PDA

View Full Version here: : TZ Mensae


sjastro
24-03-2017, 04:22 PM
TZ Mensae is an Algol type spectroscopic binary with an orbital period of 8.569 days.

The components are of spectral classes A1III and B9V.
The ideal scenario would have been to image the star when each component was in transit but I had no information on this.
Instead the two set of spectral images were taken five days apart.
I would have preferred to have taken the images at half the orbital period but the weather didn't cooperate.

As I am still a beginner at this, I am rapidly learning of the limitations of low resolution spectroscopy.
Since eclipsing binaries are in the line of sight of the observer, the conditions for Doppler shift are at their most favourable.
Even though the attachment shows a "shift" between the spectra, the magnitude of the shift is only around 1 nm which probably reflects a calibration error than any Doppler effect.
At a dispersion of 1.1 nm/pixel, I suppose a minimum shift of 5 nm at 500 nm based on optics and seeing conditions is required to be confident that the shift is real.

The difference in the spectra is the more dominant hydrogen absorption lines
in the 23/03 image perhaps indicating the A1 star was near transit.
What is puzzling however is the Planck curves for blackbody temperature looks remarkably the same for both spectra.
I'd be interested in hearing comments on this apparent discrepancy.

Steven

Merlin66
24-03-2017, 05:29 PM
Steven,
Since we communicated I've found this reference:
http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1987A%26A...175.. .60A&db_key=AST&high=3f2ea4260a27320

The info on the period should help determine the phase.....

sjastro
25-03-2017, 01:42 PM
Thanks for the paper Ken.

A back of the envelope calculation using the ESO phase data reveals the spectra gathered on 18th March is about 2 days out when the transit of the primary star occurred.

A worksheet illustrating the transit dates for the primary and secondary star for the next few months will provide a window of opportunity for taking spectra.

Taking spectra on the primary and secondary transit dates might take care of of the Planck curve dilemma.

Steven

robin_astro
27-03-2017, 04:37 AM
Hi Steve,

A1 and B9 class stars are very close in temperature. Add the fact that the primary eclipse is only partial and the change in the continuum shape with temperature during eclipse will be very small.

From Ken's reference the range of RV is +-100km/s so the maximum shift from quadrature to eclipse (if total) would be 100km/s or 0.2nm at 600nm. Because the eclipse is partial though the shift in the line centre will be less than that. (There is a light curve in the paper so it would be possible to calculate the net contribution from each star during eclipse but I have not calculated it) This would be very tough to see with the Star Analyser due to the low resolution and difficulty in getting a high precision calibration. It is straightforward though with stable slit spectrographs to measure line centres to better than 1/10 of the resolution so the shift might be measurable with a resolution of say 0.5nm

Cheers
Robin

robin_astro
27-03-2017, 06:37 AM
Scrub that. If the two stars are the same luminosity, there will be no shift in the line centre of the blended lines (which will always be the case at Star Analyser resolution) at any phase, independent of what happens at eclipse. Shifts with phase would only be seen due to differences in luminosity between the two stars and would be less than +-100km/s. The best times to see this would be by comparing spectra taken at the two quadrature positions. I think you will still need something with higher resolution than the Star Analyser to see anything though.

Robin