Floyd E. Bettis, Brian J. Schwanitz
Schlumberger Well Services
Anchorage
Lowell Crane
ARCO Alaska Inc.
Anchorage
A field study, involving logging runs with an ultrasonic imaging tool, of several wells drilled in Alaska indicates casing wear from drill pipe rotation can be reduced if the dogleg severity is kept less than 50/100 ft.
The dogleg severity must be kept very low to limit casing wear if the drill pipe rotating time is expected to be more than a few days.
Factors related to casing pipe wear include drill pipe hardbanding, hole deviation, and directional dogleg severity. Excessively worn casing may not withstand pressure tests, may develop holes, or may cause operational problems.
The Ultra Sonic Imaging (USI) tool measures the extent of casing wear from normal drillstring rotation. USI logs are run to evaluate cement and measure casing thickness (results are presented in terms of metal loss).12
In Alaska, more than 90% of the oil and gas wells are drilled directionally, with deviations ranging from a few degrees to greater than 900. Under these conditions, casing wear during drilling has always been a potential problem.
Also, the use of new directional drilling technology has the potential to increase the severity of the casing wear problem. The steerable bent-housing/mud-motor assembly has allowed increased drilling rates while keeping the well bore very close to the projected course. During steering operations to correct the well bore course, however, the drilling penetration rate is low. Thus, the well course is often rapidly brought back to the projected course, resulting in a high dogleg severity, often greater than 101/100 ft.
Field examples clearly illustrate the strong correlation between increased dogleg severity and increased casing wear.
The USI uses a rotating transducer for 100% coverage of the casing. Four measurements are made from the echo received by the transducer: casing thickness, rugosity (wrinkles), cement acoustic impedance, and internal radius. In a casing wear study the internal radius and rugosity are of greatest importance. The casing thickness measurement obtained from casing resonance is also valuable under some conditions.
The following field examples studied the effects of drill pipe hardbanding, dogleg severity, well bore deviation, casing grade, rubber protectors on the drill pipe, and rotating time on the extent of casing wear.
WELL A
During drilling operations below the 7 1/8-in. intermediate casing, metal filings were found in the shale shaker. The drilling contractor had recently renewed the hardbanding on the drillstring tool joints; thus, casing wear was probably occurring.
The pressure test of the recently cemented 5 1/2-in. liner failed, so a leak at the 5 1/2in. liner top was suspected.
The USI log was run to evaluate the cement quality behind the liner and simultaneously inspect the 7 1/8-in. casing for wear. Fig. 1 is the USI log of the interval from 2,050 to 2,250 ft, and the casing wear is shown very clearly. Track 2 is the amplitude of the energy returning from the internal casing surface. Roughness caused by corrosion or abrasion (a groove cut by drill pipe or wire line operations) reduces the energy returning to the sensor.
Track 4 displays the average variance radiuses of the casing. The red coding indicates the radius is greater than the nominal internal radius of the casing. The blue coding indicates less than the nominal internal radius of the casing.
Track 5 presents the maximum, minimum, and average internal radius of the casing. The external nominal radius is also shown.
Track 6 is the casing thickness computed from the casing resonance. The maximum thickness loss shown is flat at 30% because the minimum thickness that can be measured by the USI log is limited by the frequency range of the transducer. With a transducer frequency range of 650-190 khz, a casing wall thickness 0.177-0.59 in. can be measured. In this example, the thickness is below the minimum range of the tool. Therefore, the internal radius measurement, determined from the first echo arrival, is used in the casing wear study.
Track 7 is the percentage total wall loss determined from the thickness measurement. (Note: In Fig. 1, Tracks 2, 5, and 7 are two-dimensional views of the measurements covering 100% of the casing circumference.)
A number of holes are present in the interval shown in Track 5. A hole is indicated if the internal maximum radius extends beyond the external radius of the casing. A distorted egg-shaped casing could have a maximum internal radius that extends beyond the average external radius of the casing; however, the minimum internal radius would also be much less than the nominal internal radius of the casing.
Note in this example that the minimum internal radius was very close to the nominal internal radius of the casing, indicating that the casing was not oval and that holes were present.
After the casing pressure test failure, a casing cutter was run, and the damaged 7 1/8-in. casing was recovered (Fig. 2). The damaged casing sections were measured with an ultrasonic tool at the surface. The results from the USI log and from the surface measurements were plotted for comparison (Fig. 3). The uncertainty in the USI log radius measurements is - 0. 008 in., and the surface ultrasonic tool method has an uncertainty of - 0. 003 in.
