Conventional production logging (PL tool strings use the spinner to measure fluid velocity. It is a ruggedised turbine type flowmeter for downhole use. Production monitoring is an important PL activity for quantifying zonal contributions to the total well production. The information is used for reservoir simulation studies, production history matching, material-balance calculations and must be accurate. Current PL interpretation assumes that the spinner is a velocity averaging sensor, but the driving torque on a spinner blade is proportional to fluid density. In multiphase systems the spinner therefore often fails to respond to the momentum of the lighter gas phase, leading to an underestimation of the flow from the deepest zones in a gas well, or failure to respond to low gas rates. This causes potential zonal production and allocation errors. An appropriate spinner response model could quantify better the commercial hydrocarbon production from the deepest zones, aiding the application of timely water shutoff technology and lead to reservoir recovery factor improvement.
The paper will show that the spinner can be classed as a momentum averaging sensor and its multiphase response modeled from this standpoint. Errors that a velocity averaging response model introduces are computed for vertical and deviated wells. Gas and liquid inflow profiles are then converted via a momentum averaging model back into raw spinner velocities to display the before and after effect of changing between the two models. The significance of the differences is discussed. Potential PL technology developments to take advantage of the new model are reviewed. Data from several production logs in gas wells are processed with this new model and changes in the computed inflow profiles are presented. The results will demonstrate that production logging measurements from wells producing gas and liquid can be improved, leading to better reservoir production management and ultimate hydrocarbon recovery improvement.