Introduction to Underbalanced Drilling

Benefits and Limitations of Underbalanced Drilling

Underbalanced drilling offers a number of important benefits:

• Maintaining wellbore pressure below the reservoir pressure allows reservoir fluids to enter the wellbore, thus avoiding formation damage. Since significant formation damage is avoided, the stimulation requirements during well completion are also reduced, leading to considerable savings.
• During underbalanced drilling there is no physical mechanism to force drilling fluid into the formation drilled. Therefore, lost circulation is kept to a minimum when fractured or high permeability zones are encountered.
• Drilling underbalanced can help in detecting potential hydrocarbon zones, even identifying zones that would have been bypassed with conventional drilling methods.
• Due to the decreased pressure at the bit head, UBD operations demonstrate superior penetration rates compared to conventional drilling techniques. Along with reduced drilling times, an increase in bit life is typically reported.
• Since there is no filter cake around the wellbore wall, the chances of differential sticking are also reduced.
• Since conventional drilling fluids are not used in underbalanced drilling applications, there is no need to worry about disposing potentially hazardous drilling mud.

A combination of all these factors can significantly improve the economics of drilling a well. UBD is often preferred if it reduces formation damage and hole problems, and reduces the cost of stimulation in fractured or moderate/high permeability formations. Moreover, with good mud logging and drilling records, UBD can provide valuable Formation Evaluation data.

Underbalanced drilling also has disadvantages that can prove detrimental to the outcome of the drilling process:

• There is a higher risk of blowout, fire or explosion
• Underbalanced drilling is still an expensive technology. Depending on the drilling fluid used, the cost can be significant, particularly for extended reach horizontal wells
• It is not always possible to maintain a continuously underbalanced condition. Since there is not a filter cake around the wellbore, any instantaneous pulse of overbalance might cause severe damage to the unprotected formation
• UBD has its own unique damage mechanisms, such as surface damage of the formation due the lack of heat conduction capacity of underbalanced drilling fluids
• It is more complicated to model and predict the behavior of compressible drilling fluids
• Borehole collapse is always a concern in UBD operations
• Conventional Measurement-While-Drilling (MWD) cannot be used unless annular aeratin is employed
• System corrosion occurs when air is used as the aerating phase
• Vibration of drill string in gas drilling poses a danger to personnel and equipment.
• Borehole erosion occurs at high annular velocities.

Wellbore stability concerns are one of the main limitations of underbalanced drilling. Borehole collapse as a result of rock stresses is one issue to be considered. The other issue is chemical stability, which is seen in shale and claystone formations. Both these issues can have serious implications in underbalanced drilling. Defining maximum drawdown and reviewing chemical compatibility with the proposed drilling fluids is a key issue in the feasibility of underbalanced drilling.

It is generally believed that borehole-induced shear stresses are responsible for borehole collapse. Low bottom hole pressures lead to an increase in shear stresses acting around the circumference of a well, hence leading to an increased risk of shear failure (Figure 1 (Near-wellbore stress environment).

Failure analysis may be performed using Mohr-Coulomb failure criterion. For deviated boreholes, the in situ stress anisotropy should be considered, and a model of linear elastic plate in plane strain may be used in failure analysis. However, the presence of steep inflow pressure gradients around a well can lead to tensile failure and spalling of the borehole wall (Figure 2 (Near-wellbore stress environment).

There is usually an optimal window of bottom hole pressure that is low enough to allow drilling underbalanced and yet high enough to prevent bore hole collapse. Mathematical models such as that presented by Guo and Ghalambor (2002) can be used to establish this bottom hole pressure window.

Well Control in UBD

Conventional well control methods are based on the constant bottom hole pressure concept, in which the idea is to maintain a constant circulating rate, and to use a choke on the annulus to keep the BHP in the wellbore at or slightly above the formation pore pressure. Well control in UBD is a different proposition, because formation fluid is entering the wellbore even as drilling is proceeding—in effect, a continuous kick is being circulated out of the hole. Surface pressure and BHP must be controlled simultaneously. The controlling factors are choke pressure, liquid volume, liquid properties, gas volume, and gas-to-liquid ratio. These parameters can be altered individually to simultaneously control surface and bottomhole pressure. The concept of controlling BHP by using the choke to modify standpipe pressure is still equally valid.

Before starting UBD, normal standpipe pressure should be observed or calculated. When gas enters the wellbore (whether injected or from the formation), standpipe pressure will drop as hydrostatic pressure in the annulus is reduced. The reduction in standpipe pressure is nearly equal to the reduction in BHP due to gas in the annulus. BHP is controlled by opening or closing the choke to raise or lower standpipe pressure. Introducing gas into the drill pipe changes the density in the drill-pipe column and introduces a small error in calculating BHP. The reason for using this technique is to control BHP near reservoir pressure. This reduces the ability of formation fluid pressure to unload the hole and limits surface pressure to a safe level.

