Deepwater Technology Report

New mud supply process saves time and expense in riserless deepwater drilling

Drilling riserless with a mud system that mixes correct-density drill fluids on demand allows surface casing to be set deeper and through shallow hazard zones

Michael B. Johnson, Deepwater Technology Manager, Fluids Technology, Baker Hughes INTEQ

escribed here are a methodology and process being employed to redefine interval limitations in deepwater operations. Discussions include limitation of conventional methods, types of riserless fluids used today, and details of a new process that supplies higher-density fluids by mixing concentrated mud onsite with seawater to avoid high-volume storage / supply limitations. Selected case histories illustrate applications in more efficiently running 20-in. casing to provide the base for more efficient final-string design.

Deepwater drilling is taking place in certain offshore areas through young sedimentary sections often characterized by narrow margins between formation pore pressure and fracture pressure. As a result, drillers are challenged to prevent the upper-hole section from collapsing.¹ The traditional approach of running an additional casing string may not be viable in deepwater wells due to engineering constraints or contingency cost.

The alternative is to extend the casing-setting depth, thus extending depths of subsequent casing strings. This methodology employs Dynamic Kill Drilling (DKD) fluids and related equipment offered by Baker Hughes INTEC, to mix weighted, pre-formulated fluids with seawater while drilling the upper hole sections (riserless interval).

In riserless drilling, the decoupling of hydrostatic pressure above and below the mud line allows for bottomhole pressure (BHP) management with increasing depth. The casing interval can be extended, due to the higher mud-pressure gradient, compared to drilling in a riser mode. This methodology has been developed and used in numerous deepwater drilling operations to extend surface-casing depths. Significant cost savings have been realized by: 1) reduced number of work boats utilized; 2) reduced storage area for kill / pad fluids; 3) greater success in running casing to bottom, and on the first attempt; and 4) fewer casing strings required to achieve desired TD.

Shallow Hazard Management

The DKD methodology was first developed to address catastrophic well loss associated with shallow-water flows coupled with low fracture gradient and increasing BHP. Pelletier, et al noted that, "Shallow-water flows may occur while drilling a shallow, over-pressured formation at deepwater sites. Penetrating an over-pressured sand without a weighted mud can lead to large washouts and caves, and formation compaction and collapse."² 

Another need for a reliable process arose from deep reservoir targets that required under-reaming to achieve needed casing sizes. Casing could be set above the hazard, allowing the mud density to be increased, while reducing the potential for lost returns. But this option was not desired in many instances as it could jeopardize reaching all the targets. Thus, a reliable process to push the 20-in. casing past the hazard and eliminate a pipe string was needed. For this process to work properly, very large fluid volumes possessing the appropriate mud properties would need to be pumped at the precise density.

Dual-Gradient Drilling

The new process employs a dual-gradient concept, consisting of seawater hydrostatic above the mud line and the ability to vary hydrostatic pressure below the mud line through mud-weight variations. Unlike mud-lift drilling, the process has no annular restriction capability to vary downhole pressure. Seawater and mud gradients below the mud line are cumulative to the BHP, Fig. 1.

Fig. 1. Dual gradient prior to running the riser.

The mud density required to maintain borehole stability is calculated with consideration to: BHP requirements, the seawater contribution and depth below mud line. The interest in dual-gradient / mud-lift drilling has been great due to the many advantages, not the least of which is shallow-hazard management and reduction in size / number of casing runs, as shown in Fig. 2. The DKD process offers similar advantages to dual-gradient / mud-lift industry initiatives, but is only applicable in the absence of a riser. Even with this limitation, the process has proven to be an effective tool.

Fig. 2. Casing setting depth comparison between a conventional and dual-gradient system. After SPE/IADC 59160.

Present-Day Options

Pumping weighted fluid down the drill pipe with returns to the mud line, located thousands of feet below sea level, presents numerous obstacles. Factors to be considered prior to pumping include fluid type, required fluid volumes, logistics and availability of a reliable mixing apparatus, as described here:

• Fluid types applicable include heavy brine and water-base fluids. When choosing between a brine or a blended water-base system, engineers must consider all factors discussed in this article. Additionally, fluid type must still address existing drilling-fluids issues such as drillstring balling, hole quality and fluid cost.
• Fluid volumes are calculated based on pump rates, ROP, hole size, interval length, drilling and cementing practices and mud weight.
• Logistical issues include the barrels of weighted fluid needed to drill the interval, rig storage capacity, boat capacity, distance from supply base and weather.
• The mixing apparatus should incorporate a metering device for blending fluids of different types / weights so that the end result is an instantaneously homogenous fluid.

