I guess this is mostly for Scott, but others may have the same question: I really enjoyed your presentations, and am particularly grateful that you recorded them! Aside from having that time on that night available, it is really handy to replay items to help understand the details.
In these presentations, you make the point (well) that the slit orientation should be perpendicular to the equatorial axis. I have been trying to get my brain wrapped around this question on implementation, and just can’t: Do you know if either that Alt-Az or GEM mount actually hold that orientation? I have searched, and can’t seem to find a great answer. My seeing nights have been very limited lately so have not had a chance to try to check it. I know you use a rotator to set it, presumably whenever it has drifted off enough to need to be adjusted, but… I ask particularly because I am aspiring for a robotic configuration. I have seen others posting on working autonomous operation, but this particular topic never seems to be mentioned.
Thoughts?
Thank you!
-Mike
1 Like
I intend to get an electronic rotator. As an interim step two bubble levels have been mounted. Refer to page 12 of this document for more info about the slit orientation - https://www.aavso.org/sites/default/files/publications_files/AAVSO_Guide_To_Getting_Started_in_Spectroscopy.pdf
Thank you for the reply! I did see that in the presentation when I went through the Q and A again, and the document discusses initial setup, but not if it needs adjustment or if there is a way to calculate this angle. I am interested in operating this remotely, with an electronic rotator to set the angle and perhaps even track to keep it at the optimum position, if the latter is even worth bothering with.
So, this breaks down into two questions: Is there a way to calculate that angle (or is it obvious in some way I am missing), and does it change as the target moves, especially for longer/multiple exposures.
I know the GEM mounts (when properly setup) keep the image frame fixed as far as the camera image is concerned without rotation, but I don’t think that translates to a fixed angle w.r.t. the horizon that the slit needs. The mention of a spreadsheet in the presentation and picking the “middle” value also makes me think not. I begin to think an alt/az mount would inherently keep this vertical (if you don’t de-rotate). Perhaps a re-rotate for the GEM?
I may be overthinking this, since there are multiple remote spectrometers out there and I don’t see it mentioned as a big issue.
Again, thank you for the response! -Mike
I have a remote setup also, in the back yard. Slit orientation automation is my ultimate goal.
Mike,
Sorry for the delayed response. To minimize the loss of flux at the blue and red ends of a spectrum obtained with a broad spectrum instrument (e.g. Alpy 600), you need to maintain the slit oriented perpendicular to the horizon. This is easy to do with an alt-az mount as you set the orientation once and you’re done. For an equatorial mount, the slit would ideally need to be rotated continuously throughout an exposure as the angle between the slit direction and the horizon perpendicular direction is changing. A reasonable compromise it to set the slit in the proper orientation for the middle of the exposure and leave it in that orientation for the entire exposure. Minimizes the loss but does not eliminate it.
For a robotic operation you would need a motorized rotator in which software would calculate the parallactic angle and set the slit orientation to that. Although a motorized rotator would allow a continuous rotation of the slit throughout the exposure, it would probably mess up your guider calibration, so again I would set the slit to the proper orientation for the mid exposure time and leave it alone for the exposure.
None of this applies so much to high resolution narrow band instruments or those using fiber optic feeds.
There is a way to calculate the parallactic angle needed to determine the optimal slit orientation. Its based on simple spherical trigonometry with just a few inputs and you can see it in the slit loss calculator tool I developed, located in the Spectroscopy Section Tools folder:
Look in the Data tab and in cell C12 is the calculation for parallactic angle. Let me know if there are any permissions problems in accessing this tool.
Final note: Slit rotation is ideal but doesn’t kill you if you don’t. The main effect is a reduction of flux at the blue and red ends of the spectrum. This affects the shape of the continuum and the signal to noise at these ends. You can correct the contiuum by application of an atmosphere/instrument response curve (i.e. relative flux calibration) using a reference star. You will still have some S/N loss at the blue and red ends, but it may not be a concern if it is still above some desired threshold.
Scott,
Thank you for the info!
I have been digging into it lately, and a few things have popped up:
First, PHD2 can look at your rotator and compensate for any rotation changes as it guides, which helps. I worried about that myself until I saw that.
I had come up with a theory that reversing the de-rotation algorithm/formula used in alt/az setups during astrophotography might be a way to compensate in real time, but now I’ll look at the reference you sent. I also had gotten a USB inclinometer a while back (leveldevelopments.com), good to .03 degrees accuracy according to the specs, and have been looking into how that might fit. I like your comments about setting the angle at the midpoint. From what I THINK I have been able to tell from at least one paper, the error over short exposures (they defined as 10 minutes or so) is pretty negligible. Having an inclinometer to level or set the midpoint angle between exposures could be a compromise on the complexity involved, which is usually safer (KISS principle). After a quick look at the doc and spreadsheets you posted, they will help a lot with getting a feel for the sensitivity.
FWIW, I am trying to create a plugin for NINA to drive the inclinometer and perhaps support scripting. I may aspire to an ASCOM driver. I have a long way to go, however, especially since the library they provide for the inclinometer has serious issues.
Thanks again!
-Mike
Good to know about PHD2 compensation for active rotation. Additional information on slit rotation in general is found in my unpublished paper on the topic.
Yes, shorter total integration time will have a small effect on the result. How small? You can use my slit rotation tool to answer that question. Follow the instructions within the tool and try out various scenarios.
