MUD RELAXATION MEASUREMENTS HELP PREDICT HOLE CLEANING ABILITY

Bert R. Bloodworth
Kelco Oil Field Group Inc.
Midland, Tex.

George J. Keely Jr.
Baroid Drilling Fluids Inc.
Midland, Tex.

Peter E. Clark
University of Alabama
Tuscaloosa, Ala.

A relaxation measurement test, easily performed in the field, can help determine the effectiveness of a drilling fluid to suspend solids and carry cuttings to the surface. Field tests have confirmed that this measurement can help the drilling fluids engineer maintain optimum rheological properties on the mud system. Laboratory measurements have shown that the relaxation test measures a property that is related to a true yield point.

Hole cleaning and solids suspension are key functions in defining the performance of drilling fluids. Typically, the ability of a fluid to perform these functions is evaluated by determining the plastic viscosity, yield point, and gel strengths with the use of a rotational viscometer. These accepted measurements, however, may not adequately predict hole cleaning and suspension characteristics of the fluid.

Rheological data, derived from the use of traditional instruments such as the Fann model 34 and 35A viscometers, have been the most common means of predicting annular performance of drilling fluids.1 2 Data generated with these viscometers may not be adequate for predicting annular performance capabilities such as solids transport and suspension.

The Bingham plastic rheological model is often used to describe the viscous behavior of drilling fluids. The Bingham plastic viscosity and yield point are determined from shear stresses taken at shear rates of 1,022 sec-1 (600 rpm) and 511 sec-1 (300 rpm). Although this model is widely accepted for determining annular fluid performance, it is also recognized that the shear rates used are too high because they fall outside typical annular shear rate ranges. The plastic viscosity is the slope of the line drawn between the 600 rpm and 300 rpm readings (Fig. 1). Yield point is the intercept on the shear stress or Y-axis of an extrapolated line drawn between these two data points.

In many cases, the power law model more closely approximates actual fluid properties even when calculated from the 600 rpm and 300 rpm data points (Fig. 2). Because this model can be applied using data from annular shear rate ranges, it will provide much greater accuracy in predicting the performance of a drilling fluid in the annulus. This is especially true for the low-solids, highly shear-thinning systems currently used.

The power law model more closely describes the viscosity of a fluid under annular conditions, but it may not accurately describe the suspension and hole cleaning potential of a drilling fluid. These functions are determined by measuring the ability of a fluid to provide viscosity at shear rates below 5.1 sec-'1.1 4 However, some fluids exhibit substantial viscosities at shear rates from 5.1 sec-1 down to 0.06 sec-1 and high Bingham yield points, yet fail to provide suspension (Table 1). These phenomena indicate that viscosity measurements or Theological models can be misleading.

Hole cleaning and suspension capabilities are more accurately predicted by determining the viscoelastic properties of a drilling fluid rather than measuring viscosities at predetermined shear rates above 1.0 sec-1.1 3 Although many drilling fluids exhibit an elastic component, the shear stress required to transform them from a semisolid state to a liquid state can vary.

In this case, viscoelasticity is defined as the degree of deformation or elastic stretch a fluid can withstand before being transformed from a quasi solid state into a viscous liquid state (Fig. 3). 3-5 Experiments to determine the yield value of a fluid are generally performed in a laboratory environment with a controlled stress rheometer or by extrapolation of the data from a controlled shear rate viscometer. Although these methods of determining yield stress are very accurate, the controlled stress rheometer is primarily a laboratory instrument which can not be easily operated under field conditions.

VISCOUS FLOW

Laminar flow of Newtonian fluids through pipes or annulses is characterized by a parabolic velocity profile. The velocity approaches zero at the pipe wall and is at a maximum in the center of the flow. The flow has been described as being similar to a series of fluid laminae flowing past each other with the one nearest the wall stationary and each succeeding laminae toward the center of the flow moving faster. 6

Non-Newtonian fluids are characterized by a velocity profile that is not necessarily parabolic. As the fluid becomes more non-Newtonian, the velocity profile becomes flatter at the center (plug flow). The diameter of the plug increases as the deviation from Newtonian behavior increases. Describing a non-Newtonian fluid with a power law model allows some generalizations to be made about the velocity profile.

