Settling and sagging of barite in inclined boreholes may lead to safety and operational problems. To study the effect of rheology on settling, a laboratory tool was designed, consisting of two connected tubes, one inclined and one vertical. The hydrostatic pressure was measured at the bottom of each pipe. Stable and unstable muds can clearly be differentiated through their pressure behaviour. Several muds were studied at simulated static and dynamic conditions. The results show that sagging is most severe during laminar flow and also indicate that the rheological parameters may be used for predicting stability problems. Introduction In weighted drilling mud barite tends to segregate slowly. In directional drilling operations the settling process is accelerated. Barite settles in the lower side of the borehole and starts sliding when the borehole has an inclination above 30 °. This phenomenon is known as barite sagging. Sagging can lead to drilling and completion problems; a density variation or non-linear hydrostatic pressure gradients which can lead to pressure control problems, while thick and tight barite beds can lead to high torque and drag, stuck pipes and plugged boreholes, and even lost circulation. The sag problem is related to the so called Boycott effect(1), first described in 1920. Hanson et al.(2) have investigated the phenomenon and found that most of the sagging occurs while the mud is circulating. The same conclusion was reached by Bern et al.,(3) the sagging tendency is highest at low annular velocities. Zamora and Jefferson(4) presented a method for tracking drilling fluid density variations, which helps to detect, but not to predict drilling mud instability. Jamison and Clements(5) developed a test method to characterise settling and sag tendencies in static drilling fluids, however their equipment was not able to distinguish between settling and sliding. They also found that their data, based upon standard API rheological parameters like PV and YP, were unsuitable for prediction of sagging behaviour. It is apparent from previous works that the most difficult part of the problem is the prediction of drilling mud instabilities for both static and flowing drilling fluids. To date there are no API test procedures for sag testing. A new simple laboratory tool was therefore designed for the purpose to study and develop a method to predict sagging. Drilling Mud Separation Settling of Particles In vertical wells the settling of weighting material is generally not a problem due to the long settling distance. In horizontal wells the distance to the lower side of the wall is only about 0.2 m, which leads to rapid generation of solids beds. The settling velocity of a single spherical particle, vs,l, in a fluid is expressed by Stoke's law(6): Equation 1 (available in full paper) A barite particle (ρp = 4,200 kg/m3) with a diameter (dp) of 20 µm in a fluid of density (ρfluid) 1,500 kg/m3 and a viscosity (µ??of 40 cP will settle at a rate of 53 mm/h.

The phenomenon of 'gas migration' in oil well cementing is believed to occur during the transition state between initial and final set of the cement. As part of an initial study to elucidate the mechanism of gas intrusion into the cementing material, the external and total chemical shrinkage of API class G cement slurries was measured for the first 48 h of cun'ng. In order to counter the disturbance of segregation and formation of bleed water during the measurement of external chemical shrinkage, precipitated calcium carbonate (:5 15 % ,fineness 18 m 2 / g) and the viscosifier polyvinylalcohol were added. Neither addition significantly influenced the hydration rate of G cement with water-cement ratio (w/c) = 0·50 during the first 48 h. The chemical shrinkage, both total and external, of G cement slurries on hardening seems to be largely independent of w/c in the first 48 h. The external and total shrinkages of a neat G cement slurry curing at 20°C and atmospheric pressure are about 1·0 mll 100 g cement or 1·2 vol. % and 2·2 mlll00 g cement or 2·6 vol. % respectively after 48 h.

Particles like barite and cuttings influence the friction properties of a mud. The standard API lubricity tester, however, cannot measure friction of fluids containing particles, and to overcome this problem it was modified with a cam setup. It was found that particles indeed alter the friction. Large beads are being used to reduce friction. They are, however, filtered out in the solids control equipment and to avoid this, we have investigated smaller polymer microbeads which will pass unhindered. The microbeads reduce the friction in water based muds with around 40% which is significantly better than four commercial lubricants.

No abstract

The phenomenon of ‘gas migration’ during oil well cementing is believed to occur during the transition state between initial and final set of the cement. In order to evaluate the importance of pore openings and total porosity in the critical time gap, a suitable experimental technique was tested on some neat oil well cement slurries. The hydration was effectively stopped every 30 min at 20°C and every 20 min at 60°C by dropping plastic tubes containing the cement slurry into liquid nitrogen (i.e. quenching), cracking the tubes open and letting the frozen bits of paste thaw in ethanol. The change in porosity and pore size distribution was determined by helium pycnometry and mercury intrusion porosimetry as a function of time in the setting period for a plain API class G cement slurry (w/c = 0·50) at both 20°C and 60°C. These data were compared with the amount of chemically hound water in the same samples, and used to predict the total porosity at a given degree of hydration. The excellent correspondance of experimental and theoretical porosities validates the experimental procedure, which can also be used in explaining the variation in gas migration between commercial oil well cement slurries.