800 4300 2500 2200

normally pressured formations, which have a self-supporting structure of solid particles (so the pore pressure depends only on the weight of the overlying pore fluids), and abnormally pressured or geopressured formations, which are not fully compacted into a self-supporting structure (so the pore fluids must bear the weight of some or all of the overlying sediments as well as the weight of the overlying fluids). The hydrostatic pressure gradient of formation fluids varies from 0.43 psi/ft to over 0.52 psi/ft (0.1 to 0.12 kg/cm2/m), depending on the salinity of the water.

The bulk density of partially compacted sediments increases with depth, but an average SG of 2.3 is usually accepted, so that the overburden (or geostatic or litholostatic) pressure gradient is about 1 psi/ft (0.23 kg/cm2/m), and the pore pressure of geopressured formations is somewhere between the normal and the overburden pressure gradients, depending on the degree of compaction.

Besides controlling pore fluids, the pressure of the mud column on the walls of the hole helps maintain borehole stability. In the case of plastic formations, such as rock salt and unconsolidated clays, the pressure of the mud is crucial

The buoyant effect of the mud on the drill cuttings increases with its density, helping transport them in the annulus, but retarding settling at the surface. Very rarely is an increase in mud density justified as a means of improving cutting carrying capacity.

In the interest of well safety, there is a natural tendency to carry a mud density well above that actually needed to control the formation fluids, but this policy has several major disadvantages. In the first place, excessive mud density may increase the pressure on the borehole walls so much that the hole fails in tension. This failure is known as induced fracturing.

In induced fracturing, mud is lost into the fracture so formed, and the level in the annulus falls until equilibrium conditions are reached. The problem of maintaining mud density high enough to control formation fluids, but not so high as to induce a fracture, becomes acute when normally pressured and geopressured formations arc exposed at the same time. Under these circumstances, it is generally necessary to set a string of casing to separate the two zones. Several methods have been developed for predicting the occurrence of geopressures.1 Knowledge of the expected pore pressure and fracture gradients: 15 usually enable casing to be set at exactly the right depth, thereby greatly reducing the number of disaster wells.

Another disadvantage of excessive mud densities is their influence on drilling rate (rate of penetration). Laboratory experiments and field experience have shown that the rate of penetration is reduced by mud overbalance pressure (the differential between the mud pressure and the pore pressure when drilling in per meable rocks)5'6 7-8 9 and by the absolute pressure of the mud column when drilling rocks of very low permeability. A high overbalance pressure also increases the risk of sticking the drill pipe (see Chapter 9).

Lastly, excessive mud densities are a disadvantage because they unnecessarily increase mud costs. Mud costs are not a very important consideration when drilling in normally pressured formations, because adequate densities are automatically obtained from the formation solids that are dispersed into the mud by the action of the bit. Mud densities greater than about 11 lb/gal (1.32 SG) cannot be obtained with formation solids because the increase in viscosity is too great Higher densities are obtained with barite which has a specific gravity of about 4.25, as compared to about 2.6 SG for formation solids, so that much less of solids by volume is required to obtain a given density. Mud costs are increased not only by the initial cost of the barite, but also, and to a greater extent, by ihc increased cost of maintaining suitable properties, particularly flow properties Because of the incorporation of drilled solids, the viscosity continuously increases as drilling proceeds, and must be reduced from time to time by the addition of water and more barite to restore the density.

Flow Properties

The flow properties of the drilling fluid play a vital role in the success of the drilling operation. These properties are primarily responsible for removal of the drill cuttings, but influence drilling progress in many other ways. Unsatisfactory performance can lead to such serious problems as bridging the hole, filling the bottom of the hole with drill cuttings, reduced penetration rate, hole enlargement, stuck pipe, loss of circulation, and even a blowout.

The flow behavior of fluids is governed by flow regimes, the relationships between pressure and velocity. There are two such flow regimes, namely laminar flow, which prevails at low flow velocities and is a function of the viscous properties of the fluid, and turbulent flow, which is governed by the iner-tial properties of the fluid and is only indirectly influenced by the viscosity. As shown in Figure 1-2, pressure increases with velocity increase much more rapidly when flow is turbulent than when it is laminar.