Lag time from choke action to standpipe pressure is variable and results from having to compress or decompress annular fluid. The speed of a pressure wave through a static fluid column equals the speed of sound through the fluid. If the pressure wave is traveling in the opposite direction of flow in a non-static fluid column, the travel time is equal to the speed of sound minus the fluid velocity. Normally, transmission time is on the order of minutes, but in some foam systems, it may take as long as an hour to come to equilibrium. Predicting lag time in a single-phase system (either all liquid or all gas) is a relatively simple calculation. However, predicting sonic velocity or lag time through a multiphase system is a function of the speed of sound through each component, including the void or quality fraction, but it is not a simple weighted average of velocities of the two components. In practice, the sonic velocity through a gas/liquid system is considerably less than the sonic velocity through either of the individual components. To complicate matters, sonic velocity is a function of density.

Similarly we could find the procedure to calculate pressure drop in the system to develop an equation for ECD for different fluids. Therefore, having an accurate model for ECD for various fluids will help to control bottom hole pressure in UBD. Increasing the ECD in a system increases the circulating BHP in a manner opposite from impressed pressure from the choke. The ECD tends to act from bottomhole to surface much like a fluid-density increase. ECD is preferable to using choke pressure in controlling BHP in wells with shallow casing strings. However, deliberate use of ECD as a control tool has the following problems when making connections and on trips:

1. ECD is a function of flow and is lost when flow stops, such as when making connections.
2. The effect of ECD can reverse when pipe is pulled out of the hole; thereby reducing BHP with a swabbing effect.
3. Liquid properties that cause increased ECD also tend to increase gel strength and can cause a pressure surge when breaking circulation.

The effect of ECD as a function of fluid properties is generally expressed in the Newtonian, Power Law, or Bingham Plastic models. Such models are at best imprecise, but are useful tools in predicting how changes in fluid properties will affect ECD. The more flow resistance, the higher ECD will be. In gaseated or conventional fluids, resistance to flow is calculated from readings taken from a viscometer as apparent viscosity, n and K, or as plastic viscosity and yield point (depending on the fluid model). At present, the equations are the best measure of the flow resistance of foam, with an assumed “viscosity” value. The key concept of well control in UBD is to have an accurate measure of ECD and profile of pressure drop in the wellbore.

Hole Cleaning Considerations

Decreased bottom hole pressure typically causes higher penetration rates. However, higher penetration rates can increase the circulating bottom hole pressure and bring the well back to overbalanced conditions. Moreover, due to the annular fluid segregation, there is an increased risk that the wellbore will pack-off, resulting in stuck pipe. In this situation, gas tends to rise while the liquid settles to the bottom of the hole. This is a major cause of increased bottom hole pressures because of the increased fluid density at the sand face.

Large cutting volumes generated by high penetration rates are also difficult to remove. Therefore, penetration rates should be carefully adjusted to ensure sufficient hole cleaning and slug removal.

Inadequate liquid flow rates can cause sticky-hole conditions that result in differential sticking. A decrease in ROP would therefore be needed for the cuttings to be transferred to the surface. A viscosified aqueous phase is an important factor in achieving better ROP.

When drilling with foam and mist, hole cleaning efficiency reaches a limit after a certain level of underbalance, and the drilling rate starts to decrease as illustrated in Figure 3 (Drilling rate as a function of underbalance).

In this situation, an increase in fluid rate is needed to increase the cleaning action and allow a higher rate of penetration.

Formation Damage in UBD

Failing to maintain underbalanced conditions on a constant basis eliminates much of the benefit of UBD. Overbalanced pulses allow whole mud and solids into fractures, filtrate and localized solids to invade the matrix, and can push whole mud and solids into macro porosity. Sequential invasion and filter cake creation, and then removal during UB flow phase may result in cumulatively “deeper” damage occurring.

Common causes for loss of underbalance while drilling are pipe connections, conventional/MWD operations, kill jobs/bit trips, localized depletion, variable/multiple pressure zones, frictional flow effects and poor hole cleaning. Rapid pressure-transient increases, even if the peak value is still less than the bulk reservoir pressure, may still result in near wellbore overbalanced invasion and damage effects. The lower the permeability/pressure of the formation, the more sensitive to rapid pressure-transient increases the system will be.
Methods of reducing overbalanced pulses include:

• Coiled tubing (avoid connections).
• Parasite or concentric string approaches.
• High fraction circulation prior to connections (reduces near-wellbore pressure and may aggravate problems with localized depletion).
• Maintaining annular flow.
• Rapid connections.
• Drill with double/triple pipe stands.