The power law model has two parameters used to describe the flow: n is the flow behavior index and typically varies between 1.0 and 0.1, and K is the consistency index which is a measure of the thickness or viscosity of the fluid. If the n value is 1.0, the fluid is defined as Newtonian. If the n value approaches 0.2, the velocity profile flattens and occupies a major part of the available area (Fig. 4). In the flat part of the velocity profile the shear rate is near zero. Fluids which exhibit a high viscosity under these near-zero shear rate conditions offer significant improvements in hole cleaning efficiency.6

Viscosity measurements with ordinary field instruments such as the Fann model 34 and 33A viscometers actually measure shear stress at the wall for a given shear rate. However, because of the narrow clearance between the bob and the rotating sleeve, the shear rate varies only slightly across this gap.

Non-Newtonian fluids with an n value of 0.2 exhibit a relatively flat velocity profile with cost of the shear occurring at the well bore wall and at the drill pipe (Fig. 5). Where annular flow laminae are traveling at approximately the same shear rate (i.e., in the plug flow region) there is virtually no shear between laminae. This contradicts a general perception that polymer solutions shear-thin across the entire annulus or well bore.1 Normally, shear rates within the boundaries of this quasi static state approach zero. It is in this region of non-Newtonian flow where the hole cleaning and suspension properties of a fluid should be determined.

VISCOSITY INVESTIGATION

Drilling fluids are subjected to a wide variety of velocity conditions while being pumped through the annular regions of the well bore. Annular wall shear rates can range from 2 sec-1 to 200 sec-1 depending upon hydraulic diameters and flow rates.

Various drilling conditions, including deviated or horizontal wells, unstable hole conditions, and abnormal pressures, can cause serious, fluid-related problems which can affect the entire drilling operation. Almost all drilling fluid-related problems can be associated in one way or another with viscosity. The identification of Theological shortcomings in drilling fluids is by no means difficult; however, tactical measures taken to correct such problems are often inadequate and can create additional hole problems.

It is logical to assume that if hole cleaning and suspension problems occur in drilling fluids, an increase in viscosity is necessary. In some cases, the products added to the system increase the apparent or high shear rate viscosity but may have little or no effect on viscosities in the low shear rate range (Fig. 6, Additive B). In other instances, the products added to the fluid increase the low shear rate viscosity (LSRV) alone, with the apparent viscosity, often resulting in a thick, unmanageable drilling fluid (Fig. 6, Additive A). (LSRV is defined as the viscosity taken below a shear rate of 1.0 sec-1.) Ideally, viscosities in the high shear rate ranges (above 170 sec-1) should remain as low as possible to minimize pressure losses while pumping. Viscosity increases should be made in the low shear rate ranges to ensure optimum solids transport (Fig. 6, Additive C).

DRILLING FLUID EVALUATION

The ideal drilling fluid is considered to have a low viscosity at high shear rates and a high viscosity at low shear rates.1 2 Although there is field equipment that adequately measures shear stress at shear rates from 1,022 sec-1 to 5.11 sec-1, there has not been a practical method to measure viscosity below 5.11 sec-1 or 3 rpm on the standard Fann model 35A viscometer.

Recent field use of the Brookfield LVTDV-II digital viscometer has made it possible to measure viscosities at 0.06 sec-1.1 2 This instrument has been of great assistance in attempting to solve hole cleaning, barite sa and suspension problems.1 8 In some instances, high Brookfield viscosities in this low shear rate range (0.06 sec-1) have misled laboratory and field personnel, resulting in no appreciable change in the ability of the drilling fluid to provide suspension (Table 1). This is evidence that viscosity measurements at shear rates of 0.06 sec-1 and above do not always portray the viscous functionality in drilling fluids.

