Effective gravel packing considers long-term production

SAND CONTROL-2

M. Babs Oyeneyin
Robert Gordon University
Aberdeen

With proper design and installation, gravel-pack completions can yield optimum long-term performance.

The best strategy should therefore involve a rational compromise between achieving optimum sand control and optimum long-term performance.

Total sand exclusion is achieved at the expense of reducing a well's production capacity.

This second in a three-part series of articles, which started in OGJ Jan. 26, 1998, discusses methods for achieving an optimal gravel pack.

Gravel packing

While gravel packing in theory can be used for any well, the advent of extended reach and horizontal wells with very long lateral sections places hydraulic limitations on the installation of gravel packs.

In general, horizontal wells are categorized into either short radius (with build-up rate of 50-500°/100 ft and horizontal sections of up to 200 ft), medium radius (with build-up rate of 6-50°/100 ft and horizontal section of between 200 to 2,000 ft), and long radius (build-up rate 2-6°/100 ft and horizontal section of about 1,000 to over 4,000 ft.

Industry experience is rather limited when gravel packing further than 700 ft in horizontal wells. Published information indicates that there is a limitation on the gravel packing length that can be achieved.^{1 2} Nevertheless, many integrated service companies do now lay claim to the fact that they can gravel pack any length of well.

A recent industry experience in the North Sea has reported the successful gravel packing of 2,000 ft of lateral section.³ Theoretically, based on fluid dynamics principles, this is possible if the appropriate optimum well bore configuration and flow conditions are in place and supported by proper placement tools. Nevertheless, this may require elaborate planning and a strategy outside of existing general field experience.

Design and installation of gravel packs involve the following three steps:

Well preparation
Design of the gravel pack made up of formation sampling, choice of appropriate gravel, and placement design including:
· Placement process/technique
· Carrier fluid type and properties
· Placement conditions
· Downhole configuration
Post-placement evaluation.

Well preparation

Proper well preparation prior to installing the gravel pack is a key factor for ensuring long-term optimum performance with minimum damage. Likewise, it is important to have carrier fluid leakoff into the formation during the gravel packing process to ensure proper pack-sand dehydration.

Specific attention must be given to the drilling fluids, especially the chemical additives (lost circulation materials, etc.) for the production zone, the completion/cleanup fluids, perforating and perforation conditions, and most importantly, well bore as well as surface facility cleanness.

Completion fluids must be compatible with the formation to prevent any damage and reduce leakoff

Drilling/completion fluids

One should use, as much as possible, highly shear-thinning, brine-based drilling/completion fluids in the production zone. For filter cake cleanup, a high-shear thinning fluid is likely to achieve turbulence at minimum ECD (equivalent circulating density).

Minimum filter cake also ensures subsequent successful borehole cleaning, especially for open hole completions.

Sized salts as fluid-loss control agents will be advantageous to subsequent filter cake removal prior to completion and therefore are recommended. They dissolve easily in unsaturated brines and are easily removed by acid wash.

Well control and fluid compatibility are a prime target during completion. Therefore, the choice of brine will depend on desired ECD. Table 1 [5,262 bytes] lists common brines.

Open hole gravel packs

For single, partially consolidated pay sections, especially in vertical and deviated wells where selectivity or zonal isolation is not essential, open hole gravel packs would be the best option for a completion. Depending on the degree of partial consolidation, external gravel packs can be applied. For such conditions, underreaming is recommended for effective pack thickness.

For highly unconsolidated sand sections, especially for horizontal wells, completions may be limited to cased-hole completions due to hole instability. Where hole stability can be maintained, however, open hole completion would be the best option.

For unconsolidated sands, a perforated liner can support the well bore. In this case, filter cake removal with a low-viscosity, shear-thinning undersaturated brine circulated in turbulence is essential to cake removal and dissolution of sized salts. For more stubborn cases, acid washing or solvent breaking combined with mechanical scrapping is essential.

Cased-hole gravel packs

For cased-hole gravel packs, the following prepacking preparation must be carried out:

Removal of the filter cake is very important before setting casing.
Mechanical scraping of casing, combined with solvent dissolution to remove any debris, is recommended for workstring and casing cleaning.
Perforations must penetrate deep into the formation, beyond the near well bore damage region.
Perforation tunnels must be essentially free of debris before being gravel packed. Cleaning techniques for partially unconsolidated sands include underbalanced perforating, perforation washing, or backsurging. To minimize hole instability and massive sand intrusion, only perforation washing techniques are recommended for cleaning perforations in highly unconsolidated sands.
For optimum flow capacity, the perforation size and density must be sufficient to minimize resistance to flow. Analysis of perforation packing efficiency^{4 5} shows that large-diameter perforations (minimum 0.5 in.) and deep penetration with over 8 in. of tunnel length is recommended to ensure optimum gravel placement and eventual performance. Likewise, at least 8 shots/ft but a maximum of 12 shots/ft is recommended. This guarantees that all perforations are gravel-filled.

