Geoscientists Provide Critical Needs for Optimal Frac Design

The critical well in the field development needs a fracture design. What do you need to know to generate a good design? How do you generate the design? This paper presents an overview of the information, where to find it, and how to use it to generate a fracture design.
Information NeededReservoirThe most important reservoir information to know is permeability. This can be obtained from pressure build-up (PBU) tests, nodal analysis matches, and core measurements. Current reservoir pressure and the original reservoir pressure can be obtained from direct measurement after perforating or measured from PBU analysis. A high-permeability well might be designed with a tip screen-out (TSO) whereas a low-permeability well would need a long, low-conductivity fracture.
Reservoir porosity can be obtained from log measurements or core testing. The geological setting of a play will yield its drainage area, pressure transitions, and potential tectonic problems. Bottomhole temperature is measured from logs. Production fluids (oil, gas, water) and their saturations can be calculated from logs and from core data.
The geology of the play has a major effect on fracture design. Faults, unconformities, natural fractures, and other geological features can have major effects on the fracture design. Ignoring this information can have disastrous results.
Reservoir Fluid Properties The reservoir wetting phase (oil or water wet) can be determined from core information or inferred from production in the same reservoir. The gas gravity and percentage of impurities such as carbon dioxide, nitrogen, and hydrogen sulfide can be obtained from a reservoir sample sent to a fluid lab. Oil gravity can be determined from laboratory measurement of fluid samples. The production yield (gas per barrel or barrel per mcf) can be measured or estimated from other known reservoirs in the area.
Rock Properties Some type of lithology log is critical to identify the formation layering (sand, silt, shale, etc.) which is usually a gamma ray, or possibly SP. The Young’s modulus and Poisson’s ratio of the rock can be determined from core laboratory measurements. Modulus can also be obtained from a calibrated dipole sonic log. Sieve analysis of core samples from the producing interval yields data about possible fines movement for “soft” formations. “toughness” (or “apparent toughness”) controls the tip pressure required for fracture propagation. This is a complex variable that must be measured from field mini-frac testing.
Wellbore and Production Information If a well is deviated, a deviation profile can be obtained from its deviation survey. If the zone is offshore, water depth is needed to calculate zone stresses. The existing casing size and drilling bit size must be known to choose perforating charges with the proper hole-size and penetration for the proppant size and concentrations planned to be pumped. The work string size must be sized for the planned pump rates for the stimulation treatment. The production string must be sized for the expected production rates as determined by a reservoir simulator, nodal analysis, or company policy. Production line pressure or the suction pressure for the compressor and the associated temperature are needed for an estimation of the production rates the reservoir will produce. Any other facilities information that could hinder the potential production, such as pipeline or separator size, must be known.
Developing a Geo-Mechanical ModelStress Profile
A stress profile can be generated from historical data or by knowing the pressure profile for the zones and using the relation in Equation 1, where sCL is the in-situ stress or the fracture closure stress, OB is the weight of the overburden, which is typically between about 0.85 and 1.1 psi/foot-of-depth, n is Poisson’s ratio, typically between 0.2 and 0.3, PRes is reservoir pressure, and T is any tectonic in-situ stress effects. (1) See hardcopy for equation
A dynamic value for n can be determined from dipole sonic logs, and thus a stress log generated. Unfortunately, the log does not measure in-situ stress due to T, thus, log data MUST be corrected with field-measured in-situ stresses.
Fluid Loss Information Leak off or fluid loss is a function of formation permeability, reservoir temperature, and the viscosity (and wall building) characteristics of the fracturing fluid. A fluid loss profile can be generated from core testing, previous field experience, or estimated from published equations. Generally, a “final” value for fluid loss in a particular formation ‚ (with a particular fluid) MUST come from field, mini-frac testing.
Fracturing Materials-Fluid and Proppant There are many choices of fracturing fluids and proppants. A fracturing fluid should be chosen on the basis of reservoir permeability as well as temperature and wettability of the formation. The fluid chosen should not yield an efficiency of less than 10%. If the efficiency is less than 10%, a system with better fluid loss control should be selected.
The fluid should be tested with a Fann 50 with the chemicals from the field area as well as the local source water used. Breaker schedules should be developed and tested in the laboratory, and confirmed in the field.
The proppant is selected based on the “effective proppant stress,” availability in the field, and price. Major considerations for proppant selection are formation permeability and the “need” for fracture conductivity. This dependence on desired fracture conductivity, kFw, and formation permeability can be seen from dimensionless conductivity, FCD: where a desirable FCD is ALWAYS greater than 2.
Proppant stress can be calculated with Equation 3 where D is incremental stress due to the propped fracture width. D is usually small (200 to 400 psi) but can be significant in some cases such as TSO treatments in a moderate permeability, “hard” rock. Pwf is bottomhole flowing pressure. (3)See hardcopy for equation
Wellbore Information Well deviation is important to know. The stress calculation is based on TVD depth. Perforations scheme for the well also depends on the wellbore deviation and fracture type. Work string size can dictate potential pumping rates, and should be considered prior to choosing a fracturing fluid.
Example A company drilled a rate-acceleration well in a partially depleted, low-water-drive gas reservoir. The reservoir is a large anticline structure and is bounded on one side with a fault. Reservoir information can be found in Table 1. A core was taken in the first well and tested, yielding a modulus of 1.0 ¥ 106 psi in the sand and 1.5 ¥ 106 psi in the shale. Poisson’s ratio in the sand was 0.25. The core showed the sand was not friable, so fines were not expected to be an issue. The well was located far from the fault, and was a non-deviated hole with 7-in. casing set to TD. It was logged with a triple combo log (see Fig. 1). A previous frac showed there were no tectonic effects (i.e., a “normal” closure stress).
The sand stress was calculated to be 4,743 psi at the top of the pay zone with a fracture gradient of 0.59 psi/ft. The shale was calculated to have a closure stress of 4,886 psi at a depth of 7,856 ft. with a fracture gradient of 0.62 psi/ft.
A 30 lb/1000 gal low-guar gel system was chosen from the results of the previous frac, yielding a leakoff of 0.003 ft3/min, thus yielding an efficiency of 20% after a 7,000-gallon mini-frac.
The proppant chosen for the job was an economical ceramic proppant. The price of the local sand and ceramic proppant were about the same, but the ceramic proppant yielded much higher conductivity numbers at the given stresses, and thus less was needed for the equivalent conductivity.
The available information was placed into an analytical production model, which determined the optimum fracture length was found t