Wrong SpectralTypes for stars

[Guckytos]

HI i just found that Celestia has problems with some of the SpectralTypes it should know.

I had asked the question in forum, which classes Celestia knows and Grant gave me the answer O,B,A,F,G,K,M,L,T,R,S,N,C,WC,WN,D (for a white dwarf),Q (for a neutron star), X (Black hole). Is that list really complete or are there some more classes, for variable stars for example?

Anyways, here are the encountered errors:
1. Celestia doesn't seem to know the C class (Carbon stars), well it does display a star, but says SpectralType "?4V" instead of i.e. "C4V".

2. If i create a N class star in my stc, Celestia makes it into a S class star on display and vice versa.

3. The Wolf-Rayet types WC and WN are seem to be not really two different types, but are just displayed as stars with SpectralType "W4V" for example instead of "WC4V" and "WN4V".

I am using 1.3.1 at the moment, but these errors could also be reproduced with 1.3.2. pre 11.

Regards,

Guckytos

[granthutchison]

Is that list really complete or are there some more classes, for variable stars for example?

There's no coding for variable stars. The list is complete as of 1.3.2.

1. Celestia doesn't seem to know the C class (Carbon stars), well it does display a star, but says SpectralType "?4V" instead of i.e. "C4V".

This works in 1.3.2, but was added later than 1.3.1 ... I can't recall which pre version made the change.

2. If i create a N class star in my stc, Celestia makes it into a S class star on display and vice versa.

Ha. Celestia has its numbers crossed. The error is also there in reading stars.dat, where Hip 1401 should be N but shows up as S, whereas Hip 14555 has the reverse error.

3. The Wolf-Rayet types WC and WN are seem to be not really two different types, but are just displayed as stars with SpectralType "W4V" for example instead of "WC4V" and "WN4V".

This one's been reported before. I think Celestia's losing track of the two-letter string internally somewhere.

Grant

[danielj]

You are wrong,Grant.I have Celestia 1.3.2 and there is no carbon star.Neither the R,N,S,C stars.The quote ?5V or someting like that still remains.
Is that a correction in CVS?

[granthutchison]

You are wrong,Grant.I have Celestia 1.3.2 and there is no carbon star.Neither the R,N,S,C stars.The quote ?5V or someting like that still remains.
Is that a correction in CVS?

No, it's a clean installation of the distribution 1.3.2, downloaded via the Celestia homepage, running on a machine with no other versions of Celestia and no add-ons installed. It contains about 100 stars of spectral class C, such as T Lyr, HD 156074, HD 16115. Spectral class C was supported from 1.3.2pre1 onwards, and stars.dat was updated for subsequent versions.
Are you saying that these stars don't appear as class C in your installation of 1.3.2? I'm a bit surprised.

You might also try to install a carbon star near the solar system with:

Code: Select all

555555 "Test"
{
   RA 0
   Dec 0
   Distance 10
   SpectralType "C8III"
   AbsMag -1.5
}

This also displays the correct spectral class in my installation. Is there a problem in yours?

R,S and N have been supported from very early in Celestia's history, so I really can't explain why you'd have a problem other than the one Guckytos has just pointed out.

Grant

[symaski62]

Code: Select all

555555 "Test"
{
   RA 0
   Dec 0
   Distance 10
   SpectralType "C8III"
   AbsMag -1.5
}

I see red

[Bob Hegwood]

Grant,

When I add your code to an STC file, I see a red star with the classification
of: C8 III T Lyr is also correctly listed with this classification.

However, I also see the star HIP 3937 listed with a classification of ?5 V.
Mayhap a bit of confusion to those of us who do not fully understand the
star classification system.

Just thought I'd ley you know since DanielJ is Looong on wind, but short
on action.

Take care, Bob

[granthutchison]

Thanks for the feedback, Bob.
Hip 3937 has no spectral information at all listed in Hipparcos, which is why it's a "?". The "5V" bit is an unfortunate leak of Celestia's internal assumptions ... no listed subclass, assume 5; no listed luminosity class, assume V. (The assumptions are required so we can pick approximate temperatures and rotation periods off a look-up table.)
You'll be glad to hear that the next version of Celestia will suppress this sort of stuff ... so all you'll see displayed will be the original data and no assumptions. If Hipparcos says just plain "G", that's all that'll appear on the screen, rather than "G5V". If there's no information at all, all you'll see is "Class: ?", which is much less confusing, I think.

Grant

[granthutchison]

Have just been playing with the Chris's latest revision to star handling. As well as the improvement in spectral class handling I described above, he has also:
1) Fixed the reversal of classes S & N
2) Made classes WC and WN display properly
3) Added support for white dwarf spectral classes

Oh, and added black-body star colours as an option ...


Grant

[Bob Hegwood]

Have just been playing with the Chris's latest revision to star handling.

Good deal, Grant...

These changes will make Christian happy, and the black body addition
should be interesting. Thanks very much for furthering my education.

Are these changes available in the next pre-release?

Thanks, Bob

[Guckytos]

Yep,

these changes will make me happy

And a Black body addition too, cool. But have you also solved the (infinite) Diameter Problem with the black holes?

And how is then the support for the White Dwarfs spectral classes realised?

Could you please send me some details, so that i can update my guide? The guide can be found at my homepage at the moment, but Adirondack will soon upload it to the motherlode.

English Version: http://home.arcor.de/christianlenz/Down ... _Guide.doc
German Version: http://home.arcor.de/christianlenz/Down ... ndbuch.doc

Regards,

Christian "Guckytos"

[granthutchison]

And how is then the support for the White Dwarfs spectral classes realised?

It's still in a bit of flux, I think ... at present Chris is supporting the more modern spectral classification for white dwarfs, and reading the temperature subclass when it's present to give a better idea of surface temperature. Fine spectral detail (strings of subsidiary letters) aren't at present supported; nor is the old DO, DB, DA, DF, DG, DK, DM sequence. These limitations may or may not change.
Can I send you details once things are finalized?

Bob, I'm sure there will be a prerelease containing all these features ... but I suspect it won't be tomorrow. Chris is doing a seriously major rewrite of star handling, to include orbits and custom overrides for star definitions.

Grant

[Bob Hegwood]

Bob, I'm sure there will be a prerelease containing all these features ... but I suspect it won't be tomorrow. Chris is doing a seriously major rewrite of star handling, to include orbits and custom overrides for star definitions.

No hurry, Grant... Was just curious. "All things come to he who waits."

Thanks again for the education.

Take care, Bob

[chris]

And a Black body addition too, cool. But have you also solved the (infinite) Diameter Problem with the black holes?

Yes, I fixed this with a checkin last night. It's also possible in an .stc file to supply a custom radius that will override the default value computed from the luminosity and temperature.

And how is then the support for the White Dwarfs spectral classes realised?

Could you please send me some details, so that i can update my guide? The guide can be found at my homepage at the moment, but Adirondack will soon upload it to the motherlode.

I've give you a full description soon; there are still a few more details to decide upon.

--Chris

[Guckytos]

Thanks Chris and Grant,

i am looking forward to this new features. Especially if i can redefine the sun radius, that is something that will probably inspire some cool new systems.
But take your time, better do it a bit slower and really good,than the Microsoft-way lol

Regards,

Guckytos