Aperture: This one is easy. The larger the aperture - the fainter you can go. However, remember
that with CCDs modest sized telescopes can do a lot of work. The vast majority of stars in the AAVSO program
are mag 16 or brighter, so purchasing a "light cannon" isn't required. However, we do have some programs
which need large aperture (High Energy Network, I'm looking in your general
direction...).
Note that bigger aperture has the downside of smaller FOV (for same f-ratio).
|
Field of View: CCDs are notorious about having small field of views (although this is changing
as technology progresses). Small fields of view make it hard to get the comp stars in the same field as
the variable. It also can make it frustratingly difficult to find the object. The smaller the f-ratio the
better.
The Mount: This is crucial. The mount's ability to find the object ("goto") and the tracking
will profoundly affect your observing experience. Also consider the setup time. A mount that isn't very accurate,
takes forever to align, and tracks poorly will leave you so frustrated that you may lose interest in the hobby.
If possible, do not skimp on the mount. Equatorials are a must because alt-az mounts cause field rotation during
medium and long exposures, which destroys photometric accuracy.
|
|
The simpler the optical system the better. Stay away from focal reducers and anything else that can
cause vignetting unless you really know what you are doing. Also stay away from color corrected systems and
in general anything that adds a nonuniform
effect to the field or changes the natural color of the incoming starlight. Celestron's Fastar system cannot
be used for photometry. Also watch out for poor baffling because it can create ghost images that look like
stars in your final image.
|
![[!]](/images/caution.gif) Warning Signs
Don't worry too much. Most telescopes are great with CCDs. Small scopes have rich fields of view and large
ones can go as deep as some professional scopes. Find a scope that you will enjoy. It needs to fit your budget
and be easy to setup and use. Remember what Clint Ford said, "The best
scope for you is the one
you use the most."
|
|
No Telescope? No Problem!
It is possible to use a CCD for photometry without a telescope. Simply attach
a 35mm camera lense to the CCD (usually with a T-adapter) and aim at the sky.
You still need to use a proper filter, though. There are many
bright variables that are very red so have scatter in their light curves. These
would be ideal targets for such a system.
|
|
|
Best Advice
The absolutely, positively, best advice you can follow when purchasing a telescope is to talk to
people in your local astronomy club. Spend some time looking through their scopes, ask for
anecdotes and advice, seek out the CCD observers and see what they use. However, remember that most club members won't know
anything about photometry so seek out those with photometric experience.
2.2 CCD Camera
|
The best CCD camera for you depends, like the telescope, a lot on your personal situation. You need to consider the cost,
your observing goals, your telescope, and then of course the reputation of various manufacturers and models.
|
|
Observing Goals
Do you plan on only doing variable star photometry or do you also want to use your camera to also take
pretty pictures, hunt for asteroids, etc? For pretty pictures, binning and antiblooming can be a big help, but
both are not recommended in the world of photometry. Also, resolution is important for pretty pictures but it
isn't in photometry. Here we are going to give you advice regarding photometry only. If you plan to use your
camera for other targets then seek the advice of groups in that area of interest and choose a good "middle-ground"
camera.
What Are You Looking At?
First, get a feel for what observing programs interest you and read up on them. There may be
different features on a CCD that will help in some programs. For example, if you are going to do a lot of
supernovae hunting or GRB work then you will want a camera with a wide field of
view. If you are doing studies of bright LPVs then you will want a camera with a good dynamic range
and large well depth.
Sampling & Pixel Size (Telescope compatibility)
|
One of the most important things you need to consider when purchasing a CCD is its pixel size. For most consumer
(front-illuminated) chips you want to spread the FWHM of the star across 2 pixels of the CCD for optimal photometry according to
"Astronomical Image Processing" by Richard Berry & James Burnell. This
will help you optimize your signal to noise ratio and improve accuracy. To get the best FWHM you will need to determine how much
of
the sky will fall on a single pixel of your CCD camera. This means making a calculation concerning the telescope's aperture,
f-ratio, and the size and number of the CCD chip's pixels.
|
|
Full-Width Half-What? FWHM stands for "Full-Width Half-Maximum" and is basically the size of the star on the chip.
The size is chosen by counting pixels that are filled to 1/2 the dynamic range between the background
and the brightest (fullest) pixel in the star's image. Most software will determine this for you and we'll discuss
it in more detail in the Photometry section.
|
|
|
To determine the amount of sky that will fall on the pixel use this formula:
Sky = (206.265/Y) * X (if focal length is millimeters)
Sky = (8.12/Y) * X (if focal length is inches)
Where X is the pixel size of the CCD chip in microns and Y is the focal length of the telescope in millimeters.
Example: A 10" f/10 telescope (2540mm focal length) and a CCD with the Sony ICX 083 chip (11.6 X 11.2 micron pixels = 11.4 avg)
will give
0.9 arc second per pixel. To see how much sky will fit on the entire CCD we just multiply that by the size of the array (752x582).
In general, 1 arc second per pixel would be too small. However, by binning the
chip you can double, triple, or quadruple the pixel size. So by triple binning we would get a pixel size of almost 3 arc seconds.
Here is a calculator we created to do the calculation for you:
Pixels vs. "Seeing"
For most amateurs living in suburbia seeing conditions limit your telescope's resolution to between 3 and 4 arc seconds,
although it may be smaller for those in great locations or on days of exceptionally good seeing. What you want to do is
come up with a telescope + ccd system that will spread this seeing out over 2-3 pixels. Example: If you get
seeing conditions on average of 3 arc seconds, you will want a telescope that puts 1.5 arc seconds on each pixel so that
the FWHM of the star will cover 2 pixels in your final image. This is called sampling.
