WELL ILLUSTRATES CHALLENGES OF HORIZONTAL PRODUCTION LOGGING

S. Evan Robertshaw
Mobil Oil Canada
Whitecourt, Alta.

Steven C. Peach
Schlumberger Production Services
Edmonton, Alta.

Special considerations need to be made for production logging in horizontal wells.

These considerations include: the logging procedures, equipment design, safety related to conveying production logging tools with coiled tubing, pressure deployment of up to eight logging tools simultaneously, and data acquisition.

The case history of Mobil OH Canada's Well Rainbow 1-14KR-110-6W6M illustrates production logging in a horizontal lateral.

PRODUCTION LOGS

Production logs can determine the source and type of fluids flowing into a well bore. The most common tools simultaneously measure fluid velocity, fluid density, temperature, pressure, and capacitance.

Typical reasons for production logging include identifying sources of undesirable gas and water inflow that may require zonal isolation, and identifying intervals of low or no inflow that may require stimulation.

Ultimately the objectives of production logging include optimizing well productivity and maximizing recovery.

Although the objectives of logging vertical and horizontal wells are generally the same, there are differences in how the information is obtained and in the factors that affect data interpretation.

In vertical wells, logging tools are conveyed on electric wire line through a full lubricator system. This permits logging while the well bore is under pressure.

In horizontal wells, the tool string is most efficiently pushed along the well bore by coiled tubing. The two unique factors affecting the interpretation of horizontal well log data are:

The length of the logged section is often more than 20 times longer than in vertical wells. At the end of a horizontal section, inflow interpretation is relatively straight forward. However, because of the additive effects of multiple fluid phases entering the well bore, interpretation becomes increasingly difficult as the tools are pulled away from total depth.
Phase separation between oil, gas, and water occurs along the horizontal plane and is more significant in large diameter liners. The effect is analogous to a very long horizontal separator or a pipeline.

As determined by the location of the fluid interfaces with respect to the tools, tool responses are affected by the changing fluid velocity and density moving past or throuah the tools.

RAINBOW 1-14KR

Well Rainbow 1-14KR was the first horizontal well drilled into the Keg River AA dolomitized limestone pinnacle reef in northwestern Alberta (Fig. 1). Drilled in February 1990, the primary purpose of the well was to evaluate the effectiveness of horizontal laterals in reducing a severe gas and solvent coning situation, common in vertical wells in this reservoir.

Initially, the average pool GOR (gas/oil ratio) was 2,500 cu m/cu m (14,000 scf/bbl) compared to the original solution GOR of 95 cu m/cu m. Several wells even exhibited GORs in excess of 5,000 cu m/cu M.

Although during the first 1'12. years the original 1-14KR well produced oil at a respectable average rate of 65 cu m/day (400 b/d), a rapid GOR increase from 400 to 2,300 cu m/cu m occurred within 8 months of well completion.

In early 1990, the Keg River AA pool was in the chase gas stage of a vertical hydrocarbon miscible enhanced oil recovery (EOR) project. Fig. 1 illustrates the pool outline. The light shading represents the thick 150 m (492 ft) reef core, which exhibits high permeability. The darker shading is the thinner reef flank.

The 1-14KR trajectory is shown in Fig. 1. The dash line is the build section and the solid line is the horizontal section.

The well intersects both the reef's flank and core and runs parallel to the assumed natural fracture system that has a north-south orientation.

The vertical well bore section was drilled south of the reef to lessen drilling problems, such as lost circulation, and cementing problems associated with the long gas and solvent-filled sections of the reef's core.

This strategy offered the added advantage of maximizing well bore distance from the three gas injectors.

The thin remaining oil bank, averaging 15 m, is illustrated in Fig. 2. The south-north well trajectory is shown in elevation view, about 4 m above the original pool's oil/water contact.

Analysis of open hole porosity and micro-electric borehole image logs reveals that this well is in a very heterogeneous reservoir characterized by localized natural fracturing and significant porosity variations (Fig. 3).

