LEAK-OFF TESTS HELP DETERMINE WELL BORE COMPRESSIBILITY
Y.A. Hazov, Y.A. Hurshudov
SevKav NIPIneft
Moscow
Well bore compressibility and hydraulically formed fractures can contribute to elastic well bore deformation in unstable shale formations.
During leak-off tests in a basin near the Terek River in eastern North Caucasus in the former Soviet Union, the mud and well bores had anomalous, high compressibilities. Subsequent analyses indicated the system Compressibility was related to elastic hydrofracture behavior, with the fracture being open without additional pressure applied at the surface.
As the pressure on the well was increased, the fracture took mud, and the fracture returned the mud with virtually no losses as the pressure was decreased.
If a well is shut-in during leak-off testing, the pressure increases linearly with pumped volume after a short transition period. With a constant pump rate, the pressure is linearly related to time (Fig. 1). Fig. 1 is a plot of pressure vs. time for the leak-off test for well 4 Hankala. The well was drilled to 5,410 m, and the casing shoe was set at 5,280 m.
The slope of this straight line is defined by the volume of the well and by the elastic properties of the well and the mud (Equation 1). The value P in Equation 1 uses the average mud compressibility for the entire well and the elastic properties for both the cased and open hole sections.
The straight line is also known as the "minimum volume" line." Many field measurements on open well bores have produced a curve with a decline slightly below the straight line. The specific compressibility of the system calculated for this curve is substantially greater (up to 2-3 times) than that calculated for similar, cased well bores.
During the early history of leak-off testing, the nature of this phenomenon was not clear. Furthermore, at that time the compressibility of the entire system was thought to be related only to mud elasticity, and most researchers and engineers did not consider the elastic properties of the well.
FLUID COMPRESSIBILITY
The analysis began with a study of the compressibility of low-salinity freshwater drilling fluids without free or dissolved gas. The specific compressibility of the mud was evaluated as a sum of the compressibilities of the liquid and solid phases proportionally to their volume concentrations (Equation 2).12
Many researchers and engineers theorize that the compressibility, of the solid phases in the fluid does not depend on the pressure (that is, in the usual ranges during drilling). One recommendation is 0.25 X 10-4 MPa-1 for carbonates and 0.3 x 10-4 MPa-1 for clays and sandstones. The compressibility of the liquid phase is far greater than that of the solid phases and varies with temperature, pressure, and salinity.
To eliminate the need to calculate the liquid and solid volume fractions, Equation 2 can be rewritten with the appropriate density terms as Equation 3.
The specific compressibilities of most liquids can be obtained from common industry tables on fluid density variations with temperature and pressure. Because the temperature and pressure change with depth in a well, the drilling fluid is not under uniform conditions from surface to total depth (Equation 4).
Thus, each section in the well will react to pressure changes differently according to its local specific compressibility. To calculate the pressure increase rate in the well, it is necessary to use an average value of the specific compressibility including the effects of temperature changes, pressure changes, and fluid distribution along the length of the well (Equation 5).
NaCl solutions with a concentration of 1% at temperatures of 20-300 C. and at pressures of 0-150 MPa were used for the density and compressibility calculations. The density curves were produced from a linear extrapolation of the data.
A temperature gradient of 0.03 C. was calculated using a linear model and a constant surface temperature of 18 C. The computer calculations were based on data from actual well measurements. An analysis of the calculations and actual measurements indicated the pressure increase rate in a shut-in well depended not only on the elastic properties of the mud but also on some other factor, possibly elastic well deformation (rock and casing).
WELL HOPE DEFORMATION
The specific compressibility equation (Equation 2) should therefore include the effects of elastic well bore deformation. To increase the well pressure by Ap, an additional volume, must be pumped (Equations 6-8).
Based on conventional equations for thick walled cylinders, the compressibility equations were modified to study three cases: An open hole, a cased and cemented well with 100% cement bond, and a cased and cemented well with no viable cement bond. The quality and degree of the cement bond, or cohesion, between the casing and the formation is often unknown. Thus, the best case (100% bond) and worst case (O% bond) scenarios were studied to determine the extremes.
In the case of no cement cohesion, all of the load from internal pressure is taken by the casing. The specific compressibility for the wells with no cement bond was significantly greater than that for wells with 100% cement bond. Table 1 lists the system specific compressibilities.
A comparison of the data shows that the influence of elastic well bore deformation was quite large. The elastic well bore deformation is considered a main reason for the difference between the measurements and the calculations.
