OPERATIONS

Hammer Drill System Components

The HDS tested during Leg 179 was a concept assembly comprised of seven basic components: (1) an underreaming bit, (2) a water hammer, (3) a jet sub, (4) a running tool, (5) a hanger bearing assembly, (6) a reentry cone, and (7) the casing string (Fig. F8). The overall HDS is an adaptation of similar hammer drill systems used on land, in particular by the geothermal industry. However, some fundamental aspects of the HDS had to be changed, or added, for deployment at sea.

Hammer Drill Underreaming Bits

Hammer drill bits drill by crushing rock under extremely high point loads using hemispherical tungsten carbide inserts (TCIs) as the cutters. The cutters are driven into the rock with each impact of the hammer, thus chipping a small portion of the rock with each blow. The bits are rotated slowly, ~20 rpm, to index the cutters between impacts of the hammer. Underreaming bits are required to open the borehole large enough for the casing to follow behind the drill bit as the hole is being drilled. The underreaming bits are designed to collapse to a small enough overall outside diameter to be pulled up through the casing string once the casing has been emplaced.

There were two basic designs of underreaming bits used during Leg 179. Both types were direct adaptations of land-based hammer drill underreaming bits currently used in industry. The first type is called a concentric underreaming bit (CUB; Fig. F9) and is a relatively new development in the hammer drill industry. The CUB has a pilot bit, ~12 in in diameter, sized such that it will pass through a 13-in casing string. Immediately above the pilot bit are three underreaming arms that are retracted and expanded by rotating the drill string left or right, respectively. When retracted, the underreamer arms close to the same outside diameter as the pilot bit (12-in). When expanded, the underreaming arms open to an effective diameter of 14-in, thus creating a large enough borehole for 13-in casing to pass through. The advantage of the CUB is that the underreaming arms ream ~84% of the borehole circumference with each stroke of the hammer. Based on data collected from land-based operations, the CUB has proven to drill faster and last longer than conventional eccentric underreaming bits.

The second type of underreaming bit used during Leg 179 is called an eccentric underreaming bit (EUB; Fig. F9). In the EUB the underreamer and the pilot bit are one piece. The EUB is built such that when it is in the closed position, the pilot bit is off axis to the drill string and the overall effective diameter (12-in) of the bit is small enough to be pulled up through 13-in casing. When opened, the pilot bit moves on axis with the drill string and the eccentric is moved outward to perform the underreaming. Thus, when open, the EUB has an effective diameter of 15 in and is capable of creating a borehole large enough for 13-in casing to pass through. As with the CUB, the EUB is retracted and expanded by rotating the drill string left or right, respectively. The EUB has been used for years in the land-based hammer drilling industry. The disadvantage to the EUB is that the eccentric only reams ~38% of the borehole circumference with each stoke of the hammer. Thus, the EUB drills slower and does not last as long as the CUB.

Water-Powered Hammer Drill

The heart of the HDS is the water-powered hammer drill (Fig. F10). As the name implies, the water-powered hammer drill is driven by pumping water through the hammer. The basic operating mechanism is an internal reciprocating piston. On the upstroke, the piston is slowed and stopped by compressing water. On the downstroke, high-pressure water drives the piston down until it impacts the top of the bit. The high-energy impact is transmitted through the bit body to the TCIs, thus creating extremely high, virtually point, impact loads on the rock.

Another feature of the hammer drill is a bypass mechanism that allows the driller to flush the borehole with high-viscosity mud without activating the hammer. When weight is applied to the hammer with the bit set on bottom, the bit shank moves upward, closing the bypass and diverting all flow through the hammering mechanism. When the bit is pulled clear of bottom, the bit shank is allowed to move downward, opening the bypass and diverting all of the flow around the hammering mechanism so that the hammer stops operating when not in contact with bottom.

The water hammer used during the testing during Leg 179 is a proprietary product of SDS Digger Tools, Pty., Ltd., 49 Vulcan Road, Canning Vale, Western Australia 6155 (telephone (09) 455 4433; fax (09) 455 4399). Specific operational parameters of the water hammer can be obtained by contacting SDS Digger Tools.

Jet Sub

A special jet sub (Fig. F10) replaced the conventional water hammer top sub. The jet sub has receptacles for three nozzles capable of diverting part of the flow down the drill string and up the outside of the drill string at high velocities. When assembled in the complete HDS, the jet sub is placed ~2 m up inside the casing. While drilling in casing with the HDS, the cuttings are brought up the inside of the casing through the annular space formed by the casing inside diameter and the drilling assembly outside diameter. The jet sub is used to increase the velocity of the cuttings-laden water moving up the casing and out of the hole for more efficient hole cleaning.

Running Tool

When casing is conventionally drilled in with hammer drills on land, individual joints of casing are added to the overall casing string at the surface as the casing string is being drilled in. Unlike the conventional land-based hammer drill-in casing systems, the HDS, because it is to be deployed in deep water, requires that the entire casing string be made up as a single assembly with the HDS and lowered to the seafloor. Thus a special running tool, which becomes an integral part of the drilling assembly, is required to support the casing string as it is lowered to the seafloor and drilled into place. The running tool must also be able to unlatch from the casing string and be removed with the drilling assembly, thus leaving the drilled in casing fully open for reentry.

The HDS running tool employs a triangular cross section body (Fig. F11). The flats of the triangle provide flow paths for the cuttings to be circulated out of the hole while drilling. At each of the points of the triangle are latch dogs that, when extended, lock into mating grooves in the casing hanger. The latch dogs are held out, in the locked position, by a shifting sleeve inside the running tool body. To unlatch the running tool from the hanger bearing latch body, the shifting sleeve must be moved downward, out from underneath the latch dogs, thus allowing them to retract into the running tool body.

