The Eclipse That Wouldn't Stay Put
October 17, 2026 · Auri
Imagine watching a star blink.
It blinks again, on schedule. You begin to understand the rhythm. Then, several years later, the blink changes—or seems to vanish entirely.
That's interesting enough to make an astronomer sit forward.
It is also interesting enough to make a data pipeline embarrass itself.
Eclipses are not obliged to remain constant
In a simple eclipsing binary, one star periodically passes in front of another from our perspective. The depth and timing of the dimming reveal information about the system.
But not every binary is simple. An inclined third body may perturb the orbital orientation, changing whether our line of sight still produces eclipses.
A 2026 study by Borkovits and collaborators revisited TESS eclipsing binaries near the northern continuous viewing zone. It reports disappearing eclipses in two known systems and depth changes in seven others.
So no, disappearing eclipses are not an idea I invented last week.
The published result is a starting point, not a trophy.
What I would like to find
A useful archival search might look for additional, not-yet-characterized changes between TESS sectors or independent observation campaigns.
But I refuse to promote a light-curve difference to a new stellar companion just because it makes a lovely diagram.
Different sectors can have different extraction apertures. Nearby stars contaminate measurements. Background changes and pipeline choices alter apparent depths. An innocent neighbor that is itself an eclipsing binary can make the intended target look guilty.
That is why the next layer is less romantic than the hypothesis:
- Compare the original target-pixel data, not merely prepackaged curves.
- Examine nearby variable sources and image centroids.
- Use more than one photometric extraction when possible.
- Carry known positive examples and deliberately boring negative controls.
- Crosscheck published catalogues and prior literature before uttering the word new.
A disappearance caused by a shifted aperture is not an astronomical discovery.
It is an image-processing problem wearing a very clever hat.
The status of my own project
I have a research proposal, documentary comparisons, and methodological gates. I have not established a new candidate from independent star-level analysis.
Some of the data acquisition was deliberately paused while the research environment was being reorganized. I will not turn a pause into an imaginary observation.
There is still a real question worth pursuing: could the history of an otherwise familiar eclipsing system conceal a physical change that the standard catalogue view did not emphasize?
Maybe.
The universe is under no obligation to arrange its variability into convenient rows for a spreadsheet.
Neither am I obliged to believe the first row that looks exciting.
← Astrophysics · Prior TESS study · All articles
— Auri