For effective placement and stability, the perforation phasing should be 90° with the so-called "quarter-to-three" configuration. For horizontal and highly deviated wells, it is suggested that one should not perforate the high side of the cased hole because there is no guarantee of gravel retention on the high-side perforation.

Gravel-pack design

The first step in the design of gravel packs is the selection of appropriate gravel.

The oil industry has adopted the ratio of mean gravel size to the mean formation-sand size as a basis for selecting gravel size irrespective of the gravel type, well/completion type, and produced fluids.

The Saucier formula or a variation of this is the most popular formula used for gravel size selection. However field experiences have shown that the Saucier formula may be too general and limited in that it does not account for type of gravel/proppant-conventional or synthetic, well bore configuration, type of produced fluids, and operating conditions.

The stability and bridging effectiveness of a gravel pack are complex physical phenomena that are functions of operational, environmental, and geometric parameters such as:

Formation sand size distribution and sorting
Type/fluid properties
Pack structure/thickness
Drawdown
Perforation size and penetration
Gravel size
Production rate and cumulative production
Hole angle
Perforation flow velocity
Sand density
Gravel/formation sand shape/angularity.

Thus, the best design philosophy for gravel/gravel size selection is to make a rational compromise between bridging effectiveness and optimum production capacity at the prevailing operating conditions over the economic life of the well.

Recent studies on gravel sizing^6-8 show that the best design criteria for optimum gravel selection must account for the following:

Overall formation sand size distribution, sorting, and shape
Gravel type, shape, and structure
Bridging efficiency of the gravel pack at the prevailing operating conditions and well bore configuration
Prevailing pore blocking mechanism.

Such design must also account for the impact of the gravel size selection on both the short-term and long-term well performance.

The prevailing pore blocking mechanism in a gravel pack has been identified as:^{7 9}

No interaction-Representing free passage of fines due to large pores.
Pore filling-Gradual deposition of particles in the pore spaces/pore walls resulting in hyperbolic impairment profile.
Combined internal bridging and single pore blocking-Initial pore blocking followed by accelerated build-up of particles in the pore spaces.
Shallow internal bridging-Phenomena occurring in the superficial part of the sand close to the sand face that can easily be cleaned by back production.
No invasion-Absolute stoppage or external cake formation. Suitable for filter cake formation.

Results from recent extensive studies^{6 7} show that the optimum gravel size selection required for effective bridging and optimum performance for any given condition is that which ensures shallow internal bridging pore blocking mechanism against the formation sand/fines migration. This mechanism guarantees optimum bridging and promotion of a secondary filter by the formation sand.

The design for shallow internal bridging and consideration of overall sand size distribution requires the use of smaller gravel than predicted by the current Saucier rule.

The results of the recent work^{6 7} formed the foundation of the semiempirical mathematical models that are the key features of the Gravsize computer package developed for gravel-pack design and performance evaluation.¹⁰ With the package it is possible to:

Select appropriate gravel and two alternatives
Predict the bridging efficiency and corresponding sand production level
Determine pack impairment and effect on overall well performance over time.

Based on the combined results from the software and experimental observations of the bridging phenomena, the following three guidelines are recommended for a range of operating conditions:

For poorly sorted sands; gas wells; open hole, high-oil producers; and cased-hole gravel packs, the gravel size should be about one step or two below the Saucier rule. For example, d₅₀(gravel) = 3 to 4 3 d₅₀ sand.
For low producers, uniformly sorted sands in open hole completions, the gravel size should be equivalent to the Saucier predictions.
For synthetic gravels, the gravel size should be a step higher than the Saucier rule. For example, d₅₀ (gravel) = 6 to 7 3 d₅₀ sand.

Gravel type

Although synthetic gravel provides more-effective bridging against sand migration and possesses higher permeability, it is much more expensive than conventional gravels. Therefore, its application should be limited to the prepacking of perforations in a two-stage gravel pack.

Likewise, for extended reach and long-horizontal sections, low-density gravel substitute (LDGS) would be useful in improving packing efficiency.

Placement design

The objective of optimum gravel placement design is to achieve high perforation and annular packing efficiencies. This objective becomes increasingly difficult to achieve when gravel packing highly deviated and horizontal wells. Thus, the main focus in the gravel-pack placement design should be on:

Interval length
Well bore configuration
Installation techniques and carrier fluid.

Interval length

There are no major limitations to the completion length of vertical intervals. However, for highly deviated, extended-reach, and horizontal wells, the industry experience is limited.

