PART I: This is the first of a series on the various applications of Pressure-While-Drilling™ (PWD) sensor measurements, focusing mainly on shallow water flows - a major deepwater problem. Subsequent articles will cover other common drilling problems, such as kicks and well control; loss/gain (ballooning); leak-off tests (LOTs) and lost circulation; hole cleaning, pack-off, and hole collapse; mud properties; and drilling practices.
Deepwater drilling is expensive and often proves troublesome due to problems such as shallow water flows (SWFs), low margins between pore and fracture pressure, and unconsolidated formations. A tool designed to detect pressures while drilling (PWD) is part of the measurement while drilling (MWD) string and records annular and sometimes drillstring pressures during the drilling process.
Pressures can be pulsed to surface while circulating to make real-time decisions. This raw pressure is normally converted to an equivalent mud weight (EMW) at surface for display and interpretation. This EMW is a calculated value derived from the measured pressure and true vertical depth at the sensor point. Over the past four years, the PWD sensor has proved to have a variety of applications and has dramatically grown in use, particularly deepwater drilling.
In deepwater drilling, SWFs are commonly observed while drilling the riserless section. The flows can occur to depths of 5,000 ft below the mud line, in water depths of more than about 1,500 ft. These are particularly common in areas of the Gulf of Mexico, but have been observed in other deepwater areas such as the Norwegian North Sea. This shallow section is normally drilled with unweighted seawater, allowing any interval penetrated with overpressure and sufficient permeability the opportunity to flow. It has been proposed that SWFs may be initiated through supercharging of the formation, inducing a loss/gain situation (see Alberty).
The sediments at this depth can be extremely unconsolidated, and the formation can quickly erode due to unrestricted flow. Because these sections are typically drilled without a riser, there are none of the usual surface indications of a flowing well (pit gain, gas, shut in pressures). The flow can only be identified with the PWD sensor and later confirmed with a remotely operated vehicle (ROV).
Three distinct fluids columns influence the PWD measurement in SWF situations.
Such SWFs can pose no problem, if they are of short duration and controlled. In an extreme case, continuing SWF problems led to the loss of the first Ursa template in the Gulf of Mexico through erosion around the structural casing and well collapse (see Eaton). After identifying the SWF situation, the PWD sensor is used to help monitor and control the methods used to mitigate the flow.
Identifying flows
Several different methods have been attempted:
- Setting casing above the flow zone and drilling ahead with weighted mud
- Allowing the well to flow while drilling and spotting heavy pills to balance the flow at total depth
- Drilling with a weighted mud system to balance the flow with no returns
- Drilling at a high rate of penetration (ROP) with the weight of solids-loading balancing the flow.
The following schematic examples show how to identify typical SWF situations with the PWD sensor and some of the methods used to distinguish between a flowing and collapsed well.
The first interval contributing to the PWD measurement is the body of seawater between sea level and the seafloor. Except for the tiny effect of waves and tides, this pressure remains constant. Next is that interval between the seafloor and the SWF source. The annular pressure contributed by this interval is a combination of the seawater drilling fluid plus the weight of drill cuttings returning up the annulus and the slurry produced from the SWF source.
Pwd measurements can correctly indicate "clean zones". Caliper measurements cinfirm this interpretation.
Besides the pressure profile itself, ROV observation while the well is static, can confirm that SWFs are not pure formation water flows, rather a combination of formation water and unconsolidated formation (sand and silt) that makes up a dense slurry. The lowermost interval is the new hole drilled below the source interval, and in this example, where the PWD sensor is situated, usually a short distance behind the bit.
The annular pressure attributed to this interval is the column of the seawater drilling fluid plus the weight of suspended drill cuttings. In addition, pressure is added below the mud line while circulating due to dynamic pressure losses (equivalent circulating density or ECD). This is usually a small contribution in these riserless sections.
Annular pressures
Changes in annular pressure, and therefore EMW, may be caused by many factors, including ECD pressures while circulating, SWF slurrys, hole collapse and restrictions, and cuttings loads. It is important to understand that the point source measurement of the PWD sensor reading is a cumulative reading of all the events occurring within the three zones previously described. Interpretation of changes in EMW must incorporate many of the variables occurring at or near the bit, as well as dynamic conditions that occur far above the bit in drilling situations.
Many times, the gamma ray tool measures a large washout and the assumption that this interval is sand may be erroneous. For the purposes of clarity, we will refer to "clean zones" as any interval in which the gamma ray measurement approaches zero. This is the base case for interpretation of EMW in riserless drilling operations.
A washout does not necessarily occur when drilling the clean zone, as evidenced by the caliper in the upper interval. The EMW is unaffected when no washout occurs. In the case of the lower clean zone, the hole washes out beyond the depth of investigation of the gamma ray sensor. A significant increase of 0.1 to 0.7 ppg EMW can occur when a clean zone washes. However, this effect lasts only for the duration of the clean zone, and the EMW returns to its baseline soon after drilling the unconsolidated interval. In this case, the well is not flowing and the section can be safely drilled.
A SWF can occur when the clean zone is charged above the normal seawater gradient. In this depth-based example, the EMW exhibits the sharp increase similar to the previous example, but does not return to the baseline after drilling through the interval. Instead, the well continues to flow while drilling ahead and the EMW remains above the seawater mud weight gradient due to the weight of entrained solids in the SWF slurry.
A SWF, continuing to flow after initial influx into the wellbore, causes a slow decline in the drilling EMW.
