Starter CCD camera

Which of those telescopes you mentioned are you most likely to be using for photometry, and what is the focal length?   Do you know the typical seeing (scintilation) for your area (arcseconds)?  

I'd suggest you read the first few chapters of the AAVSO Photometry Guide.  It explains how to pick a CCD camera that matches your telescope and seeing conditions.

Phil

Here are some suggestions for you.

Match your CCD pixel size with your focal length.  I made the mistake of trying to do photometry while undersampling using a CCD camera with 9-µm pixels with a telescope with a 740mm focal length.  For a target with no other nearby stars, it's fine if I defocus it, but if there is light from adjacent stars bleeding into a target or comparison star, there can be accuracy issues (though less for companion stars so long there is enough of them to average out any bleed over).  I had issues taking measurements of the recent supernova in NGC 4457 because the supernova is  so close to the bright galactic core.  The target and NGC 4457 was barely discernable with the CCD with the 9-µm pixel.  So I switched it out with a CMOS camera with a smaller pixel size (ASI183MM-Pro with 2.4-µm pixels) even though the CMOS was only 12-bits as opposed to most CCDs with 16-bits. 

For monochrome camera using photometric filters

The optimal pixel size will be 1/2 to 1/3 FWHM.  So, if the seeing conditions in your area 4 arc-seconds FWHM, then the optimal pixel resolution would be 1.33 to 2 arc-seconds per pixel.  If seeing conditions are better, you will probably have to go down to 1 arc-second per pixel in resolution.  When it comes to photometry, oversampling is better than undersampling.  ZWO used to sell a cooled monochrome CMOS with a bit depth of 14-bits (ASI178MM-Pro) but it has been discontinued.  However, QHY has a cooled monochrome camera with the same IMX178 chip as the ASI178MM.  The pixel size is 2.4-µm.  So the focal length of the lens/telescope can go all the way down to 250mm so long as seeing conditions are never better than 4 arc-seconds per FWHM.  So going a big longer (~400 mm) would be prudent.  However, if you have a telescope with a longer focal length, then the individual pixels don't have to be so small.

For cameras with bayer filter (one-shot astronomy CCD/CMOS and DSLRs)

You will have to defocus unless your images are heavily oversampled because the FWHM must spread across multiple pixels in the green channel.  The Bayer filters will be arranged over the pixels in a checkboard pattern only with three colors with green making 2 out of every 4 filters.

From what I read from the DSLR manual, the telescope or lens has to defocus until the light spreads across a diameter between 8 to 30 pixels.  From what I can tell from my limited experience, the DSLR+lens combo is good for bright variable stars where the target brightness is so substantially higher than the adjacent stars but extra care must be taken with dimmer target stars as their brightness is not that much different from the other stars.  Defocusing is not necessary for a monochrome astronomy camera paired with the right telescope so long as the stars FWHM is at least two pixels across.

"I think about buying a "starter" CCD + Johnson V,B filters."

I agree, this is what you should do.  You just need to pick the camera that works best with your focal length and seeing conditions. 

Knowledgeable people at the nearby observatory can tell you the average seeing for that site.  Pick the telescope you will use, then choose a monochrome CCD camera, with regulated cooling and a filter wheel, which gives the right image scale and field of view for the telescope you will use.

Photometry is different from astro-photography for pretty pictures.  CMOS technology for photometry is changing rapidly, and there is no single reliable book or document which can tell you how best to do photometry using the CMOS camera which you may buy.  For CCD photometry there are many books, courses, documents that can help you along the way.

Phil

Another bit of advice.  Go to astrobin.com and do a search for any potential purchases with the "deep sky object" filter.  I might be wrong here, but any telescope/mount/camera combination that can take images of deep sky objects will be abled to be usable for photometry.  If anything, it should be easier to do photometry than astrophotography so it should be able to do the job with one minor exception.  As spp mentioned, use a cooled camera.  Yes, AAVSO will accept data from an uncooled DSLR but it's better to use a cooled camera.

But if you really want to use a DSLR, they do make special coolers for DSLR cameras if you want to go that route.

Hi Mathias--

You'll be looking for a monochrome CCD camera, a filter wheel, and V nd B filters. We;re in a transition now as makers phase out making CCDs and switch over to CMOS sensors. At this time, a CCD sensor is more conservative and predictable option.