Laminar Flow. Laminar flow in a round pipe may be visualized as infinitely thin cylinders sliding over each other. The velocity of the cylinders increases from zero at the pipe wall to a maximum at the axis of the pipe. The difference in velocity between any two such cylinders, divided by the distance between them, defines the shear rate. The axial force divided by the surface area of a cylinder def ines the shear stress. The ratio of shear stress to shear rate is called the viscosity, and is a measure of the resistance to flow of the fluid. The unit of viscosity is the poise; the shear stress in dynes/cm2 divided by the shear rate in reciprocal seconds gives the viscosity in poises. The unit employed in mud viscometry is the centipoise (cp), which is one hundredth of a poise. (See Table A-l for the SI unit.)

A plot of shear stress versus shear rate is known as a consistency curve, or flow model, the shape of which depends on the nature of the fluid being tested: with fluids that contain no particles larger than a molecule (e.g., water, salt solu-

Flow Velocity

Figure 1-2. Schematic diagram of laminar and turbulent flow regimes.

Flow Velocity

Figure 1-2. Schematic diagram of laminar and turbulent flow regimes.

tions, oil, glycerine), the consistency curves are straight lines passing through the origin. Such fluids are called Newtonian because their behavior follows the laws first laid down by Sir Isaac Newton. The viscosity of a Newtonian fluid is defined by the slope of its consistency curve (see Figure 1-3). Since the viscosity of a Newtonian fluid does not change with shear rate, a viscosity determined at a single shear rate may be used in hydraulic calculations involving flow at any other shear rate.

Suspensions such as drilling muds that contain particles larger than molecules in significant quantities do not conform to Newton's laws, and thus are classified under the general title of non-Newtonian fluids. The shear stress/shear rate relationship of non-Newtonian fluids depends on the composition of the fluid. Clay muds having a high solids content behave approximately in accordance with the Bingham theory of plastic flow, which postulates that a finite stress must be applied to initiate flow, and that at greater stresses the flow will be Newtonian. The consistency curve of a Bingham plastic must therefore be described by two parameters, the yield point and the plastic viscosity, as shown in Figure 1-3. The shear stress divided by the shear rate (at any given rate of shear) is known as the effective or apparent viscosity. Figure 1-4 shows that effective viscosity decreases with increase in shear rate, and is therefore a valid parameter for hydrau lie calculations only at the shear rate at which it was measured. Indeed, as shown by Figure 1-5, effective viscosity is not even a reliable parameter for comparing the behavior of two different muds.

The decrease in effective viscosity with increase in shear rate is known us shear thinning, and normally is a desirable property, because the effective viscosity will be relatively low at the high shear rates prevailing in the drill pipe, thereby reducing pumping pressures, and relatively high at the low shear rates prevailing in the annulus, thereby increasing cutting carrying capacity. The ratio of the yield point to the plastic viscosity (known as the YP/PV ratio) is a measure of thinning: the higher the ratio the greater the shear thinning.

Drilling muds which consist of polymers and little or no particulate solids behave at high shear rates as though they had a yield point, but actually the consis-

Shear Rate

Figure 1-3. Ideal consistency curves for common flow models.

Shear Rate

Figure 1-3. Ideal consistency curves for common flow models.

Shear Rate

Figure 1-5. Comparison of effective viscosities at two shear rates. At shear rate 1, mud A has the higher viscosity, but the order is reversed at shear rate 2.

Shear Rate

Figure 1-5. Comparison of effective viscosities at two shear rates. At shear rate 1, mud A has the higher viscosity, but the order is reversed at shear rate 2.

stant at low rates of shear. Muds have a rather indefinite yield point which is less than would be predicted by extrapolation of shear stresses measured at high shear rates. Figure 1-3 compares the consistency curves of the three flow mod-

The fact that the consistency curve of clay muds intercepts the stress axis at a value greater than zero indicates the development of a gel structure. This structure results from the tendency of the clay platelets to align themselves so as to bring their positively charged edges towards their negatively charged basal surfaces. This interaction between the charges on the platelets also increases the effective viscosity at low rates of shear, thereby influencing the values of n and K, and is responsible for the formation of a gel when agitation ceases.