The correlation was strong (0.93). One location (2,166.52,167.5 ft) on joint No. 52 had the most variance. The USI log indicated 100% metal loss, while the surface method indicated a 77% metal loss. Visual inspection of the casing found a yield where the casing stretch exceeded the elastic limit of the steel, which allowed the maximum internal radius to exceed the external radius of the casing. Therefore, the USI log maximum radius method showed this area as a hole.
The casing wear occurred only over a small interval of the angle-building section of the borehole. Also, only a low percentage of the highly deviated section, where angles reach 600, had casing wear.
Additional well data were reviewed, including dogleg severity, azimuth, and hole deviation from the Guidance Continuous Tool (GCT). Table 1 is the GCT log data from the interval where the casing failed during the pressure testing of the top of the 5 1/2-in. liner. During the drilling of this interval, hole angle was built while the hole azimuth changed from 95.180 to 89.730, causing a sharp dogleg to the left.
Drilling operations are normally performed with the drill pipe under tension. In these conditions, while building hole angle and with the drill pipe in tension, it was expected that the casing wear would occur at or very near the high side of the casing, which is the 0-3600 location on the USI log plot.
Surprisingly, the casing wear, as shown on the USI log, occurred near 900 from the top side of the casing as shown in Fig 1. Combining the GCT log data with the USI log data indicated clearly that the casing wear occurred on the inside of the sharp dogleg to the left.
This correlation was of interest, so the study of casing wear and dogleg severity was extended to additional wells where USI log data and GCT log data were available.
(Note: Drill pipe rotating time on this well was very short. The drill pipe was also round-tripped in the well twice.)
WELL D
After the identification of serious casing wear in a well where the 9 5/8-in. casing had failed during drilling of the 8 1/2-in. borehole, a number of changes in operating procedures were suggested to reduce the casing wear.
One of the proposed operating procedure changes was to set the 9 1/8-in. casing in tension. Because the 9/8in. casing was being set with cement covering a few hundred feet at total depth with an additional 2,000 ft of arctic-pack cement covering the permafrost interval, the expansion of the casing as it reached formation temperature could result in increasing the dogleg severity in the uncemented interval of the borehole.
Well D was selected to test the results of setting the 9/8in. casing in tension. Well D was logged with the Ultra Sonic Imaging tool (USI) after total depth was reached and the 7-in. liner was cemented in place. The 9/8-in. casing was exposed to wear from drill pipe during drilling of the 8 1/2-in. borehole and during cementing and clean-out operations as the 7-in. liner was being set.
Fig. 4 is a plot of dogleg severity computed from the GCT log data. The greatest dogleg occurred at 1,500 ft true vertical depth (1,502 ft measured depth).
Fig. 5 is the USI log in this interval. Casing wear occurred near the high side of the casing, confirming that a change in deviation, not a change in azimuth, caused the dogleg severity (Table 2). The deviation increased from 11.10 at 1,505 ft to 23.60 at 1,700 ft. Borehole azimuth changed less than 20 in the same interval.
A less serious dogleg occurred from 1,929 to 2,186 ft.
Azimuth changed from 15.690 to 9.820 while deviation increased from 28.190 to 34.370.
Because the borehole course turned to the left as the deviation increased, the casing wear should have occurred on the inside of the turn (the left side of the casing). The USI log in this interval confirms the casing wear occurred at 80-900 from the high side of the casing.
The correlation of metal loss with doglegs remained very strong in Well D. The combination of the relatively short rotating time, the lower dogleg severity (maximum of 3.10/100 ft), and the setting of the casing in tension were expected to eliminate most of the casing wear. Unfortunately, these changes did not reduce the casing wear, so additional operating changes were made.
WELL C
The casing in Well D was exposed to drilling operations for only a short time, whereas the casing in Well C was exposed to drilling operation for a much longer period. Well C had hole problems, and drill rates were slow after the 9 1/8-in. protection string was set.
The interval from 4,110 to 4,260 ft is in the angle-building section. Azimuth information is not available on Well C because the GCT was not run. Table 3 lists the deviation data from the general purpose inclination tool, which was included in the logging tool string.
The USI log showed the wear occurred on the high side of the casing in this interval, indicating an increase in deviation, not azimuth, caused the wear. The maximum metal loss in this interval exceeded 50%.