As the earlier writers have suggested, a key part of choosing a sutiable camera is the focal length of the telescope, so that the size of the star image covers a reasonable number of pixels and the field of view is larger neough to include at least one comp stars. The middle of the range would be 1,5 arcsec/pixel and a minimum of 15 arcminutes field of view. 

Now examine the range of products from a vendor such at Atik (https://www.atik-cameras.com/specification-tables/). These range from very basic to quite expensive. You'll also want to do the same for other vendor's product lines.

Let's look at two telescope options: a 200 mm f/5 Newtonian (f =1000 mm) and a 200 mm f/10 SCT (f = 2000 mm). Scanning the list for candidates, the 414EX and the 460EX are CCD cameras, cooled, with 16-bit readout, full well depth >15,000 e-, and a M42x0.75mm thread on front. 

In the 414EX you have a Sony ICX825 CCD that's 9.0 x 6.7mm with 6.45 um pixels. 
On the Newtonian, the field is 30 arcmin wide and 1.2 arcsec/pixel
In the SCT, the field of 15 arcmin wide and 0.6 arcsec/pixel.
The Newtonian and the 414EX would be a pretty good match.
In the 460EX, you have a Sony ICX694 CCD that's 14.8 x 9.8 mm with 4,54 um pixels.
On the Newtonian, the field is 0.84 x 0.56 degrees and 0.9 arcsec/pixel.
On the SCT, the field is 0.42 x 0.28 degrees and 0.45 arcsec/pixel.

You can carry out this exercise for other products in this and other vendors.

I suggest you get the filter wheeel from the same compnay you buy the camera from, and that you get one that is computer controlled. Before long, you'll want to operate the telescope, camera, and filter wheel without the need to touch the telescope. Not only does this make operation efficient, but if the telescope you're using has a shaky mounting, you won't bump it every time you switch the filter.

Rest assured that you can well away from the otimum configuration and still do excellent photometry.

I hope this helps. If you have more questions, please ask....

--Richard
 

Thanks to all for your suggestions.

I will check in detail the options at the observatory once we start observing again, hopefully already next Saturday.

I do have somne experience with DSLR photometry now and given the limitations I am quite satisfied about the results.

On CMOS vs. CCD: You mean over time CCD will "dissapear" or limited to high-end applications? 

But preference is always a black/white, right? As I also do a bit spectroscopy at the observatory, we jave a camera for that purpose. Maybe I start with this one as trial. 

I am sure I will come back with more questions later.

Clear skies,

Matthias

Hi Mathias--

Yes, you want monochrome. The Bayer array in color cameras absorbs light and changs the spectral response of the sensor. With a monochrome camera, the photometric filters control the spectral respsonse. 

Since you can still buy a 16-bit monochrome camera with a CCD sensor, that is a sensible and conservative option. Several years from now, as the CMOS technology improves, you may be able to buy a CMOS camera with the desired characteristics.

And since you already have a monochrome (specs?) camera for your spectroscopy, why not try it for photometry?

--Richard

Hi Matthias,

All good advice above...I've been doing photometry for only about a year now and so consider myself in the starter group. My camera is an Atik 314L+ mono, 16-bit. The pixels are 6.45 microns square. The telescope I use is a 102mm f/7 refractor resulting in a 1.86 arcsec/pixel. The chip is 6.7mm X 8.97mm giving a field of view of .72 X .54 degrees. This is wide enough to usually find a number of comp stars around the LPVs I target.

This is a discontinued camera by Atik, but can be found on the used market reasonably priced. For a starter you might consider this or a similar camera. Good Luck!!

Jamey Jenkins

Here's a question I would like to pose to the group: I once asked if the read noise nullified the advantage of having a 16-bit system.  The Atik 314L+ has a read noise of 7 e- and a full well capacity of 26,000 e-.  So the dynamic range would be 71.4 dB or 11.8 stops.  Going by the math and dynamic range, there should be no benefit to having a bit depth greater than 12.  Now, someone did respond to my question saying that an extra 2 bits would be needed for the noise, but they didn't go into detail why.  The QHY178 cooled camera is a 14-bit CMOS that is sold either as a one-shot or monochrome.  At the gain setting of 0, the gain is 2.05 e- and has a full well capacity of 15,000 e-. or 12.8 stops and would be a good match to its bit depth unless the 2 extra bits for noise is included.  If anyone could go into detail on matching the dynamic range of a camera with its bit depth, it would be greatly appreciated.