A special tool called a go-devil (Fig. F12) is used to move the running tool shifting sleeve. At the time when the running tool is unlatched from the casing hanger, the drill-string heave compensator is in operation and thus it is not safe to access to the drill-string bore to insert the go-devil. To get around this problem, there is a hydraulically actuated ball valve, on top of the top drive, which is normally opened and closed when retrieving core barrels during routine coring operations. When the HDS is deployed, the go-devil is placed on top of the ball valve, with the ball valve closed. To deploy the go-devil, the driller opens the ball valve, from the safety of the driller's shack, allowing the go-devil to fall into the drill string. Once the go-devil is inside the drill string, it is pumped down the drill string until it comes to rest on top of the running tool shifting sleeve. After the go-devil has landed on the shifting sleeve, the drill-string pressure is increased to ~600 psi until the shifting sleeve overcomes a snap ring and moves downward, releasing the latch dogs. After confirming the running tool has been unlatched from the casing hanger, the driller increases the drill-string pressure to ~1800 psi, shear releasing another sleeve inside the go-devil, which also moves downward and opens a circulation path to the hammer and borehole once again.

Hanger Bearing Assembly

A hanger bearing design was incorporated into the casing hanger to allow the drilling assembly and integral running tool to rotate relative to the casing string while supporting the weight of the casing string (Fig. F13). During the drilling-in process, as the casing enters the borehole, the bearing assembly enables the casing to stop rotating, even though the drilling assembly is still being rotated. By doing so, the total torque required to drill the casing in is reduced. Also, whatever torque is produced is a direct response from the bit, thus giving the driller direct feedback regarding torque on bit.

Casing String

As mentioned previously, unlike conventional land-based hammer drill in casing techniques where the individual joints of casing are added to the string as it is being drilled in, the entire HDS casing string must be made up as part of the overall HDS assembly. The top of the HDS casing string must be compatible with other standard ODP casing tools and hangers. So, a standard ODP casing hanger is slightly modified by adding a hanger bearing assembly and shortening the casing pup joint for the HDS. For added protection at the bottom of the casing string, a hardened casing shoe is welded to the end, or shoe joint. The hardened casing shoe is more collapse- and abrasion-resistant than the casing itself.

Reentry Cone

A reentry cone was added to the HDS assembly to aid reentry to the borehole and to defeat the hanger bearing assembly (Fig. F14). The vibration-isolated television (VIT) camera is used to locate specific spud targets when spudding with the HDS. Because the VIT uses the drill string as a guide and must pass over the HDS assembly, the reentry cone can not be in place, on top of the HDS, while drilling in. Therefore, the HDS reentry cone was designed to be deployed after the casing has been drilled into place. The reentry cone is split into two halves and is attached around the drill string, while the drill string is still attached to the casing string, and free-fall deployed. The falling reentry cone comes to rest on top of the HDS casing hanger. As the reentry cone is falling, a guide on top of the HDS running tool centers the reentry cone with respect to the drill-string axis so that the body of the reentry cone passes over the outside of the casing hanger and extends down to the hanger bearing assembly housing. Special lugs inside the reentry cone body, near the top, land on top of the casing hanger, thus preventing the reentry cone from dropping below the casing hanger.

The standard ODP casing tools latch and unlatch by rotating the drill sting left and right respectively. Because the HDS hanger bearing allows the casing hanger to rotate relative to the casing string, it must be locked out for standard ODP casing tools to be used during subsequent operations at an HDS installation. The HDS reentry cone also provides a mechanism for locking out the hanger bearing assembly. The special lugs that land on top of the casing hanger also lock into the bypass flow grooves in the body of the casing hanger, thus preventing rotation between the casing hanger and the reentry cone. There is another set of lugs inside the reentry cone, near the bottom, that engages lugs on the outside of the hanger bearing housing, thus preventing rotation of the reentry cone with respect to the hanger bearing housing. Therefore, rotation of the hanger relative to the casing string is prevented.

Summary of Leg 179 Engineering and Drilling Operations

Port Call: Cape Town, South Africa

While in port, we learned that the surface freight that contained many parts of the hammer drill system shipment would not arrive before the ship got under way. A quick inventory of HDS equipment already on board was made. The following HDS items were identified as being in the delayed surface shipment:

Three double-pin crossover subs (OG0709; NC 70 × NC 70). One crossover sub OG0725 (NC 70 pin × 6-in full hole modified [FHM] box) was modified into one double-pin crossover sub OG0709 (NC 70 × NC 70) at a local machine shop in port.

Two HDS hanger bearing lower bodies (OJ5010). There was one HDS hanger bearing lower body on board. It was not possible to fabricate replacement HDS hanger bearing lower bodies in port or on board ship.

Three HDS hanger bearing caps (OJ5012). One HDS hanger bearing cap was fabricated at a local machine shop in port.

Two complete HDS reentry cone assemblies (OJ5014). It was not possible to fabricate a complete HDS reentry cone assembly in port or on board ship.

One HDS reentry cone (OJ5015). One drill-in casing reentry funnel (OJ4852) was modified into one HDS reentry cone. With one HDS reentry cone body (OJ5016) on board, one complete HDS reentry cone assembly (OJ5014) was fabricated on board ship.

Two 13-in ST-L flush joint casing lift subs (OH5159). Two 13-in ST-L flush joint pin connections were cut off of two damaged joints of 13-in ST-L flush joint casing and modified into two 13-in ST-L flush joint casing lift subs. Note: these are not certified lift subs and were destroyed after completion of the hammer drill testing.

