The sand production philosophy for a particular field and production system must be formulated after careful evaluation of a number of interacting factors which range from the well and surface facilities design to the operating procedures for the production system.

1. Sand production philosophy

In general, sand influx cannot be tolerated whenever the overall safety of production operations is jeopardised or when uneconomic measures need to be implemented in order to sustain production.

The optimal integration of the different aspects will be a function of economic, operational, safety and environmental considerations. Close cooperation between the Petroleum Engineering, Engineering and Production Operations department is essential when debating this issue because of the variety of subjects that need to be addressed:

1. Prevention of sand production: In certain cases, sand production can be prevented or minimised by limiting the applied drawdown, by a suitable selection of the perforated intervals and by implementing appropriate operating practices.
2. Well and surface facilities design: Most of the problems that can be experienced stem from the abrasive effect of sand, possibly aggravated by a corrosive environment, or from sand accumulations in the wellbore and surface facilities. If during a development, sand deposition and/or erosion rates are likely to present a potential system constraint, suitable measures with respect to well design material/component selection, sand detection (shut down systems) and monitoring should be addressed.
3. The availability of production operations support will be a function of the field location, its remoteness and the planned manning philosophy. Clearly it is essential that potential sand production problems are documented in the field development plan to highlight to process engineers the need for sand handling capability in the detailed process design. In principle if sand control is deferred, facilities should be designed to handle sand which will necessitate the early involvement of Production Operations staff, to assist in the development of a specific operations philosophy over the life of the venture. This will provide a list of preferred options on key operational and facility requirements, for example: unmanned versus manned, local versus remote, manual versus automatic etc. More generally this will provide guidance on various issues, whilst giving due regard to factors such as local legislation, and company policy.
4. A monitoring programme is necessary to allow early detection of significant sand production and trigger prompt remedial action, hence optimise production. Erosion of equipment at critical locations needs to be carefully monitored. A preventive maintenance programme is required to ensure critical items of the production system are functioning (e.g. safety valves) and to allow early removal of sand accumulations
5. Handling and disposal of produced sand may become a problem especially if disposal in an environmentally acceptable manner causes problems. If continuous sand production is experienced, dedicated sand separation and disposal facilities may be required. In general the preferred option is on-line sand removal systems to minimise intervention, carry over and protect down stream equipment. Cleaning sand prior to disposal is normally undertaken by specialised service companies unless sand cleaning is incorporated into the process.
6. Finally the potential risks that can be incurred because of sand production need to be carefully evaluated. Equipment failures may result in safety and environmental hazards while unscheduled production interruptions may have unfavourable economic consequences such as failing to meet contractual deliveries. The above issues form the basis of a sand production philosophy, which is incorporated into a more general operations philosophy and development plan to achieve the prime objective: optimise cash flow whilst safeguarding the technical integrity of the facilities. The appropriate involvement of all technical disciplines is considered a prerequisite in this process.

1.1 Methods of minimising sand production

Some sandstone reservoirs are competent enough to initially allow sand free production, but changing producing conditions may induce sand production in a later stage of the life of the field. When the sandstone is of such marginal competence, sand production may be minimised by one or a combination of the following:

• ·suitable selection of the perforated horizons;
• ·limiting the drawdown;
• ·controlling the reservoir pressure depletion;
• ·implementation of specific operating guidelines.

These practices are commonly referred to as "passive sand control methods" as they usually do not require additional effort during the completion operation in comparison with a conventional completion. Their applicability is however limited as they only qualify for a certain type of reservoir, are not always practical and are frequently uneconomic because of reduced deliverability.

Based on empirical and rock mechanical considerations, initial sand failure is related to the collapse of perforation tunnels and occurs when the effective stress acting on the near wellbore region exceeds a critical value, which is a function of rock strength. Although the conditions leading to initial sand failure are relatively well understood, the likelihood of massive sand failure is very difficult to predict because of the phenomenon of post failure stabilisation. This may be caused by structural sand arches which form in the vicinity of a perforation, effectively stopping sand production. Transient (bean up rate) and cyclic loading (production/shut in cycles) effects will have a significant impact on both initial failure and post failure stabilisation.

