S Sagitta,+Oct,Nov.

[Malpi12]

Using my Canon 60d to observe the variation of U Sagitta (an Algol type eclipsing binary) was quite interesting 
https://stargazerslounge.com/topic/398978-u-sge/
but so far I have only caught it going in to minimum, not out the other side ! Trying to time minima with British  weather I have decided is a mug's game ! !

I found, in Burnham's Celestial Handbook, S Sagitta, a continuosly varying candidate nearby.  Much more convenient for occasional viewing between clouds. Bright and  a good  1  magnitude range.

A month of obs  later I had this not very smooth set:

However, with a hint by @IB20  and @robin_astro  in
https://stargazerslounge.com/topic/395015-algol-variables/?do=findComment&comment=4277498 I delved into the rabbit hole of Phase Folding, learned how to code it in LibreOffice, and produced this quite interesting phase diagram.

A wiki page
https://en.wikipedia.org/wiki/S_Sagittae
 on S Sag  has a plot which, when resized and overlaid in  Gimp, is a nice (I think !) match.

 

Some brief details of my home-brew method  (a  full description would be a bit tldr!)
images were variously stacks of 10 to 25 subs in DSS, camera fixed on  tripod, vintage lens 135mm @ f2.8, exp 2sec, 800iso
magnitudes estimated in Gimp using a 10x10pixel aperture in histogram mean, then in Gnuplot I fitted the data with a 1/x^2 curve to match the aavso comparison stars between m3.8 and m7.8.
 

Edited November 27, 2022 by Malpi12

[Malpi12]

Things to do when it is cloudy/raining outside - ponder upon the little bit of data one already has
Are you sitting comfortably :-  

More reading about this phase folding lark : the period does not need to be known in advance, it will produce a smooth plot with guesses near the correct interval, and a scattered plot at other guesses, more about that lower down.
First we look at the linear time plot,, ,

Even with the very limited data that I have so far I think we can have a guess at an approx. period. (hindsight is a helpful thing as well )
It looks like my first obs. (27 Aug) was near a min, then three more, near days 15-19, 24-25 and 33-34 (pink)
The first long, cloudy, interval is about twice as long as the following two, so assume that there were at least 4 minima in this 33 to 34 day period - the red line with an uncertain start point but a fairly well defined end between day 33 to 34.

I say 'at least', but there could have been eight or 16 ! However there is no indication of a missed minimum during the reasonably filled last period.

So, a best guess of between 8 (=32 days) and 9 days (=36 days) for the red seems reasonable ???

It turns out that I need not have learned spreadsheets, my fav. Gnuplot also has a Modulo function., which makes it easy to generate many phase plots with guesses at the folding period

I first chose 8, 8.2, 8.4, 8.6, 8.8 and 9 days. 

and Lo! the plots magically come together best at a 8.4 day period, how satisfyingly close to an actual 8.38 that is , , , or have I just introduced some severe observer bias with afore-knowledge   I show a multiplot but can upload individual larger ones if anyone wants them.

Trying to narrow it down with 8.35, 8.4, 8.45 etc. plots was less convincing - more data needed - I wish!

[DaveNicko]

Interesting, thanks for the write-up!

Indeed, more data would be good 🙂

Dave

[Malpi12]

Apologies for not updating this topic earlier - every time I was about to update I got a few more lucky breaks in the clouds and  more obs to process !
Anyway here we go with Oct and Nov additions to nicely fill in the gaps in the Sep lot.

Of particular interest is the 26 Sep (green) obs half way down the decline that seemed displaced from the other greens in the slope. I thought at the time it was a rogue or my bad, but a month later (30 Oct, pink) suggested it was a real bump. 4 (blue) in Nov confirm the wiggle !

A google on "bump cepheid"  is quite interesting.