Sand Control

Sand control in horizontal wells

The evolution of sand control from historical water wells to recommendations for 15,000 ft+ horizontal oil/gas wells.

Derry D. Sparlin Sr., Consulting Engineer

This summary article backgrounds the need for formation sand entry prevention in downhole producing wells from man’s first dumping of rocks into water wells drilled with rock or iron tools. Thousands of years later, the oil/gas industry invented basic gravel pack and sand screen methods to prevent inflow of unconsolidated formation sands. Now the drilling of horizontal boreholes over 15,000-ft long has created ever new challenges and need for sand control in extremely long sand intervals.

The discussion covers this story with: 1) A historical review; 2) Basic barefoot, gravel pack and stand-alone screen completions; 3) Selecting sand control type for a horizontal well; and 4) Innovation for future use.

HISTORY OF SAND CONTROL

Way, way back, when early humans needed water, they dug water wells with their hands. Then they needed more water, so they began using tools to drill into sand beds. They found that they could work faster and go deeper by using a heavy rock or iron as a percussion tool, like a yo-yo, to cut through to the water bearing sand.

Water-well technology. Sand problems were born. Loose sand was coming up with the water or falling to the bottom of the well, stopping the water flow. They were able to prevent the problem by filling a small part of the well with large rocks.

Much beyond the rock-dumping era, better ways to stop sand production and maintain high water production rates were developed. By the late 19th and early 20th centuries, they cut holes in liners to stop sand production, and later found they could drop gravel down the annulus to form gravel packs that gave even better results. Oil companies quickly modified sand control techniques to handle deeper and dirtier wells.

Oil industry advances. Perforated liners were popular in the early 20th century, without much concern about the sizes of perforations to be used. Then academia jumped on the oil wagon with such people as C.J. Coberly, who published his work on “Selection of screen openings for unconsolidated sands” in 1937.¹ This was a starting point for technical means of designing sand control tools.

Great improvements were developed by the oil/gas industry despite politics and prejudices. Some of the major sand control techniques developed during the second half of the 20th century were: sand consolidation, slotted liners, wire wrapped screens, prepacked screens, premium screens, gravel packs, frac packs, high-angle-hole gravel packing and horizontal-well gravel packing.

The most challenging of these has been controlling sand in horizontal wells, as drilling technology is now able to drill to distances of more than 15,000 ft from heel to toe. For the uninitiated reader, “heel” is the beginning of the horizontal wellbore and “toe” is the end of the horizontal wellbore.

CONTROLLING SAND IN HORIZONTAL WELLBORES

Regarding barefoot completions, the simplest and least expensive completion is to put a well on production without anything left in the productive interval. Without casing, liner, screen or tools therein, we rely on the sandstone being strong enough, and shale beds stable enough, to resist changing pressures as the well is depleted. This works very well in strong sandstones, dolomites and limestones, but will fail in weaker sandstone, or where there is tectonic activity.

Sometimes it is impossible to understand why sand begins to be produced from a very strong sandstone, but the most common cause is the onset of water production. There may be a chemical reaction between extraneous water and natural cementing agents of the oil bearing formation. This should always be tested and anticipated by core tests.

One barefoot completion in a horizontal well off Western Australia began producing sand after a year or so of production, and it was found that the heel had collapsed. When the operator attempted a clean-out it couldn’t get into the hole, and no explanation could be found. In another case, a vertical well near the Gulf of Mexico stopped producing, and a downhole camera found that the casing was sheared at one point. The conclusion was that some tectonic activity caused two layers to slide across each other.

Stand-alone-screen completions. A Stand-Alone-Screen (SAS) completion simply employs some type of screen or liner positioned inside a productive interval. Prepacked screens are expensive, so most stand-alone-screen completions are done with slotted liner, wire-wrapped screen or premium screen. One of the first SAS horizontal-well completions was in the Helder field offshore the Netherlands. It was done with a large-diameter prepacked screen, resulting in negligible skin factor and successful sand control.