The gel strength of some muds, notably fresh water clay muds, increases with time after agitation has ceased, a phenomenon that is known as thixotropy. Fur-

thermore, if after standing quiescent the mud is subjected to a constant rate of shear, its viscosity decreases with time as its gel structure is broken up, until an equilibrium viscosity is reached. Thus the effective viscosity of a thixotropic mud is time-dependent as well as shear-dependent.

Turbulent Flow. Flow in a pipe changes from laminar to turbulent when the flow velocity exceeds a certain critical value. Instead of layers of water sliding smoothly over each other, flow changes locally in velocity and direction, while maintaining an overall direction parallel to the axis of the pipe. Laminar flow may be compared to a river flowing smoothly over a plain, and turbulent flow to flow over rapids where interaction with irregularities on the bottom causes vortices and eddies.

The critical velocity for the onset of turbulence decreases with increase in pipe diameter, with increase in density, and with decrease in viscosity, and is expressed by a dimensionless number known as Reynolds number. With most drilling muds the critical value of the Reynolds number lies between 2000 and 3000.

The pressure loss of a fluid in turbulent flow through a given length of pipe depends on inertial factors, and is little influenced by the viscosity of the fluid. The pressure loss increases with the square of the velocity, with the density, and with a dimensionless number known as the Fanning friction factor, which is a function of the Reynolds number and the roughness of the pipe wall.

Control of Flow Properties at the Well

It is comparatively easy to formulate a mud with suitable properties; it is much more difficult to maintain those properties while drilling, because of dispersion of drilled solids into the mud, adsorption of treating agents by drilled solids, and contamination by formation fluids. Maintaining suitable properties is the job of the mud engineer, who should visit the well at least once a day, check the flow properties and other properties, and recommend a suitable treatment.

The influence of drilling fluids on well performance is most critical in the pipe/hole annulus; therefore, mud samples are taken directly from the flow line, and tested immediately before any thixotropic change takes place.

For routine viscosity measurements, mud engineers use a two-speed concentric-cylinder viscometer, such as the Fann VG meter. This instrument enables the plastic viscosity (PV), the yield point (YP), and the apparent viscosity (AV) at 600 rpm, to be quickly obtained from three elementary calculations. Gel strengths and the power law constants, n and K, may also be calculated (see Chapter 5). The gel strength is usually measured 10 seconds {initial gel strength) and 10 minutes after agitation ceases. Because of the indefinite nature of the yield point at low rates of shear (see curve for typical drilling mud in Figure 1-3), the initial gel strength is often used instead of the yield point for flow properties in the annulus.

Knowledge of these parameters provides the information necessary for day to day control of the mud rheology. The PV and K depend largely on the bulk volume of solids in the mud and on the viscosity of the suspending liquid, whereas the YP and the gel strengths depend more on the presence of colloidal clays, and on contamination by inorganic salts. Either the YP/PV ratio, or the flow index n, may be used to characterize the shear thinning properties of the mud. The difference between the initial gel strength and that taken after a 10-minute rest period may be used to judge how thick the mud will get during round trips

The plastic viscosity is reduced, if necessary, by the addition of water, or by the mechanical separation of excess solids. Excessively high yield points or gel strengths are reduced by the addition of certain high molecular weight compounds, known as thinners. The most commonly used thinner nowadays is chromelignosulfonate solubilized with caustic soda, but solubilized lignite and polyphosphates may also be used.

If the gel strength is too low, it may be increased by adding bentonite. Ideally, the gel strength should be just high enough to suspend barite and drill cuttings when circulation is stopped. Higher gel strengths are undesirable because they retard the separation of cuttings and of entrained gas at the surface, and also because they raise the pressure required to re-establish circulation after changing bits. Furthermore, when pulling pipe, a high gel strength may reduce the pressure of the mud column beneath the bit because of a swabbing action. If the reduction in pressure exceeds the differential pressure between the mud and the formation fluids, the fluids will enter the hole, and possibly cause a blowout. Similarly, when running pipe into the hole, the downward motion of the pipe causes a pressure surge which may, when conditions are critical, cause induced fracturing with consequent loss of circulation. Methods have been developed for the calculation of the magnitude of these pressure surges.10 " 12

Thinners used to reduce the gel strength of fresh-water or low-salinity muds have an unfortunate secondary effect: The replacement of calcium—or other polyvalent cations on the clay cuttings—by the sodium used to solubilize the thinner tends to disperse the clay into small particles. Some of these particles are not removed at the surface, and are recycled again and again until they are reduced to colloidal size. This action makes the control of viscosity very difficult and expensive when drilling through colloidal clay formations with a fresh water mud. Because of this adverse reaction, it is imperative that the mud engineer make pilot tests on a small sample of mud to ensure that no more thinner than absolutely necessary is added to the mud.

