Kick detection, well control, and formation "flow-back" interpretation now more effective

July 1, 1999
Measurements address deepwater drilling problems

PART II: Part II of this multi-part series on sensor measurement of downhole pressures in deepwater focuses on kick detection, well control, and formation "flow-back" interpretation. The next part will focus on the role of downhole pressure sensors in detecting lost circulation.

Pressure responses in both time and depth vary greatly from one well control situation to another. However, similarities in the type of kick may result in a similar PWD® (Pressure While Drilling trademark) log. For example, in a typical salt-water kick, the influx occurs after the last connection is made and both a reduction in equivalent mud weight (EMW) and an increase in the active pit totals are observed.

This kick is recognized, drilling activities are stopped, and the well is shut in. The wellbore becomes a closed system and the downhole pressure sensor records a pressure buildup curve that reaches equilibrium at the pore pressure of the invading formation.

This recorded data has been used to dissect well kill procedures and refine techniques used by rig crews. Normally this information is not available in real time due to slow circulation rates. However, the complete record of the kill from shut-in to circulating out of the kick is available when the downhole pressure sensor is retrieved.

Many times, determining well control problems from real-time, pumps-on, downhole pressure data is difficult in the absence of other data. However, when these data are combined with conventional mud logging data, a clear picture develops.

In a typical saltwater kick, with formation pressure greater than ECD, the influx occurs after the last connection is made and both a reduction in equivalent mud weight (EMW) and an increase in the active pit totals are observed.

Click here to enlarge image

Pit volumes and gas-in-mud percentages indicate that the well is underbalanced over the three historical connections. The length of time taken to complete each of the first two connections could account for the increase in flow-back pit volumes and connection gas. Ambiguous surface data over several connections are common, especially in loss/gain situations. In fact, the rig crew did not raise the mud weight until after the third connection produced a sustained pit gain of approximately 10 bbl. An earlier response would have prevented this influx.

Pumps off conditions

Downhole pressure sensing provides not only pumps-on data but also three critical points during pumps-off conditions. These points are the maximum, minimum, and average pressure measurements, which provide solutions for missing data when the pumps are off. It should be remembered that most measurement-while-drilling (MWD) systems require some type of fluid movement or pressure drop to transmit hydraulic pressure pulses from the bottom hole assembly (BHA) to the surface. By providing a limited amount of pumps-off information, the downhole pressure service identifies the trend of lower static EMW at each connection.

Well control issues can appear during any rig operation, but especially dangerous is an influx while tripping. The accompanying log example illustrates a kick occurring while tripping out of the hole. After the incident, the pore pressure measured 16.05 ppg. This value was taken after shut-in from the stabilized downhole pressure build-up curve. Looking back, both the circulating EMW and the static EMW just before the short trip were greater than pore pressure. However, swab pressures from the short trip lowered the EMW below pore pressure and the well took a fluid influx. While tripping back in the hole, a decrease was noted in the downhole pressure sensor as it entered the swabbed-in light gas.

The downhole pressure service only records these kinds of events. There is no circulation throughout the short trip and none during the shut-in period of the well. Also, the slow pump rates required to kill wells normally are not great enough to allow for real-time information to be transmitted. Future enhancements to this service will incorporate solutions to acquiring real-time data throughout all rig operations, whether circulating or not.

The use of gas hydrocarbon averages (%) and pit volume totals (bbl) in addition to PWD data can more effectively identify influxes with formation pressure between mud weight and ECD.

Click here to enlarge image

Loss/gain

The phenomenon called borehole ballooning, breathing, or loss/gain can result from drilling close to the fracture pressure. Slow mud losses are observed while drilling ahead, followed by mud returns after the pumps have been turned off, such as during a connection or flow check. Usually any flows during these periods are cause for concern because they may be mistaken as an influx of formation water, liquid hydrocarbons, or gas.

