Previous   |    Next

doi:10.2204/iodp.proc.336.103.2012

Operations

Port call and transit to North Pond

Prior to Expedition 336, the JOIDES Resolution was berthed in Curaçao. Although Expedition 336 officially did not begin until 15 September 2011 in Barbados, all of the CORK hardware and experiments were sent to Curaçao in advance of the expedition. In addition, a few scientists and engineers boarded the ship in Curaçao and used the 2 day transit to Barbados to start preparing the CORK observatories and a new in situ tool for detecting microbial life in ocean floor boreholes—the DEBI-t.

Expedition 336 officially began with the first line ashore in Bridgetown, Barbados, at 0948 h on 15 September. Other than personnel transfers, only minor port call activities were planned; these began immediately after the early arrival of the ship. On Day 2 of the port call, the remaining scientists, engineers, technical staff, and crew boarded the ship in Barbados on 16 September. The vessel was then secured for sea and departed Bridgetown, Barbados, at 0742 h on 17 September. The pilot departed the vessel at 0812 h, and the vessel started the 986 nmi transit to Hole 395A (all times are local ship time, Universal Time Coordinated [UTC] – 3 h). We arrived over Hole 395A at 2330 h on 20 September, having averaged 11.4 nmi/h. At 0012 h on 21 September, the vessel was placed in dynamic positioning mode over Hole 395A and operations began. All operational tasks, along with task start and end times, are listed in Table T1.

Hole 395A

Operations for Hole 395A began by picking up the CORK pulling tool for the Hole 395A CORK and making it up to a pony drill collar. The rest of the bottom-hole assembly (BHA) was attached and the drill string was tripped to just above the seabed. At 2166 meters below rig floor (mbrf), tripping operations were temporarily halted to install the vibration-isolated television (VIT) camera system to begin running the subsea camera. It quickly became apparent that there was a problem with the sonar system on the camera system, and it had to be retrieved for repair. Tripping continued to 2824 mbrf, where operations were again suspended to install the camera system. Tripping operations resumed as the camera system was carefully lowered toward the seafloor. Periodically, the camera system winch was stopped and the hoisting function was engaged. At ~3700 mbrf, it became clear that the winch did not have sufficient power to retrieve the camera system. Drill pipe tripping operations continued while mechanics and engineers diagnosed the winch problem. When tripping operations were complete, the top drive was installed. However, from 1630 h on 21 September until 1630 h on 22 September, diagnostics and repair were performed on the camera system winch. After both the hydraulic motor and the hydraulic pump were changed, the winch was restored to working condition. The camera system was deployed and lowered carefully to the seafloor. The Hole 395A reentry cone and CORK were located, and the pulling tool was lowered over the CORK and latched on. The CORK was then picked up ~7 m, and the core line was deployed to retrieve the data logger, which was attached to a 600 m long thermistor string. After running an overshot three times and attempting to jar onto the top of the data logger, attempts to pull the thermistor string were suspended and the CORK was pulled to the surface along with the data logger/​thermistor string. After the CORK was landed and secured in the moonpool, the recess where the top of the data logger was located was cleaned and the overshot was installed. The data logger was then jarred loose and all 600 m of the thermistor string, including 10 thermistors and the sinker bar, was removed at the rig floor. The thermistors were cut out of the string, and portions of the string were sampled for microbiological analysis. A lifting sub was installed on top of the CORK, and it was pulled through the rig floor and then moved to the starboard aft main deck for sampling. Next, the stinger, made up of three joints of 5.5 inch drill pipe, was broken down and laid out for additional microbiological sampling. After the rig floor was cleared of CORK pulling tools, the logging bit and BHA were made up and the drill string was tripped back to seafloor. There was a break in tripping operations after running Stand 70 to deploy the camera system, which was lowered to the bottom, following the bit. Hole 395A was reentered after 19 min of maneuvering. After rigging up to log with the DEBI-t microbiology string (DEBI-t, natural spectral gamma ray, and temperature), the tool string was run into the hole to log Hole 395A. After a 45 min interval to repair the logging winch, the tools were run down into the hole. The logging tool was unable to pass a section of the hole at 4670 mbrf (~186 mbsf), and after repeated attempts the logging string was pulled back to the surface and rigged down. On the basis of previous logs, we inferred that the tool string could not pass this section because it was hanging up on a ledge. After the logging bit was lowered ~21 m below the ledge, we were able to make two runs with the microbiology logging string all the way to the depth objective of 600 mbsf. The logging string was not lowered farther to keep it from landing on the bottom of the hole and causing potential damage to the tools.