EMW typically reduces slowly with depth. Several theories partially explain this gradual reduction in EMW:
- As drilling continues, the interval between the flow zone and the bit accounts for more and more of the total annular pressure. The average density of the annular fluid in this interval is usually less than the combined density of the flowing slurry/drilling fluid mixture.
- The flow itself may be gradually diminishing with time, yet no known measurement of a SWF has been done to confirm this theory.
Another explanation may be that the nature of the flow is changing and that decreasing amounts of solids are contained in the slurry. This particular signature of a SWF is consistent in the examples seen throughout the Gulf of Mexico and the North Sea.
The previous examples all related annular pressure and EMW to depth. Operators derive the most benefit from PWD sensor information when the data are presented in a time format since much of the time is spent off bottom. The presentation of PWD sensor data combined with synchronized surface data in a time format allows a drilling engineer to correlate downhole events with surface drilling data.
Standard surface data measurements include stand-pipe pressure, flow in, ROP, rotary revolutions per minute (RPM), and torque. The quantity of data measured and displayed can overwhelm even the most experienced in PWD measurement interpretation, so great care must be taken to present only the most relevant data in real time.
EMW increase
Notice the abrupt increase in EMW when encountering the SWF. The sudden increase can occur in less than a minute and can exceed safety margins at the previous casing shoe. The gradual tapering of EMW is evident in time as well as depth. In the time presentation, however, EMW changes in different drilling operations help explain the hole conditions.
For example, during a connection, both the pumps and rotary stop. This results in the elimination of ECD or the increase in mud density due to frictional forces as the fluid circulates up the hole. Typically, the ECD is higher during the SWF due to the increased flow and rheology of the slurry. The frictional pressure losses from the flowing zone will continue to exert on the sensor. (In this example as well as others, the authors have expanded and simplified the PWD sensor signature for clarity).
A SWF, continuing to flow, can result in hole destabilization, as indicated by excessive drilling torques.
Well conditions are not static while a connection is made, as evidenced by the steady decrease of EMW during these intervals. This decrease may be attributed to solids fallout from the seawater drilling fluid, which has relatively little carrying capacity. When circulating and rotary drilling resume, the character of the EMW confirms the suspicion that cuttings were falling out of suspension during the connection. Cuttings are re-suspended and the EMW increases to a maximum value due to the increased cuttings load.
This load is quickly transported out of the hole and onto the seafloor, and the EMW begins to decrease as the annulus returns to a steady state condition. The next connection starts the process all over again. The SWF initially aids in the removal of the cuttings by adding to the annular fluid velocity. Eventually the zone may wash out or the flow diminishes to such a degree that the trend may reverse and the EMW signature increases. Careful real-time observation with PWD sensor data can minimize problems by alerting the operator to these conditions.
Borehole cleaning
In this example, a flow is in progress, but the combined lifting capacity of the circulating drilling fluid and the flowing slurry are not enough to clean the hole properly. The resulting increasing EMW trend is typical in a continuing SWF situation.
As the well is deepened below the flow zone, it becomes increasingly difficult to clean the well. The hole can often destabilize at this time, bridging and packing off the wellbore, typically when circulation is stopped. At this time, the flow of slurry and drilled cuttings exceeds the cleaning capacity of the well.
Close EMW monitoring allows higher flow rates and better hole cleaning efficiency without formation damage.
At approximately 3:35 on the accompanying figure, immediately after a connection, the EMW increases and becomes erratic and spiky due to the restriction. This is correlated with a sharp increase in surface torque. At this stage, it is very difficult to drill ahead and common for the drillstring to become stuck.
The previous examples have described problems associated with SWFs occurring at a relatively shallow depth in reference to the subsea mud line. Not all problems in these intervals are associated with SWFs. At no time is the well flowing in this example, but both the EMW and torque indicator curves show dramatic increases in a section of the hole that is predominantly shale.
This apparent hole failure results in a barely manageable hole-cleaning predicament. In this case, the collapse occurs during each connection. The saw-tooth shape of the curves indicates that conditions improve while drilling the stand of pipe but immediately deteriorate at a connection. However, the increased pump rate at X300 aids in hole cleaning and reduces the overall torque and EMW.
SWFs have proved to be a serious hazard when drilling the uppermost riserless sections of many deepwater wells. Historically, these flows have been difficult to deal with, due to lack of surface and downhole information. Over the past four years, PWD sensor information has increasingly been applied to this problem, leading to reduced risk in drilling these sections through more informed decision-making.
References
- Alberty, M., "Shallow Waterflows: History, Mechanisms, and Intervention," Shallow Water Flow Forum, Jun 11-12, 1998, Houston.
- Eaton, L.F., "Drilling Through Shallow Water Flow Zones at Ursa," SPE/IADC 52780. SPE/IADC Drilling Conference, Mar 9-11, 1998, Amsterdam.
Author
Chris Ward is a global drilling optimization advisor with Sperry-Sun in Houston, where he specializes in MWD drilling tools and applications. He previously worked as a geologist for Arco in London before moving to the Norwegian operations of Sperry-Sun. He holds a BS degree in geology and a PhD degree in geochemistry from the University of London.
Mitch Beique is a drilling engineer with Sperry-Sun, assigned to Halliburton's global deepwater solutions team. He has 21 years of drilling engineering experience in North America, seven years specializing in deepwater drilling. He holds BS degrees in petroleum and electrical engineering from Texas A&M.