LEAK-OFF TESTS
Table 2 lists the specific compressibilities from leak-off tests and from calculations for several wells. The system specific compressibilities from the leak-off test data agree well with the calculated results. Fig. 2 is a plot of the calculated specific compressibilities and the actual system compressibilities from the leak-off test results.
For the 171 Rightbanked well, the specific compressibility defined from leak-off test data is considerably different from the calculated value (Table 2). On several other wells, abnormally large specific compressibilities were measured during the leak-off tests. There are several hypotheses to account for this phenomenon:
- During compression (or pumping up), the presence of free gas produces a nonlinear pressure profile. In the leak-off tests analyzed, only the linear part of the curve was considered. Thus the strong influence of free gas would take place at the beginning of the pressure increase, before the transition period ended. For a linear pressure increase, the free gas is compressed such that its effect or interference is not greater than any measurement errors. Also, dissolved gas can only change the specific compressibility of water-based fluids by a small percentage.
- Fluid loss and filtration also contribute to nonlinear curves. The liquid losses increase as time and pressure increase.
- One possible explanation of this phenomenon is a plastic deformation (increase in well diameter) of the well, particularly during the drilling of thick shale formations at pressures greater than pore pressure 3 4 On each well studied, after circulation began the pit volume decreased by about 4-16 cu M. Once circulation stopped, all this volume came back to the pits from the well in a matter of minutes. If a well was shut-in immediately after the pump was stopped, the well would have a slight pressure (about 2 MPa). The mud logging found considerable gas entry (1,000-4,000 units) at the moment the pumps were shut off.34.
MAIKOP SHALES
Similar problems with fluid losses in the pits occurred during the drilling of the Maikop shales in several wells in the eastern North Caucasus. Especially in the Low Maikop shales (for example, Well N 80 N. Braguny), the slight mud losses (10-12 cu m) during drilling were "found" once circulation stopped.
It is thought that the drilling problems in the shales and the anomalous high compressibility of the system (mud and well bore) are related.
Plastic deformation of shales takes considerable time. 5 Shale deformation occurs at a slow rate and may take up to 10-20 hr. The pressure increase and decrease during the leak-off tests took only 5-10 min, however, and there was no residual deformation. If plastic deformation is occurring, it may not be a main factor here.
The explanation of the flow back may lie in the theory of hydraulic fracturing (elastic hydrofractures).6 For practical purposes, the fracture is considered open even without additional pressure applied to the well. As pressure is increased, the fracture will take in mud. The fracture returns the mud, virtually without losses, as the pressure is reduced at the surface.
This fracturing phenomenon probably occurred in the Maikop whales. Fracturing of the Maikop shales during drilling has not always led to lost circulation. in such cases, drilling could continue. The mud was usually found to be overbalanced in these wells.
Although drilling with mud densities greater than the fracturing pressure is generally not a good practice, it helped provide the conditions necessary to evaluate elastic deformation.
In abnormally pressured intervals, the overweighted mud can approach the pore and fracture pressures. Small errors in determining pore pressure can lead to high mud densities when static pressure is greater than fracturing pressure.
In most areas, the pore pressure in shales is determined from log data. However, the pore pressure data come from logs run after an interval is drilled, if offset logs are unavailable. If an open fracture appears during drilling, the subsequent logs may indicate increased porosity and pore pressure. Other methods of predicting the pore pressure should be used.
When there are no fractures in the drilled interval, only relatively small volumes of rock take part in the elastic deformation. The result is a low system compressibility and a high rate of pressure increase. If open fractures are present and if they are somewhat large, however, the elastic deformation will be quite large because of the greater volume of rock exposed. The system compressibility will be high with a corresponding low rate of pressure increase.
REFERENCES
- Asnidei, B., et al., Leak off tests, Forages, No. 80, July/September 1978.
- Chenevert, M.E., and McClure, L.J., "How to run casing and open hole pressure tests," OGJ, Mar. 6, 1978, pp. 66-76.
- Gill, J.A., "How borehole ballooning alters drilling responses," OGJ, Mar. 13, 1989, pp. 43-52.
- Gill, J.A., "Well logs reveal true pressure where drilling responses fail," OGJ, Mar. 16, 1987, pp. 41-45.
- Johnston, D.H., "Physical properties of shale at temperature and pressure," Geophysics, Vol. 52, No. 10, October 1987, pp. 13951401.
- Nolte, K.G., and Smith, M.B., "Interpretation of fracturing responses," Journal of Petroleum Technology, September 1981, pp. 1767-1775.
Copyright 1993 Oil & Gas Journal. All Rights Reserved.