Near-UV observation

VV Cep 2017-2019 Campaign
Moderator for this forum: Ernst Pollmann
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: Near-UV observation

Post by Thilo Bauer »

Paolo,

thanks for the work. Good hint to try this using mu Cep.

Weather is cloudy here and waiting for new adapter to use my focal reducer. I'll give it a try with both stars VV and mu Cep with my Alpy.

Thilo
Phil Bennett
Posts: 10
Joined: Mon Aug 24, 2015 7:40 am

Re: Near-UV observation

Post by Phil Bennett »

The spectrum of VV Cep in the near UV is composite, with the hot component dominating shortward of 4000 A. The hot component is highly variable and is not just the spectrum of the early B-type star; it also includes a substantial (and sometimes dominant) contribution from Balmer continuum emission. (The Balmer continuum flux varies by about a factor of 3).

The Balmer continuum appears only at wavelengths shorter than Balmer limit (in practice, shorter than about 3700 A).

In terms of M star subtraction, alpha Ori provides an excellent spectral match and is of the same spectral type (M2 Iab), although of course, they are not at all close in the sky. But these spectra can be reduced separately (and you only need one good spectrum of alpha Ori for use in subtraction).

Mu Cep is a reasonable spectral match, although probably not as good as alpha Ori.

If anyone can reach down to below 3700 A, spectrophotometric monitoring of the UV spectrum of VV Cep would be very useful. Low spectral resolution is fine for this project.

When I figure out how to do this on this board, I will attach a postscript figure (Fig 3.9 from the Giants of Eclipse book) showing the intrinsic (dereddened) UV spectrum of the hot component, as observed by HST/STIS. In this figure, the mid-eclipse spectrum has been subtracted from the observed UV spectrum at three egress phases (at optical egress: Epoch 9, 7 months later, Epoch 14, and 3 years later near quadrature, Epoch 19). The subtracted spectra have then been corrected for interstellar extinction using a standard (CCM) reddening law.

What remains should be the intrinsic spectrum of the hot component. It is obviously variable. Note that the UV spectrum is still in eclipse at Epoch 9.

Phil
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: Near-UV observation

Post by Thilo Bauer »

Phil,

Thanks for the hint. I'll wait a little while for alpha Ori. Air mass and seeing could be a mess, however.

Looking onto my first raw spectra taken for alpha Lyr with the Alpy, Canon DSLR and Astronomik UV/IR Filter, the setup should go down to 3800 A. But intensities to be expected here are not very bright. From earlier experiments the setup could go beyond that and without filter in the UV, but I don't have a suitable filter to avoid second oder spectra overlapping the red.

Please note, that the image of the spectrum is enhanced by gain of a cubic non-linearity to show the lines in the UV, while the uncalibrated spectrum shown is linear.

Think thats ok to start, anyway.

Thilo
Attachments
alphaLyrSpectrum.jpg
Paolo Berardi
Posts: 578
Joined: Thu Sep 29, 2011 10:51 pm

Re: Near-UV observation

Post by Paolo Berardi »

Thanks a lot for all your comments. So, it's not so simple spectrum. When possibile I'll observe alpha Ori. If I have understood correctly, I should correct VV Cep spectrum for interstellar extinction before subtracting alpha Ori (or MU Cep) one. For VV Cep I read about a reddening of E(B-V)=0.4 (Hack et al. 1989).

I see on Google-book the dereddened spectra of the hot companion at three epochs. During quadrature phase the spectrum (low-res representation) from 3800-4000A shows a lot of emission lines (many of them should be Balmer lines) with a blue continuum. Below 3700A there is a "hump" that should be the Balmer continuum emission.

Concerning lines, from old literature (The VV Cephei Stars, Anne P. Cowley 1969PASP...81..297C) I read about an emission line due to [NiII] at 3993A, very close to a line present in the observed spectrum.

Paolo
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: Near-UV observation

Post by Thilo Bauer »

Paolo,

thank you for your thoughts. This points into similar direction, what I was thinking about.

Much of the work for interstellar reddening has been done by using integral photometry. It doesn't seem to be easy, however, to use the integral photometric values E(B-V) to obtain a correction in the spectral domain.