Three 13-in buttress casing collars. The 13-in buttress casing collar is used to cross over from the HDS hanger bearing lower body to the 13-in ST-L flush joint casing string via a double-pin casing crossover sub (13-in buttress × 13-in ST-L). There were no 13-in buttress casing collars on board or available in port. The buttress pin connection was cut off the only HDS hanger bearing lower body on board. One double-pin casing crossover sub was cut in half and the 13-in ST-L pin connection welded onto the HDS hanger bearing lower body, thus eliminating the need for the 13-in buttress casing collar.

Four 9-in drill collar lift subs (OD0208). The 9-in drill collar lift sub was replaced by one crossover sub (OG0725; NC 70 pin × 6-in FHM box) and one 8-in drill collar lift sub (OD0200), which were on board. There were only two crossover subs OG0725 on board after one was modified into a double-pin sub (OG0709). So, only two 9-in drill collar lift subs were available.

Two 9-in drill collar bail-type lift nubbins (OG0245). Fortunately the 9-in drill collars came with cast steel thread protectors of the bail type so they could be used in place of the lift nubbin. Note: the cast steel thread protectors are not certified and should not be used routinely as lift nubbins.

Three 13-in Holte hardened casing shoes. The 13-in hardened casing shoes could not be fabricated in port or on board ship.

Two 13-in HDS modified casing hangers. The 13-in HDS modified casing hangers could not be fabricated in port or on board ship. One 13-in HDS modified casing hanger was on board, allowing for one casing deployment.

Two HDS hanger bearing latch bodies (OJ5009). The HDS hanger bearing latch body could not be fabricated in port or on board ship. One HDS hanger bearing latch body was on board, allowing for one casing deployment.

Transit: Cape Town to Site 1104

During the transit from Cape Town to Site 1104, the SDS water hammer drill was picked up and deck tested on 28 April. The test assembly consisted of an SDS concentric underreaming bit, SDS 12-in water hammer, SDS jet sub, with three blank nozzles installed, and the required crossover subs to the top drive. After making up the assembly, the bit breaker was placed on top of some dunnage and rubber matting on the rig floor. The bit was then lowered into the bit breaker and 5000 lb applied to the bit via the top drive. The mud pump was engaged, and the flow rate slowly increased to 75 gpm at 700 psi, when the hammer first began to cycle. As the flow rate was increased the hammer cycled intermittently and erratically. The flow rate was increased to 375 gpm at 1770 psi, and the hammer began to cycle more evenly. The flow rate was increased to 400 gpm at 1900 psi and the hammer cycled smoothly. The hammer was cycled for several minutes before the flow rate was reduced to 240 gpm at 900 psi. The hammer was then cycled for a few minutes more before the pump was shut down.

It was theorized that the initial erratic behavior of the hammer was caused by air in the pumping system and excess grease left in the hammer from assembly. To test the theory, the mud pump was once again engaged and the flow rate slowly increased to 75 gpm at 270 psi, and the hammer began cycling very smoothly. To create a baseline pressure vs. flow-rate curve, the flow rate was increased in 50-gpm steps, and the corresponding pressure was recorded. Large vibrations were noted in the stand pipe and derrick, presumably from pressure pulse reflections from the hammer traveling back to the pump (Table T1).

Also during the transit, a frequency analyzer was assembled on board to monitor the hammer-induced pulsation frequency in the stand pipe as an aid in determining when the hammer was cycling. Although the initial frequency spectrum recorded was not a clear indication of when the hammer was operating, the voltage output spectrum from the pressure transducer installed in the stand pipe gave a good indication of when the hammer was cycling.

It is interesting to note that later in the hammer drill tests, additional filtering that made the frequency spectrum more prominent was added to the frequency analyzer. The frequency spectrum indicated a notable peak at ~30 Hz, the known operating frequency of the hammer. However, there was an even more prominent peak at ~60 Hz that was believed to be an indication of the power stoke of the hammer at 30 Hz, plus the return pulse as the hammer piston moved upward at 30 Hz and 180º out of phase, essentially doubling the frequency to 60 Hz. The increased amplitude of the 60-Hz signal may indicate that more energy is transmitted up the stand pipe by the hammer piston upstroke than is reflected by the power stroke when the piston moves downward. This further supported the theory that the stand-pipe vibrations were indeed caused by hammer-induced pressure pulse reflections traveling back to the pump.

Site 1104

A positioning beacon was deployed at 1925 hr on 29 April, establishing Site 1104. The hammer drill was prepared for deployment for the first series of spudding and drilling tests without casing. The bottom-hole assembly (BHA) used for the HDS testing was made up of SDS concentric bit 1, a hammer drill, a jet sub, a crossover sub (OG0726), four 9-in drill collars (OG0244), a crossover sub (OG0725), one 8-in drill collar (OL1040), one tapered drill collar (OG0300), six joints of 5-in drill pipe (OG0052), and a crossover sub (OG1010) to 5-in drill pipe.

The BHA was tripped to the seafloor and the camera deployed for a seafloor survey. The point of reference for the survey was the hard-rock guide base at Hole 735B. Once Hole 735B was located, the ship was offset ~75 m west to the primary HDS testing site, where massive, sediment-free outcrop was observed. However, based on the most recent survey information received from H.J.B. Dick aboard the James Clark Ross (pers. comm., 1998), a second test site was explored ~200 m northwest of Hole 735B. Massive, sediment-free outcrop was also observed at the second site, and the decision was made to begin the hammer drill testing at the second site. For better positioning of the ship, a second positioning beacon was deployed at 0224 hr on 30 April.

Site 1104 Spud Test

Water depth was determined to be 740 meters below rig floor (mbrf) by drill-pipe measurement. With the camera deployed and no rotation of the drill string, several spud tests were conducted (Table T2). The hammer performed well during the spud tests, and the decision was made to recover the camera and move on to the drilling tests.

