March 2008
Special Focus

Microflux control MPD improves understanding of downhole events

MPD is solving critical problems, reducing downtime, lowering risks of drilling challenging wells and reaching TD where previously impossible.


In standard mode, MFC offers a simple way to identify complex drilling conditions. 

Helio Santos and Erdem Catak, Secure Drilling; Paul Sonnemann, Chevron

In the oil industry, a simple solution typically is considered insufficient to solve complex problems. There are two main reasons for this thinking:

1. The more complex the solution, the more expensive it is, so service companies can charge higher prices.

2. The more complex the solution, the fewer people that are capable of understanding and delivering it, thus protecting the service companies’ advantage.

As the problems faced are thousands of feet below the surface, the correct understanding and interpretation of the problems are difficult. However, the results obtained from more than 46,000 ft drilled with the Micro-Flux Control (MFC) method indicate that a simple solution may provide better insight than a complex one.

The industry has made significant progress in the last few years with Managed Pressure Drilling (MPD). Early users succeeded at solving critical problems, reducing downtime, lowering the risks of drilling challenging wells, and reaching TD when it was not possible using conventional drilling.

Since the first well was drilled with MFC MPD in August 2006, the method has been used on many wells in both the standard (when the mud weight is hydrostatically overbalanced) and special (when the mud weight is hydrostatically underbalanced) modes. The wells were drilled with water- and oil-based fluids with densities up to 18 ppg, offshore and onshore, for both exploratorion and development. The flexibility and simplicity to change from one mode to the other allows the operator to select the proper configuration depending on well conditions, well problems, rig capability, crew competency and other conditions. One interesting finding is that the standard mode can provide unique value in understanding more accurately the downhole events, leading to a clearer identification of the problems faced.

MFC METHOD

The MFC MPD system uses a Rotating Control Device (RCD) to keep the well closed to the atmosphere at all times, and a specialized manifold with a very small footprint that includes redundant chokes, a flowmeter, and data acquisition and control electronics. The simplicity of this standard MPD system makes it attractive for use on many wells.

Standard mode. The system can be used to drill on any well and rig in the standard mode, the simplest configuration, because there is no change to any safety or design criteria. The choke is always run fully open to apply the minimum backpressure possible at the surface. The system detects influxes and losses very early and controls any influx automatically, keeping the total volume to less than 5 bbl. This configuration can also help the user identify other normal drilling problems including washouts, mud pump problems, leak-off test inaccuracy, casing test inaccuracy and trip gas.

Special mode. The special mode typically requires additional surface equipment. The system is versatile enough to allow the operator to switch between standard and special modes providing that the additional equipment for drilling in the special mode is available and set up. With this ability, the operator can drill a portion of the well in the standard mode while problematic segments can be drilled in the special mode.

When drilling in the special mode, surface backpressure can be controlled to drill the well under various conditions; for example, keeping a desired wellbore pressure. The driller can compensate for the continuously changing friction loss by applying backpressure at the surface. This is typically beneficial when the well is drilled hydrostatically underbalanced.

SELECTING THE METHOD

The first step is always to decide which methods and equipment will be required for a specific well. Lower ECD might be required, for example, or continuous circulation during connection, better characterization of kicks and mud losses, smaller kick sizes to reduce the number of casing strings, better mud weight management, or Constant BottomHole Pressure (CBHP).

The vast majority of wells will benefit from one of the above specifications, or from a combination thereof. If the MPD solution is simple enough and affordable, it can be deployed on a wide variety of wells. However, if the solution is complex and very expensive, only a few wells will be able to benefit.

SOLVING PROBLEMS WITH MPD

A few years ago, the industry thought that MPD would be useful only on narrow-margin wells where mud weight is below the pore pressure. On many narrow-margin wells, mud losses begin as soon as the mud pumps are turned on. One option to avoid the losses is to have a hydrostatically underbalanced mud weight such that, with the friction generated when the fluid is in circulation, the final pressure inside the wellbore would be smaller than the fracture gradient. This MPD application is called the constant bottomhole pressure variation, as there is a need to compensate the hydrostatically reduced mud weight to avoid influx when the pumps are off. The MFC special mode provides this drilling option.

