While post-frac erosion and Fracture Driven Interactions (FDI) can be good indicators of a successful frac, getting long-term performance data has mainly been limited to full wellbores of a single design and waiting months for the results. The Permian Asset team was debating the impact of stage architecture on production. Did different designs come on strong, but then fall off over time? Could a doubling of flow area from the standard design lead to a doubling of performance? An experiment was designed to answer these questions with a toolbox of emerging production diagnostics. By combining these latest techniques with some geologic characterization from in-bit accelerometer data, performance of a particular design can be examined on a single wellbore, saving time and money. Three different stage architecture designs were alternated along a Wolfcamp wellbore. These designs were selected to look at the impact of cluster and stage spacing and perforation orientation. A novel methodology for oil tracers (AB testing) was deployed in the test well. Oil tracers that had previously underwent calibration testing for an affinity for each other and the reservoir were pumped during frac. A unique oil tracer was pumped per design. Additionally, an oil tracer was pumped in stages along a greenfield portion of the lateral and another in brownfield stages next to a parent well. A carbon fiber rod was run to get an initial look at production allocation by cluster, stage, and design type, allowing a comparison to oil tracers at that snapshot in time.
Depletion analysis was calculated by stage along the entirety of the lateral. Depletion values ranged from 0 up to 500 psi. Despite lateral variability, when averaged by design type, the depletion values were within the margin of error. The oil tracer is showing that an extended stage length 0 degree phased stage (300’ Top Shots) is the top performer versus an extended stage length 90-270 deg phased design (300’ 90/270) and a design with half the stage length and cluster spacing to double the flow area (150’ Top Shots). Initially, the greenfield portion of the lateral (3600’ in the toe) was outperforming the next 3600’ of lateral adjacent to a parent well, but this trend reversed over time. The carbon fiber rod provided a cluster level analysis with the majority of the clusters contributing to the overall production. There is a heel bias associated with the well that is in directional agreement with the greenfield and brownfield tracers at the time of deployment. The rod has the most production attributed to the 150’ Top Shot design, followed by the 300’ Top Shots, with the 300’ 90/270 design coming in last. There is a discrepancy between the top design with the oil tracer and fiber rod data. This could be a normalization issue or a water allocation problem. Continued work is needed in this space to find the root cause. Overall, the stacked diagnostics provided actionable insights on completion designs that can inform future deployments.
