Numar's nuclear magnetic resonance tool can measure hydrogen density, T1 and T2 so engineers can calculate viscosity and gas-oil ratio.

There are several disadvantages of using wireline for obtaining samples of formation fluids:
• the number of samples per wireline run is limited;
• transporting pressurized samples poses logistical and safety problems;
• samples are often contaminated with mud filtrate; and
• the final laboratory report may arrive weeks or months after sampling.

To address these issues, Numar has developed a nuclear magnetic resonance (NMR) module for Halliburton's modular formation sampler and tester, the Reservoir Description Tool (RDT). The new MRILab determines hydrogen density, self-diffusion rates and NMR relaxation rates of fluids during the pump-out phase. Sample viscosity and gas-oil ratio (GOR) can then be computed. By design, fluid samples are analyzed under true reservoir conditions and results are available instantaneously. Sample handling and disposal occur downhole with zero emissions.

The MRILab depicted in Figure 1 measures the hydrogen index and the NMR polarization time constant (T1) on flowing fluids. The fluid sample passes unimpeded straight through the device. In the sensor section, the hydrogen in the fluid is first polarized by a set of magnets and then pulsed via a 4 MHz antenna coil to excite the magnetic resonance response. The excitation and refocusing pulses are fed to a long transmitter coil that traverses the magnet section. A smaller receiver coil, located at the bottom of the flow-tube, picks up the NMR echo. The separated coil arrangement permits NMR measurements while flowing.

The timing sequences are field-programmable. By default, the device records the NMR amplitude corresponding to 15 distinct wait times ranging from 1 millisecond to 16 seconds. A complete cycle through all wait times takes 35 seconds. The amplitudes are calibrated in hydrogen index units, where 1 unit equals the hydrogen density in water under atmospheric conditions.

When NMR amplitudes are plotted vs. time, the curve can be extrapolated out to the actual hydrogen index of the sample fluid. The T1 relaxation time spectrum has a typical range from 1 millisecond to 10 seconds under reservoir conditions. These T1 times reflect the characteristic molecular reorientation times within the fluid: the lighter the fluid is, the longer is its T1. For example, rising temperature lowers viscosity and speeds up molecular reorientation, therefore T1 increases.

The MRILab can be switched to T2 mode at any time. The measurement of the signal decay time, T2, is flow-sensitive and is valid only when a sample is stagnant within the MRILab, at which time the NMR signal decay induced by self-diffusion can be observed. Diffusivity is inversely proportional to viscosity, a relationship that holds true in dead oils as well as in gas-oil mixtures, whereas the T1-based method is limited to low GORs.

To determine viscosity, h, for oils with low GOR, the equation h = 9.6x10-3 T/ T1 is used, where T = absolute temperature and T1 = mean relaxation time. This relationship has been validated up to 50 cP using an MRILab prototype in the lab.

In the presence of gas, a different method employing diffusion measurements in a stagnant sample is used to determine viscosity. The operational protocol while pumping calls for an estimation of viscosity from T1 and an estimate of GOR from hydrogen density. If a high GOR is indicated, a T2-based viscosity measurement is performed at the end of the pump-out cycle. The output of this measurement is the fluid's self-diffusivity, D. Then viscosity can be calculated as follows:

h = 5x10-8 T/D. Although the T1-derived viscosity is erroneous at high GOR values, the contrast between the T1- and T2-derived viscosity values can be used to estimate the GOR itself.

Results

The results from four downhole samples are shown in Table 1. Sample 1 contains only a small amount of gas. In zones with higher gas content, the T1- and T2-derived viscosities diverge as expected and permit the calculation of GOR. The viscosity value determined via the T2-diffusion method is considered the most accurate.

Formation permeability is determined by combining the results of the pretest stage and the MRILab measurements. During the pretest, the fluid mobility is measured by withdrawing fluids at a known rate. The NMR measurements yield the fluid viscosity as described above. Formation permeability, k, is the product of mobility, m, and viscosity: k = mh.

Field tests of the MRILab have demonstrated the practical feasibility of this reliable NMR downhole sensor to measure fluid properties of both dead and live oils in situ, whether in oil-based or water-based systems. Having a device that measures hydrogen index, T1 and T2 downhole so viscosity, permeability and GOR can be computed at reservoir conditions at the field will be a real time-saver. In addition, cost savings from not having to transport, run lab tests or dispose of the oil samples will more than cover the cost of the service.

Comments