RELAXATION MEASUREMENTS

Relaxation measurements are made after the standard viscosity determinations have been compiled from the Farm or Brookfield viscometer. Following the last viscosity determination, preferably at a shear rate of 5.11 sec-1 or less, the instrument's motor is turned off. The indicator dial or display is then monitored as the spindle or bob attempts to return to zero. Fluids which exhibit an extended relaxation time indicate improved hole cleaning and suspension characteristics.

Viscosity measurements were performed on two different polymer solutions mixed in 2% KCI water. One solution was hydroxyethyl cellulose (HEC) at a concentration of 0.7% by weight (2.5 lb/bbl), and the other solution was clarified xanthan gum at a concentration of 0.45% by weight (1.5 lb/bbl). A volume of 1,750 cc of each solution was prepared to obtain funnel viscosities. Additional rheological data were obtained with a Farm model 35A viscometer.

The HEC solution exhibited a much higher funnel viscosity, plastic viscosity, yield point, and Farm 35A 3-rpm value than that of the clarified xanthan gum solution. After the 3 rpm reading, the machine was turned off to obtain a 10 sec gel strength. The viscometer dial went to zero rapidly for the HEC solution, but the dial did not reach zero for over 1 min for the clarified xanthan gum solution. This experiment was repeated numerous times to ensure that this was a reproducible phenomenon.

This was also observed using a Brookfield LVTDV-II digital viscometer, which allowed viscosity measurements to be made below 5.11 sec-1 (Farm 35A, 3 rpm) on each solution. The HEC solution yielded a considerably lower viscosity at 0.06 sec-1 than did the clarified xanthan gum solution. This measurement contradicted the findings from the Farm 35A viscometer, which indicated that the HEC solution should be a superior suspending fluid based on its elevated yield point and 3 rpm reading.

Computer software was used with the Brookfield viscometer to monitor shear stress decay and to examine the relaxation measurement comparisons more accurately between the two solutions. Immediately following the 0.06 sec-1 readings, the Brookfield viscometer was turned off and the electronic display monitored to record the ability of each solution to build a gel structure. As the spindle attempted to return to zero, the clarified xanthan gum solution exhibited a substantially higher resting plateau than did the HEC solution.4 8 The resting plateau for the HEC solution was a reading of 700, and the clarified xanthan gum solution produced a reading of 7,000. Because neither solution permitted the spindle to return to zero within the time frame of the experiment, both appeared to exhibit a yield value (Fig. 7).

To correlate the above findings with suspension properties, settling tests were performed with a coarse ground flint. To 500 cc samples of each solution were added 15 g of the flint. The HEC solution did not provide adequate suspension, whereas the clarified xanthan gum solution suspended the material for an extended period of time. This indicated that even though the HEC solution had a substantially higher yield point and 3 rpm reading than the clarified xanthan gum solution, it did not provide sufficient viscosity to suspend the flint chips.

Therefore, the Farm 3-rpm dial reading is not always a reliable indicator for the determination of hole cleaning and suspension capabilities. Even though a fluid may exhibit a low 3 rpm shear stress, an extended relaxation measurement provides a better assessment of these functions.

This measurement is important for attempting to determine hole cleaning and suspension capabilities of a fluid. The following were determined in the development of an improved method of qualifying the suspension capacity of a drilling fluid:

Polymeric and clay viscosifiers used to increase LSRV have different effects on 3 rpm (5.11 sec-1) viscosity
Polymeric and clay viscosifiers used to increase LSRV can also have dramatic effects in the high to moderate shear rate ranges (above 170 sec-1)
Relaxation measurements provide evidence that the standard viscosity measurements may not fully predict hole cleaning and solids suspension properties
With selected viscosifiers, it is possible to increase LSRV and relaxation measurements while not adversely affecting high to moderate shear rate viscosity (plastic and funnel viscosity).