Meticulous planning, coupled with fit-for-purpose mixing equipment, is critical to "pushing" the 20-in. to the desired depth.

A typical well employing this process, in combination with a weighted slurry-fluid system, will require from 8,000 to over 20,000 bbl of heavy-weight fluid. Blending these heavy fluids with seawater, at ratios dependent on final mud weight, will yield between 16,000 and 40,000 bbl of mixed fluid. These high volumes are too large for conventional means. Only the industry’s newest drilling vessels are equipped to efficiently handle such volumes. Therefore, the process must minimize fluid volume stored on the vessel. This process must also include a system capable of reliably delivering high mixing rates and mud weight flexibility. One such unit is illustrated in Fig. 3.

Fig. 3. The Dynamic Kill Drilling (DKD) process utilizing IMM – third-generation equipment.

DKD Process & Equipment

To deliver reliable accuracy and meet mixing requirements, the process incorporates: preplanning software, the mixing manifold (IMM) and real-time surveillance software.

   Preplanning software. Many uncertainties may exist during preplanning of a well with a high potential for a shallow hazard. Well designed preplanning software can be invaluable in generating scenarios associated with the uncertainty of hazard depth, pore pressure, ROP and associated pumping rates (GPM). It should be capable of converting pore pressure estimates into downhole equivalent mud weight at regular intervals below the mud line. It should also calculate fluid volumes at different ROPs and GPMs and, finally, total volume requirement.

   Mixing manifold. The IMM comprises flexible quick-connect hoses, ball valves, magnetic flow meters and a mixing chamber. The equipment was developed over a six-year period and represents third-generation equipment. Early prototypes, while functional, had some undesirable characteristics. The first-generation mixing chamber was large and unwieldy for quick installation. The second generation employed a mixing chamber as a dual-function device. An eductor provided the necessary blending function during the DKD process and subsequently functioned as a high-performance shearing unit. It also saw improvements made to the metering component.

The current (third-generation) IMM, incorporates all of the previous improvements plus some additional modifications. The modern system: 1) is shipped to location in a toolbox; 2) can be rigged up in 4 to 6 hr; 3) uses a third-generation mixing chamber; 4) incorporates specifically engineered flow-tube extenders for increased meter accuracy; and 5) can blend three different fluids simultaneously.

   Real-time surveillance software. During the execution phase, differences between preplanning assumptions and actual conditions must be analyzed. Surveillance software should track volume depletion at present and projected rate of use, volume required to short-trip, and minimum volume required to set pad mud and pump cement. Critical information can be obtained, such as: how much longer can I drill at the present ROP/GPM; how much longer if I alter GPM with the same ROP; and what if GPM is increased to some value, resulting in a higher ROP?

Additional System Advantages

Equivalent circulating density (ECD) management, which has always been an important planning component, has taken on new importance in ultra-deep, deepwater wells (defined as those drilled below 20,000 ft TVD). Many ultra-deep wells have, at TD, most of the following: smaller casing, smaller drillstring and hole sizes, higher mud weights and a much higher ECD.³ 

In many cases, the DKD process can impact small-hole and ECD problems associated with ultra-deep wells by eliminating a casing string early in the well construction. This approach has led to the production section gaining one hole size. Landing the 20-in. in a higher pore pressure environment also has implications for the rest of the well, permitting the operator to set subsequent strings at greater depth.

   Wellbore quality. Drilling substantial portions of the 26-in. hole section with formulated drilling fluids, in place of more conventional brine or seawater and sweeps, has improved wellbore quality. The initial sections of deepwater wells are conventionally drilled with seawater, combined with high-viscosity sweeps. Drilling with seawater and sweeps to the 20-in. casing point can result in need to squeeze cement to achieve the maximum leak off test (LOT). Benefits of a formulated fluid include simplified 20-in. landing, better casing centralization, improved zonal isolation when cementing, and fewer remedial cement squeezes.

   Costs. Thousands of barrels of drilling fluid may be utilized during the DKD process. Therefore, fluid composition must be economical, while still meeting established criteria. A dense fluid must be formulated to maintain both filtration control and viscosity when blended with seawater. This task can be accomplished using a low-salinity water / mud slurry. On occasion, higher-density, more shale-inhibitive calcium chloride brines are used due to storage and weather issues. These fluids can be slightly more technically challenging to formulate and may cost substantially more per barrel due to the limited products to build viscosity and filtration control in a concentrated divalent environment.

Case Histories

The case histories summarized below were chosen to demonstrate the wide-ranging applications of the DKD process.