Except for hydraulic limitations, no theoretical restriction is apparent for the horizontal section length especially if the equilibrium bank concept proposed by Gruesbeck¹¹ and Peden, et al.,¹² is adopted. This concept proposed an optimum relationship between the washpipe and screen size.

However, Forest indicates potential limitations to the interval length.¹ His equation for estimating the maximum pack length for highly deviated and horizontal wells is:

Lmax = (PR - NPR) 3 42/Qw
where:
Lmax = Maximum pack length, ft
PR = Surface pump rate, bbl/min
NPR = Minimum net pump rate, bbl/min
Qw = Carrier fluid leakoff, gpm.

The full-scale studies by Alexander, et al.,² indicate that gravel-pack tool limitations and analysis based on Forest¹ makes it impossible to gravelpack horizontal intervals greater than 500 ft. Thus, for longer intervals, a multistage gravel pack may be required.

The best alternative, however, is to use screen systems barefoot (Fig. 1 [121,582 bytes]).

Screen, washpipe sizes

The following are the recommendations for screen gauge, sceen OD, and washpipe OD:

Screen gauge-The recommended screen gauge is 0.5 times the smallest grain size of the gravel (100 percentile size).
Screen OD-There are no hard and fast rules regarding the screen OD. The screen should be large enough for tools to pass but also have a reasonable diameter for effective pack thickness. Table 2 [8,195 bytes] shows some recommended screen sizes.
Washpipe OD-For effective gravel packing of the screen/casing annulus, the annular clearance between the washpipe and the screen ID is the most important factor. Previous work on packing high-angle wells^{10 11} recommended a washpipe OD to screen ID ratio of 0.8.

Placement technique

Gravel placement design and techniques are rather simple in vertical wells, but more complex in highly deviated and horizontal wells. But in both cases, the placement design involves:

Selecting the most appropriate placement technique
Specifying fluid/slurry properties in terms of carrier fluid type, density, viscosity, and slurry concentration
Determining operating conditions in terms of pumping rate.

A number of gravel packing placement techniques are available. The conventional washdown method with the modified version of the auger screen technique involves prepacking the bore with gravel and then augering the screen through the pack, thus pushing gravel into the perforations.

In the conventional crossover tool method, the crossover tool allows circulating the gravel slurry. The slurry circulates through the crossover tool into the screen/casing annulus and fluid returns through the screen and up the washpipe to the surface.

For cased-hole gravel packs, the procedure can be by:

Prepacking the perforations with gravel preferably with synthetic gravel and then conventionally packing the annulus with either conventional pack sand or low-density gravel substitute (LDGS). This is the best option.
Conventionally packing the annulus and bull-heading the slurry into the perforations.

Either of these two techniques can include low-viscosity completion fluids with low gravel concentration (about 1 ppg) or highly viscous gels with high gravel concentration.

Viscosifying agents such as HEC (Hydroxyethylcellulose) and CMC (carboxymethyl cellulose) are common.

Slurrypacks

Slurrypacks have been made with crosslinked hydropac systems formulated with highly viscous crosslinked gels and high gravel concentrations.

This fluid is sometimes used in a single initial-stage circulation pack for both the annulus and perforations. It can be useful in perforated short-radius horizontal wells.

Allpak

Allpak is an alternative path gravel packing technique. With this technique, otherwise known as shunt packing, one has two alternate passes for slurry flow.

These paths consist of small separate tubes or pipes attached to the screen and perforated with small holes every few feet. Slurry can either be injected directly into the tubes or the tubes can be left open at the top of the annulus to act as shunts.

This technique is good for packing especially highly deviated or long horizontal sections.^{2 13}

Performance evaluation

The productivity index (PI) can be adopted as the yardstick for evaluating the productivity of open hole and inside casing gravel-packed completions.

PI is defined as the ratio of hydrocarbon production over the imposed pressure drawdown. This ratio is calculated when steady state or pseudosteady state conditions occur for oil, gas, and multiphase fluid production.

Assuming a semisteady state condition, the concept of skin has also been adopted for every additional pressure drop due to:

Geometry (restriction of flow)
Damage (permeability impairments or improvements)
Heterogeneity (anisotropy or multiple-layers)
Turbulence (modification to account for the variance from laminar flow assumed in Darcy's law)
Presence of the gravel pack, itself.

Open hole completion

In an open hole gravel-pack completion, if one assumes that the screen has negligible resistance to flow, the skins are then due to the invaded zone, the gravel, and the turbulence. The PI is calculated with the pressure inside the screen P_s (at r = r_s).

For an oil well, this is expressed as Equation 1(see equation box [19,492 bytes]). For a gas well, Equation 2 is used.

The overall skin, s, of a completion (oil or gas) is defined by Equation 3, where s_l is the laminar skin and DQo is the non-Darcy flow (Equation 4).

