Affiliation
American Association of Variable Star Observers (AAVSO)
Wed, 03/12/2014 - 17:35
I have recently received my first CCD camera. (Orion G3 Monochrome, 1/2" sensor, 752 x 582 pixel array)
I came up with a pixel size of 9.5 microns in ISIS using the Alpy 600 with the 3mm hole, darks, bias, no flats.
I used my small Newt, 130mm obj. at F/5 and came up with a resolution of 105 in ISIS.
Does this sound normal?
Also, as there are still glaciers in my backyard here in Minnesota, USA, -0 degrees C., is it necessary to use CCD cooling?
I will try to attach an image of Sirius.
Thank you for any advice.
John
Hi John,
Your camera has 8.6um wide pixels so the dispersion should be roughly 4.5A/pixel though it will vary to some degree with wavelength.
The resolution of the ALPY in slitless mode (which using the large hole effectively gives) will depend on the star image size but this should be nice and small with your scope. This means you should be able to get around R=500-600 (~10A) provided the spectrum is in focus and you do not have significant drift during the exposure. I am not sure how ISIS estimates the resolution in slitless mode so the number you get may not be reliable. If you can post your spectrum this will give a better idea if what you are getting is ok. (You should be able to see a nice string of sharp Balmer lines getting closer and closer towards the blue end)
An uncooled camera will be fine for bright targets like this but cooling will be needed for the faintest objects (My nova Cep spectrum was a series of 7x 10min exposures for example using an ATIK 314L camera running at -10C)
Cheers
Robin
Robin, thanks for the info, that helps.
This is the link to a .1 second exposure of Sirius.
http://users.cloudnet.com/jfb/image_spectra.html
John
Hi John,
That looks good. I did a quick measure on the Balmer lines and they have a FWHM of about 20A which is about what the width is in any case for Sirius so your resolution (R) is certainly much higher than 100 and probably the expected ~500-600. The blue end is slightly truncated at around 3900A (due to the size of your CCD?). With the ATK314 I can get down to the limit of the CCD's sensitivity at ~3600A. for comparison I have attached an example of the A2v reference star HD114330 I used for nova Cep which has a resolution of 10A.
The limitation with the ALPY in slitless mode with the large hole is that it is difficult to get a good wavelength calibration on unknown stars as there is no fixed reference. If you run using the slit, this is in a fixed position. it is difficult to get the star on the slit though unless you have the guiding module. With bright objects one way is to defocus to give a larger star which is easier to find and keep on the slit. You still get good resolution because of the slit but you do lose a lot of light that way. The mirror slit guider makes life much easier and is a must for slit spectroscopy of faint objects.
Cheers
Robin
One other thing I did notice though is that the focus deteriorates significantly at the red end (See how the spectrum widens and compare the sharpness of the H alpha and telluric lines in your spectrum with mine) Since you are using a reflector this is not going to be chromatic aberration so I suspect scintillation due to the poor seeing at low elevation and very short exposure. (If you take a series of short exposures and compare them you will probably see the spectrum changing dramatically from frame to frame.) For bright targets like this I would recommend averaging a number of exposures and summing them to reduce the effect. To maintain the resolution you can also align the images like planetary imagers do.
Cheers
Robin
I'm not sure that I agree with you, Robin. John's spectrum is 10x0.1sec, so would be spaced over a fair amount of time so that the multiple spectra would average out most scintillation effects. In addition, scintillation is inversely related to wavelength: it is stronger at blue wavelengths.
My guess is that the CCD or some optical element is tilted with respect to the optical axis, or John does have some optical element, like a coma corrector, in the light path.
Arne
Hi Arne,
I see that John has changed the image originally posted at that address. (In the original post it was just the spectrum image, stated to be a single exposure) , John can you confirm if the image of the spectrum shown there now in your poster is the sum of 10 exposures or a single exposure?
Another possibility is that the camera focus is not set correctly leading to collimation error in the spectrograph. This can easily happen with collimated slitless setups and is not immediately obvious but could most easily be tested by checking the daylight sky spectrum using the slit mode.
Cheers,
Robin
Thanks Robin and Arne,
The exposure is a single, .1 second of the raw image.
The profile is from 10 x .1 seconds.
Sirius was indeed scintillating GREATLY that night. Seeing was not good.
I also used my lowest quality scope with 1.25" focuser, as the weather was unfriendly.
John
Hi John,
Can you confirm that you get a sharp spectrum of the daylight sky with the slit in position rather than the 3mm aperture? The potential problem when operating slitless is that you can unwittingly compensate for an out of focus camera by using the telescope focus. The result is you get a sort of in focus spectrum but the beam through the grism is not parallel and you can end up with the sort of aberrations that you get for example in the simple Star Analyser setup.
Cheers
Robin
Thanks Ken,
I tried the 25 micron slit against sunlight and it is indeed out of focus.
I have nice weather to straighten it out.
John
Thanks Robin,
I was just reading in a book by Ken Harrison and his name stuck.
John
That's OK, John! ;-)
Hi Ken,
It's a good book!
I'm reading it for the second time.
John
Robin,
I got Sirius again last night (that doesn't sound good)
through some clouds but I think it turned out much sharper.
Stacked several images. I can see Hyd. Alpha without straining my eyes.
http://users.cloudnet.com/jfb/spectra_images.html
Thanks again,
John