Two-dimensional flow towards a guarded downhole sampling probe: An experimental study

Flow towards a guarded sampling probe: Modelling of a 2D flow cell

“…Theory (Sherwood, 2005) predicts that the level of contamination in fluid arriving at the central sampling probe should drop to zero in finite time, and this is confirmed by field experience (Weinheber and Vasques, 2006;O'Keefe et al, 2006). Tarvin et al (2008) discuss experiments that investigate 2-dimensional flow towards a guarded downhole sampling probe. The experiments were compared to the predictions of a finite-difference numerical reservoir simulator.…”

“…However, pumping rates during sampling are sufficiently high that velocities due to buoyancy are small compared to those due to pumping in the region of interest around the sampling probe. Gravitational effects were negligible in the experiments of Tarvin et al (2008): in particular, in a bounded flow cell the velocities due to fluid withdrawal do not decay far from the probe and therefore remain greater than velocities due to buoyancy everywhere within the cell. We therefore take the densities of the two fluids to be that of water, so that gravity plays no role in the simulations: results can therefore be compared directly to results of the complex variable analysis of Sections 3-5 in which gravity was similarly absent.…”

“…Fig. 10 shows the initial position of the interface between filtrate and original pore fluid, shown in the same orientation as the experiments of Tarvin et al (2008) i.e. with the x axis vertically upwards and with the probe situated at the midpoint of the lower boundary.…”

“…Intermediate saturations are depicted as intermediate shades of grey. The interface is assumed to be sharp initially, as in the experiments of Tarvin et al (2008). Fig.…”

Fluid Sampling From Porous Rock to Determine Filtrate Contamination Profiles Around a Well Bore

“…Filtrate continues to arrive at the probe, from near-well bore regions increasingly far from the probe, but this filtrate enters the guard annulus and is discarded. Analysis of idealised axisymmetric and plane geometries (Sherwood 2005;Tarvin et al 2008;Sherwood 2009) has been confirmed by laboratory experiment (Tarvin et al 2008) and by field tests (Weinheber and Vasques 2006;O'Keefe et al 2006). We consider flow towards a guarded probe in a well bore and assume that a fixed fraction λ of the flow Q goes into the central sampling probe, with the remaining fraction 1 − λ going into the guard probe.…”