Any screen in an open hole will scrape along the low side of the drilled hole and pick up drilling fluid solids, cuttings, fluid loss agents and formation sand that can reduce screen permeability. Such screen damage may have only a minor effect on well productivity, but may force fluid to flow through short screen sections, eroding holes through the screen, commonly called “hot spots.” Centralizers or shrouds help prevent this but add cost and bulk.

A stand-alone-screen is likely to have non-uniform inflow, as it is easier for fluid to flow into the screen that is nearest to the heel of the wellbore. This should not be a concern in most wells, but it may cause screen erosion and premature failure. Such problems are most likely to occur in wells where watercut is high, and fluid flow friction from the toe of a well restricts production more from the far end of the borehole than from near the heel. This problem can be reduced by using flow restriction systems, such as the ResFlow system, trade marked by Reslink, that evens inflow velocity throughout the length of the horizontal wellbore.²

Perforations in a cased hole should only be toward the high side of the casing, because low-side perforation tunnels are likely to be loaded with mud solids, fluid loss solids, and sand/ dirt, as affected by gravity. High-side perforations will be somewhat self-cleaning and are less likely to collapse.

The velocity of producing fluid entering radially into a screen may be so low that sand will remain in the annulus, and sand dunes may completely fill some sections of the annulus. This adds noticeable skin and may stop all production from the wellbore beyond the point where sand fills the annulus.

Gravel packed completions. These comprise gravel sized to stop movement through the wellbore of formation sand packed around a screen or liner. There have been many studies and publications that help determine proper gravel size, but the main problem with all of them is that we rarely ever have enough full-bore cores to provide the necessary data.

Ideally, one formation sand sample should be selected and studied from each foot of potentially productive formation sand. Sand grains in each sample must be separated (perhaps “crushed” is a better word) into the sizes that will be likely to move into the wellbore when the well is put on production. Then a gravel size is selected that will prevent the smallest grained, productive sand from invading the gravel pores. It is unlikely that this has ever, or will ever, be done, because it is not practical for oil/gas wells to cut enough cores and analyze samples from each foot.

Many years ago, Chinese successfully gravel packed municipal water wells without any concern about gravel sizes.³ Their success was obtained by first circulating a huge amount of sand out of the open hole, lowering a slotted or perforated liner into the hole and filling the annulus with any size of gravel that was available. They calculated that the gravel fills 30 in. or more radially around the liner. When such a well was on production, the fluid velocity flowing into the periphery of the gravel pack was so low that water would not carry sand into the gravel.

Advantages of gravel packing include: 1) Gravel in the screen/ casing or screen/ openhole annulus serves as a filter-aid that reduces solids invasion into the screen; 2) Well productivity increases as gravel packing replaces some low-permeability sand with high-permeability gravel; and 3) Screen damage is reduced.

The most likely situation where gravel packing should be used is in a very weak formation where sand is likely to fill the annulus around the screen. Gravel helps support the formation in stronger sandstones, reducing the effect of sand movement, screen plugging and erosion.

Horizontal well packing. It is more difficult to successfully pack the annulus around a screen in a horizontal hole, than in a vertical hole, because gravity pulls gravel to the low side of the hole. The gravel can form dunes that prematurely stop the gravel packing operation. The number of gravel pack failures in horizontal wells has increased over the last several years as wellbore lengths increase. Horizontal wellbores are now reaching 20 to 30 thousand feet long, and this is a challenge to gravel packing, in addition to being more expensive than a stand-alone-screen or bare-foot completion.

Water pack system . One way to gravel pack horizontal wellbores is the water pack system, often called the alpha/ beta pack. This technique uses low-viscosity brine to circulate gravel into the annulus at high enough rates to prevent gravel dunes from prematurely filling the annulus. Failure to push gravel all of the way to the toe and back, accounts for some failures. This potential problem may be circumvented by using special screens or prepacked screens that serve as downhole filters, so that any section of the annulus that is not fully packed with gravel will have a back-up to stop sand movement.