To avoid the pressure oscillations due to stop/start of the mud pumps, continuous circulation devices have been developed and are already in use. These devices can also be used for CBHP, as there is no need to stop the circulation during connections. However, it is crucial to have a contingency plan in case problems arise with mud pumps or with the equipment itself.

Applying CBHP is relatively easy using any of several options. The main challenge is to define the correct bottomhole pressure to be applied. This is where the accuracy and automatic response of the MFC standard mode enters the equation.

WHEN TO USE EACH MODE

Wells are drilled based on predicted pore and fracture pressures. The mud weight program is based on these estimated curves and adjusted as drilling progresses. No matter how well the estimated pressures have been defined, reality will always deviate from the estimated curves. Casing depth and the mud weight programs are based on the pore and fracture pressures, and to optimize drilling it is essential to know the actual values.

Problems associated with drilling conventionally include the risk involved in taking kicks and the inaccuracy on flows, mud volumes and pressure measurements. Based on offset well data from conventionally drilled wells, many wells today are classified as narrow margin due to indications of kicks and losses within a very narrow pressure window. In these cases, the MFC special mode has been used to achieve the CBHP variation.

From experience on other wells using the MFC standard mode, it was observed that rather than using the special mode at all times on a section, the optimal solution would often be to begin in the standard mode. With the mud weight as close as possible to the estimated pore pressure, or even slightly below if possible, commencing drilling with the MFC standard mode would allow confirmation of whether the mud weight is hydrostatically underbalanced. Based on the leak-off test obtained at the shoe, and periodic dynamic leak-off tests or dynamic Formation Integrity Tests (FITs), the hydrostatic mud weight and the bottomhole pressure to be used can now be more accurately defined.

Rather than applying backpressure from the beginning of the section, pore pressure can be checked if the section begins with the standard mode. Figure 1 shows a connection made without applying backpressure in the beginning of the 8½-in. section of an exploratory well drilled onshore Texas with 15.5-ppg oil-base mud, at a 9,242-ft vertical depth. The 9 5/8-in. casing shoe was set at 8,305 ft. In this case, the mud weight was hydrostatically overbalanced, as the connection was normal. Flow-out decreased very quickly to zero after the pumps were turned off, and flow-out matched flow-in very quickly after starting the pumps. There was no sign of ballooning or other problems.

Fig. 1

Fig. 1. In a connection made without applying backpressure in the beginning of the 8½-in. section of an exploratory well, flow-out went to zero very quickly after the pumps were turned off, and matched flow-in very quickly after starting the pumps. 

Drilling in this section progressed without problems so far, including no losses or kicks detected. At this point, there was no need to change the mud weight or apply backpressure. This section was considered to have a very narrow mud-weight window from estimated values, and the special mode was considered necessary in this section.

The well began to show the beginning of a change in behavior at 10,278 ft, with flow-out taking longer to match flow-in after pumping began compared with previous connections, Fig. 2. When drilling conventionally, this would be interpreted as mud loss, and it would give a possible false indication of exceeding the fracture gradient. However, when using MFC, it could be seen that fluid was being stored in the formation, as flow-out was converging to flow-in.

Fig. 2

Fig. 2. Starting at 10,278 ft, flow-out took longer to match flow-in after pumping began. 

When the pumps are shut off, the fluid stored in the formation will be returned to the wellbore, as can be seen during a connection in the same well at 10,561 ft, Fig. 3. The fact that there is flow-out of the well after the mud pumps are shut off does not necessarily mean that the mud weight is hydrostatically underbalanced. In this case, it was simply the fluid stored in the formation being returned to the wellbore. Following conventional wisdom, the mud weight was unnecessarily increased to 16.1 ppg.

Fig. 3

Fig. 3. The fluid stored in the formation was returned to the wellbore during a connection at 10,561 ft when the pumps were shut off. 

Very often this fluid return is misinterpreted as a kick, and would lead the well to be shut in, pressures to be measured and mud weight to be increased. This action only makes the problem worse, causing more fluid to be stored in the formation after turning the pumps on, and, therefore, more fluid to be returned to the well when the pumps are off. Misinterpretation can lead to a further increase in mud weight until losses occur as the fracture gradient is reached.