As noted previously, any influx from the formation can result in a well control problem, the magnitude of which is dependent on its volume and composition. However, if the flow is due to mud returns, well control is not an issue. The question then is: "How does one know unequivocally if it is an influx, or if it is mud that was lost while drilling, flowing back into the wellbore?" If the well is shut in, both situations typically show a pressure build up (see Ward).

This loss/gain situation has often been mistaken in the field for an influx of formation fluid. Misdiagnosis often leads to unwarranted well control procedures that can be costly. One way to identify such a situation is with downhole pressure profile during pumps-off periods. A normal connection is typically square shaped when the pumps are stopped and started.

When the pumps are off, the EMW is equivalent to the mud weight in the annular column, and in this case, about 14.5 ppg. During circulation, the EMW is quickly established at a level equivalent circulating density (ECD), in this example close to 15.5 ppg. As the loss/gain develops, the downhole pressure connection profile changes. When the pumps are turned off, the EMW slowly decays to the static mud weight as mud bleeds back from the formation in a manner similar to an LOT. When circulation is re-established, the EMW slowly builds up to the ECD level as fractures are slowly re-filled.

Loss/gain is a relatively common problem in deepwater wells due to the low overburden. If a loss/gain situation is misidentified as an influx, the normal response is to increase the mud weight. This soon leads to a total loss. The correct response is to decrease the mud weight, decrease the ECD (reduce flow rate), or live with the losses and gains.

Recorded downhole pressure data can effectively monitor influxes with formation pressure between mud weight and swab pressure while tripping out of the hole.
Click here to enlarge image

Leak-off test

Normally, an LOT is performed at the start of each hole section after drilling out a few feet below the casing shoe. The LOT is designed to assess the cement and formation integrity. The casing shoe is the weakest part of a gas-filled wellbore and is essential in determining the kick tolerance for the next section.

PWD data can be used to interpret loss/gain situations like simple formation flow-back and formation influx.

Click here to enlarge image

The LOT can also give some information on the formation strength as an upper limit for the ECD to prevent lost circulation. However, casing shoes are often placed in relatively strong formations, and much weaker zones can be drilled into below. Occasionally open hole LOTs are performed after drilling such zones.

A LOT is performed by shutting the well in and pumping at a slow rate, normally with the cement pumps. The pressure is allowed to increase until a slope-break in the pressure increase curve is observed. Pumping is stopped and pressure held for a few minutes to observe the fracture closure and then bled off. A formation integrity test (FIT), in which the pressure is increased until a set limit is reached, is sometimes preferred.

In most cases, the downhole and surface measurements differ. The downhole EMW may be lower or higher, and differ by as much as 0.5 ppg or more. The downhole pressure sensor measures the LOT pressure at the formation and is as accurate as the pressure gauge. LOT measurement errors at the surface are due to a number of factors, including uneven annular mud weight, pressure loss in surface lines, mud compressibility, breaking gels, and poorly calibrated surface gauges.

Downhole pressure measurements are taken at the formation and can elimiate potential errors observed in surface-measured LOTs.
Click here to enlarge image

In deepwater wells, the pressure and temperature effects on mud in the long cold riser usually mean that the downhole mud density is higher than the surface mud weight. This often leads to an under-estimation of the LOT at the surface, resulting in lower kick tolerances and shortened distances between casing strings. The downhole pressure sensor records, and later transmits the pumps-off, LOT measurements upon circulation startup. Traditionally, the well is circulated for an hour or more before each test to establish an even mud weight for the surface test interpretation. Because the downhole pressure sensor measures the formation pressure directly, this procedure is not necessary.;

References

Ward, C., Clark, R., "Anatomy of a Ballooning Borehole Using PWD," In Mitchell, A., Grauls, D. (eds), "Overpressures in Petroleum Exploration," Workshop, Pau, April 7-8. 1999.

Authors


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 BSc 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 degrees in petroleum and electrical engineering from Texas A&M.