Case History 1 – Increased fracture gradient. This example addresses the process used to push the 20-in. beyond the depth normally achieved with seawater and sweeps, to obtain a higher LOT. In this manner, all subsequent casing depths can be extended, minimizing number / size of strings. This prospect was located in 2,167-ft water and the casing program was designed for a rapidly increasing pore-pressure profile, to reach TD with a 9 7/8-in. minimum hole size. This required either pushing the 20-in. shoe deeper than is possible with seawater and sweeps or using a "big bore" well configuration.

The decision was made to use the DKD process to push the 20-in. an additional 430 ft, requiring an 11.0-ppg fluid over this interval. The process equipment was installed on the mud pits. Prior to drilling, the blending system was tested to determine maximum flowrates on the seawater and mud lines. Pilot tests were conducted on blending the 16.0-ppg mud with seawater to the required yield point. Tests showed that a cut-back of two barrels seawater to one barrel of 16.0-ppg fluid would produce adequate rheology and would not require further polymer treatments. However, additional monitoring of the yield point was planned during the riserless portion of the interval.

Next, boats containing the 16.0-ppg base mud were tested to determine the maximum pump rate to the rig. The 30-in. casing was jetted 320 ft below the mud line, pumping pre-hydrated bentonite sweeps as needed. A 26-in. drilling assembly was run and drilling began. Pre-hydrated 100-bbl bentonite sweeps were pumped at every connection. Drilling continued to 3,600 ft, where the DKD process was initiated using an 11.0-ppg mud system. The mud pumps were lined up on Pits 1 and 2, equalized through the centrifugal suction, with the 16-ppg base mud in Pits 3 and 4. This configuration provided a consistent mud-weight supply for the IMM.

Drilling progressed, averaging 1,100 to 1,200 gpm downhole while pumping base fluid from the boats, feeding the mixing manifold and blending seawater to a final 11.0 ppg. Rheology was monitored and, no polymer additions were required to maintain an adequate yield point. At TD of 4,030 ft, bottoms up was circulated, and the bit was pulled to the shoe. The well was flow-checked and the bit was run to bottom with no fill.

The well was displaced to an 11.5-ppg pad mud by isolating 250 bbl of 11.0 ppg in Pit 2 – on the last 250 bbl of having bottoms up pumped – and weighting up Pit 1 to 11.5 ppg by adding 50 bbl of 16.0 ppg base fluid. By using the IMM to pump the weighted fluid to Pit 1, only 4 min. was needed to increase mud weight. Suctions were changed on-the-fly back to Pit 1 and the seawater-to-mud ratio was changed from 2:1 to 1.5:1, thereby maintaining a constant 11.5-ppg mud. This was all achieved within minutes without shutting down the mud pumps. Once the hole was displaced to 11.5 ppg, the drilling assembly was pulled into the 30-in. casing and the well was monitored for 15 min. The bit was then pulled, and the 20-in. was run to bottom and cemented without incident. The shoe test was 1.0 ppg higher than anticipated.

Case history 2 – No hazard. This application did not involve a geological hazard. Instead, the process was used to simply push the 20-in. farther than conventional seawater and sweeps would allow due to wellbore stability issues.

This well was located in 4,746 ft of water. Initial plans were to initiate the DKD process at the normal 20-in. shoe depth of ±2,200 ft BML and extend the casing shoe an additional 1,000 ft. Seawater and sweeps were utilized to 2,200 ft. Average ROPs during this portion ranged from 100 to 150 ft/hr.

The process was initiated at 2,200 ft BML due to indications of hole closure. Starting with a 10.0 ppg mud, weight was increased, due to hole indications, to a maximum of 11.0 ppg. Operations progressed smoothly to the point that it was decided to push the shoe farther than originally planned. The "push" ended up totaling 1,334 ft and was stopped due to decreased ROP and a lack of mud volume to continue. A 13.0-ppg pad fluid was spotted prior to running pipe and cementing. Notable factors for this section included: 1) no indications of hole closure; 2) no required cement squeezes, and 3) a higher than expected LOT.

Case History 3 – Shallow hazard mitigation. A major operator in the Gulf of Mexico identified a potential shallow hazard at 1,418 ft below the mud line on a prospect located in 6,200+ ft water. Substantial time and costs would be saved if the hazard could be drilled while drilling the riserless section below the structure pipe. If accomplished, a 20-in. casing string could be run across the hazard, eliminating the need for a 26-in. string. Pore pressure data indicated that a 12.0-ppg fluid would be needed to control flow within the hazard. The length of the hazard was projected as 1,480 ft.

After detailed pre-planning, structure pipe was washed down, and 1,418 ft of 26-in. hole was drilled using seawater and sweeps. The downhole pressure tool confirmed a suspected shallow flow at this depth, requiring initiation of the DKD process. A 16.0-ppg fluid at 850 gpm was blended with seawater at a rate of 730 gpm to yield a 12.0-ppg fluid. This mud weight was held constant over the entire hazard section with no indication of water flow. Slight variations in the weight of the mud being blended with seawater necessitated a slight adjustment to the blend. This was accomplished easily be adjusting the meters controlling the flow tubes.

