A fixed stream model gives no optical detection in eight TESS sectors of HAT-P-67

HAT-P-67 b has a known helium stream ahead of the planet. We test for a constant optical drop at its phases in eight TESS sectors from 2020, 2022, and 2024. We use a fixed ramp from phase -0.30 to -0.05 and exclude transit. The primary fit uses 6991 bins and simultaneous terms for stellar and orbital light changes. The PDCSAP depth is -12.9 ppm, with a standard error of 119.7 ppm for orbit blocks. Positive depth means loss of light. The nominal two-sided 95% interval is -254.4 to +228.6 ppm. Ten nuisance models and SAP flux give no detection of a constant positive pattern. The primary nuisance projection leaves 9.35% of the raw weighted template information. Separate fits by year do not have a stable positive sign. Independent bin and full-matrix checks reproduce the main calculation. These results test one shape with a fixed ephemeris. The intervals require assumptions about independent blocks and the stellar model. They are not calibrated physical upper limits. The study cannot test the roughly 4 ppm helium contribution estimated in earlier work. We preserve the raw light curves, source hashes, code, full fit grid, and figures.

HAT-P-67 b · TESS · atmospheric escape · optical photometry · stellar variability · fixed signal template · reproducible analysis