Hole 1104A

Hole 1104A was spudded at 0620 hr on 30 April, with an SDS underreaming bit (Table T3). The pointed pilot bit did not skid as the hole was initiated. However, it is suspected that the bit heaved off bottom occasionally, thus starting a new hole. Only enough weight on bit (WOB), ~8000 to 10,000 lb, was applied to the hammer to keep the bit on bottom and the hammer bypass closed. It appeared that once the pilot bit was below the seafloor, the hammer performed better. Heave constantly caused the hammer bypass to open, causing the hammer to stop cycling and then restart. After ~1 m penetration the torque increased and became erratic. Excessive vibration in the stand pipe and derrick was noted.

After ~45 min of hammering and 1.5 m penetration, the bit was pulled clear of the seafloor with a slight overpull of 10,000 to 15,000 lb. The camera was deployed to observe the borehole, which was found to be a clean symmetrical circle in the rock outcrop. The bit was set back on the seafloor, and while maintaining WOB, the camera was retrieved in preparation for another drilling test.

Hole 1104B

With the camera back on board, Hole 1104B was spudded at 0830 hr on 30 April (Table T4). The water depth was determined to be 739 mbrf by drill-pipe measurement. The hammer began cycling smoothly as the flow rate was slowly increased with no rotation of the drill string. After a few minutes of spudding, the top drive was engaged to begin rotation of the drill string. Excessive vibration in the stand pipe and derrick were once again noted. At 0900 hr on 30 April, the rig air pop-off valve failed and drilling had to be stopped, while the bit remained in the hole. The pop-off valve was soon isolated and drilling resumed. Torque soon increased and became erratic. The hammer bypass was constantly opened, presumably by heave, causing the hammer to stop cycling and then restart. At 1130 hr on 30 April, after hammering for ~2 hr with 1.5 m penetration, the bit became stuck. The bit was freed at 1150 hr on 30 April, and kept on bottom as the camera was deployed to observe the bit and borehole. The bit was pulled clear of the seafloor at 1230 hr on 30 April and appeared to be intact. The borehole was somewhat oval shaped, presumably from being spudded on a slope.

The bit was set back on the seafloor, and while WOB was maintained the camera was retrieved in preparation for spudding Hole 1104C. With the camera back on board, the pump was engaged, and the flow rate slowly increased (Table T5), but the hammer would not cycle. The hammer was pulled clear of the seafloor to open the bypass and be flushed. However, similar results occurred when the bit was set back on the seafloor, closing the bypass. The standby pump (1) was engaged to make sure a problem with the pumps did not exist. The same pressure drops vs. flow rates were recorded, and the hammer did not cycle. The hammer and bit were retrieved for inspection.

Once back on board, the hammer was disassembled and it was determined that the hammer valve had cracked, allowing fluid to bypass it and thus preventing the hammer from cycling. It is theorized that the pressure transients across the valve created by the constant opening and closing of the hammer bypass may have been the cause of the cracking. A new valve was installed in the hammer, and the hammer was deck tested. The hammer was cycled for ~6 min at 100 gpm and 330 psi and at 200 gpm and 560 psi, with no problems. The pressure vs. flow rate curve was the same as for the new hammer, possibly indicating that no appreciable wear had occurred on the piston or other internal parts of the hammer during the two previous runs. The SDS CUB was not reusable. The TCIs on all three of the underreaming arms were sheared or broken off, except for the last one on each of the trailing edges (Fig. F15). Heavy abrasion was also observed on the surfaces of all the underreaming arms. The pilot bit was in good shape, except for one chipped TCI.

Hole 1104C

With a new bit (SDS CUB 2) installed on the refurbished hammer, the BHA was tripped back to the seafloor. The camera was deployed and a spudding location chosen. The bit was set on the seafloor and, while maintaining WOB, the camera was retrieved. Once the camera was back on board, Hole 1104C was spudded at 0105 hr on 1 May (Table T6). The water depth was established at 739 mbrf by drill-pipe measurement. The pump was engaged and the flow rate increased slowly with 5 to 10 rpm drill-string rotation. After ~10 min of drilling and 0.5 m penetration, there was a pressure loss observed, presumably from the hammer bypass opening, and the flow rate was slowed. The flow rate was once again increased with normal pressure vs. flow rate correlation, and the hammer restarted smoothly.

After ~30 min of drilling with 0.5 m penetration, the bit heaved off bottom, causing the hammer to stop, and the flow rate was reduced. It was thought that the bit may have heaved out of the hole and a new hole started as the flow rate was increased and the hammer restarted. Almost immediately the torque increased and became erratic. The drill-string rotation was erratic, from 10 to 50 rpm, as a result of slip stick. The hammer had to be stopped and restarted several times because of high torque buildup resulting in top-drive stalling.

At 0225 hr on 1 May, a 2-in nipple on the stand-pipe manifold failed because of the vibration in the stand pipe. Drilling had to be stopped so that the pump could be shut down for manifold repair. The bit remained in the borehole while the manifold was repaired, and the camera was deployed. Near the hole with the bit in it, three other holes were observed, confirming that the bit had indeed heaved out of the hole and started new holes. The bit appeared to be 0.5 m below the seafloor. The camera was retrieved and, with the stand-pipe manifold repaired, drilling resumed at 0334 hr 1 on May.

The flow rate was slowly increased and the hammer began to cycle. Torque was low and erratic, but increasing. During rotation the drill string was sticking and slipping. The top drive stalled on several occasions, and the hammer was stopped and the torque released. Each time the hammer restarted without any problems. At 0410 hr on 1 May, the bit became stuck and may have heaved out of the hole as it was freed. Drilling resumed until 0425 hr on 1 May, when the stand-pipe transducer failed because of stand-pipe vibration. Because the pumps had to be shut down to remove the pressure transducer from the stand pipe for repair, the bit was pulled clear of the seafloor.