There is a paper describing the curve of interstellar reddening in the visible wavelength range. The curvature of interstellar reddening in the spectral domain follows a discontinuous linear curvature in the visible range. It is a function of 1/lambda of the wavelength lambda. Two segments with different slope exist for 3500-4600 A and 4600-9000 A (see graphics on page 32 of the work). The correction probably should be solved by the task of spectro-photometry. Unfortunately, the curvature is given as a relationship of magnitudes, not intensities.

Undershill, Anne B.; Walker, G. A. H. 1966. The shape of the interstellar reddening curve. MNRAS, Vol. 131, p.475
http://adsabs.harvard.edu/abs/1966MNRAS.131..475U

Not sure, if there are newer contributions how to manage this using spectroscopy with modern detectors, like CCD. Anyway, the correction would assume a lot about the root cause of the extinction (or reddening), like size of dust grains.

Any advise or recommendation on how to manage this?

Thilo
Firefly
Posts: 51
Joined: Wed Sep 28, 2011 8:03 am
Location: Tianjin

Re: Near-UV observation

Post by Firefly »

Hi all:

I read all your talks here. And great work!!! 保罗. Now please allow me to ask a question:

Since only data shorter than 370nm is much more useful & meaningful, then if only HST can do this job perfectly? Because it is in the universe where there is no atmosphere obstructing UV. For us standing on the earth, Ha or Hb regions are easy to work at?

One diagram here. Hope for your comments and detailed advice

Dongdong
Attachments
pic.PNG
pic.PNG (22.14 KiB) Viewed 7942 times
Paolo Berardi
Posts: 578
Joined: Thu Sep 29, 2011 10:51 pm

Re: Near-UV observation

Post by Paolo Berardi »

Dear 李冬 thank you! I agree, I have almost no signal below 3700A (also severe chromatic aberration at the begin of spectrum).

Waiting a reply from experts to your questions...

See you!
Paolo
Ernst Pollmann
Posts: 461
Joined: Mon Sep 26, 2011 7:16 pm

Re: Near-UV observation

Post by Ernst Pollmann »

Hi Dong Dong,
to describe the different times of contacts, would be fairly comprehensive.
So, that's why I would recommend the paper of Moellenhof & Schaifers:

Spectroscopic Observations of VV Cep (C. Moellenhoff, K. Schaifers, 1981)

which is to find in my website:

http://astrospectroscopy.de/literatur.html

The study of that paper will give you a impression of that, what we have to expect in 2017 to 2019.

Ernst
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: Near-UV observation

Post by Thilo Bauer »

Paolo,
Paolo Berardi wrote:... I have almost no signal below 3700A ....
I agree.

As posted in another thread: Even above wavelengths of 3800 nm luminosity isn't as bright, as one should expect. Given a B-type companion of the suggested mass ratio of both stars of the binary system should be a hot giant star, which should provide more signal in spectral range 3800-5000 A. More than what I've found in my spectra: almost nothing here.

Thilo
Peter Somogyi
Posts: 420
Joined: Sun Jul 13, 2014 8:56 am

Re: Near-UV observation

Post by Peter Somogyi »

After 1 year, fresh result from this weekend - with the same LHires + 2400/mm grating - the only change is a new 12" Newton (the old was a 10"):
pso_vvcep_20160911_007.png
pso_vvcep_20160911_007.png (50.26 KiB) Viewed 7428 times
Emissions have changed somwhat, would be interesting to know why.

Used 2x2 binning, pixel intensity at approx. 2200 ADU / 10 minute. (Typical conversion for a 1x1 bin case to divide the ADU value per 4 above bias which is 300 ADU).
The new observation took 4 x 10 minute (with a Mu Cep 2 x 10 minute in between). Sky was not perfect, but was able to shoot close to zenith.

I do not have any ripple on the flat (in fact, flat patterns not even dependent on any wavelength...seeing the same low-order humps at the same place in red and green as well, just the dusts are much more defined). Hope the flat is valid (hence attaching 7 cep in the zip - it looks to be smooth), made the flat 3 x 1 minute average with the internal lamp.

Only the left 20% of the new spectrum gets less focused here by calibration.

Attaching a zip file with results and a Mu Cep, as I read it might be useful for experiments.

Cheers,
Peter
Attachments
pso_vvcep_upload.zip
(28.77 KiB) Downloaded 367 times
Post Reply