Sand production may be prevented to a certain extent by either limiting the effective stress acting on the reservoir and/or by selecting those zones in the reservoir which will be strong enough to support the near wellbore stress during production. Depletion plays an important role as it increases the effective stress acting on the reservoir rock. Massive sand failure is also often triggered by a change in the wellstream composition such as the onset of water production.

Practical guidelines for limiting sand production are derived from the sand influx prediction methods. The considerable degree of uncertainty inherent in the current sand prediction methods is directly reflected in their practical reliability. Local field experience may however increase the confidence level with which these methods can be applied.

1.1.1 Drawdown limitation

The first action to take when a well starts to produce sand at high levels is to reduce the drawdown by reducing the production rate. In view of the often erratic nature of sand sampling methods, it is general practice to insist on a confirmation of the high sand reading before beaming the well back This should however be done quickly to avoid a catastrophic sand failure. Experience shows that in some cases there is a critical flowrate below which little or no sand production is experienced. This is probably the oldest form of sand control, typified by "beaming down", but is not always successful and is frequently uneconomic.

Alternatively, the maximum allowable drawdown can be established by dedicated production tests.

1.1.2 Selective perforation

The concept of avoiding weaker reservoir material to avoid sand production is well known. In practice, the selection of the perforated intervals is based on the value of the sonic wave transit time, as given by the sonic log or on a log-derived formation strength index.

1.1.3 Control of pressure depletion

Pressure maintenance can serve to limit the maximum effective stress to which the formation is subjected. In practice, pressure maintenance schemes are only implemented in view of ultimate recovery considerations. However, the economic assessment of pressure maintenance schemes should include the possible savings by not having to install sand exclusion and the consequent benefit of higher well productivity.

1.2 Erosion of production equipment by sand

Erosion of production equipment can be caused by cavitation of produced fluids or impingement of liquid droplets and solid particles. It is difficult to accurately predict erosion rates because of the number and complexity of the variables involved i.e.:

• ·sand particle characteristics (size distribution, hardness, density, angularity...);
• ·sand concentration;
• ·mechanical properties of tubulars and facilities;
• ·fluid properties (viscosity, density, liquid to gas ratio...);
• ·flow parameters (velocities, flow regime...);
• ·pipework geometry (local turbulence effects...).

In a corrosive environment, the material loss process can be significantly accelerated by the combined chemical and mechanical action of the produced fluids and solids.

Many studies have been conducted to better understand and combat equipment erosion by sand. Results of these studies and field experience show that erosional damage due to sand can be prevented provided flow velocities and sand rates are kept below a critical level and the geometry of the facilities is optimised.

Whenever high erosion rates are expected, flow velocities should be limited by increasing the tubing or flowline diameter or limiting the flowrate. This is particularly true for gas or high GOR oil wells where high velocities due to gas expansion can be encountered.

The tolerance of subsurface equipment to sand may become a limiting factor, especially if artificial lift is required. For example, downhole pumps which rely on a fast sliding motion in contact with the produced fluids, or on high velocity moving parts are most susceptible to sand ingress. Gas lift systems, jet pumps or low speed pumps ("beam or Moineau" type) are the most tolerant to sand production.

1.3 Sand transport

If sand production is expected, flow velocities should be high enough to transport sand to the point where it can be separated from the produced fluids. A balance has to be found in sizing tubulars and flowlines as they are usually designed to minimise pressure rather than optimise transport velocities.

The minimum fluid velocity at which sand will be transported is a function of many variables. Such as:

• sand particle size, shape and density;
• fluid density and viscosity;
• flow characteristics.

Sand settling in the wellbore can also be prevented by emphasising simple completions (i.e. minimum number of packers and accessories), a smooth flow path and minimum deviation. Some allowance for settling of initial sand bursts can be provided by large ratholes. Efficient removal of sand fill from the wellbore by wireline or coiled tubing should also be catered for. The frequency of pigging operations should be adapted to prevent the build-up of sand accumulations in the surface flowlines.

