Mario Zamora, Dan Jefferson
M I Drilling Fluids Co.
Houston
A new method for tracking drilling fluid density variations helps detect barite sag, which may contribute to drilling problems.
The method is based in part on continuously measuring fluid density during the first circulation after the fluid has been static for some time.
In deviated wells or wells with weighted fluids, barite sag has aggravated or caused drilling problems such as lost circulation, stuck pipe, high torque and drag, poor cement jobs, logging difficulties, and well control difficulties. Sag is defined as a significant variation in mud density measured during the first bottoms up circulation after a weighted mud has remained static for some time in a directional well.
The density variations are caused in part by slumping of beds formed when weight material settles to the low side of the hole. Furthermore, the bed formation occurs while the fluid is circulating and not just during static conditions.
Fig. 1 is an example of barite sag measured after a trip on a Gulf of Mexico well. The roughly sinusoidal shape of the graph is the characteristic "fingerprint" of barite sag: light mud, followed by heavy mud, and then the original mud weight.
As expected, the heaviest mud weight usually occurs at bottoms up. The maximum mud weight difference from sag on this well was 2.4 ppg. Mud weight differences as large as 4 ppg and 7 ppg have been recorded in the Gulf of Mexico and North Sea, respectively.
Sag was initially thought to occur more often in oil based muds, but it has now been found in all types of muds ranging in mud weight from 12 to 20 ppg. In the past, sag was often ignored or simply tolerated, presumably because the link to drilling problems was not clearly established. Heavy mud off bottom usually was attributed to slugs or mud dehydration. Many companies now willingly measure mud weight continually during a bottoms up circulation. This practice has confirmed the existence of sag and increased awareness of its potential consequences.
Sag is difficult to prevent, but its consequences can be minimized by using proper drilling practices and appropriate mud properties. Barite sag is not entirely a mud problem and must be considered from the perspective of the overall drilling process. Perhaps sag management should be applied in the same manner as the task force approach developed for reducing stuck pipe problems.1
SAG MECHANISMS
In the sag process, a bed of weight material is deposited on the low side of an inclined well bore. (Barite is the most common weight material, but hematite and calcium carbonate also are used.) The higher the specific gravity, the greater the tendency to form a bed. Under the right conditions, the bed can slide or slump toward the bottom of the hole. This slumping causes the sag fingerprint shown in Fig. 1 when the mud is circulated bottoms up.
Despite the association of sag with operations involving static mud, sag is basically a dynamic settling problem. The attempts to minimize sag by treating it as a static problem, for the most part, have been unsuccessful. Laboratory tests and field experience have proven that most of the barite bed is formed while the mud is being circulated, especially at low to moderate flow rates. This phenomenon was first reported in a study of cementing problems in deviated well bores and was later confirmed by an extensive study devoted to barite sag.2 3 Additional settling and most of the slumping occurs during static periods.
Despite several reports of sag from flow rates normally used during drilling, density variations usually are not apparent during normal circulation for several reasons:
A bed has the greatest tendency to slump at intermediate inclination angles of 30 600. The most severe slumping can occur at angles of 40 500. Disturbances from tripping drill pipe or logging tools sometimes may be sufficient to initiate slumping.
Surprisingly, slumping beds can cause changes in hydrostatic pressure. This pressure change was recently observed in an S shaped Gulf Coast well. It was thought that an increase in hydrostatic pressure measured at the bottom of the well was caused by the slumping of a heavy sag bed from a 600 interval into a lower 470 interval. The increase in the length of the vertical component of the heavy bed is one possible explanation for the pressure change.
COMPLEX FLOW
Barite sag in directional wells is aggravated by complex flow patterns, which differ dramatically from those in vertical holes. Skewed velocity profiles caused by eccentric drill pipe contribute to bed formation. Other mechanisms, however, can also cause a significant increase in settling rate.
Hindered settling is predominant in vertical wells. Settling particles displace fluid that provides an upward force to neighboring particles. The net effect is an overall decrease in settling rate. Settling is further reduced if gel structures develop while the mud is static or if the combination of viscosity and annular velocity overcomes the effects of gravity.
An orderly flow pattern known as "Boycott settling" (named after a researcher) forms when the hole is inclined.4 The settling rate can increase by three to five times under these conditions. Fig. 2 illustrates Boycott settling under static (not circulating) conditions.