Having completed his tests, the mud engineer recommends a treatment to be added to the mud every time the Marsh funnel viscosity rises above a specified value. The crew measures the viscosity with a Marsh funnel by the time in seconds for one quart (or one liter) to be discharged from a full funnel. The values obtained are only relative to the mud being tested, but are adequate for the crew to carry out the instructions of the mud engineer, and for him to observe the results of his treatment the next day.

Filtration Properties

The ability of the mud to seal permeable formations exposed by the bit with a thin, low-permeability filter cake is another major requirement for successful completion of the hole. Because the pressure of the mud column must be greater than the formation pore pressure in order to prevent the inflow of formation fluids, the mud would continuously invade permeable formations if a filter cake were not formed.

For a filter cake to form, it is essential that the mud contain some particles of a size only slightly smaller than that of the pore openings of the formation. These particles, which are known as bridging particles, are trapped in the surface pores, while the finer particles are, at first, carried deeper into the formation. The bridged zone in the surface pores begins to trap successively smaller particles, and, in a few seconds, only liquid invades the formation. The suspension of fine particles that enters the formation while the cake is being established is known as the mud spurt. The liquid that enters subsequently is known as the filtrate.

The rate of filtration and the increase in cake thickness depend on whether or not the surface of the cake is being subjected to fluid or mechanical erosion during the filtration process. When the mud is static, the filtrate volume and the cake thickness increase in proportion to the square root of time (hence, at a decreasing rate). Under dynamic conditions, the surface of the cake is subjected to erosion at a constant rate, and when the rate of growth of the filter cake becomes equal to the rate of erosion, the thickness of the cake and the rate of filtration remain constant. In the well, because of erosion by the mud and because of mechanical wear by the drill string, filtration is dynamic while drilling is proceeding; however, it is static during round trips. All routine testing of filtration properties is made under static conditions because dynamic tests are time-consuming and require elaborate equipment. Thus, filtration rates and cake thicknesses measured in surface tests correlate only approximately to those prevailing down-hole and can be grossly misleading. The permeability of the filter cake—which may readily be calculated from static test data—is a better criterion because it is the fundamental factor controlling both static and dynamic filtration.

The permeability of the filter cake depends on the particle size distribution in the mud and on the electrochemical conditions. In general, the more particles there are in the colloidal size range, the lower the cake permeability. The presence of soluble salts in clay muds increases the permeability of the filter cake sharply, but certain organic colloids enable low cake permeabilities to be obtained even in saturated salt solutions. Thinners usually decrease cake permeabilities because they disperse clay aggregates to smaller particles.

The filtration properties required for the successful completion of a well depend largely on the nature of the formations to be drilled. Stable formations with low permeabilities, such as dense carbonates, sandstones, and lithified shales, can usually be drilled with little or no control of filtration properties. But many shales are water-sensitive, i.e., on contact with water, they develop swelling pressures which cause caving and hole enlargement. Sealing of incipient fractures by mud filter cake will help control the caving, but the type of mud used and the chemical composition of its filtrate are more important factors. Superior hole stabilization is obtained with oil-base mud when the salinity of the filtrate is adjusted to prevent swelling pressures from developing in the shales.

