Three kinds of observations refine the orbit and masses of V819 Herculis

One system, three views of motion

V819 Herculis contains a close pair of F-type stars that eclipse each other every 2.23 days, together with a brighter, evolved third star on a much wider orbit. The outer period is about 900 times the inner period, making this a strongly hierarchical triple: a tight binary and a distant companion, rather than three stars with comparable orbital separations.

In their preprint, Furkan Akar and Özgür Baştürk combine three observational channels within a single Bayesian model. Earlier studies had combined different pairs of these channels. The new analysis connects all three through the same Keplerian outer orbit, allowing each measurement type to help resolve ambiguities in the others.

How the measurements work together

Astrometry measures the changing angular separation and direction between the inner pair and the third star. It constrains the projected orbit and its tilt, but angular measurements alone do not establish the physical size without a distance scale.

Eclipse timings supply another dimension. As the inner binary travels around the triple system’s centre of mass, its light has a changing distance to travel. Eclipses therefore arrive earlier or later than expected from a constant-period clock. This light-travel-time effect measures the binary’s motion along our line of sight.

Spectroscopic radial velocities measure the third star’s motion toward or away from us. Together, the timing and velocity signals constrain physical orbital sizes. Matching those sizes to the angular orbit then constrains distance. Kepler’s third law supplies the total mass, while the relative motions determine how that mass is divided between the inner pair and the third star.

The final data set contains 83 astrometric measurements from 1980.5–2018.7, 215 eclipse timings spanning 1982–2024, and 226 archival radial velocities from 1975–1992. The timings include measurements derived from five sectors of TESS observations.

The inferred orbit and masses

The model estimates an outer period of 2016.97 days, with an uncertainty of approximately 0.27 days, and an eccentricity of 0.6800 ± 0.0016. An eccentricity of zero would describe a circle; this value indicates a substantially elongated orbit. Its inclination to the sky plane is 55.65 ± 0.20 degrees.

The inferred combined mass of the eclipsing pair is 2.721 solar masses, with a reported uncertainty of about 0.055 solar masses. The third star’s mass is 1.954 ± 0.070 solar masses. These are dynamical estimates, not masses imposed from stellar classifications. The analysis does not separately determine the masses of the two eclipsing stars.

The model’s light-travel-time amplitude parameter is approximately 0.0104 days, or 15 minutes. This describes the timing signal’s characteristic scale, not an eclipse duration. The third star’s radial-velocity semi-amplitude is about 18.46 kilometres per second. Reported parameter uncertainties correspond to the 16th–84th percentile ranges of the Bayesian posterior distribution.

Testing the distance scale

The inferred orbital parallax is 14.08 ± 0.13 milliarcseconds. Although the main fit uses an informative Gaia DR3 parallax prior, replacing it with a broad uniform prior changes the result to only 14.10 ± 0.13 milliarcseconds. That test supports the authors’ conclusion that the combined observations largely determine the distance scale. Alternative eccentricity priors also leave the solution essentially unchanged.

The parallax agrees with Gaia DR2, Gaia DR3 and the Hipparcos New Reduction within their uncertainties. However, Gaia DR3’s single-star fit has a high diagnostic error statistic, RUWE = 11.85, and no fitted orbital solution for this system. Catalogue agreement therefore requires caution: unresolved orbital motion can affect a single-star astrometric fit.

What remains uncertain

The precision is conditional on the adopted model and treatment of measurement errors. Many older astrometric measurements lack published uncertainties; the authors assign conservative values and fit additional scatter terms. They also exclude measurements affected by eclipses or flagged as outliers, and correct two position angles for a 180-degree ambiguity.

Separate fits provide useful but imperfect agreement. Astrometry-only and timing-plus-velocity fits recover consistent periastron dates, but their periods differ by 2.3 combined standard deviations. The joint period also differs substantially from a 2008 determination, illustrating how updated data and modelling can shift apparently precise estimates.

The third star’s mass exceeds expectations for an ordinary G8 main-sequence star, supporting an evolved-star interpretation rather than establishing a newly discovered evolutionary state; earlier work already described it as evolved. The model does not remeasure the angle between the inner and outer orbital planes, and a previously reported faint possible fourth component remains unconfirmed.

Predictions for future observations

The paper supplies predicted relative positions through 2031.5. High-resolution observations near the next outer periastron, around 2030.5, could test the orbit where motion changes rapidly. These positions and their uncertainty ranges are model predictions, not measurements; new imaging, spectroscopy and eclipse timings can test whether the present precision holds.