Sand settling in the surface facilities is a function of the residence time, fluid properties and local velocities. The design of separators, pressure vessels and all potential sand traps should cater for the easy removal of sand fill, by providing drain points and jetting systems where sand can be flushed out. Additional features such as manways, hatches, flush and drain lines should be built in all major tanks and vessels, where most of the produced sand will settle due to the long residence times.

Sand will inevitably cause production interruptions due to the sand accumulations in various parts of the production system. The monitoring and shut down system should be tailored for this situation.

1.4 Sand production and cavity growth

Experience shows that in some fields, hydrocarbons can be produced together with continuous, manageable sand production. This suggests either the "sloughing" of fluidised formation or the creation of cavities in the reservoir in direct communication with the wellbore. In general, a gradual increase of the well productivity is often noticed.

There is presently no method available to measure the true shape of a cavity produced by sand production behind a production casing. Current production logging tools can only measure the height of a cavity. Hence only an account of the cumulative amount of sand produced can give an idea of the total equivalent cavity volume, assuming the formation does not have a 'creeping' behaviour. Keeping an accurate historical record of produced sand is therefore important to monitor global cavity growth in terms of volume and shape. Cavity growth limits may be inferred from the following considerations:

• ·Prevention of zonal isolation problems i.e. to prevent the cavity from growing too close to a reservoir fluid interface or an adjacent zone which needs to be kept separated.
• ·The larger the cavity, the more difficult control of a remedial sand control job will become, if required. However no specific guidelines can be given on this topic.
• ·Buckling/collapse of unsupported casing due to axial compressive loads, resulting from formation compaction or temperature effects. Of these two effects, normally only formation compaction should be of concern at reservoir level. It is important to note that in the case of reservoir compaction, a length change will be imposed on the casing string, independently of the presence of a cavity. It is difficult to predict the effect of a cavity on the stability of the casing under those conditions.

1.5 Allowable sand production levels

The concept of a unique sand production cut-off level that can be applied to a wide variety of wells and production facilities under differing producing conditions does not exist. The adoption of a rigid arbitrary rule is likely to lead to premature remedial operations or unwarranted production restrictions.

2 Framework for a sand control strategy

Sand control is obviously required whenever evidence exists that unmanageable sand production will occur initially or soon after a well is put on stream. This is typically the case for shallow, unconsolidated sandstones. The remaining problem will then be to select the most suitable sand control method.

In many cases however, it is not certain when sand production will become a problem. For the more consolidated sandstones where only indications of possible sand influx exist, an approach common to operations is to base the decision for sand control on a simple rock strength indicator such as formation depth or the measured sonic wave transit time. It may also be policy for certain fields to apply sand control indiscriminately as an insurance against any sand production.

Decisions based on these simple approaches may result in higher initial completion costs and deferred production. The economic penalty of sand exclusion should be a strong incentive for a more critical approach.

2.1 Objectives to be considered

When considering a sand control strategy, two objectives must be pursued:

• ·minimise sand production;
• ·maximise hydrocarbon production.

Some degree of compromise between these objectives will be required since prevention of the movement of sand is generally incompatible with the unrestricted flow of fluids. An "engineered" sand control strategy should integrate technical, operational, economic and safety considerations in order to identify the need for sand control and define the optimal timing if required. A framework to support the decision making process is presented in this section.

There should be no preconceived ideas on which method will be used if sand control proves to be required. The identification of the most appropriate sand control method usually requires a much deeper and detailed level of investigation than is required to define the optimum sand control strategy. It is important at this stage to avoid getting involved in too much detail so as to keep a clear overview of the problem and an open mind to alternatives.

In other words, the requirement for sand control is a strategic issue that needs to be addressed when preparing the field development plan. However both this requirement and the selected method should be constantly reviewed in the light of the experience gained when actually developing the field.

