Different values of period of peculiar hydrogen-deficient binary Upsilon Sagittarii are known:
Regrettably the star is too bright for mass surveys, but oversaturated ASASSNSP photometry gives a hint on period about 29.5d (or about 59d with two minimums). Other noted above periods have not been confirmed.
Hi Mikhail,
upsilon Sgr is a very bright system that has been already studied in the past.
When you mention that there are different orbital periods, that’s not correct, the orbital period is only one and is well-known because it is not only based on photometry but on spectroscopic observations.
The shorter period of 21 d. likely comes from pulsations (PVTELI-type, typical of hydrogen-deficient binaries), although the system is so complex and unique that this can’t be confirmed.
I have revised the star entry in VSX to include the short period pulsations that you found in TESS data and spectral information from a very recent (2023) paper.
About the ASAS-SN light curve that you attached: you have mentioned it yourself, the data are saturated, ASAS-SN can’t observe a 4.5 mag. star. What you plotted is the Moon cycle not a period related to the star. So keep this in mind for the future: just ignore ASAS-SN for stars brighter than 9-10 mag.
Cheers,
Sebastian
TESS data shows fast low-amplitude oscillations of peculiar hydrogen-deficient binary Upsilon Sagittarii with period about 0.6d. If they are real pulsations, it will be interesting to unterstand their nature.
Mikhail, I have moved your other post here, do not publish posts on the same object in different forums, it becomes difficult to follow, it is better to have all in one place.
Cheers,
Sebastian
OK Sebastian, thank you for ups Sgr revision with TESS data!
Indeed I did not think about Moon, thank you for the explanation!
This is another good example of not trusting robo survey data simply because it is new, hi-tech, blah-blah. A good starting point is “assume it’s wrong”, including with one’s own data. Just because data are taken from a spacecraft where it’s never cloudy doesn’t mean it is not loaded with multiple inscrutable systematic problems of an instrumental nature. You can’t simply download data and get results out, cuz the raw data are crap.
Perhaps the AAVSO could arrange a talk/webinar/other with someone like Anne-Marie Cody, who has written/published a lot of the extraction schemes for TESS and K2 data. To be on the themes of the general amazing-ness of those data, but also how full of problems they are — “why these data need work, and what to do about it”. I would like to hear, for instance, about how the more-or-less unfiltered data affect the astrophysical interpretations. It seems to me that, OK you can get periods out, but often the physical meaning of the data is uninterpretable for many types of stars from lack of color-indices, spectra, etc. RR Lyraes and Cepheids are examples, where the colors change significantly during the pulsations, so the phase changes as a function of color/temperature. This seems to be glossed over in the literature.
\Brian
Thank you, Brian, webinar is excellent idea!
Regarding pulsator’s color changes: indeed there is analysis in the literature, for example Preston’s observation of RV indexes “loops”. But you are right, color indexes behaviour is special question and them calculation is not trivial task. It maybe theme for other webinar 
