The problem of solid cleanout in horizontal wellbores is studied experimentally. The special case of drilling fluid circulation with no inner pipe rotation is considered. This case is similar to Coiled tubing drilling where often hole cleanout must be performed. Sand sized cuttings (ranging from 260 micron to 1240 micron) are used. Critical velocity and wall shear stress required for initiating bed erosion are measured. Water and viscous polymer base fluids with 3 different polymer concentrations are used.Results have shown that water always initiate cuttings movement at lower flow rates than polymer solutions. Fluids with higher polymer concentration and hence with higher viscosity required higher flow rates to start eroding the bed. Critical wall shear stress is also determined from pressure loss measurements. Analyzing the data revealed that water initiate cuttings removal at lower pressure loss than more viscous fluids. Higher viscosity fluids always showed higher pressure loss at the initiation of bed erosion.For the range of cuttings size studied, results show that intermediate cuttings size is easier to remove. Smallest and largest cuttings are more difficult to move requiring higher flow rates and higher pressure losses.Dimensionless analysis of relevant parameters to the process of cuttings movement is performed. It is shown that dimensionless wall shear stress (in the forms of Shield's stress and also ratio of shear velocity to settling velocity) correlates well with particles Reynolds number. Based on this finding two correlations are developed to predict wall shear stress required for initiating cuttings movement under different conditions.

Fully developed turbulent flow of water through a horizontal flow loop with concentric annular geometry was investigated using high resolution particle image velocimetry (PIV). Reynolds number range varied from 17,700 to 66,900. Axial mean velocity profile was found to be following the universal wall law (u+ = y+) in the viscous sublayer (y+ < 10) and log law away from the wall (y+> 30). Radial position of zero shear stress and maximum velocity were found to be slightly different (2%). Root mean square values of velocity fluctuations velocity, Reynolds stresses, vorticity, and turbulent kinetic energy budget were also analyzed.