Fig. 6 shows the USI log in a lower interval of Well C. Well C was drilled with an S profile, with the borehole deviation reduced in this lower interval. The casing wear here occurred on the low side of the casing, indicating that the casing pipe wear was caused by a change in deviation, and not a change in azimuth.
The change in deviation from 31.50 at 6,800 ft to 27.40 at 7,100 ft caused a severe dogleg (Table 3). Casing metal loss exceeded 50% in these intervals: 6,896-6,910 ft, 6,959-6,964 ft, and 6,987-6,989 ft. The maximum metal loss in the section was 63%.
The burst strength of 9 5/8-in., 47-lb/ft N-80 casing is 6,870 psi. Well test procedures required applying 4,200 psi to the 9 5/8-in. string. With the large loss of metal shown by the USI log, the test procedure was modified to apply only 3,200 psi to the 9 5/8-in. string. During the test, however, the 9 5/8-in. casing failed at 2,575 psi. After the 9/8-in. casing failed, the burst strength of the casing was recalculated using a wall loss of 63% in the internal yield equation. The calculated burst strength of 2,382 psi agrees well with the actual failure of 2,575 psi.
WELL B
In Well B, casing protectors were installed on the drill pipe to reduce casing wear. Ideally the rubber protectors would have been placed on the drill pipe such that the rubber protectors would not enter the uncased section of the borehole. Because the 9 5/8-in. casing extended to only 5,946 ft and total depth was below 10,700 ft, it was not possible to keep the rubber bumpers in the cased section of the well without a large number of drill pipe trips.
The danger of losing an excessive amount of rubber in the well during drilling greatly increased. Also, a large number of rubber protectors on the drillstring were expected to reduce the mud circulation rate, increasing cleaning difficulty. Furthermore, when the rubber protectors entered the open hole section of the borehole the rotating torque increased sharply.
Fig. 7 is an interval of Well B's USI log. Dogleg severity was acquired with the GCT log survey shown in Track 1. A severe dogleg was present from 1,795 to 2,050 ft. Borehole deviation increased from 27.130 to 44.880, while azimuth changed from 105.410 to 103.540 (Table 4). The small change in azimuth to the left caused casing wear on the inside of the turn.
The wear occurred about 450 to the left of the high side of the casing (Fig. 7). Here the loss is 40% or less. The rubber protectors on the drillstring were not very successful in preventing casing wear where the dogleg severity was extreme.
Total rotating time on this well was 41.5 hr, plus two round trips. Even with the limited rotating time, the casing wear was still excessive with the casing protectors installed.
The USI log found casing wear in another section from 4,900 to 5,020 ft, where the dogleg was caused by a change in deviation from 55' to 590. Again, there was a strong correlation between high dogleg severity and significant casing wear. A 10 turn to the left also occurred in this interval and was reflected in the wear again being to the left of the high side of the casing (Table 4).
In the extended reach section of a borehole, doglegs are not usually present, and the drill pipe rests on the low side of the casing. Casing wear therefore occurs on the low side of the casing, very near 1800 from the high side of the casing.
In the extended reach section of this well, the wear occurred near 1800, as expected (Fig. 8). With dogleg severity less than 20/100 ft, metal loss was very low in the interval from 5,170 to 5,270 ft and was caused by the drill pipe resting on the low side of the borehole in the extended reach interval.
WELL E
The casing run in Well E was upgraded from J-55 to L-80 to determine the effect of casing grade on wear in the high dogleg severity intervals. Fig. 9 shows the USI log from 500 to 700 ft, which corresponds to the top of the kick-off interval.
The maximum metal loss of 19% occurred at D-51 ft. The maximum dogleg severity in this interval was 6.60/100 ft, which was not as severe as the dogleg severity found in some of the other field examples (Table 5).
Upgrading the casing reduced the amount of metal loss that occurs in both the well kick-off interval and intervals of increased dogleg severity.
WELL F
During this study on casing wear, the rig changed to using nonhardbanded drill pipe on Well F.
Casing wear occurred at the shallow kick-off point in this directional well. Metal loss peaked at 31% in the kick-off interval. In the kickoff interval a maximum dogleg severity of 4.260/100 ft Was computed from GCT data (Table 6).
The maximum dogleg severity in Well F occurred from 1,250 to 1,400 ft. Dogleg severity reached 6.60/100 ft. Fig. 10 is the USI log from this interval. Maximum metal loss of 33% occurred at 1,345 ft.
