Since I lost my observing site I have switched to a fixed DSLR and wide-angle lens on my back deck. I’m using 15 to 20s exposures which is great for stars brighter than 8th or brighter. Been observing Gamma Cas and Mira so far.
The Green channel transformed to V looks very good and the closure of the two check stars are also very good.
However, when solving for B the closure is off by several tenths of a magnitude or more. The VSP catalog errors given are quite large. I understand the BSC stars were observed using PEP and UBVRI, Johnson or Johnson and Johnson-Cousins not sure yet. I’m using the 23, 27 primary, and one of the 34 comparisons for Gamma Cas.
Using the 23 as the unknown and the 27 as primary I get for the 23 B magnitude observed - catalog (O-C) of -0.136 magn. (brighter) than the B catalog value 2.610.
Using the 34 as the unknown and the 27 as primary I get for the 34 O-C -0.194 magn. (brighter) in B than the catalog value of 4.010.
A bit confused about B at this point.
In V the 23 O-C is +0.053 and the 34 O-C is +0.027 magn. These are quite acceptable.
The comparison stars in VSP reference the Bright Star Catalog as the source of the magnitudes. Anyone have any comments or info on the accuracy of the photometry of the Bright Star Catalog especially in B?
…since only a few stars are involved, you might dig into the literature more directly. Start with the Mermilliod compilation of individual data, which has bibcode links, and is VizieR item II/122B. There is also the merged ‘homogeneous means’ catalogue, item II/168.
Thanks, kinda what I gathered. The catalog errors are very large on VSP. I understand bright stars are hard… Don’t understand the B observed - B catalog biases. Should I be worried or just assume the catalog errors in B are not very good and just report what I get.
I’ve not tried to look at catalogue values of bright stars critically, but I do have a general comment about DSLR photometry for B-V and B.
In my experience also, B-V and B measured values can be rather different from catalogue, and I therefore I do not report DSLR transformed B to the AAVSO, only transformed V. For example, one night not so long ago a check star B-V was different from catalogue by 0.188 mag units, yet the Check V was almost spot on, the difference from catalog being only 0.006 mag. I don’t understand why that should be, because I use the target (measured) B-V in the formula that calculates target V.
I’ve reported many DSLR-based transformed RGB to RcVB over the years since I started observing again. Never had biases like what I’m observing in the Gamma Cas comps. Red to Rc is usually the problem I have using DSLRs for photometry. I have a few more things to change and test in my observing before blaming the BSC. The VSP comparison star errors are very large and if those quoted in the chart sequence table are 1-sigma errors then the biases perhaps aren’t that bad! Bright stars are hard. The three comps I’m using and their associated (B errors) are the 23 (0.100), 27 (0.100) and what I call the 34-1 (0.141) which is Eta.
The 34-1 comparison is Eta Cas, HD 4614 and 4614A a blend, HR219.
The 23 comparison is Alpha Cas is HD 3712 and HR 168.
The 27 comparison is Delta Cas HD 8538 and HR 403.
The 34-1 and 23 photometry comes from "Stellar Photometry in Johnson’s 11-color system (Ducati, 2002). VizierR gives this roll up info: “This catalogue gives photometric data in the UBVRIJHKLMN colors and some additional information. Data for the JHKLMN bands come from NASA Ref. Pub. 1294 (1993). More complete information is found in Wisconsin Astrophysics 504 (1993).
The effective wavelength of the Johnson filters are: U = 360 nm B = 450 nm V = 555 nm R = 670 nm I = 870 nm J = 1.2 um H = 1.62um K = 2.2 um L = 3.5 um M = 5.0 um N = 9.0 um.”
The V and B errors given in Simbad are given as (~).
The 27 comparison photometry comes from Oja.T Astronomy and Astrophysics, Suppl. Ser., Vol. 89, p. 415 (1991). The roll up info in VizieR is "Results are presented from UBV photometric observations of 1000 stars of the Bright Star Catalogue and the faint extension of the FK5. Observations were carried out between July 1987 and December 1990 with the 40-cm Cassegrain telescope of the Kvistaberg Observatory. The 27 is the only star with errors given in Simbad and are V 2.680 (0.009) and B 2.810 (0.007).
So depending on the comps two different observing updates to the YBSC are at work. Also were the Ducati data using non-standard filters?
I was going to try more defocus to try to sample the Bayer matrix more tonight but alas the next snow-maker high clouds are rollin’ in. Maybe next week.
Probably the best one-stop source for the UBV data on the bright stars is the Mermilliod ‘means’ catalogue, which is VizieR item II/168. This includes data up to 1989 and in principle has the various zero-points and color terms corrected to the fundamental standard system. Cousins V-R in the north is more problematic for naked-eye stars, since really the only source is the 1991 Kornilov catalogue.
My look at these suggests the following:
Name V B-V V-Rc
HR 219 = eta Cas AB 3.44 0.58 0.357
0.02 0.02 0.012
HR 403 = del Cas 2.678 0.130 0.072
0.004 0.004 0.012
…where V and B-V come from Mermilliod (‘mermio’), including his mean errors on the second line of each entry; (V-R)c is from the Kornilov et al WBVR catalogue with their Johnson V-R transformed to Cousins V-R using the Ben Taylor relation, which is in 2005ApJS..159..100T. He says the errors in the Kornilov V-R data are 0.012 mag, which I’ve simply added to each star above. Taylor implies that there are no systematic errors in the Kornilov V-R as a function of RA. (The original Johnson VRI standards have a significant RA error, which has caused all sorts of problems down the line.)