A clarified (lighter density) layer forms along the high side of the hole and tends to move upwards. A sediment (heavier density) layer forms concurrently along the low side and tends to move downwards. A cross sectional pressure gradient develops which, in effect, causes a flow that should be counteracted by the mud gel strength. Even high gel strengths, however, may be ineffective in preventing existing beds from slumping.
Boycott settling can be easily demonstrated in a zag tube, which consists of three segments of clear acrylic tubing connected by 450 elbows (Fig. 3). Colored glitter particles are dispersed in biopolymer viscosified water to show the complex flow patterns during seemingly static conditions. When the zag tube is placed as shown on the left in Fig. 3, hindered settling occurs in the vertical sections and Boycott settling occurs in the inclined section. When the zag tube is inclined as shown on the right in Fig. 3, each segment undergoes Boycott settling, and the flow patterns in the elbows are especially interesting.
Low to moderate flow rates can effectively accelerate Boycott settling. Circulating mud will flow along the high side, forcing a corresponding increase in the slump rate of the barite bed along the low side. Fig. 4, taken from a video recording of a lab experiment, demonstrates the three flow zones possible in a tube under dynamic conditions. The suspension zone, between the yellow lines in the center of the tube, simultaneously accepted particles from the flowing zone above and fed particles to the slump zone sliding downwards. The flow pattern clearly would be further complicated had the experiment been conducted in an eccentric annulus, similar to drill pipe in a well. Adequate models describing this behavior have yet to be developed.
Higher flow rates, particularly when coupled with pipe rotation, disturb Boycott settling. The velocity profile may still be asymmetrical, but the formation of barite beds may be significantly, reduced.
FIELD MEASUREMENTS
Sag can be identified and corrected only if mud weight variations are documented in trip reports and analyzed at the well site. The mud should be weighed at regular intervals during the bottoms up circulation. Ideally, the mud weight should be measured continuously if automatic equipment is available or every 5 10 min otherwise. The extent of the density variations indicates the severity of barite bed deposition and movement.
Currently, there is no convenient technique to track and compare trip reports. The following equation, called the sag register, should help track barite sag.
Sr = e(10 WdfWc)
In this equation, Sr is the sag register (dimensionless), Wd is the maximum mud weight difference (ppg), and Wc is the circulating mud weight (ppg). Wd is determined by subtracting the minimum mud weight from the maximum mud weight recorded during the bottoms up circulation.
If appropriate, the maximum mud weight may be adjusted for a known barite slug. Wc is the baseline or normal mud weight.
If no sag is detected, Sr = 1. Limited data suggest minimal sag problems should be expected for 1
Trip reports are sometimes difficult to interpret. For example, gas cut mud encountered during bottoms up circulations may skew data. A pressurized mud balance, instead of a rig balance, may be necessary to obtain accurate data.
Fig. 5 is a plot of mud weight measurements using a rig mud balance and a pressurized mud balance during a bottoms up circulation on an offshore well. The pressurized mud balance gave more accurate sag data because the mud was gas cut. The Sr values were 3.96 and 2.12 for the pressurized and rig balances, respectively. Also, unusual or irregular circulation schedules and hole geometries can create complex mud columns. If a complex mud column forms, delineating downhole conditions may become difficult or impossible.
LAB STUDIES
Reproducing field trip report results in the laboratory is difficult. Duplication of downhole conditions and field pumping schedules are perhaps the biggest obstacles. Nevertheless, laboratory studies have been effective in evaluating critical parameters affecting sag and helping suggest guidelines to minimize its occurrence.3 The need to perform tests under dynamic conditions is paramount.
The investigations to date have examined the dynamic and static behavior of more than 75 field muds. For most of the muds tested, barite bed formation began as soon as circulation was initiated. The bed shown in Fig. 6 was formed in the laboratory at 450. Mud was drained from the 4 in. test section at the end of the test. Throughout the testing, the dense beds remained fluid and began to slump once their mass increased above a threshold level, regardless of whether the fluid was being circulated. In some cases, the barite bed could be removed by increasing the mud velocity.
Static settling was minimal in most cases. Low flow rates dynamically enhanced the Boycott effect and accelerated settling. The resulting density stratifications were often clearly visible: clarified fluid along the upper side of the tube, a fluid bed along the lower side of the tube, and a suspension zone in between.