HORIZONTAL OBJECTIVES

While the Rainbow 1-14KR was drilled primarily to evaluate horizontal technology's ability to reduce seN,ere gas and solvent coning, the purpose was also to address the thin remaining oil bank and the heterogeneous, fractured nature of the reservoir. Specific production logging objectives for 1-14KR were to:

Evaluate completion effectiveness by determining which intervals were contributing to the inflow, and then which of these were contributing to the high GOR and miscible solvent production. To determine the impact of the fracture system on this and future completions, flow and fluid density profiles were to be compared with naturally fractured intervals.
Use the logging results to improve well performance. Current facility limitations on gas compression capacity and a ceiling on average pool solvent/oil ratio makes remedial action a necessity. Intervals of low or no flow were to be identified for acid stimulation to increase oil inflow and oil recovery.
Provide input for designing a flexible completion that minimizes future workover requirements.
Provide an indication of a practical, optimum horizontal well length in this pool.
Prove horizontal well technology in this application can be successful and increase confidence to pursue other applications.

COMPLETION

As shown in Fig. 4a, Rainbow 1-14KR has a medium-radius build section of 15/30 m and is cased with a 178-mm (7-in.) cemented liner, The top of the liner is at 1,716 m measured depth with an inclination of 580.

The build section encountered the reef at a measured depth of 1,782 m and at an inclination of 74. The total measured depth of 2,250 m corresponds to a true vertical depth of 1,728 m.

The completion included perforating 252 m over a 443 m gross interval (Fig. 3) with 86-mm (3-3/8 in.) tubing-conveyed guns at 7 MPa (million Pascal) (1,000 psi) underbalanced. Perforating was completed in one run. No acidizing was performed.

The interval from 1,847 to 1,915 m was perforated with a shot density of 13 spm (shots/m) (4 shots/ft) to optimize inflow over the section of the reef flank where porosities are lower.

The other intervals in the reef's core where porosity is higher were perforated at 6 spm.

WELL PERFORMANCE

The well has produced oil at an average rate of 65 cu m/day since the original completion in April of 1990. Cumulative production, to September 1991, is 35,000 cu m (220,150 bbl).

The GOR increased from 400 to 2,300 cu m/cu m in the first 8 months of production.

After 1 year the well was observed to have asphaltene deposition in the well bore and the measured pressure drawdown was 2.3 MPa 14% of reservoir pressure, while producing oil at a rate of 100 cu m/day.

In January 1991, a wash of 5 cu m toluene followed by 15 cu m of 15% hydrochloric (HCl) acid was performed using coiled tubing. Drawdown was reduced to 1% of reservoir pressure. The well continued to produce oil at a rate of 100 cu m/day. GOR remained the same.

LOGGING CONSIDERATIONS

The importance of safe and efficient deployment, running, and retrieval of the logging tools cannot be underestimated. The logging tool string is deployed in the well with full shut-in tubing pressure at surface. In the case of Rainbow 1-14KR, gas pressure was 14 MPa with 0.5% H2S. Of major importance is an on site service company coordinator, trained and rested personnel, minimum on site support personnel, and avoidance of night operations.

Specialized procedures are required to work in a well under pressure and run and retrieve production logging tools. The use of coiled tubing to convey the tool string requires a large hydraulic injector head to push (and pull) the coiled tubing and tool string.

It is impractical to use a long lubricator section to contain the tools between the blowout preventers (BOPS) and the injector head while logging. This would result in the injector head being high aboveground.

A technique for deploying the tools into the well and removing the lubricator prior to logging was developed. The technique is a safe and efficient method of pressure deployment of logging tools (Fig. 5) and is listed in the box.

Fig. 6 illustrates the final surface equipment configuration for conveying production logging tools in the hole.

A reliable shear release or weak point is an important design consideration. This avoids premature separation of the logging tools from the coiled tubing during the makeup of the injector head and while running and retrieving the logging tools.

The release is for emergency purposes in the event tools become stuck in the hole. The release is located above the deployment bar, as shown in Fig. 5, and cannot be activated by compression.

There are two design features that reduce the risk of the shear release. The production tubing should have a bell-shaped wire line reentry guide complete with recess for the pin end of the tubing.

Additionally, tapered shoulders of the deployment bar eliminate hanging up on any hardware in the well as the tools move. In the event the tools are released from the coiled tubing, the fishing neck below the shear release has a chamfered edge to help engage an overshot and grapple.

Job planning includes a coiled tubing stress analysis. This is performed to ensure that the production logging tools can be pushed into and retrieved from the horizontal well bore with the selected coiled tubing size.

Primary inputs into the stress analysis program are directional survey data, coiled tubing size and weight specifications, and well bore diameter. Fig. 7 shows the profile of the horizontal section of the Rainbow 1-14KR well bore.