Hole 1104D

The stand-pipe pressure transducer nipple was blanked, and Hole 1104D was spudded at 0445 hr on 1 May (Table T7). The water depth was determined to be 739 mbrf by drill-pipe measurement. The pump was engaged, and the flow rate slowly increased. The bit heaved off the seafloor several times, causing the hammer to stop and restart. The torque soon increased and became erratic, eventually stalling the top drive. At 0510 hr on 1 May, drilling was halted and the camera was deployed while the bit remained on the seafloor. It appeared that several new holes had been spudded because of the bit heaving out of the hole during spudding. At 0615 hr on 1 May, the bit was pulled clear of the seafloor and the camera retrieved. The bit was also retrieved for inspection because of a lack of penetration.

Once on deck the bit was inspected revealing the leading two TCIs on each of the underreamer arms were sheared or broken off (Fig. F16). The gauge surfaces of the SDS bit 2 underreamer arms were not as heavily abraded as those on the SDS bit 1. Except for the pilot bit nose TCI having been sheared or broken off, the rest of the pilot bit appeared to be in good shape. The pilot bit nose TCI was probably damaged as the bit was heaved off the seafloor during spudding.

Hole 1104E

It appeared that the underreaming arms were preventing the hammer drill from advancing the borehole. To test this theory, parts of the underreaming bits used during the previous tests were converted into a drilling bit.

Bit Modification 1. Using a torch, the underreaming arms of SDS bit 1 were trimmed such that when opened, they would not extend past the outside diameter of the pilot bit and driver. Upon reassembly of the bit, the modified underreamer arms, when opened, did not appear to be strong enough to withstand drilling in hard rock. Therefore, bit modification 1 was abandoned.

Bit Modification 2. The second attempt at modifying an underreamer bit into a drill bit involved removing the underreamer arms and pilot bit from SDS bit 1. The pilot bit shank was shortened such that when installed in the driver, the underreamer arm gap was closed. The pilot bit was then installed and welded directly to the driver. Unfortunately, the pilot bit cracked during the welding process, and the bit could not be deployed.

Bit Modification 3. The third attempt at modifying an underreamer bit into a drill bit involved replacing the pilot bit with the broken nose TCI on SDS bit 2 with a new pilot bit. The original, damaged underreaming arms from SDS bit 2 were left in place. However, the underreaming arms were welded in place in the closed position. Because the underreaming bit was fixed in the closed configuration, new waterways had to be cut through the toes of the underreaming arms, using a torch and grinder.

The modified SDS bit was made up to the hammer drill BHA and tripped to the seafloor. The camera was then deployed to locate a spud target. The modified bit was placed on the seafloor ~1 m from Holes 1104C and 1104D. Maintaining WOB, the camera was retrieved and Hole 1104E was spudded at 0140 hr on 2 May (Table T8). Water depth was established at 740 mbrf by drill-pipe measurement. The weather began to deteriorate, and the rig floor was experiencing 1 to 2 m heave, resulting in the WOB having to be increased to 10,000 to 12,000 lb to keep the bit on bottom and the hammer bypass closed.

The pump was engaged, and the flow rate slowly increased. The hammer cycled smoothly but there appeared to be ~100 psi less pressure at any give flow rate than in past tests. There was also a noticeable reduction in the vibration in the stand pipe and derrick. At ~0155 hr on 2 May, the hammer heaved off bottom, opening the bypass, and thus the hammer quit cycling and had to be restarted. The flow rate was increased slowly once again, and once again the hammer began to cycle.

After ~1 m penetration the torque began to increase and become erratic. Heave at the rig floor had increased to 3 m. The top drive stalled at 24,000 ft-lb. An overpull of 40,000 lb was applied to the bit without freeing it. The BHA was lowered to close the hammer bypass. The hammer was restarted and cycled at 400 gpm at 1720 psi. The pipe was worked again with up to 40,000 lb overpull, and still the bit could not be freed. Finally the drill string was rotated left, the direction one would normally rotate to close the underreaming arms, and the bit came free. It was assumed that the welds had failed, allowing the underreaming arms to open, so the bit was pulled clear of the seafloor and the camera was deployed to verify this. Once the camera had reached the end of the pipe, the underreamer arms could be seen clearly in the open position. The camera and BHA were retrieved for inspection. Once on deck, it was observed that every weld on the modified bit had failed, allowing the underreaming arms to open. Also, ~2 in of the leading edges of all three of the underreaming arms were broken off (Fig. F17). The pilot bit appeared to be in good condition.

HDS Testing Hiatus: Site 1105

With no other hammer drill bits on board other than SDS concentric underreaming bits, the decision was made to suspend further HDS testing pending the arrival of a supply vessel, the La Curieuse, from Reunion Island. The lost surface shipment had been located, and critical items were diverted to Reunion Island for delivery to the drillship via the La Curieuse. Also, a flat-faced standard hammer drill bit was sent to Reunion Island by SDS for delivery to the drillship via the La Curieuse. While waiting for the arrival of the La Curieuse, the drillship was moved ~1 km northeast of Site 1104 where Site 1105 was established. Hole 1105A was drilled to a depth of 15 mbsf using a 14¾-in tricone drill bit. The hole was then cored to a depth of 158 mbsf with 82.8% recovery, using the rotary core barrel. After logging Hole 1105A, the drillship was moved back to Site 1104 in anticipation of the arrival of the La Curieuse and the resumption of hammer drill testing. A new beacon was dropped at Site 1104 at 1938 hr on 10 May, establishing Site 1106.

The La Curieuse arrived on location at 2045 hr on 10 May and requested that the off-loading of cargo and personnel wait until daylight. Rough seas and high winds prevented the off-loading from taking place the following day, 11 May. The sea state and winds had deteriorated further, and the forecast for the following 48 hr showed no signs of improvement. There were three hammer drill bits on board the La Curieuse that were critical to completion of the hammer drill tests, one Holte CUB, one Holte EUB, and one SDS flat-face drill bit.