Samples of drilling fluids were obtained from several drilling sites. In some cases, annular performance indicators such as yield point and gel strength appeared to be adequate to ensure timely delivery of cuttings to surface. However, some wells had poor delivery of drilled cuttings to the surface. Brookfield Theological data taken at 0.06 sec-1 also indicated that these fluids should have been capable of cleaning the hole.

Relaxation measurements were made on each fluid, which were correlated to hole conditions. Those wells with hole cleaning problems revealed reduced relaxation times; those wells drilled without problems exhibited extended relaxation times. These results verified the need to further study the use of relaxation techniques on drilling locations.

FIELD INVESTIGATION

The following two field examples are taken from wells drilled in southeastern New Mexico.

On one location, the drilling operation was halted for the running of electric log surveys. However, the logging tools could not be run to bottom, and subsequently the viscosity of the fluid was increased to the following levels: Funnel viscosity 140-150 sec/qt, plastic viscosity 30 cp, yield point 55-60 lb/100 sq ft, and gel strengths 12/15 lb/100 sq ft.

The concern was that there were not many cuttings coming out of the hole, yet on short trips there was always considerable fill.

A Farm model 35A viscometer and a Brookfield LVTDV-11 rheometer were used to obtain Theological data on the fluid. The Brookfield indicated that the viscosity at 0.06 sec-1 was 38,000 cp, which seemed to be adequate to remove drilled cuttings from the well. Relaxation measurements were performed with both the Farm Model 35 and the Brookfield LVTDV-II:

Farm 35A viscometer: When the motor on the instrument was turned off (from a 3 rpm reading of 12), it took 9 sec for the instrument to reach zero.
Brookfield LVTDV-II: A 38,000 cp viscosity at a shear rate of 0.06 sec was obtained on the sample, and the relaxation measurement indicated a resting dial reading of 3,500.

Relaxation measurements indicated that the fluid was not capable of suspending debris nor of effectively transporting solids to the surface. Even though the fluid's yield point and gel strength seemed sufficient for hole cleaning, relaxation measurement indicated otherwise. Pilot tests were initiated.

The plan was to thin the fluid with respect to high shear rate properties, such as funnel viscosity, plastic viscosity, and yield point. This was accomplished by the addition of water to a pilot sample. The second phase of the plan was to maintain an initial gel strength of 12-13 lb/100 sq ft. This was accomplished by the addition of xanthan gum to increase or maintain low shear rate viscosity based on the 3 rpm dial reading from the Farm model 35A.

Relaxation measurements were run to detect any increase in yield value. Pilot tests with xanthan gum indicated that this remedial treatment was beneficial, so xanthan gum. additions to the system were started. Table 2 is a comparison of the drilling fluid properties before and after biopolymer treatment.

Based on the reductions in funnel viscosity plastic viscosity, yield point, and Brookfield measurements, it can be concluded that the drilling fluid was thinned considerably. However, the low shear rate viscosity in the form of a yield value has markedly improved, as illustrated by the increase in relaxation measurement. Without the implementation of the relaxation measurement procedure, the functional merit of the changes in fluid properties could not have been readily assessed. The well bore responded positively to the fluid alteration by unloading a large volume of cuttings. Logging operations proceeded without further problems.

To illustrate the advantage of the use of xanthan gum to optimize fluid rheology, pilot testing with nontreated bentonite was conducted to increase yield value on the same base sample. Although the relaxation measurement was increased, high shear rate properties (funnel viscosity, plastic viscosity, and apparent viscosity became excessive and would have resulted in pumping problems.

After the total interval depth was reached on the second well in New Mexico, attempts were made to run electric logs. The logging tools failed to go to bottom, stopping over 300 ft from the total measured depth. The following is a Theological overview of the drilling fluid system in use at the time: Funnel viscosity 38 sec/qt, plastic viscosity 11 cp, yield point 19 lb/100 sq ft, gel strengths 5/8 lb/100 sq ft, n value 0.585, and K value 3.0 lb-sec-n/sq cm.

