Celestia Forums

I was working on another bug and using Alpha Auriga (Capella) as my test system, when I noticed that the magnitudes of the B component seems wrong. SIMBAD gives it's Vmag as 0.96, but the value in spectbins.ssc is 5.0. Also, the magnitude of ALF Aur A is given as 0.08, which I think is the combined magnitude of both components.

--Chris

This is I think a consequence of the parameters of the secondaries not being given in the data source Fridger used to generate the database. The result is a lot of secondaries with spectral type ? and apparent magnitude 5.

chris wrote:I was working on another bug and using Alpha Auriga (Capella) as my test system, when I noticed that the magnitudes of the B component seems wrong. SIMBAD gives it's Vmag as 0.96, but the value in spectbins.ssc is 5.0. Also, the magnitude of ALF Aur A is given as 0.08, which I think is the combined magnitude of both components.

--Chris

This I have documented in spectbins.pl ... and that's why it is so crucial to include these PERL files into the Celestia distribution!

The magnitude of the second component of alpha AUR is missing in Pourbaix' paper. Since I will NOT start to /hand-insert/ numbers from other unnamed sources, and since /we do need a definite value here/, I arbitrarily have assigned to all these cases mv=5.0.

via this PERL statement:

$mvB=$mvB eq ""?"5.0":"$mvB";

There must have been a good reason why Pourbaix did not use that value which you found in SIMBAD. He certainly knows as well how to exploit this data base . Lets look a little closer why:

SIMBAD gives only 2 external sources for Capella B = HD 34029 -----------------------------------------------------------------

CCDM J05168+4559P:
http://vizier.u-strasbg.fr/viz-bin/Vizi ... 68%2b4559P

here m_V = 2.1 [Edit:] actually m_V=? for the P component

GJ 194B: http://vizier.u-strasbg.fr/viz-bin/Vizi ... 194%20BBye

here m_V=0.71

So how did you convince yourself in person that -- in view of this LARGE spread of m_V for just TWO estimates --, there is ANY credibility of the quoted SIMBAD value m_V=0.96??

One day, one might consider merging Pourbaix' catalog with a batch readout of SIMBAD, which is at least an acceptable, quotable algorithm. However, SIMBAD just lists/collects values that have been obtained somehow without providing an evaluation of their credibility!

I could additionally print a warning in the *.stc data files, whenever some lacking data were encountered...that's easy, but I did not want to vast file space after it's all in the PERL scripts anyway.

Fridger

Actually, the CCDM page is the page for all components of CCDM J05168+4559, it lists m_V=2.1 for CCDM J05168+4559A - it doesn't list a magnitude for CCDM J05168+4559P...

A bit more searching through the references gives Hummel et al. 1994

M_V (Aa) = 0.29?±0.04
M_V (Ab) = 0.14?±0.04

The distance modulus used in the paper is 0.62?±0.01, which gives

m_V (Aa) = 0.91?±0.04
m_V (Ab) = 0.76?±0.04

According to this paper therefore, the spectroscopic secondary is visually brighter, which seems odd, though this apparently confirms the work of other observers (see the text on p1865)

In any case it is obvious that the magnitude of Capella B is not a number that just had to be filled in (SIMBAD), as it rather appeared from Chris' mail above.

The available mag values look strange and thus are essentially meaningless since there is very little solid evidence. That's what apparently also Pourbaix has concluded leaving the entry empty in his table.

Bye Fridger

This actually brings up an important difference between visual and spectroscopic binaries.

In visual binaries, the primary is the brighter star.
In spectroscopic binaries, the primary is often the star which gives the most prominent spectral lines.

Griffin & Griffin (1986)
Bagnuolo & Sowell (1988)

ajtribick wrote:This actually brings up an important difference between visual and spectroscopic binaries.

In visual binaries, the primary is the brighter star.
In spectroscopic binaries, the primary is often the star which gives the most prominent spectral lines.

Griffin & Griffin (1986)
Bagnuolo & Sowell (1988)

Andrew,

many thanks for these informative (and admittedly also somewhat entertaining) papers. The entertaining component rather refers to the fact that > 20 years ago, papers were written in a much more "personal" style. You would more directly address your scientific opponent by name in a publication, for example...etc.

So indeed there seems to be some evidence for spectroscopic binaries that the secondary, less massive component can well be brighter than the A component.

Bye Fridger

t00fri wrote:So how did you convince yourself in person that -- in view of this LARGE spread of m_V for just TWO estimates --, there is ANY credibility of the quoted SIMBAD value m_V=0.96??

I didn't go so far as to look at the original sources. I just recalled reading somewhere that the components of Capellar were roughly similar stars, and checking on SIMBAD confirmed this. With such great undercertainty in the magnitude of the second component, I can understand why it wasn't included in Pourbaix's catalog.

--Chris