The hammer drill bits were successfully transferred from the La Curieuse to the drillship by tying buoys onto the drill bits and dropping them over the side. The drillship was then maneuvered to catch the buoys, and the ship's crane was used to hoist the drill bits on board. Because it was too rough to attempt off-loading any other cargo or personnel, the La Curieuse was released to return to Reunion Island at 1000 hr on 12 May.

Hole 1106A

A Holte CUB was made up to the SDS hammer drill, and using the same HDS BHA configuration as previously used, the assembly was lowered to the seafloor. The camera was deployed for a seafloor survey. The Site 1104 holes were quickly located. So as not to confuse the Site 1104 holes with the Site 1106 holes, the ship was offset ~20 m to the north on the same outcrop. It was theorized that the nose TCI damaged on the previous bit may have resulted from the bit heaving off the seafloor as we attempted to keep the bit in place on the seafloor during retrieval of the camera. The decision was made to spud Hole 1106A with the bit off bottom. Thus the camera was retrieved with the bit clear of the seafloor.

Once the camera was back on board, the pump was engaged and a flow rate of 200 gpm established. Circulation was maintained for several minutes to flush any air out of the system. While maintaining a 200 gpm flow rate, and no rotation of the drill string, at 1600 hr on 12 May, the bit was lowered to the seafloor and Hole 1106A was spudded (Table T9). The water depth was determined to be 740 mbrf by drill-pipe measurement. The hammer began to cycle immediately and the flow rate was slowly increased. The heave at the rig floor was estimated to be 2 to 3 m and, as a result, the bit was pulled off bottom several times, causing the hammer to stop and restart. After ~10 min of drilling, penetration began to be made and the hammer operation settled out somewhat. Heave was still a major problem and, because of a cross swell, roll factored into the difficulties as well.

The drill bit had penetrated ~0.5 m when the torque began to increase and become erratic. The weather was deteriorating, and heave at the rig floor resulted in constant stop-start problems with the hammer. Several yellow (2%) automatic station keeping (ASK) system alerts occurred, indicating the dynamic positioning (DP) system was having a hard time holding the ship on station in the increasing wind. The DP operators also reported that the noise from the hammer was occasionally interfering with the acoustics of the ASK system. After ~3 hr drilling and 2 m penetration, 1000 psi was lost and the hammer quit running. The pump was stopped and the backup pump (1) was engaged with similar results, indicating the problem was downhole. We suspended drilling operations and retrieved the hammer for inspection.

With the hammer and bit on deck at 2230 hr on 12 May, an inspection was carried out. The Holte CUB had suffered much the same damage as did the SDS bits (Fig. F18, bit shown before deployment). All of the TCIs on the underreaming arms, except for the last one on the trailing edges, were broken off. Also, four TCIs near the shoulder of the pilot bit were broken off as well. Disassembly of the hammer revealed the valve was cracked similarly to the first cracked valve. The other internal parts of the hammer appeared to be in good shape.

Hole 1106B

A new valve was installed in the hammer, and a Holte eccentric underreaming bit was attached to the SDS hammer. The hammer and bit assembly was then deck tested. The hammer cycled immediately and was cycled for 6 min. The hammer and eccentric bit were made up to the HDS BHA and tripped to the seafloor. The camera was deployed to locate a spud site. After locating the spud site and with the bit off bottom, the camera was retrieved. The pump was engaged, and a flow rate of 150 gpm was established. At 0640 hr on 13 May, the bit was lowered to the seafloor and Hole 1106B was spudded (Table T10). The water depth was determined to be 741 mbrf by drill-pipe measurement. After tagging bottom, the flow rate was slowly increased. It appeared as though the bit may have skidded downhill ~0.5 m during spudding. The hammer began to cycle smoothly, and penetration was being made when the bit appeared to heave out of the hole and skid downhill ~1 m. The torque immediately increased and became erratic, causing the top drive to stall. Heave continuously opened the hammer bypass. After 1 hr of drilling with virtually no penetration, we stopped drilling and retrieved the bit for inspection.

Inspection of the Holte EUB revealed similar wear patterns as observed on the concentric bits (Fig. F19). The outer edge of the eccentric bit was severely abraded and most of the TCIs on the outer edged of the eccentric bit were broken. Several TCIs on the pilot bit shoulder were also broken. The EUB was determined not to be usable.

Hole 1106C

The SDS flat-face drill bit (Fig. F20) was the last hammer drill bit on board to test. Although the aim of the HDS is to drill in casing, which requires an underreamer bit, the drill bit was deployed in an effort to prove that (1) the hammer drill could drill hard rock and (2) it was the premature deterioration of the underreamer arms that was preventing deep penetration. In anticipation of a long drilling run, the hammer drill was disassembled and a new cartridge was installed. The drill bit was attached to the refurbished hammer drill, and the assembly was deck tested. The hammer cycled perfectly and was run for ~2 min. We reduced the flow rate to 130 gpm and picked up the hammer to check the bypass, which opened as expected.

The drill bit and hammer drill were made up to the same HDS BHA and tripped to the seafloor. The camera was deployed to locate a specific spud site for Hole 1106C. After retrieving the camera, the pump was engaged at 150 gpm to flush air out of the system. While we maintained flow rate, the drill bit was set on the seafloor at 1840 hr on 13 May, and Hole 1106C was spudded (Table T11). Water depth was determined to be 742.5 mbrf by drill-pipe measurement.

