Sodium doublet to calculate the resolution of ALPY?

Ask your questions, show your results
James Ley
Posts: 70
Joined: Tue Jan 23, 2018 6:57 pm

Sodium doublet to calculate the resolution of ALPY?

Post by James Ley »

When I calculate the resolution power with ISIS, I typically get values around 520 -560.

I'm sure that I don't completely understand the concept of resolution, but I was asked how, with such low resolution, I could observe small values and small changes in the EW of Balmer lines.

I also don't think that if my resolution was 500, that I should be able to resolve the sodium doublet. Is that correct? Here is the sodium doublet that I observed in SS Vir
13 Vir sodium doublet.png
13 Vir sodium doublet.png (14.73 KiB) Viewed 9007 times
Can I use the sodium doublet to calculate the resolution? Are there any other lines that I could use?
Thanks.
Shelyak Alpy 600 (spectroscope, guiding module and calibration module) | GSO 8" f/4 | Orion Atlas Pro | ZWO ASI178MM-Cool (mono) | ZWO ASI290 Mini (mono)
"think like a photon" -- Steve Shore
Andrew Smith
Posts: 321
Joined: Fri Sep 30, 2011 12:23 pm

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Andrew Smith »

James, one of the advantages of EW measurements is that they are insensitive to resolution. While the line is broader at lower resolution it is also less deep keeping the area constant. Provided the line is not blended with other lines (the risk being higher at lower resolution) then EW is, at least to first order, independent of resolution.

You should use lines from a calibration lamp to calculate resolution not the Na doublet which will is a blend of the two lines.

Regards Andrew
Ken Harrison
Posts: 197
Joined: Thu Oct 25, 2012 5:45 am
Location: St Leonards, Australia
Contact:

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Ken Harrison »

+1 for the calibration lamp. All the processing software(s) available to the amateur include the facility to show the FWHM of a calibration lamp line.....
"Astronomical Spectroscopy - The Final Frontier" - to boldly go where few amateurs have gone before....
"Imaging Sunlight - Using a digital Spectroheliograph" - Springer
http://www.astronomicalspectroscopy.com
Robin Leadbeater
Posts: 1926
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Robin Leadbeater »

The calibration lamp line FWHM can give a good indication of the resolution of the system but is only accurate if the illumination from the lamp approximately mimics the same focal ratio as that of the telescope. If not it can exaggerate or underestimate the effect of aberrations in the spectrograph optics. As an extreme example, a small lamp placed in front of the telescope aperture will give a lower FWHM as it is effectively an infinitely high FR source. (In this situation you can even completely defocus the spectrograph and still get the same FWHM !)

Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Robin Leadbeater
Posts: 1926
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Robin Leadbeater »

James Ley wrote:
I also don't think that if my resolution was 500, that I should be able to resolve the sodium doublet. Is that correct? Here is the sodium doublet that I observed in SS Vir
This is an interesting question. Theoretically you should not be able to separate these 2 lines with the ALPY at a resolving power of 500 (12A resolution at the Na D lines) but there is some sign of a split in your spectrum. We can simulate this by taking generating two synthetic gaussian line profiles with a 12A FWHM using ISIS and summing them at different separations.
resolution_5A.png
resolution_5A.png (8.9 KiB) Viewed 8904 times
resolution_12A.png
resolution_12A.png (8.79 KiB) Viewed 8904 times
At 5A separation there is clearly no split but at 12A separation we just see a split, confirming that the FWHM of the line is a good indication of the resolution. (The separation at which two lines can be resolved) This does not explain though why there is apparently signs of the D lines having been split in your spectrum, which is surprising. Perhaps it is just noise, perhaps the lines are not gaussian ? Can anyone offer any other explanation ?

Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Benjamin Mauclaire
Posts: 258
Joined: Thu Sep 29, 2011 10:14 am

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Benjamin Mauclaire »

Hello James, Robin,

I've made a small Python code to study this question. But I've studied it with an other point of view from Robin:
from which FWHM the doublet of fixed lines is separated ?
We then are able to explore when Rayleigh criterion is checked or not (line observed are separated or not). Resolution power (RP) are computed for each line FWHM at medium wavelength of lines studied.

- First example is the Na D1-D2 line doublet fixed at 5889.950 and 5895.924 A:

Image

Rayleigh criterion appears to be verifed for FWHM<=4 A.
Corresponding RP~1450.
So in James spectrum made with Alpy600 (RP~600), the two apparent absorption components aren't real lines.
Moreover continuum noise around Na lines has same signal that the two absorption components.
Thus, what James saw in his spectrum aren't Na D1 and D2 lines.

- Then, the study of the well known PN [SII] lines fixed at 6716 and 6731 A:

Image

These lines seem easier to resolve with Alpy600. Rayleigh criterion appears to be verifed for FWHM<=9.5 A, ie RP~700.
Of course, spectroscope internal focus on wavelength zone and slit used may impact on RP.