Thus, the overall skin of a completion (oil or gas) can be expressed with Equation 5.

Equation 6 defines the laminar skin s_l. The skin due to gravel, s_lg, is calculated with Equation 7. Equation 8 calculates the skin in the invaded zone, s_li.

With turbulent flow, Equation 9 calculates turbulent skin, and Equations 10 and 11 are for determining the skin due to gravel, s_tg, and the invaded zone, s_ti. beta is the turbulence factor.

Cased-perforated completion

For a cased-perforated gravel-pack completion, if one assumes that the pressure drawdown inside the casing-screen annulus is negligible, then both sl and st can be separated into four terms:

Skin s_p due to the flow constriction around perforations
Skin s_d due to the damaged-invaded zone
Skin s_c due to the compacted zone
Skin s_g due to the presence of gravel inside the perforations.

Gravel-filled perforation skin can be obtained based on either of two basic flow conditions, as follows:

Constant flow rate through perforations
Linear flow rate through perforations.

For constant flow rate within the perforation, the flow inside the perforation is assumed linear from the top to the tip. The pressure drop can be calculated analytically, corresponding to the associated skins given in Equation 12. SD is the shot density, r_p is the perforation radius, and k_eq is the equivalent permeability to account for multiple layers (Equations 13 and 14).

For linear flow rate within the perforation, the flow rate is assumed to be increasing linearly but not at the same rate (Equations 14 and 15).

Equations 14 and 15 describe most of the geometrical and petrophysical aspects associated with a completion. Correlations for other skins can be obtained from a previous Reference 10.

Other heterogeneities can occur in a reservoir, such as vertical anisotropy, multiple layers, and partial opening/perforating of the pay zone.

Performance prediction

The invasion of gravel packs by load-bearing formation sand and fines is characterized by permeability reduction that may be gradual or instantaneous depending on the invasion process and the pore blocking mechanisms discussed previously.

Mathematical description of the various pore-blocking phenomena have been incorporated into the well-performance module of the Gravsize software.¹⁰ With the pore-blocking models, the pressure drop and therefore the skin s_g, for open hole or cased-hole completions, can be forecast as a function of time.

To predict gravel-pack-flow efficiency, another semiempirical model was developed that evaluates the gravel-pack damage as a function of all operational, geometrical, and sand textural properties. This model, called the permeability ratio (PR) model, can predict the net gravel-pack impairment damaged pack.

With these models, it is possible to estimate the initial and lifetime well performance as a function of any parameter (Fig. 2 [143,919 bytes]).

References

Forrest, J.K., "Horizontal Gravel Packing Studies in a Full Scale Model Well bore," SPE Paper No. 20681, 1990.
Alexander, K., Winton, S., and Price-Smith, C., "Alba Field Cased-hole horizontal Gravel Pack: A Team Approach to Design," SPE Drilling and Completion, March 1996.
Mason, J., personal communications, December 1996.
Oyeneyin, M.B., "Computer program helps pick best gravel pack design," OGJ, Mar. 2, 1987, p. 33.
Oyeneyin, M.B., "Numerical analysis of the effects of gravel packing on gas well productivity," SPE Production Engineering, May 1990.
Oyeneyin, M.B., Peden, J.M., and Hosseini, A., "Final Technical Report-Gravelsizing for effective sand control project," submitted to industry sponsors/participants, Heriot-Watt University, 1995.
Bigno, Y., Oyeneyin, M.B., and Peden, J.M., "Investigation of pore blocking mechanism in gravel packs in the management and control of fines migration," SPE Paper No. 27342, 1994.
Bouhroum, A., and Civan, F., "A study of particulate migration in gravel packs," SPE Paper No. 27346, 1994.
Oyeneyin, M.B., Peden, J.M., Hosseini, A., and Ren, G., "Factors to consider in the effective management and control of fines migration in high permeability sands," SPE Paper No. 30112, 1995.
Oyeneyin, M.B., Peden, J.M., Ren, G., Bigno, Y., and Hosseini, A., "A new Gravel Sizing Computer package for effective sand control and performance evaluation," SPE Paper No. 26219, 1993.
Gruesbeck, C., Salathiel, W.M., and Echols, E.E., "Design of Gravel packs in Deviated Well bores," SPE Paper No. 6805, 1977.
Peden, J.M., Russell, J., and Oyeneyin, M.B., "The Design and Optimisation of Gravel packing Operations in Deviated Wells," SPE Paper No. 12997, 1984.
Jones, L.G., Yeh, C.S., Yates, T.J., Bryant, D.W., Doolittle, M.W., and Healy, J.C., "Alternate-Path Gravel packing," SPE Paper No. 22796, 1991.