A quadratic time and pointing fit gives positive raw count coefficients for 26 selected HD 139139 K2 dips

We compare the stored raw counts and calibrated flux for known HD 139139 dips in public K2 Campaign 15 pixels. We use 26 fixed published events, a quadratic time and pointing fit, the same calibrated normalization, and fixed published weights. The weighted raw-count proxy scale is 1.06956. The calibrated scale is 1.02845. Their ratio is 1.03997, and all 26 raw event coefficients are positive. The net conversion scale is -0.04111; the stored background scale is 0.00455. The net change lies mainly in saturated stamp column 7. Across seven settings with pointing terms, the ratio is 1.014 to 1.061. Without pointing terms, it is 0.790 and only 20 raw coefficients are positive. Paired resampling of fixed event estimates gives a central 95 percent ratio range of 0.968 to 1.105 under an event independence assumption. Thus, the small amplitude difference is not established as a nonzero conversion effect. Prior work already used raw cadence files and independent reductions. This study adds an exact signed component comparison and reproducible event and pixel tables. Event selection used these same data. Shifted schedules are descriptive controls, not discovery tests. The supplied pointing values and normalization come from delivered products. The result does not exclude a detector effect before calibration or identify a planet. We preserve the original input, provenance, code, operators, checks, and figures.

HD 139139 · EPIC 249706694 · Random Transiter · K2 Campaign 15 · raw counts · pixel calibration · detector effects · pointing correction · saturated photometry · reproducible astronomy