Local noise tests do not show a new timing anomaly in the July 2024 TESS transit of Kepler-51d

We test two published four-planet orbit models for Kepler-51 with public TESS data. We select six expected transit events from 2024 by time and data coverage. These events are absent from the published timing table. Two events occur after the last time in that table. For the July transit of planet d, a fit with free duration gives BJD_TDB 2460512.34011. This time is 39 to 45 minutes before the two model means. The formal error is 9.2 minutes. Tests with residual blocks give an error scale of 15.5 minutes. Tests with nearby artificial transits give larger errors. For 15 tests with an adopted 1.18% depth, 8.448-hour duration, and matched baseline coverage, the root mean square timing error is 61 minutes. Tests at the fitted shape give 34 minutes. A fit with fixed duration shifts the time by 55 minutes. These data do not give sufficient evidence for a new timing anomaly. They do not select between the two tested orbit models. The later transit of planet b reaches a duration bound and gives no primary timing test. We release the raw light curves, orbit inputs, code, figures, and result tables. The injection tests have overlapping windows. Their errors are not a calibrated confidence limit.

Kepler-51 · TESS · transit timing variations · correlated noise · injection recovery · public data