Research paper On the end-effect correction for Couette type oil-field direct-indicating viscometers for Newtonian and non-Newtonian fluids V.C. Kelessidis a, ⁎, R. Maglione b , G. Bandelis a a Department of Mineral Resources Engineering, Technical University of Crete, Polytechnic City, 73100 Chania, Greece b Consultant, Vercelli, Italy abstract article info Article history: Received 5 November 2008 Accepted 4 January 2010 Keywords: end-effects Newtonian non-Newtonian Couette viscometer Analysis and experimental results are presented which quantify the magnitude of end-effects for direct- indicating, narrow gap, rotating cylinder viscometers used extensively in oil-field, for Newtonian and non- Newtonian fluids. Such viscometers have already embedded at manufacturing stage an end-effect correction and give directly the shear stress which does not allow end-effect quantification in a straight forward manner. The presented analysis resolves this problem by estimating the embedded correction, which has been determined as 6.69% of the torque developed along the cylindrical part of the stationary bob of this viscometer. Experimental results with Newtonian and non-Newtonian fluids show that there is no additional end-effect from the bottom of the stationary bob. However, there is additional end-effect contribution from the top section of the bob, which has not been taken into account in the design and manufacturing of the instrument. Its magnitude ranges from 5 to 6% for the Newtonian fluid and the high shear rate range for the non- Newtonian fluid, to 12% for the low shear rate range for the non-Newtonian fluid tested, while the impact on drilling situations is also discussed. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Data analysis and rheological parameter estimation for Newtonian and non-Newtonian fluids using rotating cylinder viscometers is normally done with the assumption that there is a narrow gap between the cylinders, in particular for the viscometers used extensively in oil- field industry (Bourgoyne et al., 1991). Implicit assumptions involve also Newtonian velocity profile for the flow between the two cylinders and that the cylinders are infinitely long. The former assumption has been questioned by several investigators for rheological measurements of non-Newtonian fluids, even for narrow gap (Joye, 2003; Kelessidis et al., 2005, 2006, 2007; Kelessidis and Maglione, 2006). The latter, known as the end-effect problem, has been questioned not only for non- Newtonian (Gucuyener et al., 2002; Savaramand et al., 2003) but also for Newtonian fluids (Lindsley and Fischer, 1947; Oka, 1957; Oka, 1960; Barnes et al., 1993). Couette geometry solutions are generally valid for infinitely long cylinders because the torque developed on the side of either the inner or outer cylinders is very large and hence, the torque developed on the top and bottom ends of the fixed cylinder can be neglected. However, when the fluids cannot sustain their own weight they must be contained in a cup thus exposing the bottom and/or top surfaces of the inner cylinder to fluid drag producing an end-effect. Accurate recovery of the flow curve from Couette viscometers depends on error-free measurements (De Hoog and Anderssen, 2006). Most of the time, though, a rheological model derived from viscometric data is required and for many slurries used in the oil-field industry, such as water–bentonite dispersions or cement. The model of choice in recent years, has been the Herschel–Bulkley equation because it describes most drilling fluid rheological data much better (Zamora and Bleier, 1977; Fordham et al., 1991; Hemphil et al., 1993; Maglione and Ferrario, 1996; Kelessidis et al., 2005; Kelessidis et al., 2006; Bern et al., 2007). Different designs of Couette viscometers, particularly with variations of the bottom of the bob, with hollow bottom bobs (Mercier, 1932; Princen, 1986) or with conical top, conical top and conical bottom end (Goodeve, 1939), have been suggested but several investigators have reported that significant errors can occur with these designs. For instance, Lindsley and Fischer (1947) found significant errors with the hollow bottom designs, while Kobayashi and Nashima (1991) reported greater end corrections for Newtonian fluids for viscometers with conical end bottoms. End-effects could be determined experimentally and the error could be absorbed in the instrument constant (Lindsley and Fischer, 1947) but this could then be material dependent, which is not desirable. Furthermore, there has been no full quantification on the end-effect correction for non-Newtonian fluids. It has been reported that end-effects are larger for non-Newtonian than for Newtonian fluids (Highgate and Whorlow, 1969). Journal of Petroleum Science and Engineering 71 (2010) 37–46 ⁎ Corresponding author. Tel.: +30 28210 37621; fax: +30 28210 37874. E-mail address: kelesidi@mred.tuc.gr (V.C. Kelessidis). 0920-4105/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.petrol.2010.01.001 Contents lists available at ScienceDirect Journal of Petroleum Science and Engineering journal homepage: www.elsevier.com/locate/petrol