In permeable formations, filtration properties must be controlled in order to prevent thick filter cakes from excessively reducing the gauge of the borehole. Furthermore, thick filter cakes may cause the drill pipe to become stuck by a mechanism known as differential sticking.1314 This phenomenon occurs when part of the drill string bears against the side of the hole while drilling, and erodes away part of the filter cake. When rotation of the pipe is stopped, the part of the pipe in contact with the cake is isolated from the pressure of the mud column, and subject only to the pore pressure of the filter cake. The differential pressure thus created may be great enough to prevent the pipe from being moved. Sometimes, the pipe can be freed by spotting oil around the stuck section, but if this procedure fails, an expensive fishing or sidetracking job is entailed. The risk of stuck drill pipe may be reduced by using a mud that lays down a thin, tough filter cake; by maintaining the lowest possible mud density so as to minimize the differential pressure, and by adding a lubricant to the mud to reduce adhesion between the pipe and the cake.1516 Stuck drill pipe is seldom experienced when oil muds are in the hole because they provide thin filter cakes and excellent lubricating qualities.

Good filtration properties are also necessary when drilling in unconsolidated sands, which will slump into the hole unless protected by the rapid formation of a filter cake.

Both filtration rate and the mud spurt must be minimized when penetrating potentially productive formations, because productivity may be reduced by any one of four mechanisms. First, the permeability of a reservoir rock containing indigenous clays may be reduced by the swelling of these clays when they come in contact with the invading filtrate, or by dispersion and transport of the clays or other indigenous fines. Particles thus transported are subsequently caught at bottlenecks in the flow channels, and greatly reduce the permeability of the rock. Second, the pressure of some reservoirs is not great enough to drive ail of an aqueous filtrate out of the pores of the rock when the well is brought into production. The filtrate remaining in the pores reduces the space available for the flow of oil or gas, thus causing what is known as waterblock. Third, fine particles from the mud, carried in during the spurt stage, may plug flow channels. Fourth, mutual precipitation may occur between salts dissolved in the filtrate and those in the interstitial formation water.

When an oilfield is first discovered, extensive laboratory testing is recommended in order to formulate a mud that will not impair productivity in subsequent wells (see Chapter 10). Such tests should be made on cores of the reservoir in question, with a petroleum distillate, e.g., diesel oil, and actual or synthetic formation interstitial brine. Impairment caused by invasion of particles from the mud may be avoided by ensuring that sufficient particles of the size required to bridge the port openings are present in the mud, or by formulating a mud whose particles arc soluble in oil or acid, or are biologically degradable. If the formation contains indigenous clays, a mud whose filtrate inhibits the swelling and dispersion of these particular clays is required. If waterblock is a problem, oil muds should be tested.

Filtration performance in the well is routinely judged by means of the standard API filtration test. In this test, the mud is subjected to static filtration throuyh filter paper for 30 minutes, and the volume of filtrate and the cake thickness arc then measured. In planning a mud program, a certain maximum API filter loss is often specified, with the thought that as long as the filter loss is kept below this figure, adequate control of downhole filtration properties will be maintained. From what has already been said in this section, it will be appreciated that reliance on API filter loss alone for control of downhole filtration performance is a highly dubious procedure, which can lead to poor drilling and production performance and greatly increased well costs. A major problem is that downhole cake thickness depends to a considerable extent on cake erodability, which does not affect the static filter loss. For example, laboratory tests have shown that emulsi-fication of diesel oil in water base muds decreases the API filter loss, but sharply increases the dynamic rate, because of cake erodability. Tests have also shown that a commercial additive that decreases the API filter loss might have little or no effect on the dynamic rate, while with another additive the opposite might be true (see Chapter 6). It is essential, therefore, that filtration control agents be evaluated at least once in local muds and under local downhole conditions in a dynamic tester.

For practical reasons, the API filter test must be used at the well site, but the results should be interpreted in the light of laboratory data. Also, the particular reason for filtrate control should be kept in mind. For example, if differential sticking is the problem, the filter cake thickness is more important than the filter loss: or if productivity impairment is the reason for filter loss control, the salinity of the filtrate or having sufficient bridging particles may the critical property

The relative acidity or alkalinity of a liquid is conveniently expressed as pH. Defined as the negative logarithm (to the base 10) of the hydrogen-ion concentration, pH units decrease with increasing acidity by a factor of 10. For example, the hydrogen-ion concentration of a solution having a pH of 3 is ten times that of a solution of pH 4. At pH of 7, the hydrogen-ion concentration is equal to the hydroxy!-ion concentration and the liquid is neutral, as with pure water. Above pH 7, the hydroxyl-ion concentration increases by a factor of 10 with each pH

unit; thus, the hydroxyl-ion concentration at pH 11 is ten times that at pH 10 (hydrogen-ion concentration is one-tenth).

