On JOIDES Resolution, two powerful pumps are available for inflating the packer and testing the formation: the "cement" pump and the "mud" pump. The mud pump is directly controlled at the rig floor by the driller and is more convenient, particularly for setting the packers. The cement pump is controlled below the rig floor and is thus much less convenient but may be more appropriate when accurately-measured, smaller volumes of higher pressure fluid are required.
1. Downhole pressure recorders
Downhole pressures can be measured using several kinds of recorders. At present, ODP maintains four self-contained mechanical model K-3 recorders made by Kuster Co. (Fig. IV-4). These are calibrated for pressure ranges of 0-9900 to 0-15,200 psi, and can be configured with Kuster clocks to record for 3, 6, or 12 hours. (An additional University of Miami clock presently available aboard JOIDES Resolution allows recording of 9, 18, or 36 hours as well.) The pressure records are scratched quite accurately onto small coated brass charts. As only one size of chart is produced, some resolution is lost if a longer recording time is chosen; in recent practice, the recorders have been set for 3 or 6 hours. The charts must be read using a calipered, microscopic chart reader, one of which is kept at the University of Miami.
During Legs 139, 146, and 148 packer scientists used electronic gauges instead of, or along with, the conventional mechanical Kuster gauges. These electronic gauges are also owned by K. Becker at the University of Miami. The electronic recorders are effectively interchangeable with the Kuster recorders, except that the electronic data are accessible immediately after the go-devil is retrieved.
The pressure gauges/recorders are housed in special carriers that attach to the go-devil. Normally, two K-3 recorders are attached to the go-devil to provide redundant data in case one recorder malfunctions. Like the Kuster recorders, the University of Miami electronic recorders are housed in special carriers attached to the go-devil. The electronic gauges have slightly different dimensions, however, so a separate set of gauge carriers must be used. ODP may consider purchase of additional electronic gauges at a later time.
2. Pressures measured during formation testing
In testing the formation using a packer, it is important to control carefully the packer inflation pressure. This is defined as the pressure (relative to hydrostatic) at which fluid is pumped into the packer element(s) when inflated, and is measured at the rig-floor pumps. The inflation pressure must be chosen carefully, based on anticipated hole conditions, formation properties, and test pressures. Given good open-hole conditions, the elements should be inflated to about 500-1000 psi, or to roughly half the planned test pressures if higher pressures are planned. If the inflation pressure is improperly chosen (either too high or too low), a "leaky" packer seal may result, which will certainly affect any measurements but will not necessarily be recognizable in the pressure data. Care must be taken not to inflate the element(s) to a pressure high enough to actually fracture the formation and thereby invalidate any testing. This upper limit on inflation pressure will depend on the lithostatic pressure and strength of the formation.
It is critical to measure and distinguish three different pressure values in the isolated section of the borehole: hydrostatic pressure, in-situ pore pressure, and test pressures. This necessity dictates the overall sequence in the testing procedure at each isolated zone:
In a pulse test, the downhole recorders monitor the decay of a short, effectively instantaneous pressure pulse applied by the rig-floor pumps. In a relatively impermeable formation, the decay of such a pulse will be long compared to the duration of the pulse, and the pressure data can be treated with the theory for an instantaneous pulse (Cooper et al., 1967; Papadopulos et al., 1973; Bredehoeft and Papadopulos, 1980). The decay of an instantaneous pulse is described by a complicated integral involving two dimensionless parameters, which depend respectively on "transmissivity" (a function of permeability and the thickness of the permeable zone) and "storage coefficient" (a function of porosity). Fitting pulse-test pressure data to this function generally resolves transmissivity much better than storage coefficient, yielding estimates of permeability but not necessarily of porosity.
In a relatively permeable formation, a pressure pulse will decay too rapidly to allow resolution of transmissivity, and constant-rate injection tests are appropriate. These involve pumping into the formation from the rig floor at a known constant rate, and monitoring the change in downhole pressure with time. The associated rate of pressure increase within the isolated zone, when plotted against log time, is proportional to permeability (Horner, 1951; Matthews and Russell, 1967). The pressure rise is analogous to the rise in temperature associated with thermal-conductivity measurements (Jaeger, 1958; Von Herzen and Maxwell, 1964). After pumping is stopped and the hole is shut in, the subsequent pressure drop can be used independently to estimate permeability, with the method of analysis closely following that for injection testing.
In oceanic sediments and crust, the ranges of permeabilities over which pulse tests and constant rate injection tests yield reasonable permeability estimates overlap, and it is often not possible to determine before testing which type of test will be more appropriate. As injection tests disturb the pressure field in the isolated formation to a greater degree than pulse tests, it is advisable to attempt pulse tests before injection tests. If several pulse and/or injection tests are to be attempted in a single formation, the pressurized system should not be allowed to "flow back" between tests and the pressure should be allowed to decay naturally as much as possible between tests. If the hole is kept "shut-in" (at the wellhead or downhole), the effects of the individual tests will be superimposed in a straightforward manner, and it will be possible to correct for the effects of previous tests.
As with temperature measurements using the WSTP or APC tool, processing pressure data from packer tests involves fitting of imperfect field data to theoretical curves. In practice, this means that data from early in the test periods tend to diverge from idealized curves due to imperfections in the test equipment and system geometry. Interpretations of transient pressure responses during formation testing generally require the assumption that the borehole is a "line-source," i.e., that the borehole is infinitely thin, while the formation extends radially to infinity. In addition, the potential complications of leaky seals, superimposed decay curves (from repeated testing in the same hole), and other factors mean that processing of packer test data requires considerable care and experience.
The programs of testing will vary from hole to hole, depending on the formation permeability. Further guidance can be gleaned from several good examples: the work done during DSDP in Hole 395A by Hickman et al. (1984) and in Hole 504B by Anderson and Zoback (1982) and Anderson et al. (1985), and the measurements made during ODP Legs 109, 111, 118, and 139 in Holes 395A, 504B, 735B, 857D and 858G (Becker, 1989; 1990a; 1991). (For the ODP work, see also relevant sections in Bryan, Juteau, Adamson, et al., 1988, Becker, Sakai, Merrill, et al., 1988, and Robinson, Von Herzen, et al., 1989; Davis, Mottl, Fisher et al., 1992, as well as ODP Technical Note 14).