Woven metal mesh sand screens, commonly known as premium screens, have been extensively used by the industry. Sand retention testing is often done to evaluate the performance of these screens and establish empirical guidelines for screen size selection. However, these tests are tedious and the results are prone to artifacts and have been used, at best, to correlate trends in sand retention performance with select sand size distribution parameters. A new method incorporating results from numerical modeling in addition to experimental data is presented to estimate the mass and size distribution of the produced solids through premium screens. This method provides a fast, reliable correlation to estimate sand production through premium mesh screens when the size distribution of the formation sand is known.

This paper presents results from a wide range of sand retention experiments. In these tests the mass of sand produced and its size distribution over time are measured. Results of three-dimensional, discrete element computer simulations of woven screen geometry placed in contact with granular sand packs of  100,000 particles are also presented. Based on both the simulations and the experiments a new method for screen selection is presented. This method is based on a correlation that allows us to use the entire sand size distribution of the formation sand and estimate the mass and size distribution of the produced sand. The method is validated by comparisons with experimental data.

A new method and new correlations for estimating the mass and size distribution of produced solids through premium screens is presented. Key differences in sand retention mechanisms between premium and wire-wrapped screens have been identified. The method uses the entire formation sand size distribution (as opposed to a single design point) and has been validated with laboratory tests. The method also helps in screening anomalous test results.