Greetings,
Recently I have been looking at all the transform coefficients and standard errors of linear L-S fits I have done over the last few years. I observe with an Atik490EX + Astrodon BVRcIc filters + SkyWatcher Esprit 100 (1.4 arc sec / pixel) usually but I also have lots of DSLR transforms done over the years. I most often use M-67, NGC 7790, SA41, SA98 and SA110 for the CCD transform fits. By far the best "quality" fits are obtained using SA110 when using the B. Skiff red extension stars. SA 110 extended sequences | aavso
However, Landolt SA20 is consistently of very poor quality for both CCD and DSLR transform coefficients estimates. At least partly a low N sampling problem and low color range, but I'm not convinced that is all of the reason for the large S.E. compared to other standard fields.
I am wondering if others who are using Landolt fields for transform coefficient determinations have had problems with SA20 as well? Any of the other standard fields that consistently produce poor fits for you?
Jim DeYoung (DEY)
Looks as though no one responded to this. The obvious answer is to observe the field together with one or more other fields to see if the zero-points and color terms are consistent. In other words, if you use those other fields to derive data for SA 20, do you get values within a few percent of Landolt's? Perhaps this is something you already have data for. It is not too late in the season to get some additional data, including, for instance, PG 0231+051 or another of the +45 Dec SA fields.
Unlike most of the equatorial standard fields, Arlo got only about half a dozen nights on SA 20. But the mean errors on the brighter stars are quite small, so ought to be reliable at the level you are seeking.
\Brian