2.2 Why and when sand control

2.2.1 What is the sand production philosophy?

It should first be established whether the local circumstances dictate a "no sand" philosophy. The consequences and risks involved by not implementing sand control may be unacceptable when:

• ·the overall safety of production operations is jeopardised (e.g. offshore locations in particular manned platforms; high GOR or high rate gas wells);
• ·adequate monitoring and/or intervention is not possible (e.g. unmanned or remote locations);
• ·remedial sand exclusion treatments are not operationally or economically attractive (e.g. subsea wells);
• ·artificial lift requirements do not tolerate sand production;
• ·the consequences of deferred production due to an unplanned shut down are unacceptable.

Production availability problems can become very acute for gas fields which occasionally have to be produced at high output levels in response to peak demand. Under those conditions, massive sand failure may occur because of the higher drawdowns and consequent higher stress levels on the formation. There may also be little spare capacity to sort out any problems occurring in the surface facilities.

In general, the level of sand that can be tolerated in the production system will be a function of operational, safety, environmental and economic considerations.

2.2.2 Can the wells be initially produced to their target level with acceptable sand rates?

Having established the sand production philosophy, a dedicated production test should establish whether sand control is required initially or not. Dedicated tests are the most diagnostic of the initial sand production potential of a well. They consist of producing at gradually increasing rates until the maximum desired rate is reached or massive sand failure is experienced.

2.2.3 Can massive sand production occur during the lifetime of the reservoir?

For friable sandstone reservoirs which can be initially produced without sand problems, the prediction of massive sand production is a crucial issue. There are currently no models available that can accurately forecast this event.

It is possible however to assess initial sand failure, i.e. the conditions that will lead to the initial collapse of perforation tunnels. This event may be followed by post failure stabilisation and lead to periods of relatively sand free production at specific producing conditions. Other factors such as a change in the composition of the produced fluids (i.e. increasing water cut) often contribute to the onset of continuous sand production. Therefore even if the initial sand failure criteria can be established, considerable uncertainty over the onset of massive sand failure is likely to remain. The confidence level will depend on the availability of field "calibration" data points, in other words sand failure events.

A sand prediction study should aim at assessing the producing conditions likely to lead to massive sand production. Together with a depletion forecast, the point in time when a well will be at risk of massively producing sand can then be estimated. Dedicated "sand influx" production tests can be carried out to simulate future operating conditions and to provide a more reliable insight into the risk of future massive sand failure.

2.2.4 When should sand control be implemented?

For reservoirs which can be initially produced sand free or with tolerable sand levels and for which the risk of sand failure as a function of time has been established, the decision when to implement sand control should depend on an economic evaluation. The merit of deferring sand exclusion will be a function of the following:

·Advantages:

• -Higher well productivity
• -Lower initial capital costs (well completion)
• ·Disadvantages:
• -Increased operating costs
• -Higher capital costs (surface facilities)
• -Deferred production if massive sand failure occurs
• -More difficult to install sand control completion when sand failure occurs

Experience shows that all sand control methods cause various degrees of productivity impairment. This will result in output restrictions if the drawdown cannot be increased to compensate for the productivity loss. Production availability assumptions should explicitly take into account impairment effects due to sand exclusion.

If sand control is deferred, the production system may have to be upgraded for sand tolerance resulting in higher initial capital expenditure. The higher operating costs will be a reflection of the additional production operations support required.

Assuming that the well can be initially produced at a higher level without sand exclusion, the economical attractiveness of this completion option will depend on how long it can be produced before requiring a remedial sand control treatment. This critical time to failure can be calculated by comparing the financial outcome of the two completion scenarios, on a Net Present Value (NPV) basis.

Simple economic models can be built to evaluate the two completion options. This evaluation relies on the following data:

• ·production forecasts for the anticipated well life for the two different scenarios;
• ·the cost aspects (capex/opex) of the two completion alternatives;
• ·basic economic parameters: discount rates, crude oil price etc.

If it is estimated that a sand failure is not likely to occur within this critical time then initial sand exclusion is not economically justified. However it should still be verified that other considerations do not in any case require control measures.