Penetration rates ranged from 90 to 130 ft/hr for the first 900 ft. ROP decreased slowly over the balance of the interval. The flowrate was reduced at a rate matching the lower ROP to ensure that sufficient supplies of weighted fluid were available to reach the projected TD. At TD, pipe was short tripped, bottoms-up circulated and a 13.0-ppg pad mud was spotted in the open hole prior to tripping out for casing. The 20-in. was run to bottom and cemented; the shoe did not require squeezing, and a higher than anticipated LOT was recorded.

Case History 4 – Batch setting 20-in. casing. This final case history addresses use of the DKD system to improve the economics of batch setting casing in deep water. A major Gulf of Mexico operator scheduled a nine-well batch setting of 20-in. pipe in 5,423-ft water. Pre-planning initiatives identified weighted pad mud as a potential bottleneck. Each of the nine wells would require 2,800 bbl of 12.5-ppg fluid with only three days between each well. The project’s drilling rig possessed a total fluid capacity of 3,200 bbl. It was possible that supply vessels could meet the mud transportation requirements, but that would require flawless execution and no weather delays.

The DKD process was identified as a means to mitigate a great deal of the risk associated with the batch-setting phase of the project. A volume of 2,200 bbl of 16.0-ppg fluid could be pre-loaded on the rig, yielding 3,955 bbl of 12.5-ppg fluid. In this manner, enough weighted fluid could be pre-loaded to service the needs of 1.5 wells and allow several days before re-supply of the drilling vessel was required.

The plan was modified to further reduce risk by loading the supply vessel with 6,000 bbl of 16.0-ppg weighted fluid that would yield 10,788 bbl of 12.5-ppg fluid. The boat would maintain the extra 3,800 bbl of 16.0 ppg as reserve volume. The batch setting on all nine wells was accomplished using 17,000 bbl of base fluid supplied by two workboats. The drilling phase was completed with zero hours of non-production time relating to fluid supply. The drilling and cementing of the nine-batch set was completed in fewer days than planned and at a reduced cost.

Conclusions

Significant advances have been made to the DKD processes, and equipment to address shallow-hazard challenges in deepwater wells. These advances cover pre-planning, equipment and installation, and job execution. Present-day equipment designs are compact and can be mobilized quickly while offering needed reliability. While first developed as a tool to mitigate shallow hazards, the process to extend 20-in. casing has been expanded to address a number of other circumstances.

Common results demonstrated in all the wells employing the process have included fewer drilling days and lower overall drilling costs. In some instances, savings were due to the cumulative cost associated with being able to eliminate: a casing string, the hole-opening run and cementing. In other cases, savings were realized due to the process eliminating the need to utilize a "big bore" well configuration. In still others, cost savings were attributed to reaching deeper horizons of interest with a larger hole size. All of the wells employing the new process have realized fewer logistical problems and maximized borehole stability. To date, all of the wells that have pushed 20-in. casing have obtained the necessary fracture gradient without the need to squeeze.

Acknowledgment

The author thanks the team members of Baker Hughes INTEQ Drilling Fluids’ DKD deployment team for making this process a success on many deepwater projects. Special thanks to Benny Silguero and Alan Giles for their collaboration and support developing the many tools necessary for preplanning and successful execution. Information in this article was originally published in paper SPE 71752, presented at the 2001 SPE Annual Technical Conference, New Orleans, Louisiana, Sept. 30 – Oct. 3, 2001.

Literature Cited

¹ Rocha, L. A. and Bourgoyne, A. T., "A new simple method to estimate fracture pressure gradient," paper SPE 28710, presented at the SPE Conference and Exhibition, Vera Cruze, Mexico, Oct. 10 – 13, 1994.

² Pelletier, J. P., et al., "Shallow water flow sands in the deepwater Gulf of Mexico: Some recent Shell experience," presented at the 1999 International Forum on Shallow Water Flows, League City, Texas, Oct. 6 – 8, 1999.

³ Barker, J. W., "Equivalent circulation density management in ultra-deep, deepwater GOM wells," World Oil’s Deepwater Technology, August 1999.

The author
	Michael B. Johnson is the Deepwater Technology Manager for Fluids Technology at Baker Hughes INTEQ. He obtained his BBA degree from Stephen F. Austin University, Texas, and joined INTEQ as a mud engineer in 1979. His experience covers the Gulf of Mexico, Latin America and all worldwide deepwater drilling basins.