After logging was concluded, the drill bit was also lowered to 600 mbsf to check for obstructions and make sure the hole was ready for the new CORK to be installed. When no obstructions were encountered, the string was pulled out of the hole and tripped back to the surface. After the drill line was slipped and cut, assembly of the new Hole 395A lateral CORK (L-CORK) commenced. The parts composing the main body of the CORK are listed in Table T2. The complete configuration of the CORK as deployed is shown in Figure F1. The configuration of the internal OsmoSampler string as deployed is documented in Table T3. Around midnight on 26 September, all casing, packers, umbilicals, and the CORK were assembled from the bottom to the top. Although the perforated casing was coated with epoxy to minimize the amount of exposed steel, the coated steel casing was washed with 10% ethanol and painted with an underwater curable epoxy paint. The deepest umbilical screens were deployed at the transition between the drill collars and 5.5 inch perforated casing. Umbilicals were secured with stabilizers (two per joint, with additional stabilizers near packers and screens) and sturdy plastic zip ties. Duct tape was used near packers to secure fittings. The CORK included a combination of steel and fiberglass casing. Details of connections, lubricants, and glues are presented in Edwards et al. (2012) and Orcutt et al. (2012).

The CORK was lowered ~10 mbrf to purge the pressure lines of air. The CORK was then raised to the moonpool and all of the purge valves were closed. All but one of the valves in the microbiology and geochemistry bays were closed. The open valve was attached to a fast-flow OsmoSampler with both standard and microbiology OsmoSamplers.

Next, we lowered the CORK to ~100 mbrf to ensure the camera system sleeve would pass freely over the CORK head. The OsmoSampler instrument string was then assembled and lowered inside the CORK. An attempt to land and latch the top plug was made, but the top plug could not be latched. The CORK was pulled back and landed in the moonpool, and an attempt was made to latch the top plug at the rig floor level. After numerous attempts with two different top plugs, it was finally decided to run the top plug without the latch being engaged; however, the top plug will still work as a gravity seal given the underpressured hydrologic system. When the CORK was pulled to the surface to check the top plug, it was apparent that the lateral valve and flow meter interface had broken off the lateral port on the CORK body. The port was then sealed with a 4 inch cap. It is unclear how the breakage occurred, and this type of valve was deployed successfully on two CORKs during IODP Expedition 327, on the CORK at ODP Site 1200, and on four cement deep-sea delivery vehicles.

Finally, we started lowering the CORK to the seafloor. The camera system was installed during the deployment and lowered to the bottom following the CORK. When the casing string stinger was just above the seabed, the drill string was spaced out for reentry and Hole 395A was reentered at 2003 h. The casing was carefully lowered into Hole 395A while observing the weight and carefully watching the string at known critical depths in the borehole (based on previous logging data). We then installed the top drive, and the drill string was spaced out to land the CORK. The CORK apparently landed, and over the next 2 h the packers were inflated to 1400 psi according to the inflation procedure and appeared to hold pressure; however, there was no slight decrease in pressure, typical of packers inflation. Simultaneously, the camera system was pulled to the surface, and we began putting together the ROV platform. At 0800 h the ROV platform had reached the CORK head and was released. The release was not smooth, with one side of the platform releasing before the other side. Eventually, the platform completely released and appeared to settle into position. An initial attempt was made to release the running tool from the CORK. After ~45 min of attempting to release the running tool, the camera was retrieved to remove the ROV platform release mechanism and slings in order to allow the camera to get a closer view of the CORK running tool. While the camera was being pulled to the surface, the driller lost the 10,000 lb of overpull that was being maintained on the CORK. At the time, we assumed that the running tool had released. However, when the camera system was lowered back to the seafloor to make a visual check of the CORK installation, we observed that the CORK head was no longer inside the reentry cone but was still attached to the running tool, offset from the reentry cone, and had broken off from the CORK casing below. We retrieved the camera system and drill string. Once the CORK head was back at the moonpool, we began to survey the damage (Fig. F2). The CORK head was then raised up to the rig floor, the running tool was removed, and the CORK head was laid down.

