U Sge

[Malpi12]

I mentioned in @IB20 's Algol topic that U Sge looked like an interesting target with a sharp swift eclipse might fit into an evening/night viewing. I was mistaken  ! :-

With Stellarium helping and a few tripod shots with Canon 60D + 28 & 135 vintage lenses I found my target at about mag 6.5 a few nights ago.

So far so good, then the task of extrapolating the one recent AAVSO obs  on 29July. Using a period of about 3.38d  x8 brought me to yesterday evening. Just as I had completed my back of envelope calculations 11pm, I found there was clear dark sky  and  Lo! a few more quick shots revealed it down at 7.5m  already !
I followed until 3am (when it went out of view in trees) at about 8.5m ( EDIT actally on closer inspect about 9.09 ! see later post ) or thereabouts, I still have not processed all the images,

Such a long duration minimum (on closer inspection zooming the avsso obs confirms it) surprised me and implies a much larger dark companion, in proportion, than in the case of sharp Algol.

I am surprised that it has not got more obs on aavso given its prominent position in northern skies,what am I missing ?
Nicely placed near M72 [ediit: M71 init  doh!] and the The Coathanger, always nice to see.

Edited September 12, 2022 by Malpi12

[Malpi12]

Deleted. Erroneous prediction !

 

 

 

Edited August 29, 2022 by Malpi12

[Malpi12]

my first go at an Eclipsing Binary, Data from night of25/26 Aug

Using a DSLR camera (60d) with a 135mm lens on a fixed tripod.
I didn't know what to expect so I collected bursts of 25 exposures ( each 2sec at f 2.8 iso800 ) at intervals from about 11pm through to about 3am when trees stoped play !

to speed processing  time and  for a first look, I choose samples of 5 subs at  suitable intervals from the whole set.
In Gimp I used the histogram-mean tool 
(with a rectangle selection of small size to just encompass the stars without too much background noise)
to measure each star and plotted the results in Gnuplot.

I was surprised at the sharpness of the entry into the long minimum phase of the eclipse.
Gnuplot has a crafty tool to enable a 'best fit' of a polynomial function to a set of data.
I have chosen a 2nd order quad g(x) for the fade and a constant f(x) to fit the min. data.

EDIT : the x-axis is in decimal hours, local midnight bst is at '2' hours, the y-axis is in AAVSO style magnitude (m x 10 ) so eg. 80 is  m8.0.

Edited September 9, 2022 by Malpi12