The 31.8-mm (1-1/4 in.) coiled tubing selected for conveying the logging tools was successful in pushing tools to total depth. But either asphaltene deposits or cement residues created some difficulties with the spinner data in the interval from 2,080 to 2,150 m.

LOGGING

The 19.9 m production logging tool string (Fig. 5) run in Rainbow 1-14KR included two velocity-measuring tools: the full-bore spinner and the diverter flowmeter.

Both tools are fully centralized for proper operation. The full-bore spinner, located on the downhole end of the tool string, measures fluid velocity by a centralized propeller. The spinner is affected by changes in fluid density and velocity, phase separation effects, tool speed, and direction of tool movement relative to flow direction.

The diverter flowmeter is used at stationary tool stops and focuses all flow through an expanded basket into a turbine meter. The tool provides a sensitive and accurate measurement of fluid and gas velocities at specific locations in the well bore. It is a valuable tool, particularly for differentiating low producing from nonproducing intervals.

Other tools in the string included the nuclear fluid density, which measures the average fluid density flowing in the well bore, pressure and temperature gauges, and a quartz pressure gauge.

The temperature gauge qualitatively indicates where gas or fluids are entering the well bore. Typically, gas entering the well bore is observed as a temperature drop due to gas expansion cooling.

The pressure gauges were run to measure drawdown, indicate when stable bottom hole flowing conditions were reached, and measure the pressure gradient along the well bore. Depth was controlled with a gamma ray and casing-collar locator tools.

Above the deployment bar is the weak point and coiled tubing head adaptor. The adaptor mates the logging tools to the electric wire line inside the coiled tubing. A conventional wire line logging truck is connected to the hub of the coiled tubing reel.

The first logging run was an up and down baseline pass with the well shut in. This pass captured velocity, temperature, and density profiles, against which subsequent flowing passes were to be compared. Observed trends of the flowing passes were later quantified for fluid and gas inflow rates and densities.

The well was initially opened to give an oil flow rate of 80 cu m/day with a GOR of 2,625 cu m/cu m and negligible water production.

After stabilization, multiple up and down logging passes were run at varying logging speeds to obtain the production logging data.

Logging was then conducted at a higher rate. Stationary stops with the diverter flowmeter were not obtained due to asphaltine plugging of its turbine.

The second higher rate was to be used to identify changes in inflow and fluid density at a higher drawdown. This would have helped substantiate the interpretation based on the first rate.

Unfortunately, the second production rate was not significantly higher than the first due to limited separator gas capacity. Both sets of production logging data were essentially identical.

RESULTS

Interpretation of the log data was based on observed trends and the use of a computerized production logging interpretation package. The computer program calculates downhole production rates on a zone-by-zone basis using an iterative technique of least squares reduction.

The solutions it offers represent a mathematical condition when the reconstructed log data best fit the actual log data.

Results from this program on the Rainbow 1-14KR data are shown in Fig. 8 and are summarized as follows:

No inflow was occurring from two of the seven intervals.
These intervals were from 2,197 to 2,225 m at the far end and from 1,945 to 1,967 m near the center of the horizontal well bore.
The majority (85%) of oil production originates from the interval 1,875 to 1,915 m, which is only 16% of the total perforated section.
The well bore below 1,967 m contributes only 8% of oil production but 62% of gas production. This section represents 109 m of perforations.
The quartz pressure gauge measured a drawdown of 0.080 MPa (0.5%) of reservoir pressure, while producing oil at a rate of 100 cu m/day with a GOR of 2,625 cu m/cu m.

ANALYSIS

Analysis of logging tool results with respect to reservoir characteristics and well bore geometry yield these conclusions:

The most useful tools in this particular logging program were the full-bore spinner and the nuclear fluid density.
The temperature tool measured very minor reductions in flowing temperature adjacent to some of the intervals suspected of contributing inflow. The low drawdown appeared not to permit sufficient gas expansion to result in a significant cooling effect.
The fluid density tool indicated that total gas/oil phase separation occurred very low in the build section, at 1,815 m.
Interpretation of log data from this well is difficult due to suspected phase separation in the 178 mm liner and the very large gas volume, compared to the oil volume. Gas volume flowing at bottom hole conditions was estimated at 1,051 cu m/day (37 Mcfd) and oil production at 106 cu m/day (667 b/d).
With only one perforated interval Contributing oil production, significant drilling and perforating damage is suspected and stimulation is required. The inflow that is dominated by one interval having a very low drawdown restricts potential inflow from tighter or damaged intervals, which in this case is a major percentage of the existing perforations.
Natural fractures, which occur below 1,882 m, appear to increase inflow particularly of gas from the core of the reef. The highest porosity interval, from 1,864 to 1,875 m (Fig. 3), is also highly fractured but does not appear to significantly contribute to gas or oil production.
The reef flank and outer core intervals over the gross interval 1,782-1,967 m (143 m of this interval is perforated) holds the highest potential for improving oil production rates while minimizing the risk of increased gas or solvent inflow.
The diverter flowmeter did not provide as useful information as expected because it was susceptible to plugging by the asphaltenes present in the well bore, and because of a flow rate limitation reached in this high GOR well. The tool could only have been used in the last half of the well bore due to its maximum fluid velocity limit of 3.3 m/min (10 fpm).