The CORK head experienced forces that bent the body through the lower part of the instrument bays. Also, the welded connection between the cup packer subassembly and the L-CORK mandrel had failed ~4 m below the top of the reentry cone (see Figs. F3, F4, F5, F6). This part of the assembly was a slip fitting that was not welded completely, and little of the connection was welded. When the 5 inch pipe was severed, the Spectra rope and umbilicals were also cut, leaving the downhole tool string in place. On the basis of recovered pieces of casing and the upper end of the remaining 5 inch diameter cup packer subassembly, the 5 inch pipe mandrel near the seafloor is not completely rounded but might not be closed enough to restrict the recovery of the downhole samplers, sensors, and experiments. Several stainless steel tubes likely extend above the cup packers and the top of the 5 inch casing (Table T4). These tubes may impede recovery of the downhole instrument string. Schematics of the parts remaining in the uppermost Hole 395A CORK installation are shown in Figure F7.

After the 10.75 inch casing in Hole U1382A was released (see “Operations” in the “Site U1382” chapter [Expedition 336 Scientists, 2012b]), a camera survey was made of the Hole 395A reentry system to provide initial data for formulating a plan to recover the downhole instrument string in 4 y with an ROV. The ROV platform was slightly offset from the center of the reentry cone, which made it nearly impossible to see down into the casing where the remainder of the CORK must be located. One of the three sonar reflectors appears to be missing (Fig. F8), and some views of the ROV platform, which rests on the cone and is not latched, are suggestive of possible damage to the platform.

The CORK pressure logging system was recovered along with the broken-off wellhead. The recorded data, shown in Figure F9, do not definitively resolve whether or not the downhole CORK packers actually inflated. The seafloor gauge and three formation gauges show very similar slight increases in pressures after the time of attempted inflation, but the similarity to the seafloor gauge could be consistent with a tidal influence without inflation of the downhole packers. However, there may be no evidence that the packers did not inflate based on analysis of the overall volume of the system, inflated packers, packer setting line, drill string, and standpipe in relation to the pressure change observed. The inflated packer volume represents only ~0.1% of the total system volume. Thus, if the overall system is truly closed (i.e., there are no leaks), then no additional pumping would be required to inflate the packers after shutting in the inflation pressure. Also, the pressure change observed during the packer inflation process indicates, assuming a closed system, a volume change equal to ~25 times that of the total volume of the inflated packers. Some of this volume change is most likely leakage in the system. Thus, based on onboard data and initial analyses, we cannot say whether or not the packers are inflated; however, we have no reason to believe that they are not inflated.

Our next objective was to install a CORK in Hole U1382A (50 m west of Hole 395A) to monitor the uppermost basaltic crust. However, this plan was put on hold when we had to leave the area because of Tropical Storm Philippe. After the ship was secured for transit at 0215 h (29 September), we departed Hole 395A and headed to the northeast to avoid the approaching storm.

Engineering summary

The first North Pond objective, pulling the old CORK and thermistor string in Hole 395A, was successfully completed.

The new Hole 395A L-CORK, a design modified to fit DSDP reentry hardware, failed during the final stages of installation in the borehole. Preliminary analysis suggests that the bottom ~0.6 m of the L-CORK hung up, indicated by damage to the leading edge of the lowermost stabilizer fins (Fig. F6, F4). This exposed the L-CORK—designed to be much longer and narrower than usual to fit the DSDP-style cone while still accommodating broader science objectives (i.e., extended height for the bays in which the scientific experiments are placed)—to much higher bending loads than anticipated. It is believed the system cracked at a weld just above the landing seat, which led to even higher loads and bending of the wellhead itself and then eventual failure at the weld. A final report will be prepared once the L-CORK, video evidence, and instrumentation data are further reviewed (see Edwards et al., 2012). However, examination of the recovered old Hole 395A CORK (Fig. F10) revealed damage to its stabilizer fins in the same area as the new Hole 395A L-CORK that failed. We also observed a noticeable lack of corrosion to its landing ring. Combine this with the inability of the old CORK to be latched-in during Leg 174B and one might assume it also landed high. The old CORK was shorter and more robust, especially in the area just above the landing ring, so it was exposed to much lower stress. As for why the CORKs landed high, we have very little information. One theory is that the interior diameter of the reentry cone’s 24 inch transition pipe, through which the stabilizer fins must pass, is narrower than indicated on DSDP engineering diagrams.

Top of page   |    Previous   |    Next