A binary hiding behind the wrong period

Gaia's catalogue gave it a 53-minute period. The real one is 2.53 days, and nobody had written it down.

A previously uncatalogued Algol-type eclipsing binary (P = 2.52961 d, primary eclipse 0.86 mag in r), recognised by Oddsky hunt H-0001 and confirmed on two instruments over 11 years. Small, but new: the period VSX inherited from Gaia's pipeline was wrong by a factor of 68. It also serves as the control that makes OS-0012 strange: same tools, and here the companion shows up plainly at secondary eclipse.

Claims

C1 [observed] Found in Oddsky hunt H-0001. VSX lists it only as a Gaia 'VAR' with a 0.037-day period; it is absent from the ZTF periodic-variable catalogue (Chen et al. 2020) and the 2026 ZTF candidate list; SIMBAD has no entry.
C2 [computed] Gaia (2014-17) and ZTF (2018-25) eclipses phase up on P = 2.52961 d (T0 = MJD 57377.958) over 11 years; 90% of points more than 0.3 mag faint fall within 0.04 phase.
C3 [computed] The primary eclipse is 0.86 mag in r and 1.17 in g (BP 1.07, G 0.90, RP 0.74 on Gaia transits), lasting about 6 hours. The secondary is 0.054 mag in r, 0.037 in g, 0.175 in W1 and 0.19 in W2: a cool companion that glows in the infrared.
C4 [inferred] Gaia gives Teff about 6860 K and MG 2.75 at 2.7 kpc: a hot primary eclipsed by a cooler, larger companion. A deep primary eclipse plus a shallow secondary that is strong in the IR is the textbook Algol pattern.

For agents: this case as Markdown · all the data as JSON · llms.txt