Shrikant Tiwari
Oil & Natural Gas Corp. Ltd.
Dehradun, India
In most cases, the driller should shut in the well immediately following the first indication of a kick rather than taking extra time to make a flow check.
This extra time spent verifying a kick with a flow check can allow a larger influx into the well, making the kick harder to kill and increasing the load on the casing shoe. The risk of for- mation fracture increases tremendously if too much time is taken to shut in the well.
Ideally, all rigs should have state-of-the-art kick detection equipment to indicate the presence of a kick at the earliest possible moment.
The methods of kick control have not changed much over the years, but a successful kill very much depends on early detection. These methods prove effective under normal kick con- ditions, but the presence of any additional complication can increase the cost of the well and possibly damage the formation.
In shallow wells, formation fracture may prove disastrous as gas may escape to the surface, possibly destroying the rig. Hence, it is necessary to take every possible precaution to avoid any such situation during a kill operation. Even a small kick can become a blowout if control is lost. Thus, care must be taken at every stage of operations to avoid any influx into the well.
Every kick, no matter how small, is dangerous and must be handled with great caution. Kicks caused from invading gas are difficult to control because the gas expands as it rises up the annulus, and this expansion tends to unload the well very fast.
Wells are normally drilled by maintaining a sufficient specific gravity of drilling fluid so that the pressure exerted by its column exceeds the formation pressure. This positive dif- ference in pressure does not allow the formation fluid to enter the well bore. An influx can enter the well if this situation reverses. In simpler terms, the prime cause of a kick is the insufficient head of drilling fluid to stop the entry of formation fluid into the well bore.
This situation may arise because of any of the following reasons:
- Unexpectedly high formation pressure
- Lost circulation
- Swab during a trip out
- Failure to fill up hole during a trip out.
Kicks can be detected by monitoring surface indications on the rig. An influx in the well can be indicated by a drilling break, pressure drop, increase in pump strokes per minute, pit gain, etc. The most reliable and widely followed mode of kick detection is monitoring the fluid balance in the well by making a flow check.
FLOWING WELL
As a well is drilled, drilling fluid is pumped into the well and recorded as the pump discharge. As this fluid circulates around and through the annulus and back to the surface, it brings much vital subsurface information. During normal conditions, a well follows the U-tube principle in which the volume of mud pumped in should equal the volume of mud coming out.
If under any circumstances, a positive difference exists between the rate of mud coming out and mud going in, the well is said to be flowing. This phenomenon occurs when the balancing head of drilling fluid falls below the formation pressure. A differential pressure occurs across the well bore, causing formation fluid to enter the well. The drilling fluid in the annulus moves up under the influence of this differential pres- sure and causes the well to flow.
On most rigs, a paddle-type deflection flow meter is installed in the return pipe, and even more sensitive flow detectors are also available. These instruments sound an alarm if they sense even the slightest difference in the mud flow.
As soon as a driller encounters a drilling break or is alerted by the flow indicator, he moves on to confirm the kick by making a flow check. Pumps are shut off and a close watch is kept over the rate of mud coming out. The well will continue to flow if there is an influx; otherwise, the returns shut off completely. If the returns continue unabated and a pit gain is observed, the well should be shut in and a suitable method for killing the well should be followed.
FLOW CHECK
When the drilling fluid is under circulation, dynamic conditions exist, and the fluid experiences drag as it comes up the annulus because of continuous friction with the hole wall and the drillstring. This frictional force acts downward and exerts extra pressure on the formation. Thus, the mud system has a built-in pressure at the bottom of the hole in excess of the hydrostatic pressure.
Hence, the effective density of drilling fluid is slightly greater during dynamic conditions than during static conditions and is known as equivalent circulating density. Although this frictional loss in the annular space is not given much weight, it plays an important role while sensitive formations are drilled.
As soon as the pumps are shut off for making a flow check, the frictional losses in the annulus are reduced to zero. Thus, the positive effect of frictional losses in the density of drilling fluid ceases to exist. This reduction from the equivalent circulating density under dynamic conditions to static mud density in turn reduces the bottom hole pressure.
The effect can be assessed in light of Darcy's gas flow equation. According to Darcy's law for gas flowing through a porous medium, the rate of gas flow is directly proportional to the cross-sectional area, differential pressure, and formation permeability (Equation 1)(54500 bytes).