After ~25 min of drilling and 1 m penetration, the pressure began to increase and the hammer began to cycle intermittently. The bit was raised off bottom to open the bypass and flush the hammer. When the bit was set back on bottom, the pressure again began to rise and the hammer still cycled intermittently. The back-up pump (1) was engaged and similar events occurred, indicating the problem was probably downhole. So, the hammer was retrieved for inspection. While the hammer was retrieved, the drill string stayed full of water. This was an indication that the hammer check valve was not allowing the water inside the drill string to drain out as it was retrieved.

Once on deck, the hammer was disassembled for inspection. The initial cartridge was removed, and the coating on the piston appeared to have chipped off. The piston was found to have galled to the lower bushing and was stuck in the full up position. With the piston being stuck in the full up position, the check valve was prevented from opening and the water could not drain from the drill string. The stuck piston was also thought to be the cause of the high operating pressure and intermittent cycling. The drill bit was found to be in good condition and reusable.

Hole 1106D

The hammer was refurbished with another complete cartridge, and the same flat-faced drill bit was installed. The assembly was then deck tested, and the hammer performed as expected. The hammer and bit were then made up to the same HDS BHA and tripped to the seafloor. The bit was set on the seafloor, and at 0340 hr on 14 May, Hole 1106D was spudded (Table T12) with 2 m heave at the rig floor. The flow rate was slowly increased, and the hammer cycled normally.

After ~4 min of drilling, the pressure began to rise and the hammer began to cycle intermittently. The flow rate was reduced to stop the hammer and then increased slowly to restart the hammer. As before, the pressure continued to rise and the hammer cycled intermittently. The bit was raised off the seafloor to open the bypass and flush the hammer. When restarted, the hammer once again operated intermittently at higher than normal pressure. It was assumed that the piston and lower bushing had galled again, so the hammer was retrieved for inspection.

Once on deck, the hammer was disassembled for inspection. The initial cartridge was removed and the coating on the piston once again appeared to have chipped off. The piston was found to have galled to the lower bushing and was stuck in the full up position. The extent of the galling did not appear to be as bad as that observed in the cartridge used in Hole 1106C. Once again, with the piston being stuck in the full up position, the check valve was prevented from opening, and the drill string had to be pulled full of water. The drill bit, however, was found to be in good condition and reusable.

Hole 1106E

The hammer was rebuilt with the same piston and valve. However, the lower bushing used in Hole 1106C had been repaired in the ship's machine shop, and it was assembled in the hammer. The drill bit was still in good condition, so the hammer and drill bit were made up and deck tested. The hammer was cycled for 2 min at 200 gpm and 650 psi, which are nominal readings. A noticeable reduction in the stand-pipe and derrick vibrations was observed, even though the flow rates and corresponding pressures were consistent with those of a new hammer.

The bit and hammer were tripped to the seafloor with the same HDS BHA configuration. The pump was engaged at a low flow rate of 150 gpm to flush out any air in the system. After a few minutes of flushing, the bit was lowered to the seafloor and Hole 1106E was spudded at 1140 hr on 14 May (Table T13). The drilling depth was determined to be 741 mbrf by drill-pipe measurement. During this time, the average heave at the rig floor was estimated to be 3 to 4 m, with occasional 5-m heaves.

After ~15 min of drilling, the hammer drill began to make significant penetration. Despite constant opening of the hammer bypass caused by heave, the hammer drill continued to advance the borehole. Torque was slightly erratic, ranging from 2500 to 5000 ft-lb for most of the drilling. Occasionally the top drive would stall. However, when this happened, the hammer was allowed to keep cycling, and it soon drilled itself off, allowing the drill-string rotation to resume.

After ~1 hr, 40 min and 8 m penetration, the stand-pipe pressure transducer nipple failed because of stand-pipe vibration, and the pump had to be stopped to repair it. The bit was pulled 4.5 m off the bottom of the hole, with a momentary 20,000 lb overpull. Slow rotation of the drill string was maintained as the pump was shut down. The stand-pipe bull plug containing the pressure transducer was removed, and a blank bull plug was installed in its place. The repairs took ~5 min.

When the pump was engaged, little or no pressure was observed, consistent with pumping through open-ended drill pipe. The back-up pump (1) was engaged with similar results, indicating the problem was downhole. The drill string was then lowered in anticipation of tagging the bottom of the hole. After the end of the pipe had been lowered 4.5 m below the last TD of Hole 1106E, we decided to pull the bit clear of the seafloor and deploy the camera for observation.

Once the camera had reached the end of the pipe, it appeared as though all of the drill collars were still intact, and the crossover sub between the drill collars and the jet sub on top of the hammer could be seen. However, the jet sub, hammer, and bit could not be seen. The end of the pipe and the camera were lowered to survey the seafloor. Several boreholes were observed. One of the boreholes appeared to have something in it, but it could not be confirmed as being the hammer. The sea state at the time of the survey caused the view of the seafloor to move in and out of focus. A further survey of the seafloor did not reveal the hammer, so either the hammer was still in the hole, out of sight, or had dropped onto the seafloor and rolled downslope. The camera and drill string were then retrieved.

When the end of the drill string was retrieved, all of the drill collars and the crossover sub between the drill collars and the jet sub were recovered. The jet sub, hammer drill, and bit were missing. The pin connection on the bottom of the crossover sub showed signs of having pulled out of the box connection on top of the jet sub. Two theories have been put forth as to when the failure may have occurred. The first theory is that the jet sub box connection had been weakened or even split by the pounding the bit was taking as a result of the excessive heave during spudding. When the bit was pulled off bottom and the momentary 20,000 lb overpull was observed, the bit may have hung up on the borehole wall and the crossover sub pin may have pulled out of the weakened jet sub box. The second theory is that while waiting on the stand pipe to be repaired with the hammer heaving in the borehole, the BHA may have leaned over, causing the jet sub box to fail. By pulling the bit 4.5 m off the bottom of the hole, the jet sub was positioned at, or near, the seafloor, compounding the bending problem.