The use of nonhardbanded drill pipe did not control the amount of metal loss in the increased dogleg severity intervals. Well F was drilled with a well controlled dogleg severity compared to other wells in the field study. The maximum dogleg severity of 6.60/100 ft at 1,379 seven ft was the only dogleg greater than 5'/100 ft in Well F.
RESULTS
In Alaska, the USI log has been a useful quantitative tool for measuring casing wear severity. The condition of the drill pipe hardbanding is very critical. If the hardband has the tungsten steel chips protruding above the weld surface, the casing can be destroyed in a short time.
- Hardbanding. Well A showed some results of incorrect hardbanding. The casing was destroyed in a very few rotating hours. Because the damaged casing was cut and removed from the borehole, direct comparisons of the USI log results and surface ultrasonic metal loss determination methods were possible, and the two methods correlated well.
- Steerable motors. The drilling procedures on directional wells are very important. With the steerable (bent housing) motors in the bottom hole assembly, the borehole course can be returned to the correct projected course very rapidly, reducing steering time, and increasing drilling rates. This procedure can cause high dogleg severity, contributing to increased casing wear in those intervals.
- Casing in tension. In drilling Well D, a correlation between dogleg severity and casing wear was found.
The casing running and cementing procedures were changed to determine if casing metal loss could be reduced by setting the casing in tension. The setting of the casing in tension could not be correlated to the less extreme casing wear in this well. Metal loss was generally low in this well. Rotating time was limited because the footage drilled below the casing shoe was lower than the average for all the wells included in the study.
- Rotating time. Well C showed the correlation between casing wear and dogleg severity in a well with a greater-than-normal rotating time. Two intervals had greater than 50% loss of metal, and during a pressure test the 9 1/8-in. casing burst with 2,575 psi applied instead of the 6,870 psi rating.
The very serious casing wear in this well correlated to both high dogleg severity and longer rotating time. This example may help explain the occurrence of holes in some older producing wells, where corrosion logging tools indicate no corrosion is occurring from electric current flow. General low corrosion, which would not cause a problem in areas of no wear, could cause holes to develop in intervals of increased casing wear.
- Rubber protectors. The data from Well B support the correlation between high dogleg severity and casing wear. The casing wear on this well was still serious, although casing protectors were added to the drillstring, and the total rotating time was limited to 41.5 hr including two round trips of the drillstring. The use of the casing protectors resulted in additional drilling problems: Reduced mud circulation rates, loss of rubber in the well bore, increased rotational torque, and additional clean up time before the 7-in. liner could be run.
The protectors were less effective than the nonhardbanded drill pipe.
- Casing grade. The 9 5/8-in. casing in Well E was upgraded from J-55 to L-80. Well E had the least casing wear of all the wells in the study. However, it should be noted that the dogleg severity only exceeded 50/100 ft in the interval where the maximum wear occurred. Reduced dogleg severity combined with the L-80 grade of casing resulted in casing wear occurring over the shortest interval in Well E compared to all the other wells in the field study.
- No hardbanding. The rig used nonhardbanded drill pipe on Well F. Less casing wear should occur during drilling rotation with pipe with no hardbanding. The nonhardbanded drill pipe, combined with reduced dogleg severity, produced little improvement in controlling casing wear.
At no point in this well did the dogleg severity exceed 6.50/100 ft. The interval drilled below the casing shoe was greater than the average of the wells in the field study, increasing the rotating time. Maximum casing wear reached 33% in Well F.
The dogleg severity must be kept very low if the drill pipe rotating time is expected to be more than a few days.
the data gathered to date suggest casing wear from the drill pipe is reduced if the dogleg severity remains below 50/100 ft of borehole.
REFERENCES
- Hayman, A.J., Hutin, R., and Wright, P.V. ' "High-Resolution Cementation and Corrosion Imaging by Ultrasound," presented at the Society of Professional Well Log Analysts Symposium, Midland, Tex., June 17-19, 1991.
- Hayman, A.J., Gai, H., and Tome, 1. "A Comparison of Cementation Logging Tools in a Full-Scale Simulator," presented at the Society of Petroleum Engineers Annual Technical Conference and Exhibition, Dallas, Oct. 6-9, 1991.
Copyright 1994 Oil & Gas Journal. All Rights Reserved.