RECOMMENDATIONS

In the interest of improving oil recovery, and reducing gas and solvent production with a simple and flexible completion, recommendations for this well based on the production logging interpretation follow.

Set a drillable bridge plug at 1,973 m, which would retain the first three perforated intervals (143 m) for production.

Production test the three remaining intervals (1,782-1,835, 1,847-1,915, 1,945-1,967 m) together, or selectively, to verify the production logging interpretation.

Selectively acidize these intervals with conventional tubing-conveyed tools.

Based on production tests after setting the bridge plug, run a dual completion (Fig. 4b) for flexibility. The dual completion allows maximum flexibility for selectively producing intervals with varying drawdowns.

Due to GOR concerns, drill future wells only in the flank. To minimize phase separation effects in the horizontal and build sections of the well bore, which adversely impact logging interpretation and vertical flow efficiency, consideration should be given to:

Optimizing the liner size (in cased completions)
Installing appropriate size production tubing, landed in or near the horizontal section.

Logging tool and procedure recommendations are:

The diverter flowmeter provides for very precise measurements of fluid velocity but is susceptible to plugging. Trends can be as important as absolute velocities so the full-bore flowmeter may be sufficient for wells where plugging problems are anticipated.
A shorter tool string with fewer tools would simplify the current tool deployment procedure.
A downhole tension sub in the tool string would indicate whether obstructions are causing observed coiled tubing slack-off and inability to push logging tools, or whether the selected coiled tubing and well geometry are the limiting factors.
A 38-mm (1.5-in.) coiled tubing with minimum 2.77-mm (0.11-in.) wall thickness should be used to convey long or heavy tool strings, especially when hole diameters are large.
A minimum of two production rates with a significant variance should be planned to both quantitatively and qualitatively observe and compare inflow trends, particularly in velocity, density, and temperature, with increasing drawdown.

Rainbow 1-14KR was drilled in a very mature, miscible pool with a thin remaining oil bank. The well is susceptible to gas and solvent breakthrough. Several different strategies are available for producing such a well bore.

One option is to complete the well in smaller sections (100 m), starting at the end. When an interval experiences unacceptable productivity, it is abandoned and another section completed, closer to the build section. Through-tubing inflatable bridge plugs run on coiled tubing would help this option.

Another strategy that may increase inflow rates is to complete the majority of the well bore with sufficient blank casing sections to permit selective zone isolation, or production with multiple packers or sliding sleeves.

A cased-hole completion provides flexibility for profile control of oil, gas, solvent, and water. Production logging provides a method of determining what these profiles are when the well is not performing to potential.

The logging results by themselves do not provide absolute answers regarding fluid inflow and density profiles. However, they provide the best available estimates, verifiable by selective interval testing with packers or packer and retrievable bridge plug configurations.

POST WORKOVER

Because of production logging results, Well 1-14KR was worked over in October 1991. In the time between the logging and the workover, the GOR ratio had increased to 6,165 cu m/cu m with 0% water cut.

Production (Table 1) shows significant gas breakthrough in the three remaining perforation intervals following the workover.

The production results and acidizing behavior indicated segregation between the three intervals.

Although the majority of the well bore has oil production potential, particularly in the more isolated reef flank, the advanced stage of the Keg River AA miscible flood and the fractured nature of the reservoir significantly reduced the effectiveness of this application.

ACKNOWLEDGMENTS

The authors would like to acknowledge the contributions and guidance provided by J. D. (Jack) Loree and Whitecourt Well Servicing personnel, Gary Hall and K. M. (Keith) Senkoe, and thank Schlumberger of Canada and Dowell Schlumberger for their analyses.