Of course the convention is not to count EVERY OTHER flash, though this comes close to reality for cylindrical bodies. We should analyze the repeating patterns, and REPORT the time between CORRESPONDING flashes, and the number of complete cycles observed. To avoid reporting an non-integer number of cycles, some understanding of the pattern is necessary. Even your "EO" flashes may turn out to be asymmetric (in time) - sometimes a slight displacement due to synodic effects, sometimes from slightly non-cylindrical, in some cases even 1/3 or 1/4 period or seemingly arbitrary, due to satellite geometry. To visualize the flash pattern (unless you have a photometric recorder) my SYNODIC.EXE or an equivalent spreadsheet should be used. This compares observed (cumulative) flash timings to multiples of the assumed period, which makes it easy to detect the pattern, and possible erroneous keypresses. Using cumulative instead of lap times "removes" the risk of a single transcription error biasing the entire series, and resulting period computed. Determining the period from ALL measured times, rather than last-first, reduces the the standard error (from accuracy/# of cycles), possibly by a factor of 1/sqrt(# of points), and the risk of having a transcription error on the first or last measurement deteriorate the reported period. Please send me some complete measurement series, and I will show you a few samples. /Björn ----- Original Message ----- Note 1: Started as eO (even flash dim, odd bright) then evolved to EO around #60, then by #90 was Eo with brightest (-1 mag) flashes around #120, but I have counted only every other flash as this seems to be the convention. ------------------------------------------------------------------------- Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive: http://www.satobs.org/seesat/seesatindex.html
This archive was generated by hypermail 2b29 : Mon Apr 21 2008 - 07:14:55 UTC