The optimum control of some mud systems is based on pH, as is the detection and treatment of certain contaminants. A mud made with bentonite and fresh water, for example, will have a pH of 8 to 9. Contamination by cement will raise the pH to 10 to 11, and treatment with an acidic polyphosphate will bring the pH back to 8 or 9. Other reasons for pH control include maintenance of lime-treated muds, mitigation of corrosion, and effective use of thinners.

Measurement of pH is routinely made by comparing the color developed on immersing a paper strip impregnated with certain dyes (indicators) with the color of reference standards. If the liquid has a high concentration of dissolved salts, or is deeply colored (such as by tannins and lignite), the colorimetric method is not satisfactory, but an electrometric method employing the glass electrode can be used to give reliable results in most muds. If the sodium-ion concentration is very high, a special glass electrode may be needed.

Alkalinity

Alkalinity measurements are made to determine the amount of lime in lime treated muds. The mud is titrated to determine the total amount of lime, soluble and insoluble, in the system (Pm) The filtrate is titrated to determine the amount of lime in solution (Pf). The amount of undissolved lime is calculated from Pm-Pf. Measurements of the alkalinity of water samples, and of filtrates of very lightly chemically treated muds, can be used to calculate the concentration of hydroxy 1 (OH), carbonate (CO?), and bicarbonate (HCO?) ions in solution.

Cation Exchange Capacity: Methylene Blue Test

The methylene blue test serves to indicate the amount of active clay in a mud system or a sample of shale. The test measures the total cation exchange capacity of the clays present and is useful in conjunction with the determination of solids content as an indication of the colloidal characteristics of the clay minerals. Similarly, shale cuttings can be characterized and some estimations can be made regarding mud-making properties and possible effects on hole stability. Organic materials, if present in the sample, are destroyed by oxidation with hydrogen peroxide. The sample is titrated with standard methylene blue solution until the adsorptive capacity is satisfied, as shown by the appearance of a blue color in the water in which the sample is suspended. If other adsorptive materials are not present in significant amounts, the bentonite content can be estimated, based on an exchange capacity of 75 milliequivalents per 100 grams of dry bentonite.

Electrical Conductivity

The resistivity of water muds is measured and controlled, whenever desired, to permit better evaluation of formation characteristics from electrical logs. Sal'

is used to lower the resistivity. Fresh water is the only means of raising resistivity. The determination of resistivity involves the measurement of resistance to the flow of electrical current through a sample of known configuration. In the direct-reading resistivity meter, the resistance measurement is converted to re sistivity in ohm meters.

The electrical stability test is used as an indication of the stability of emulsions of water in oil (oil muds): A probe fitted with electrodes is immersed in the sample; the voltage imposed across the electrodes is increased until a predetermined amount of current flows; and the voltage at this breakdown point is reported as the emulsion stability.

Lubricity

One of the functions of the drilling fluid is to lubricate the drill string. The requirement for lubrication is especially critical in directional and crooked holes, and in avoidance of wall-sticking.

The Tiinken lubricant tester has been modified and a method devised to furnish comparative results on which to base recommendations for treatment of the mud.

Corrosivity

Corrosion has been found to be the principal cause of drill pipe failures. Corrosion of the surface of the drill pipe is monitored by placing steel rings in the tool-joint box recess at the end of the pin and determining the loss in weight after a selected time of exposure to the drilling fluid. Observation of the type of corrosive attack is frequently more significant than observation of the loss in weight. For example, pitting may result in a relatively small weight loss in comparison to generalized corrosion, but it may be responsible for far more instances of drill pipe failure. Monitoring by steel rings reveals how pitting may relate to failure, but gives no information on hydrogen embrittlement, stress corrosion cracking, or other forms of fracture.

In planning the drilling fluids program, attention must be given not only to the possible corrosive effects of the drilling fluid, but also to the effects of corrosion inhibitors on the drilling fluid itself. Some corrosion inhibitors, for example, may severely affect the properties of water muds. Sources of corrosive agents, their composition, and methods of counteracting them are factors to be considered in selecting the drilling fluid.

0 0

Post a comment