Friction effects during UBD are controlled by the length of the well, borehole and drill string geometry, fluid types/ratios/viscosities, hole cleaning effectiveness, circulation rates of fluids, amount of inflow from the wellbore, stability of flow (dispersed vs. slug), and surface backpressure. Regarding bottomhole pressure (BHP) versus gas flow rate, there are two regimes. In the hydrostatic-dominated region (lower gas flow rates), as gas rate is increased, the reduction in overall system density reduces the hydrostatic head significantly, which reduces the effective BHP. In the friction-dominated region (higher gas flow rates), the gas flow rate becomes so high that increases in gas rate do not reduce overall fluid density enough to counteract the increased friction due to the higher rate. Most operators prefer to operate slightly into the friction-dominated region as this tends to reduce BHP sensitivity to minor oscillations in gas and liquid rate. Other potential damage issues include:

• Countercurrent imbibition: This occurs when imposed underbalanced pressure gradient is not large enough to overcome natural capillary pressure suction forces. This is common in subnormally-saturated tight gas reservoirs or when there is a large vertical standoff from free water contact. To avoid countercurrent imbibition, use a non-wetting base fluid, add an interfacial-tension reducing additive in water-based fluid to reduce trapping potential, and maintain a suitable level of UB pressure to counteract imbibition effects if drilling at the top of a transition zone.
• Macroporous invasion/drainage: This can occur in macroporous features (large fractures) on the low side of a horizontal well during fluid-based drilling when there is insufficient velocity to counteract gravity drainage.
• Glazing and mashing: Glazing is generally only an issue in homogeneous low perm sandstones or with air drilling/pure gas drilling. Glazing issues can be avoided with mist drilling or air hammer/percussion drilling. Mashing is caused by working action of rotating/ sliding drill string and non flush joints on the wellbore surface. It is aggravated by poor hole cleaning and a poorly centralized drill string.
• Trapped fluid saturation evolution: The point can be made that condensate accumulation or critical gas saturation creation will unavoidably occur in any event once normal production operations commence.

Fluid-fluid compatibility concerns that must be addressed include incompatible water-water reactions (scales and precipitates), incompatible oil-oil reactions (sludges, asphaltenes), water-oil emulsions and paraffin deposition.

UBD Pressure Differential

For a water-wet reservoir, the capillary pressure at the sand face always causes the water in the drilling fluid to imbibe into the reservoir. This water imbibition increases water saturation in the reservoir and reduces the effective permeability to the hydrocarbon phase a condition referred to as filtration-induced formation damage. Figure 4 shows an imbibition capillary pressure curve for typical sandstone. Initial water saturation exists in the sandstone at the moment when the rock is being drilled (Point A). The capillary pressure at the initial condition is the highest, which is the potential to cause fast water imbibition into the rock. However, as the water saturation increases in the rock, the capillary pressure drops rapidly (Point B). The water invasion at point B may not cause significant damage to the effective permeability to oil because the water has only occupied the narrow pore space “corners” in the rock. As imbibition continues, water will take over all the small pore space and begin to occupy the larger pore space (Point C). Point C should be considered as a critical point because significant formation damage due to water invasion should occur beyond this point. The capillary pressure at Point C is referred to as the critical capillary pressure. It is this critical capillary pressure that should be balanced by the UBD pressure differential to prevent further water imbibition. If the UBD pressure differential is less than the critical capillary pressure, water imbibition will continue and significant formation damage is expected to occur (Points D and E). Guo and Ghalambor (2006) presented a method to determine the critical capillary pressure for reducing formation damage in underbalanced drilling.

Limiting Technical Factors

Major technical factors that restrict the application of underbalanced drilling techniques are as follows:

• Reduced wellbore pressure gradients can cause hole stability problems.
• Formation of mud rings can block air flow, leading to downhole fires.
• Water causes cuttings to accumulate, possibly causing the drill string to stick. If aerated mud is used rather than air, differential underbalance can be reduced.
• HC’s and air often mix to achieve a flammable range. With a small spark, which can be generated by the contact between the drill string and hard minerals, the risk of fire increases.
• Stable foam condition is not easy to achieve.

Depending on the drilling site location, logistical and economical constraints can be substantial. Similarly, the need for specialized drilling equipment can also render underbalanced operations uneconomical.

“Even though the cost of drilling underbalanced could be more expensive than conventional overbalanced drilling, due to the increased ROP and reduced formation damage, it often turns out to be the more cost-effective drilling technique”.