some of the factors which can influence test results will be discussed.
As well as drawing attention to some of the experimental problems, this paper also reports the results of continuing work on screen selection criteria initially developed in SPE 822441. In particular more sands have been tested and the study was extended to cover sintered weaves and wire wrap screens. Attempting to update the guidelines for screen selection proved more problematic than expected.
The authors have performed multiple sand retention tests on more than a hundred different reservoir sands. A lot of the problems identified with the test methods are discovered during routine tests and identified during further testing when trying to understand anomalous results.
Extensive sand retention testing was used to define the limits of retention of the dutch twill weaves used as filter media for expandable sand screens. The work was reported in SPE 82244 and is briefly summarized herein. The work reported in this paper is a continuation of that work. Sand retention tests were performed with a variety of reservoir sands, and the amount of sand passing through for each test on a particular weave was plotted against the size of the sand in terms of d5, d10, d50 etc. In this way any correlation between sand size and retention performance could be ascertained and potentially used as a sizing criterion. Figure 1 shows one of these crossplots for a 150 µm DTW. Here the d10 of the sands tested are plotted against the sand passing through the weave and each point on the graph represents a separate sand retention test. Figure 2 shows the sand retention data from the 150µm weave but this time plotted against the d50 of the sands tested. These crossplots were originally presented in SPE 82244 but are updated here to include data obtained since 2003.