Prediction of Laminar and Turbulent Friction Pressures of Cement Slurries in Pipes and Centered Annuli

Abstract

Abstract Friction pressures of different cement slurry formulations with properly characterized rheology have been measured in a wide variety of flow situations, including laminar (20 to 1000 s-1) and turbulent (2000 less than Re less than 10000) flow regimes. Tests were performed in pipe and concentric annular geometries, using a pilot-scale rig. In laminar flow, it was established in a previous publication that pipe flow friction losses of cement publication that pipe flow friction losses of cement slurries can be reliably predicted from rotational viscometer measurements made in the shear rate range of the actual pipe flow. In this paper it is shown that valuable predictions can also be obtained for concentric annuli, at least when the radius ratio (inner/outer) is greater than 0.3, if the geometry is approximated by a rectangular slot. Pipe flow experimental data obtained in the turbulent Pipe flow experimental data obtained in the turbulent flow regime are compared to several existing models. A modified Dodge and Metzner approach to the problem was found to give the best agreement with the measurements. For concentric annuli, a similar procedure is shown to be acceptable when used with the rectangular slot approximation. Reasonable agreement is found between model predictions and experimental results. For both flow regimes, the proposed method is no more complicated and is proved, for most field situations, to be much more accurate than the API procedure in predicting pressure drops. pressure drops. Introduction Estimating the Reynolds number, flow regime and friction pressure of different fluids involved in a primary cement pressure of different fluids involved in a primary cement job - mud, spacer, cement - can be important for the success of the operation. Prediction of the pressure and temperature profiles in the wellbore, the control of the return flow rate and the optimization of mud removal all depend, among other things, on these three parameters. In a previous publication, the authors have shown that using a specific apparatus and procedure, it is possible to predict, from coaxial cylinder viscometer measurements, predict, from coaxial cylinder viscometer measurements, the laminar friction pressures for the pipe tow of nondispersed cement slurries. The purpose of the present paper is to extend this result to dispersed formulations and paper is to extend this result to dispersed formulations and to the turbulent flow regime. The majority of the experiments were performed in a pilot-scale pipe flow loop, but some results were also obtained in a small annular cell. Measured friction pressures are compared to the predictions of different models, including those recommended in API Spec. 10 [2]. Experimental set-up: The experimental set-up consists of three different parts. - A mixing area where about 150 L of slurry is prepared in 200 L drums with two paddle agitators of 1 k maximum power each. power each. - Two "moineau" pumps capable of delivering up to 11 m/h at a maximum head pressure of 12 bar. Pressure pulsations are damped to less than 1% of the head pulsations are damped to less than 1% of the head pressure with a compressible chamber; the flow rate is pressure with a compressible chamber; the flow rate is measured with an electromagnetic flowmeter to within +- 5 L/h. - Two separate lines, connected to the pump: the pipe flow loop includes a 16 and a 20 mm ID pipe connected serially. The return line is made of 40 mm ID plastic tubing in which the flow regime was laminar in most cases. A differential pressure transducer is mounted at least 50 diameters away from the inlet and outlet of each section to ensure that the flow is fully established. The separation between the two pressure taps is 4.4, 7.0 and 10.0 m for the 16, 20 and 40 mm ID sections, respectively. The annular flow loop consists of a 5 cm ID perspex pipe which is concentric with a 4 cm OD pipe. This vertical section is 3 m long and the pressure drop is measured across a 2 m section in the middle of the cell with a differential pressure transducer. pressure transducer. P. 379