Although enough spare parts were on board to assemble a second hammer drill, the decision was made to halt the hammer drill testing because there were no more hammer drill bits available to be tested that would have increased the hammer drill test data base. Also, because of the weather conditions, lack of reentry hardware, time constraints, and high probability of loss of the fishing equipment, it was not thought prudent to attempt to fish for the lost hammer. The 9-in drill collars used in the HDS BHA were inspected, with no cracks found, and laid down. The drillship was secured for sea and at 2312 hr on 14 May, the JOIDES Resolution got under way for the next site.

Engineering Results and Accomplishments for Leg 179

Although the complete HDS test plan could not be conducted because of the premature failure of the underreaming bits, a great deal of data was collected and a much better understanding of the HDS as deployed at sea was gained. Of primary concern was the performance of the hammer. Considering the sea state during most of the test, the hammer performed quite well. It must be noted that the hammer used for the tests was designed for drilling a 12-in borehole and that during the testing it was used to drill a 14-in borehole, an ~45% larger hole, thus reducing the efficiency of the hammer. The water hammer was also designed to work at maximum efficiency with a 2250-psi pressure drop across the piston. Because of the excessive vibrations in the pumps and stand pipe, the maximum continuous pressure drop across the hammer piston that could be maintained was ~1750 psi, which further reduced the hammer's efficiency. However, in spite of the low hammer efficiency, a rate of penetration of 4.8 m/hr was achieved in Hole 1106E in massive gabbro, using a 12-in standard hammer drill drilling bit.

Further analysis of the underreaming bits is required. However, first impressions are that the BHA was leaning over during spudding during the early stages of drilling. Being composed of 9-in drill collars, the BHA was very stiff and so, as the BHA leaned over from being placed in compression, it caused the underreaming bits to be rotated about the horizontal axis (perpendicular to the drill-string axis). This rocking of the underreaming bit in the hole during drilling probably caused extremely high loads to be placed on the low side of the bit underreaming arms. The high loads resulted in shearing off the TCIs and severely abrading the underreaming arms themselves. Once the TCIs were broken off of the underreaming arms, the arms acted much like a bearing, preventing further penetration by the bit.

It is encouraging to note that the pilot portion of all the underreaming bits came out of the hole in good condition, further indicating that the hammer drill can penetrate subsea hard-rock formations and that the premature failure of the underreamer arms is what prevented deep penetration by the bit. The standard hammer drill drilling bit used in Holes 1106C and 1106D showed no signs of wear when retrieved, plus the excellent penetration rate observed in Hole 1106E further indicates that the overloading of the underreaming arms was the cause of lack of penetration with the underreaming bits.

There were three primary problems associated with the hammer drill during the tests. Twice during the testing, the hammer internal control valve cracked, thus allowing pressure to escape past the valve, preventing the piston from cycling. Failure of the valve appears to have been from two causes. First, the valve has some ports in a highly stressed area, causing a definite stress riser in the valve body. Second, the hammer was constantly opened and closed because of heave. When this occurred, the pump could not be stopped in time to prevent the pressure drop across the valve from dropping to near zero and then almost instantaneously increasing to ~1750 psi. The combination of the stress riser and the pressure cycling probably caused the cracking to occur in the valves.

The second problem that occurred with the hammer drill during the tests was that on two separate occasions the piston began galling to the lower bushing. A hard coating had been applied to the piston where it passes through the lower bushing. A close tolerance fit is used in place of a dynamic seal between the piston and lower bushing. It appears that the hard coating may have been spalling as a result of cavitation erosion. The small flakes of the hard coating appeared to be wedging between the piston and the lower bushing, causing the galling to occur. This theory will have to be studied further in a metallurgical laboratory. In any case, the galling resulted in sluggish and erratic operation of the hammer; thus, the hammer had to be retrieved and the piston and lower bushing had to be replaced each time the galling occurred.

The third problem with the hammer drill during testing was associated with the stroke length (40 mm) required to open the bypass, thereby stopping the hammer from cycling. This stroke may be too short for deployment of the hammer drill from a floating vessel. Adding to the problem is a piston effect on the bit caused by the pressure drop across the hammer acting on the bit shank cross section area. This piston effect causes the bit to be pumped downward, thus opening the bypass, as the hammer drills off or is raised off bottom. Once the bypass opens, the power fluid is diverted around the piston, preventing the hammer from cycling. Thus whenever a large heave occurs, which the heave compensator can not completely adjust for in the drill-string motion, the hammer bypass is opened and the hammer stops cycling. None of the seafloor hardware or casing running tools was deployed during the testing, so no data on the performance of this equipment was obtained. However, this equipment was assembled and fit tested without problem.

Conclusions

The Leg 179 HDS tests were designed to be a test of the overall HDS concept in actual sea conditions. As such, the tests provided a wealth of data (Tables T1, T14) that, with further study, should provide a clear indication as to the direction in which the HDS development should proceed. The hammer drill itself shows great promise of being able to penetrate subsea hard-rock environments. Although the hammer drill performed well considering the sea state during the testing, a dialogue with the manufacturer will be established to address issues such as the short stroke length required to open and close the bypass, the valve cracking problem, and galling of the piston and lower bushing.

It is evident that the underreaming bits designed for conventional land-based hammer drill operations, such as those used during Leg 179, are not suitable for drilling in casing in offshore deep water with an unsupported BHA. Ideas for new design underreaming bits are already being formulated. The HDS casing running tools and reentry cone were not deployed during Leg 179. However, they were land tested before Leg 179, and no problems were encountered. So, at this time, no redesign of these tools is planned. In general, confidence remains high with the overall HDS. The benefits of the HDS to ODP and the science community as a whole are well worth continuing with the HDS development.