Talk:HD 54893

Effective Temperature of HD 54893
@Lithopsian, you gathered some stellar parameters from Gaia Data Release 3. Most of them are accurate, except for the effective temperature. The temperature on the starbox is 15,974 K, which is too cool for a B2 star. The only other value I found was 21,150 K, from Hohle et al. (2010) but that was from 12 years ago, which is why you reverted my edits. Is there any other source with an official value? Speed doesn't always mean quality. 400Weir (talk) 21:20, 29 September 2022 (UTC)
 * I saw the discrepancy and searched the literature for other values. Many published temperatures, most in the 16,000-17,000 K range, so I stuck with the Gaia value for convenience.  Deliberately picking the single outlier just because I think it is the "right" value would be wrong by Wikipedia standards.  Who knows, maybe the spectral class is a but funky, not for us to say.  Hohle et al is full of automated and sometimes stupid results, and based on data a decade out of date, so I avoid it if there is a semi-decent newer alternative.  If you read the paper, they derive the effective temperature by a table lookup based on the adopted spectral class and luminosity class, which is why it is the value you expect for a B2 star.  If you want to know why the discrepancy, most sources work on a B3 spectral class (or corresponding colour indices), and that has a "table lookup" temperature around 17,000 K.  In Wikipedia terms, you just have to go with it, or change the spectral class if it is too grating and you can find a reliable alternative source.  Should be easy enough, Simbad lists a much newer source and a B3IV/V type.  Problem solved?  Lithopsian (talk) 10:22, 30 September 2022 (UTC)
 * Wow! I didn't know most of the results from the paper are automated. We should stick within the range you suggested, along with the newer temperatures. Speed doesn't always mean quality. 400Weir (talk) 16:28, 30 September 2022 (UTC)
 * All these large tables and databases are at least semi-automated. Papers dedicated to one or a small number of objects have most likely been checked for sanity and ideally the authors have already been through the literature and what they come up with amounts to a consensus.  I can't remember now what some of the really odd results were from it, might have been some of the red supergiants.  HIP 27465 and HIP 23868 certainly look odd.  Whatever else is going on, it is hamstrung now by using out-of-date distances and there are so many newer databases with very comprehensive coverage.  Lithopsian (talk) 18:35, 30 September 2022 (UTC)