Affiliation
American Association of Variable Star Observers (AAVSO)
Wed, 04/20/2022 - 15:13
I just got an LHires ES0002 spectrometer.
I have it mounted on my c11 with an asi174 guide cam and a starlight express 814 ccd image cam.
I am using Maxim DL to guide and image with.
I have two challenges art the moment.
I cannot plate solve the guide cam image due to the low resolution image.
When I take a ten minute image of a star on the slit, all I get is a dim white line and no spectra.
Can anyone advise me as to how to get past this point and actually capture some spectra with this instrument?
Thank you.
Befuddled
I am long time…
Hi Jeff,
I am long time LHIRES owner (back to 2006, one of the original kit versions) used with a C11
>When I take a ten minute image of a star on the slit, all I get is a dim white line and no spectra.
What grating are you using, what wavelength region are you looking at and what star are you trying to measure ?
A horizontal line ? That sounds like it is the spectrum. At maximum resolution with the 2400l/mm grating the wavelength coverage is small and unless your target is bright, 10 minutes is actually quite short. (A high resolution spectrum with the LHIRES is approximately 20,000x (11 magnitudes) fainter compared with an exposure of the same star in an image for photometry for example). What lines you see though will depend on the star of course. eg hot stars generally only have faint lines in the spectrum other than the Hydrogen Balmer lines so there can be regions in the spectrum where you might see very little.
First step is to check everything is set up ok
Do you get a well focussed image of the lines from the calibration lamp and can you see what wavelength range you are at from the lines?
Do you get a clear spectrum of the sunlight if you point it at the daylight sky ?
Is the image of the slit in good focus in the guide camera when you point at the daylight sky?
Once you have a good sharp lamp spectrum, solar spectrum, slit image, if you then focus the telescope on a bright star to bring it to focus in the guider you should then get a good spectrum with the star on the slit (a narrow horizontal line, rotating the camera to make it horizontal) .
(The guider image should be sharp enough to plate solve on but the usual problem is it is a rather small field of view so there are often not enough stars. I just use old fashioned star hopping, resyncing the goto as I go)
Cheers
Robin
What grating are you using, what wavelength region are you looking at and what star are you trying to measure ?
I have only the 2400/mm grating that came with it. I had the wavelength set to Ha emission band at 6562. I was hoping it was Arcturus.
Do you get a well focused image of the lines from the calibration lamp and can you see what wavelength range you are at from the lines?
yes, and yes
Do you get a clear spectrum of the sunlight if you point it at the daylight sky ?
Yes
Is the image of the slit in good focus in the guide camera when you point at the daylight sky?
Yes
I started all over today and got everything focused and set up again. It looks like rain but maybe I will get a short chance to try again.
Thank you for your help
Jeff
OK, lots of lines in Arcturus and nice and bright so you definitely should be seeing something. Here is that region with a LHIRES 2400 in 142 sec (by John Foster)
https://britastro.org/specdb/data_graph.php?obs_id=7048
Robin
Thank you for the link.
I am looking forward to my next opportunity to test out the spectrometer.
I am convinced that I did not have the correct star lined up in the slit but only time will tell.
Here is a link to one of my ten minute subs of what I thought was Arcturus.
https://drive.google.com/file/d/1At2fsFf710LgU0t1l6T3euhxdJrL0s_K/view?usp=sharing
I suspect you are right. This look more like mag 10 to me. When you find it, Arcturus should be blindingly bright in the guider even in a fraction of a second guide camera exposure. I tried extracting the spectrum. Although very noisy, H alpha is there but I would say it looks too broad for Arcturus, more like a main sequence star
http://www.threehillsobservatory.co.uk/astro/temp/not_arcturus_lhires2400.png
Cheers
Robin
I captured 11 ten minute subs, darks, flats, bias and neon cal files both pre and post.
Can you tell me if this data would have been usable if I had captured Arcturus?
https://drive.google.com/file/d/1-6TogolbSHGyPlhm0OpcQF99ZMdqZ30p/view?usp=sharing
Thank you for all of your assistance, it has been very helpful.
Jeff
You've got everything you need. If you let me have the pixel size of your camera I can run it through my ISIS pipeline.
Some comments on the data.
Although very faint, your spectrum looks good (horizontal and telescope well focused)
When you do find Arcturus, 600sec exposures may saturate.