Fluid loss control is perhaps the most difficult problem with the water pack system in long, horizontal wells because low-viscosity brine can leak off too fast to maintain a high enough circulation rate. This is not a problem in low-permeability sand or shale zones because there is low fluid loss into the formation. The main problem is where there is a large variation of fluid loss into high- and low-permeability sections. This makes it difficult to control the fluid velocity necessary to push gravel to the end of the screen and back to the heel.

The “shunt-pack” system. This is another way to gravel pack horizontal wellbores using special tools in combination with viscous fluid. Gravel is transported more efficiently by this fluid with less risk of gravel dunes and better chance of fully packing the entire annulus from toe to heel. Variation of fluid loss into high- and low-permeability sands is not as much of a problem as with low-viscosity brine, but the tools are more complex.

Gelled fluids can be more difficult to control under deep, high-temperature conditions, as chemical breakers add to the complexity of controlling the gel breaking time. Gelled brines are more likely to cause some formation damage than ungelled brine, but should not affect as much depth into the formation.

Great strides have been made over the last few years to improve these gravel packing systems, both tools and fluids, and reports of failures have diminished in horizontal zones less than 2,000 ft. Longer zones are more of a challenge.

SELECTING TYPE OF SAND CONTROL FOR HORIZONTAL WELLS

Formation strength and completion length are the two most critical parameters to consider when selecting a sand control technique for horizontal wells.

Gravel packing. Totally unconsolidated sands, common along the US Gulf Coast, onshore Trinidad, Eastern Alberta, Baku and Sumatra, should always be gravel packed to give maximum stability and productivity. Gravel filling the annulus of an openhole completion or filling perforations of a cased hole, should yield the best short- and long-term results in such weak sands.

Gravel packing weak formations should not be used where: 1) Completion length is so long that there is not a good chance of packing the entire annulus; or 2) The formation is so weak that the drilled hole will not stay open long enough to get a good pack; or 3) A gravel pack is too expensive.

Gravel packing can also be successful in stronger zones, but it becomes more expensive and more technically difficult in longer horizontal wellbores.

Stand-alone-screen completions are currently the only economical option for long, horizontal wells that cannot be successfully gravel packed.

The claim that stand-alone-screen completions form “natural sand packs” is only valid in formations that are totally unconsolidated and have little or no natural cementing agents. This type of sand will fall into the wellbore when production begins. Some fines and small grained sands will flow through the screen, but larger grains will be stopped and they, in turn, will stop the smaller grains. We got this idea from the water-well people because they have done this for centuries, but their aquifers are generally shallow, weak, short intervals, and in vertical wellbores.

Stand-alone-screens work well in “quicksand” type formations, but not in stronger formations, because they interfere with wellbore cleanup. The concern for “hot spots” increases with longer wellbores because the greater overall production rate increases fluid velocity as the screen plugs to a critical point for erosion to occur.

The decision of which type of screen to use is usually based on which will fit the budget. Slotted liners are the least expensive, but have the poorest filtering capability. Large diameter prepacked screens may be the best filters, depending on the design, but are expensive and easily plugged. Wire wrapped screens are normally best for gravel packing when wire spacing is sized to stop gravel; and premium screens, with their micron-size openings, are designed as downhole filters to stop all of the formation sand.

Stand-alone-screen completions have best results in wells where the formation is strong enough that sand control is not needed.

