A direct TESS Sector 91 search misses most translated K2 dip models for HD 139139

HD 139139 showed 28 isolated dips in the 2017 K2 data. We test whether a direct aperture search of the public 2025 TESS Sector 91 data can find models with the published depths and durations. The primary search uses 8,315 exposures with QUALITY = 0, eight durations from 0.7 to 8.2 hours, a local quadratic baseline, and a fixed score threshold of 6. No primary dip reaches the threshold. At the original depths, the search recovers 111 of 14,336 isolated trapezoid injections, or 0.77%. Only 1.28% recover among trials that pass the coverage test. We also inject all events from 512 translated sections of the finite K2 sequence. Of these trials, 469 give no recovery at the original depths. Only 2 give no recovery at four times those depths. The TESS null result is therefore expected under this specified original-depth model. It does not establish that the original dips stopped. Secondary peaks depend on aperture and flux correction and do not support a new planet claim. The test assumes equal numerical aperture depths in the two instruments and fixed empirical noise scales. It does not measure upstream calibration losses or the sensitivity of other photometry methods. The package preserves the public FITS inputs, source records, event table, code, complete results, and figures for offline reproduction.

HD 139139 · EPIC 249706694 · TIC 70652803 · TESS Sector 91 · isolated dips · injection and recovery · selection function · public data · Random Transiter