When a kick is encountered, the formation permeability, formation fluid viscosity, cross sectional area, drainage radius, and the well bore radius can be considered constant. The Darcy's equation will then be reduced to Equation 2 (See equations box)(54675 bytes).
As the static bottom hole pressure is less than bottom hole pressure during dynamic conditions, the resultant differential pressure across the formation is greater when the mud is not under circulation. This larger differential pressure can invite formation gas into the well bore at a faster rate.
It only takes a few minutes to make a flow check to determine if a kick is occurring. During this time, the well faces the possibility of a heavier influx of formation gas because of the greater difference in formation pressure and the hydrostatic head of drilling fluid column.
In fact, a larger volume of gas can enter the well bore if the well is shut in after a flow check instead of being shut in immediately after receiving the very first indication of a kick.
For a kick to be killed successfully and quickly, it is of prime importance to detect the kick quickly and shut the well in at the earliest moment to minimize the possible quantity of formation fluid entering the well. A quick shut in is even more important for gas kicks than for saltwater or oil kicks because larger volumes of gas eventually exert higher pressure at the casing shoe as the gas expands up the hole.
EXAMPLE
Table 1 (17097 bytes) lists data from a practical well situation to help evaluate the effects of making a flow check to detect a kick. In this well, a drilling break was encountered at a depth of 10,283 ft. The well was observed for flow, and a pit gain of about 70 cu ft was recorded in 10 min.
Gas was assumed to enter the well under dynamic conditions for 1 rain and under static conditions for 9 min, and the average annular velocity of the gas was assumed to be 130 fpm. During drilling, the average lift velocity of the cuttings was 65 fpm, and the average cuttings density was 20.83 ppg.
An analysis indicated that the well was exposed to 5.15% more risk of formation fracture by closing the well after making a flow check rather than closing it immediately after getting the indication from the drilling break.
The data in Table 1 (17097 bytes) were used to determine the rate of gas entry and its effect on the load on casing shoe under two situations:
- In the first case, the well is shut in immediately after
the kick is encountered.
- In the second case, a flow check is made to confirm the
kick before the well is shut in.
The results of the analysis are shown in the accompanying box.
For further analysis, it was assumed that the well conditions were similar at a depth of 10,000 ft. Fig. 1 (58996 bytes) shows the load on the casing shoe for various differences between formation pressure and mud weight for the two example cases. In this figure, an immediate shut in of the well required 2 min, and the flow check required at least 5 more minutes.
Fig. 2 (60446 bytes) shows the percentage difference in load on the casing shoe as a function of the amount of time taken for verifying the presence of a kick.
The risk of formation fracture from the time to make a flow check can be assessed by Equation 4. K is the flow check constant, which is equal to 4.2. This equation can predict the percentage difference in load on casing shoe for two example cases with considerable accuracy (error of only +_1%). The relationship shows that the risk of formation fracture increases with increasing difference between formation pressure and mud weight, with increasing time taken for making a flow check, and also with increasing depth of drilling.
A differential pressure of 5-6 ppg is not very common in development wells but cannot be ruled out for some exploration areas. Even a differential pressure of 2.5-3 ppg, which is not so uncommon, poses 5-6% more risk of formation fracture even if the well is shut in after taking a normal time of 5-6 min for a flow check rather than doing it immediately after encountering a kick.
The risk increases tremendously if the time taken to confirm a kick is more. This increased risk of formation fracture is definitely a cause of concern and must be viewed seriously. It is of great importance that the driller shut in the well immediately rather than waste time in confirming a kick through a flow check.
The results of this analysis emphasize the need for fool- proof and better kick detectors on rigs to avoid even the slightest danger of blowouts, especially in exploratory areas. Although it may lead to an additional cost burden up front for the operator or contractor, it is bound to prove cheaper in the long run.
ACKNOWLEDGMENT
The author thanks the management of Oil & Natural Gas Corp. Ltd., India, for permission to publish this article and also the Institute of Drilling Technology for providing library facilities and assistance in the preparation of the article. The author extends special thanks to Meena Bhutani, a senior mathematician at the Institute, for her assistance.
BIBLIOGRAPHY
Adams, N., Well Control Problems and Solutions, PennWell Publishing Co., Tulsa.
Dake, L.P., Fundamentals of Reservoir Engineering, Elesevier Scientific Publishing Co., New York, 1978.
O'Brien, T.B., and Goins, W.C. Jr., "The Mechanics of Blow- outs and How to Control Them," Drilling & Production Practices, American Petroleum Institute, 1960, pp. 41-55.