Fixed sky models with SPOC corrections predict QLP–SPOC time differences within 0.09 seconds in 32 stars from four later TESS sectors

A known time correction fault affects public TESS QLP files. An earlier SCI study measured its sky dependence at shared times. We test whether fixed models transfer to later sectors. We reuse its frozen sample of 112 stars and 224 native column extracts. No stellar brightness enters the test. We add the SPOC barycentric correction to each QLP–SPOC time difference. We fit three sky terms and a constant to eight stars. The first model uses 328 cadences in sector 99 after a public ephemeris endpoint. The second uses sector 102. Both models are fixed before the later-sector test. Predictions require the target position and its contemporaneous SPOC correction. In sectors 100, 101, 103, and 104, the models predict 118,304 differences in 32 unused test stars. The RMS error is 0.0319 seconds. The largest absolute error is 0.08874 seconds. A model that uses the external ephemeris endpoint and one fitted offset predicts sectors 100 and 101 with a maximum error of 0.0326 seconds. This supports an endpoint clamp for the first regime. The second regime retains a fixed nonzero sky vector. These are relative archive time tests. They do not establish absolute clock accuracy or detect a planet. The package preserves source code, extracts, source hashes, full headers, query records, model choices, figures, and offline reproduction checks.

TESS · QLP · SPOC · barycentric correction · time calibration · archive audit · sector transfer · reproducible astronomy