The bit was run back into the hole to circulate out debris. Only a small amount of debris was brought to the surface during circulating operations, even though the Theological profile of the fluid indicated adequate properties for efficient hole cleaning. Annular velocities were in the 100 fpm range, which should have been sufficient to transport debris to the surface.

A relaxation measurement was run on the fluid using a two-speed, hand-crank viscometer. The measurement was recorded as 3 sec to zero from an initial gel strength of 5 lb/100 sq ft. Even though the drilling fluid seemed to have adequate viscosity and velocity to suspend and transport cuttings based on normal viscosity indicators, the relaxation measurement suggested otherwise.

Pilot tests were performed on the fluid. Xanthan gum was added to the test sample at a concentration of 0.5 lb/bbl. Table 3 shows a rheological comparison of the fluid before and after the xanthan addition.

The relaxation measurement was dramatically increased, indicating improved suspension and hole cleaning. The prognosis was clear, and 0.5 lb/bbl xanthan gum was added to the circulating system. After the xanthan addition was completed and the fluid circulated, a marked increase in debris appeared at the shale shaker.

This verified the improved carrying capacity provided with xanthan. Logging operations were then resumed and all tools were successfully run to bottom.

CONCLUSIONS

Conventional field rheological measurement devices do not always accurately indicate hole cleaning and suspension capabilities for drilling fluids.
Brookfield data in the 0.06 sec-1 shear rate range are not always an accurate portrayal of suspension and hole cleaning capacity.
Relaxation measurements can assist the drilling fluids engineer in determining whether a drilling fluid is capable of effectively suspending solids (barite/cuttings) or transporting drill cuttings to the surface.
There are many products which elevate the yield value or low shear rate viscosity of drilling fluids. However, a number of these products also increase high shear rate viscosities which can cause additional problems.
Materials that provide highly shear-thinning viscosity, low n values, and elevated low shear rate viscosity are recommended to balance the suspension and hole cleaning capacity of a fluid with its ability to be pumped.

REFERENCES

Powell, J.W., Parks, C.F., and Seheult, J.M., "Xanthan and Welan: The Effects of Critical Polymer Concentration on Rheology and Fluid Performance," SPE paper No. 22066, presented at the International Arctic Technology Conference, Anchorage, Alas., May 29, 1991.
Engel, H.R., Powell, J.W., and Sigurdson, S.R., "553-ft Gravel Pack Uses Clarified XC Polymer: Development and Application in Beluga River, Alaska," SPE paper No. 19750, presented at the SPE 64th Annual Technical Conference and Exhibition, San Antonio, Tex., Oct. 8-11, 1989.
Carnali, J.O., "A Dispersed Anisotropic Phase as the Origin of the Weak-Gel Properties of Aqueous Xanthan Gum," Unilever Research U.S. Inc., Edgewater, N.J.
Hannote, M, Flores, F., Torres L., and Gallindo, E. "Apparent Yield Stress Estimation in Xanthan Gum Solutions and Fermentation Broths Using a Low-cost viscometer," The Chemical Engineering Journal, Vol. 45, 1991.
Schurz, J., "Rheology Of Polymer Solutions Of The Network Type," Prog. Polym. Science, Vol. 16, pp. 1-53, 1991.
Grey, G.R., and Darley, H.C.H., "Composition and Properties of Oil Well Drilling Fluids," 4th Edition, Gulf Publishing Co., Houston.
Saasen, A., Marken, C., Sterri, N., and Jakobsen, J., "Monitoring of Barite Sag Important in Deviated Drilling," Oil and Gas Journal, pp. 43-50, Aug. 26, 1991.
Estela, B., Rha, C., and Huang, F., "Rheological Properties of Hydrocolloids," Food and Agricultural Engineering Dept., University of Massachusetts, Amhurst, Mass.