Python code is available here:
http://wsdiscovery.free.fr/forumimgs/rayleigh_find.py

Benji
Spcaudace spectroscopy software: saving you hundred hours of frustration.
James Ley
Posts: 70
Joined: Tue Jan 23, 2018 6:57 pm

Re: Sodium doublet to calculate the resolution of ALPY?

Post by James Ley »

Thanks to all for your answers to my questions.
I do use the calibration lamp. The resulting calibration spectrum was used when I processed the spectrum with ISIS. I observed the star 2 days later and calculated a resolution power of 551.5. Here is the image from that day, with the data points labeled in red. I think that the data is well sampled and the peak between the two minima is not just due to noise.
SS Vir data points.png
SS Vir data points.png (15.68 KiB) Viewed 8786 times
The calibration seems to be off by 0.3 nm. The H-alpha and H-Beta lines (shown in blue) are off by the same amount. This may be my mistake in selecting the H-alpha reference in ISIS. I don't know how to shift the whole spectrum.
SS Vir Balmer.png
SS Vir Balmer.png (15.36 KiB) Viewed 8786 times
My question is that when I use the calibration lamp, the lamp fully illuminates my 23 um slit and I presume this impacts the resolution calculation. However, given my 2.4 um pixels, my 6" f/6 telescope and the (sub)arcsecond seeing that I can get here in Florida, often a star is maybe 1/3 the width of the slit. Could it be that the the calibration lamp, with its wider lines, is underestimating the resolution of my smaller star images? Could the sodium D1 and D2 lines provide an estimate of the resolution of the system? Thanks again.
Shelyak Alpy 600 (spectroscope, guiding module and calibration module) | GSO 8" f/4 | Orion Atlas Pro | ZWO ASI178MM-Cool (mono) | ZWO ASI290 Mini (mono)
"think like a photon" -- Steve Shore
etienne bertrand
Posts: 1040
Joined: Mon Aug 31, 2015 11:26 am

Re: Sodium doublet to calculate the resolution of ALPY?

Post by etienne bertrand »

Hello James,
On your spectr, i think you have double pic absorption on the absorption ray at the right of the NaI, No ?
Possible it's the small pixels of the 178MM amelior the resolution ? but with a slit of ALpy 23µm and resolution of 551 i dont know if it's possible to separat NaI?

When i try the same CMOS178MM on a Lhires with slit 15µm i have the sodium doublet but the résolution if near R~1224
Image
Image
Image

@ Benji, exellent your simulation on Python !
Robin Leadbeater
Posts: 1926
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Robin Leadbeater »

Hi James,
James Ley wrote: My question is that when I use the calibration lamp, the lamp fully illuminates my 23 um slit and I presume this impacts the resolution calculation. However, given my 2.4 um pixels, my 6" f/6 telescope and the (sub)arcsecond seeing that I can get here in Florida, often a star is maybe 1/3 the width of the slit. Could it be that the the calibration lamp, with its wider lines, is underestimating the resolution of my smaller star images? Could the sodium D1 and D2 lines provide an estimate of the resolution of the system? Thanks again.
Yes I think this is the answer. If your guiding is good, you could get a higher resolution than the width of the slit. I now remember someone else seeing this (You are effectively running slitless). The problem is the resolution will be unreliable as it will depend on seeing and guiding. You will also get small shifts in calibration depending on exactly where in the slit the star is located. You should probably consider using a narrower slit which will give you a higher, more stable resolution and a stable calibration (Or a larger aperture scope of course :D ) I am very envious of your sub arcsec seeing !

Cheers
Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Benjamin Mauclaire
Posts: 258
Joined: Thu Sep 29, 2011 10:14 am

Re: Sodium doublet to calculate the resolution of ALPY?

Post by Benjamin Mauclaire »

Hi,

Anyway, some of us have an uncommom seeing.

Bertrand, I'm very suprised about your RP=1224 measured with Lhires3 150 l/mm.
ETCL spreadsheet returns RP=834 for this grating+15 um slit and computed at 5890 A.
You too have a terrific seeing?

These observations lead me to give a try to an other computation.
As spectrographs involved in this thread are low resolution, one can suposed that instrumental function (mainly from the slit) may have a greater impact than with hr spectrographs. Resulting line shape is the convolution of stellar line and instrumental function, here sinc(x), which may be close to a Lorenztian function. To test this assumption, I ploted Na D1-D2 doublet to find the FWHM at which line doublet is observed:

Image

It seems that Na doublet is visible when FWHM<=6.9, ie RP>=853.
This result looks closer to observations but doesn't let us to choose between line shape modelisation or seeing as an origin of the visiblity of Na doublet at such RP, ie spectrographs.

Benji
Spcaudace spectroscopy software: saving you hundred hours of frustration.
Post Reply