If at a certain point an influx occurs during a connection, the system will detect it and alert the operator. Figure 4 shows a connection in an 8½-in. section on another well onshore Texas at 12,970 ft, with the 9 5/8-in. casing set at 9,774 ft. At this point, the oil-based mud weight was 17.4 ppg. The flow-out increased after the pumps were shut off, and a message immediately alerted the operator of the underbalanced situation. Knowing there was margin available, a decision was made to increase the mud weight to restore the overbalanced condition, and drilling proceeded to TD.

Fig. 4

Fig. 4. The MFC system will detect an influx during a connection and alert the operator. 

When no margin is available to increase the mud weight, the other option is to switch to the special mode so that backpressure at surface can be applied to compensate for the loss of friction pressure inside the well when the mud pumps are off. In another case, a connection was made in a 6½-in. section at 15,181 ft of an onshore well drilled in Texas, Fig. 5. OBM was used with a weight of 17.5 ppg at this point. The 7 5/8-in. liner was set at 12,650 ft. As can be seen, when using the special mode the opportunity to detect a hydrostatically underbalanced situation (and, therefore, the pore pressure) is gone. If a section is drilled using the special mode from the beginning, it will not be possible to determine how narrow the mud weight window is and whether the application of backpressure is really necessary.

Fig. 5

Fig. 5. When using the special mode, the opportunity to detect a hydrostatically underbalanced situation (and the pore pressure) is eliminated. 

After evaluation of the data collected and following a long learning period, it was concluded that the section on this well could have been drilled without applying backpressure, and that the narrow margin predicted prior to drilling was much wider. The reason for the false assumption came from the previously described misinterpretation of kicks and losses. In other words, overreaction to harmless events that, when drilling conventionally, are not clearly and accurately interpreted.

OPTIMIZING DRILLING AND EXTENDING CASING DEPTHS

Many wells drilled today are considered to have very narrow mud weight windows. This is typically based on estimated pore and fracture pressures, and offset well data information indicating kick or loss situations. However, in many cases these kick or loss situations are just misinterpretation of ballooning events.

On all of the wells drilled so far using MFC, the method allowed a more accurate determination of the mud weight window, showing in all of the cases that the window available for drilling was wider than expected during the planning stages. In some wells, the special mode was replaced by the standard mode to check the actual pore pressure value. After it was concluded that the margin was wider than initially thought, the section was drilled to TD in the standard mode.

NEXT STEPS

Wells can be significantly optimized compared with drilling conventionally by knowing the correct margin available in real time, and permanently adjusting the margins as indications of kicks and losses are automatically confirmed by MFC. Casing strings can be eliminated, as contingency casings will be necessary only when it is determined that they are actually necessary. Mud weight will need to be increased only when there is a confirmation of increased pore pressure, rather than by the fear factor when drilling conventionally. Sections will be drilled to the limit safely, providing a substantial reduction in cost and risk. WO 

ACKNOWLEDGMENT

We would like to thank Cypress E&P and Chevron for permission to share the well data as well as the excellent interaction that allowed a substantial amount of data to be exchanged and, consequently, increased the learnings gained from the operations.


THE AUTHORS

Santos

Helio Santos earned BS and MS degrees in civil engineering from Catholic University in Rio de Janeiro and a PhD in geological engineering from the University of Oklahoma. He joined Petrobras as a drilling engineer in 1983. In 1986, Santos was transferred to Petrobras’ Research Center. In 2001, he joined Impact Engineering Solutions as vice president of technology, and he became president of Impact Solutions Group in 2004. He also is a director of Secure Drilling, a joint venture with the Expro Group.


Santos

Erdem Catak earned an honors degree in petroleum and natural gas engineering from Istanbul Technical University and an MS in petroleum engineering from Louisiana State University. He taught drilling, drilling fluids, well control and well completion classes at ITU and LSU for four years. In 2006, he joined Impact Solutions Group as a project engineer for Secure Drilling. He is responsible for assisting the development and introduction of Secure MPD in the field, supervising field applications and reviewing potential well candidates with clients.


Santos

Paul Sonnemann has been involved in well control training for 27 years, beginning with Sedco in 1981 and moving to Unocal in the early 1990s. He is Chevron’s industry liaison on various well control issues and is involved in the evaluation and development of MPD technologies.



      

Related Articles
Connect with World Oil
Connect with World Oil, the upstream industry's most trusted source of forecast data, industry trends, and insights into operational and technological advances.