In reality, the occurrence of a sand failure is an uncertain event. However the likelihood of failure generally increases (with advancing depletion and development of water production) during the life of the well. Assuming that the risk of sand failure can be quantified as a function of time, this risk function can be integrated in the economic model. The effect of the risk of sand failure on the expected economic return of deferred sand exclusion can then be quantified.

2.2.5 Other considerations

Flexibility in reservoir management: Installation of sand control equipment may limit the ability to carry out remedial treatments such as shutting off depleted and gas or water producing zones.

Complexity of well completion: This may become an important consideration in the case of multiple completions. Some sand control methods have a tendency to congest the wellbore with a lot of hardware. This may limit the access to adjacent zones or limit the number of zones that can be completed in the same well. It will also be more difficult to workover these wells.

Well history: Is it a new well or a workover? In case of a remedial treatment, the prime issue may become the economic viability of a sand control job. Also the chances of successful remedial sand control will be lower if the well has already produced much sand.

2.3 What if initial sand control is not justified?

If initial sand control is not justified, it must be ensured that production operations will not be jeopardised when sand production is experienced. The following points need to be addressed:

·design of the production system (well and surface facilities) for the expected sand production levels;

·operating practices to prevent sand production;

·monitoring and operating procedures for the production system;

·review of planned subsurface operations for their possible influence on sand production (e.g. acid stimulation, additional perforation);

·sand prediction studies in the light of future production experience and development plans. This may require laboratory studies or dedicated production tests;

·contingency plans for the event that massive sand production occurs in one or more wells. For instance when failure is related to the onset of water production, many wells may fail over a short period of time.

3 Basic sand control methods

Sand control methods may be classified as mechanical exclusion, chemical consolidation methods or a combination of the two. These are briefly discussed below together with their respective merits, limitations and applicability.

3.1 Mechanical methods

These methods rely on a downhole filter which stops the formation sand while allowing fluids to flow from or into the formation.

Downhole screen - Historically this was the first method of sand control used. Different types of screens can be used e.g. slotted liners or wire wrapped screens. The openings in the screen are generally sized such that the formation sand will bridge on these openings. The "prepacked screen" is a more elaborate type of screen which incorporates a consolidated gravel sheath to control the formation sand.

Gravel pack - Properly sized gravel is placed against the formation and held in place by a screen which is normally of the wire wrapped type. This gravel is sized to stop the formation sand. The gravel pack can be placed either in perforated casing (IGP- Internal Gravel Pack ) or open hole (EGP- External Gravel Pack). The open hole is generally underreamed prior to placing the gravel. This can also be done after milling a section of the casing away (MCUGP- Milled Casing Underreamed Gravelpack).

Gravel packing is the most widely used sand control method and many different systems have been developed which differ in the design of the mechanical equipment, the properties of the completion fluids used and the way the gravel is placed.

3.2 Chemical consolidation

Sand consolidation is a sand control method whereby fluids containing a cementing agent are injected into the formation to provide a bond between the sand grains after curing. As the sand grains are coated and bonded together by the cementing compound, some permeability will be lost following the treatment. The successful treatment must provide the additional strength required while preserving as much formation permeability as possible. Various types of chemical consolidation systems are commercially available. Sand consolidation systems are sophisticated chemical processes which require strict adherence to quality control for successful application.

3.3 Technical developments

Alternative techniques which are finding greater application include:

1. Selective placement: The EPOSAND formulation has been modified (Welfix DP 2000) to be compatible with the rubber elements of a selective placement tool (e.g. Nowsco's SPT tool) which is run on coiled tubing. This will allow a greater control of the placement of the resin in the formation.
2. Chemical shut-off: The objective of this method is to shut off a sand producing zone with a plugging chemical placed with a selective placement tool.
3. Expandable screens: The main idea is to run a screen in a collapsed state which can be subsequently expanded against the borehole and effectively stop formation sand.

The Production Technologist selects the appropriate technique with input from Production Operations (Well Service) Engineers as discussed in formulating a sand production philosophy.