At tip to decide on exposure and number of subs:- If it is a target I am not familiar with I sometimes take a quick single short exposure to check for focus and decide on the exposure without saturation. I then quickly produce a rough spectrum from it using Visual Spec which can subtract the background and produce a spectrum in a couple of clicks. I then check the electron counts per bin in the spectrum (counts*camera gain) The SNR will be ~sqrt no of counts so a SNR of 100 say needs a total of 10000 electron counts per bin. I can then decide on the sub exposure length and number of exposures needed
Probably overkill but I normally take more subs (~20) for the calibration frames (using cloudy nights for the darks and bias, 1200sec darks, the longest exposure I use, scaled to other exposures by ISIS)
Flats taken on the night at the wavelength used, normally when pointed in the target direction. There are a couple of strange light patches in your flat that I do not see in mine. (Reflections or a light leak? The case of the LHIRES is not very light tight. I have mine in a black bag)
Neons for each target always when pointed at the target pre and post (and even during) a set of target exposures as you have done because of the thermal and mechanical instability in all scope mounted spectrographs but particularly evident and annoying in the LHIRES. This has caused a ~2 pixel shift in your neons between start and end, not atypical for the LHIRES unfortunately but worthwhile making sure everything is tight. (I assume this is a modern unit with electronic control of the lamps not and old one with the mechanical lever?) The movement if not dealt with also degrades the resolution. There are techniques for dealing with this but it is still a significant issue. (Spectrographs should be better designed and constructed to avoid this as far as possible eg like the much more stable ALPY 600). Some of your neons are over exposed. Make sure the lines you are going to use for calibration are not saturated. (Not applicable here but if you have to use weak lines along with strong lines, making a stack of well exposed neons improves the signal on the weak ones without overexposing the strong ones)
I hope this is useful, you will get good results once you are on target
Cheers
Robin
Last night I was able to gather spectra from Arcturus.
I tried several exposures before I settled on 600 seconds again.
Unfortunately I got this outstanding explanation this morning after my night of data collection.
Here is a link to my data from last night.
https://drive.google.com/file/d/1lSEfM-WQeKbHzzf7vRwY_0F3Y8dzaWND/view?usp=sharing
I will study your response and see what I can make of it. It is very good advice, thank you for taking the time to look at the data.
Thank you so much for the help, It is a good feeling to make some headway.
Pixels are 3.69 x 3.69 microns.
Thanks again,
Jeff
I'll have a look at those too. From a quick look that certainly looks more likely to be Arcturus, though I am still a bit surprised about the weakness of the signal. The star focus looks rather soft this time though so perhaps there was a lot of light lost overspilling the slit. (What slit width are you running ? I use the 35um slit with the C11 which matches my typical 3 arcsec seeing)
I noticed a couple of other things:-
It looks like from the position of the neon lines, the camera was upside down in the first run (longer wavelengths on the left) correct in this run
There is a significant loss of focus in one of the neon lines compared to the other two (the one on the right in the latest run) and the lines in the star spectrum itself are distorted with curves and slants depending on position eg in 600_sec_v1.fit . This may point to a possible miscollimation of the spectrograph optics. With a Littrow design the distance of the camera is critical to ensure the beam is collimated at the grating when the spectrum is in focus and different cameras need different spacers to get the correct position. Do you have the correct spacing for your camera ? According to the original drawings the sensor should be 55 mm from the back plate
http://astrosurf.com/thizy/lhires3/d_Bague-T%20Montage%20CCD-3%20-%20Feuille.jpg
from
http://astrosurf.com/thizy/lhires3/index-en.html
There is also a page on Christian Buil's website about this somewhere. I will see if I can dig it out
Cheers
Robin
>There is also a page on Christian Buil's website about this somewhere. I will see if I can dig it out
Found it ! (It was on the ARAS website)
http://astrosurf.com/aras/lhires_eval/optique.htm
Cheers
Robin
Hi Jeff,
Here is your spectrum (reduced using ISIS) compared with James Foster's from the BAA database
http://www.threehillsobservatory.co.uk/astro/temp/Arcturus_Jeff.png
The continuum shape is a bit different as yours is raw without correction for instrument response or atmospheric extinction. You can either correct it using a measurement of a standard star as reference or just rectify the spectrum, normalising the continuum to 1 everywhere
Cheers
Robin
Robin:
Could you please provide more detail about how you create flats for spectroscopy. In the comment above, you state: "Flats taken on the night at the wavelength used, normally when pointed in the target direction". I really do not understand this. Why pointed in the direction of the target? I understand why to have the correct wavelength range, but why 'taken on the night"? How do you illuminate the scope at night for a flat? Do you do it at twilight? Does it matter if there are clouds? Is it better to do it when there is an overcast sky? Please assume that I know nothing about the procedure for taking flats for spectroscopy, and provide a short and simple set of steps. Thanks so much for being willing to help those of us who are less experienced.
Rick
ps. I have an EdgeHD8", with a LowSpec with several gratings and an ASI183MM-Pro. I have a panel that I can place over my dew shield opening that can accommodate a tungsten bulb for flats or a CFL for calibration. I also have neon lamps in the plate to use when I am interested only in the Ha region.
Hello hrickdiz, I am new to this also but I do have a couple of suggestions for you.
first off, Robin was explaining to me how to take flats with a LHires spectrometer. I just set a switch to the correct position while I have the rig set up to gather spectra. It is just s switch position, the same with taking a calibration frame.
My first suggestion is to read through Christian Buil website for help in calibrating your spectra. I know I saw somewhere on his site how to take a flat.
http://www.astrosurf.com/buil/index.html
Second, try and take a look at Tom Fields work at his site https://www.rspec-astro.com/ . Tomn has a lot of great information there some of which might be helpful to you.
I know when I take flats for my deep shy imaging, I need to get the histogram to nearly 50% and use a white t-shirt to help make the light source flat across the frame.
I hope this helps a little, I think Robin should be able to point you in the right direction.
You might want to start your own thread in case other resources have stopped reading this thread.
Jeff