Achieving successful zone isolation in horizontal wells

Overcoming high failure rates in casing annulus packers

C.O. "Doc" Stokley
TAM International

As horizontal well drilling activities increased, technological developments in associated drilling tools and methods expanded rapidly. In the first few years of major horizontal drilling activity, a majority of wells were drilled into carbonate reservoirs, thus completed open hole. As drilling technology developed, long reach and horizontal wells penetrating sandstone reservoirs, increased requirements for horizontal completion tools and zone isolation.

Tools such as the inflatable casing annulus packers (CAP), which for many years had exhibited high success rates in vertical wells were suddenly applied to horizontal wells for zone isolation. As verified by two independent petroleum industry studies, success rate for these types of tools was not acceptable.

One study reported that overall success of achieving zone isolation was less than 60%. Failures resulted from a combination of inflatable packer ruptures and inability to maintain seals, setting tool malfunctions and poor application selection.

Several major oil companies funded projects to evaluate available tools, field applications, and suppliers. In one case, casing was cut below a failed CAP, retrieved, and sent to Houston for analysis. In another case, failed CAPs were retrieved by pulling the liner and sent back to the manufacturer for failure analysis.

These were rare opportunities. CAPs are generally cemented into the well or inflated with cement. A full scale laboratory test was conducted on a CAP in Norway and the product returned after testing.

Prior techniques used to determine tool outside diameter versus borehole diameter (clearance) and tool length to pass through maximum dogleg was a simple geometric calculation. This analysis assumes no bending of the CAP and does not include any analysis of drag forces applied to the tool.

The basic construction of CAPs falls into two categories:

Reinforcement extending full length of expandable section thus requiring that one end sub slides on the casing mandrel while maintaining a seal as it expands
Reinforcement extending only partially from each end of expandable section, no reinforcement in middle section, and both ends fixed to the casing mandrel.

Failure analysis

Engineering evaluations of returned, failed products was conducted to establish probable failure causes. A series of laboratory tests were conducted to verify probable causes and establish design criteria and operating procedures.

In excess of 200 field applications were reviewed in an expanded database to classify type of tools and wellbore conditions versus failure rate. This phase also defined additional laboratory testing required for verification. Conclusions after failure analysis were:

Inflatable packers without efficient bonding of elastomer to metal reinforcing and/or casing mandrel were highly susceptible to damage when running into wells with doglegs in excess of 8!/100 ft and when passing through windows.
Partial damage to the elastomer cover resulted in near 100% failure rate.
Small diameter inflation valves were susceptible to failure due to high velocity erosion at initial opening, especially where opening pressures exceeded 2,000 psi. Engineering analysis was performed and found that initial flow rates in the valve exceeded 1,000 mph with 3,000 psi opening pressure.
Plugging of small diameter valves during inflation was detected in a few cases but was difficult to define as most CAPs had redundant valve systems.
Long packer mandrels were more successful than short mandrels.
Elliptical boreholes with axis ratios in excess of 1.4:1 cause high failure rates in both ability to inflate as well as sealing capacity, especially when using fully reinforced type CAPs.
Fully reinforced CAPs exhibited poor success when inflated in build sections or doglegs greater than 12!/100 ft.

Failure mechanisms

By applying side loads and dragging a CAP through a concrete trough, damage due to force and distance was quantified. As was anticipated, fully reinforced CAPs, which have minimal elastomer to metal reinforcement bonding, exhibited low resistance to drag loads without damage.

High shear pin opening pressure was tested using nitrogen to simulate compressed fluid storage. After opening and inflating, all CAPs were disassembled and inspected for effects of flow cutting. Severe flow cutting occurred after less than five seconds and prior to proper CAP inflation, resulting in high failure rates.

Drilling mud with variations in density, solids content and viscosity were tested in an attempt to simulate valve plugging. No plugging was achieved as long as fluids designed with high particle transport efficiency continued to move through the valve flow paths.

Engineering analysis was performed, using a computer model, to determine side load forces required to bend a tubular member and model output verified with lab testing of loads, length and deflection.

A full scale test in Norway verified that previous CAP design did not achieve expected pressure capability in highly elliptical hole or when inflated under high bend load conditions. Elliptical axis ratio tested was 1.5:1.

Design changes

In order to eliminate the failure mechanisms, a drastic change in application analysis was required in addition to tool design changes.

As determined from side load testing, higher bond strength would be required in conjunction with a reduction in side loads to bend the CAP mandrel. As TAM standard CAP design utilizes a one piece mandrel, longer mandrel length was a simple design change which only effects manufacturing processes and cost. Numerous commercially available elastomer to metal bonding agents and cleaning techniques were tested to determine a method of improving bonding efficiency while maintaining manufacturing flexibility.

In order to reduce flow-cutting, a multiplicity of flow path changes were implemented and tested under the same conditions as above to verify validity of changes. In addition to flow path changes, a new concept of a manually operated inflation control valve system (port collar) with flow area 20 times greater than original valves was developed, tested and patented.

Using a manually operated valve eliminates shear pins required to control inflation process thus eliminating high opening pressure shock loads and flow velocity through valves. This system is ideally suited for horizontal well completions where multiple CAPs are run and individually inflated. For harsh environment applications, this port collar valve system was designed with metal to metal seals on closure.

To eliminate potential damage to an elastomer cover over the partial reinforcement ends, the product was designed with steel reinforcing at each end exposed (no elastomer cover). This type construction has long been a standard for bridge plugs where steel reinforcement acts as "slips" when expanded to contact casing or borehole ID. New product designs exhibit higher differential pressure capability and can function with high reliability in 1.7:1 axis ratio borehole conditions.

Negotiating doglegs

The major change in application analysis was to evaluate well conditions versus bending required for the CAP to conform to borehole build angle or maximum dogleg severity section.

Having defined drag load capability of each product type, conversion of bending loads to drag forces can be used to determine product type and mandrel length to run without product damage. A proprietary computer program was then written for ease of application analysis.

In addition to design changes, sufficient field application data was available to establish guidelines for selection of seal length, type of inflation fluid, elastomer type and product construction type.

Re-entry

In wells sidetracked from existing producers, near wellbore region isolation from lateral section is becoming a normal procedure. Where build sections and window are not isolated from the lateral, a multiplicity of problems may be encountered during production.

The major problem resulting from near wellbore region is undesirable production from formations penetrated in the build section, drained reservoir conditions, and poor cement bond quality between casing and formations near the window. A reliable method of achieving isolation of the build section from lateral utilizes a CAP inflated at the "heel" with a port collar or stage tool immediately above used to cement build section and liner overlap.

Isolation of highly fractured segments or fault cuts encountered in laterals can be achieved by inflating CAPs on each side of the section to be isolated with a port collar between for squeeze cementing the interval.

Summary

When utilizing this tool design in conjunction with above selection process, successful zone isolation has been achieved in 50 of 53 applications for a 94% success rate. This rate is within range of TAM's 95% success rate target established prior to undertaking CAP redesign and selection criteria changes.

Maintenance of a detailed data base on applications and assistance provided to the producing company personnel in design of drilling programs to enhance completion option flexibility is a critical component of achieving continual improvement in zone isolation success rate.

Author

C. O. Stokleyis the technical director for TAM International in Houston. After obtaining a BS in electrical engineering from Tennessee Tech, he went to work for Schlumberger. He subsequently worked for Exxon Production Research, before joining TAM in 1982. He is a member of SPE and a registered professional engineer in Louisiana and Texas.