INNOVATION FOR FUTURE USE

Have you wasted money on a screen when none was needed? The following are two examples of some stand-alone-screen completions in horizontal wells where screens were probably not needed.⁴

1. Eight wells offshore West Africa were completed with 10 gauge, 5-1/2-in. wire-wrapped screens inside 8-1/2-in. drilled holes. PIs ranged from 61 to 102 bpd/psi with skins of -1. The wells did not produce sand during eight years and had natural production decline rates. All jobs were successful, but the successes may have been due to enough formation strength that the screens were not needed. 
2. An offshore horizontal well in the Far East was an openhole completion with a 6-1/2-in. premium screen in a 1,350-ft horizontal section. Initial production was 93 bpd/psi that declined with a slight productivity decline to 60 bpd/psi. This decline was thought to be caused by failure to adequately clean up the filter cake. Subsequently, an enzyme treatment was done to attack the starch, polymer and acid-soluble solids used in the drill-in fluid. 

This treatment successfully raised the productivity to 90 bpd/psi. There is no mention of sand production. If sand had collapsed around the screen, it would have interfered with the remedial work and should have prevented the clean-up that occurred. If this well had been produced as a bare-foot completion, the mud cake would have been easily removed, perhaps without any kind of special clean-up treatment. Chances are good that no sand would have been produced.

Critical Fluidization Velocities. A barefoot completion of a long, horizontal wellbore should be used for borderline-strength, or stronger, sandstones where sand production is unlikely. Table 1 gives Critical Fluidization Velocities of sand, which supports this recommendation, because fluid velocity is not likely to be high enough to move sand into the wellbore.

For instance, if a horizontal wellbore is 10,000 ft long, average fluid entering the wellbore must exceed 117,000 bpd to lift a 100-micron sand grain. This assumes that the produced fluid is water, all of the sand is totally unconsolidated, all sand grains are 100-micron diameter, and water is entering every foot length of the wellbore at the same velocity. To put this in perspective, 45-micron particles are about the smallest particles that can be seen by the unaided human eye.

This analogy is much too conservative. Natural formation cementing materials, grain packing arrangements, grain shapes and knitting of grains together are factors that will resist movement of sand, with production rates many times higher than the critical fluidization velocity predicts.

SUMMARY: TO ENHANCE SUCCESS.

Innovation for long, horizontal wellbores turns out to be old technology used in new ways. We borrowed some of this from old water-well technology that maintains a low flux into a well to prevent sand movement. The Chinese did it successfully by enlarging wellbores to 6 ft diameter or more and supporting formations with rocks. The differences are that new technologies achieve low flux by opening huge surface areas in long wellbores.

The following conditions should enhance success:

• Formation must have at least borderline strength.
• Production should not be surged.
• Drill as large a diameter hole as practical. 
• Use good drill-in fluid to minimize invasion. 
• Clean out mud cake carefully. 
• Isolate shale sections. 

Yes, there is risk of failure, but risk is an ancient problem that has been accepted by the oil/gas industry since Colonel Drake. Successful bare-foot completions in long, horizontal wells should yield faster completion times, earlier returns on investments, better cleanup of drill-in fluids and higher well productivities. 

LITERATURE CITED

¹  Coberly, C.J., “Selection of screen openings for unconsolidated sands,” API’s Drilling and Production Practices, 1937.

²  Augustine, J.R., “An investigation of the economic benefit of inflow control devices on horizontal well completions using a reservoir wellbore coupled model,” Paper SPE 78293, 13th European Petroleum Conference, Aberdeen Scotland, UK.

³  Zhang Yu-Xiang, “How the Chinese design gravel packs,” World Oil, April 1981, pp. 154 – 162.

⁴  Price-Smith, C., et al, “Design methodology for selection of horizontal openhole sand-control completions supported by field case histories,” SPE Drilling & Completion, September 2003.

THE AUTHOR
	Derry D. Sparlin Sr. studied sand control technology in the Production Research Division of Conoco for 17 years. He then spent three years as vice president-research for Baker Sand Control and, in 1980 formed International Completion Consultants Inc. (ICCI). He retired in Tucson, Arizona, does some consulting projects from time to time and continues to support ICCI as consultant emeritus. Mr. Sparlin is inventor, or co-inventor, of 34 US patents, served on numerous SPE committees, authored many technical papers and articles, and has visited more than 65 countries.