The majority of the lab experiments have been conducted in a flow loop with a tubular test section placed at 450, although some tests have used other inclinations and geometries.3 The data are taken while the fluid is circulated at decreasing flow rates for three 20 min periods followed by a 3 hr static period in a tube placed at 450. Cross sectional and longitudinal mud density differences are measured using nine ports in the test section. At the end of the static period, the mud is drained in regular intervals and weighed.
SAG INDEX
A well site method (sag index) for determining sag was developed to help quantify or pilot test the impact of potential fluid adjustments in the field.5 A Fann viscometer is used to shear the mud for dynamic test conditions. Mud samples taken from the bottom of the heat cup are weighed before and after shearing the mud at 100 rpm (170 sec 1) for 30 min at 1200 F.
The ability of a mud to suspend barite under dynamic conditions certainly is important; however, other factors may be equally significant. The sag index is the product of the mud weight difference measured in the viscometer and four constants based on key operational factors affecting sag: well bore inclination, flow regime, hole diameter, and length of the inclined well bore.
The sag index is useful during well planning because it can estimate sag potential qualitatively. Studies are currently under way to correlate the sag register and the sag index.
MINIMIZING SAG
Perhaps the biggest obstacle to minimizing sag is to change the belief that sag is solely a mud problem. Although mud properties certainly play key roles, the failure to use proper drilling practices can easily override certain mud contributions. The possibility of sag problems should be anticipated during the planning stages of directional wells that require weighted drilling fluids.
The physics of barite sag is such that even muds with ideal properties cannot fully suspend barite under all conditions. For example, weighted muds circulated at low flow rates for extended periods usually form barite beds regardless of the mud properties.
Key mud adjustments include increasing the low shear rate viscosity and improving the suspension properties. These properties are most commonly achieved in water based muds by using biopolymers and in od based and synthetic based muds by using rheology modifiers. Fann viscometer values at low shear rates (3 and 6 rpm) typically are used as guidelines to achieve the desired mud properties. The Bingham plastic yield point is not a reliable indicator of low shear rate viscosity.
High gel strengths are necessary for static suspension. The benefits of elevated gel strengths are not realized in dynamic situations, however. Furthermore, the gel strengths should be relatively flat to prevent an adverse impact on drilling.
Excessive mud thinning and flocculation, which create free water and eliminate low shear rate viscosities, should be avoided. Improper barite wetting in nonaqueous muds promotes hard setting and greatly intensifies sag.
Annular velocity is a key parameter in minimizing sag. High annular velocities provide energy to minimize bed deposition and help remove existing beds. Even short lapses in velocity levels, however, may induce bed formation that cannot be prevented from slumping. Drill pipe rotation and reciprocation enhance the benefits of high annular velocity.
Staging to bottom (circulating periodically as the pipe is tripped in the hole) during trips can systematically remove density variations that may have developed before and during the trip. Staging reduces the length of the heavy mud column in the annulus, thus decreasing the likelihood of lost returns during the bottoms up circulation. Fig. 7 shows an improved mud density, profile achieved by staging in the hole on sequential bit trips in a Gulf of Mexico well. The staging reduced the Sr value from 5.41 to 1.45. Barite sag and its consequences can only be minimized when all related factors are under adequate control. Fig. 8 is a trip report demonstrating the benefits of proper sag management (Sr = 1.51). Although barite sag was not eliminated, it was effectively controlled to minimize adverse effects on the drilling.
REFERENCES
- Bradley, W.B., et al., "Task force reduce stuck pipe costs," OGJ, May 27, 1991, pp. 84 89
- Keller, S.R., et al., "Deviated Wellbore Cementing: Part 1 Problems," Journal of Petroleum Technology, August 1987, pp. 955 60.
- Hanson, P.M., Trigg, T.K., Rachal, G., and Zamora, M., "Investigation of Barite 'Sag' in Weighted Drilling Fluids in Highly Deviated Wells," SPE paper 20423, presented at the Society of Petroleum Engineers 65th Annual Technical Conference, New Orleans, Sept. 23 26, 1990.
- Boycott, A.E., "Sedimentation of Blood Corpuscles," Nature, Vol. 104, 1920, p. 532.
- Jefferson, D.T., "New Procedure Helps Monitor Sag in the Field," ASME 91 PET 3, presented at the American Society of Mechanical Engineers Energy Sources Technology Conference and Exhibition, Houston, Jan. 20 24, 1991.
Copyright 1994 Oil & Gas Journal. All Rights Reserved.