Binning
This is also where binning can help. If your conditions are so bad one night that your seeing blurs the star across 4 pixels, you
can do 2x2 sampling to get it back down to 2 pixels. Binning refers to the combining of a group of adjacent
pixels into one
large pixel. 2x2 binning, for example, takes an array of 4 pixels and makes them act as 1. The downside to this for photometry is
that some of the starlight will be landing on the gates dividing the pixels and will be lost. This causes a slight hit in accuracy so
binning should be avoided when possible.
More help:
Antiblooming (ABG)
The "anti-blooming gate" is a feature on some CCDs to
prevent electrons from leaking from one pixel to another.
This happens when a star is bright enough to "saturate" the pixels on the camera, filling up the wells. What is left
over "leaks" onto adjacent pixels. This causes spikes and other artifacts in the image. ABG mitigates this effect by
putting "gates" between the pixels. However, this can destroy linearity of the chip and your photometric accuracy
will go along with it.
It is highly recommended that you purchase a camera without ABG. Many manufacturers sell cameras with ABG removed or
else have documentation on how you can remove it yourself (if you are brave and mechanically inclined). Another advantage to
not having ABG is your CCD sensitivity will be greater.
However, if you end up with ABG in your camera all is not lost. In that case, a good rule of thumb is to limit your pixel saturation to 50% of your well depth.
Most cameras with ABG stay linear in this range. For example, most 16 bit cameras stay linear to 32,000 units or less per pixel.
Your software will be able to tell you the saturation of a pixel and your CCD documentation will tell you
the well depth of your pixels.
2.3 Filters
Different models of CCD chips have different spectral responses. Because of this,
it is vital that one use proper photometric filters when observing variable stars.
|
Take a look at the diagram on the right. The 3 lines represent the spectral
response of 3 popular CCD chips. The highlighted area represents the passband of
a photometric filter.
Notice how each CCD chip has a different spectral response in the region of
the star's light in the filter's passband. If each user imaged the star unfiltered they would all get
wildly different magnitudes!
But don't think that using filters is a bad thing. In fact, this is one of the
strengths of CCDs. When you use a filter you equalize the passbands of various CCDs.
This means that your data can be added to other CCD observers without the scatter
that visual observations have (because visual observations are unfiltered).
|  |
The standard filters most often used for photometric work are:
- Johnson U (U): This is in the near ultraviolet range. Very few consumer level CCDs are sensitive in this range.
- Johnson B (B or Bu): This is the blue range. Cataclysmic variables (CV's) tend to
be bright in B, especially when in outburst.
- Johnson V (V): This is close to what the human eye sees so the V stands for Visual. It is roughly centered in the
green.
- Cousins R (R or Rc): This is red. Mira's are bright in R.
- Cousins I (I, Iz or Ic): This is very near infrared. The human eye cannot see this portion of the spectrum so an Ic
filter may look opaque when held up to the
light. There are many I bands (Is, Iz...) but Ic is the most common.
Photometric filters can usually be purchased from the same vendor that sells you the CCD. The most popular manufacturers
are Optec and
Schuler (via Astrovid). Prices can range
from around $50 to $150 per filter.
When using a system with a low f-ratio telescope you may want to get 48mm filters instead of the standard 1.25" filters to account for
vignetting at the edges of the filter.
Start With V.
To start off, you only need to purchase the V filter. The is the most common filter used and for most
stars in
the AAVSO program it is all you will need. Among the complications you run into with the other filters is that most AAVSO charts only list
V mags (or close approximations) so it can be difficult to find appropriate comp stars for the other filters. Only use other filters when
your observing program specifically requires it.
|
Some
suppliers call their filter sets various combinations of Johnson, Cousins and
Kron. They all have the same V response and the alternate names refer to the
designation of filters other than the V filter in their series. The filter is
different from a Tri-color (RGB) green filter, in that it is more narrow in bandwidth
and centered at a different wavelength. The proper filter response is
sometimes referred to as a "Bessell" response which refers back to a paper by
Mike Bessell, who published curves on filters in the IAU Colloquium 136 on Stellar
Photometry.
|
|
Anecdote: One of the authors spent hours one night trying to
get a flat field using an I filter. Frustrated and dejected, he gave up only to realize a week later that the light
source he was using was flourescent, which doesn't radiate IR!
|
|
|
IR Blocking
When using one of the standard UBVRI filters you don't need to worry about using an IR Blocking filter since they are built into the
standards for those filters.
Unfiltered Observations
Unfiltered CCD observations are not very useful to the AAVSO. In fact, it is almost preferable to do visual work instead of
unfiltered CCD work because at least the human eye (the "Mark I Eyeball") has a somewhat similar spectral response when compared
to unfiltered CCDs. Unfiltered CCD observations can be really off. For example, an unfiltered CCD obs of a red mag 10 star could
show up as mag 3 because of leaks of red light in the CCD. With that said, there are a few observing programs where unfiltered
observations are acceptable, but never assume it unless you have checked with the AAVSO first.
2.4 Computers & Software
Hardware
CCDs do not tend to require powerful computers. However, some new USB and TCP-IP supported CCD systems may require
newer machines with USB ports and properly configured ethernet cards. You should have a good working knowledge of
computers before you purchase a CCD. In fact, the more you know about computer "stuff" the easier CCD work will be.
Especially useful is knowledge about spreadsheets, databases, and some basic scripting skills. You will likely spend more
time manipulating your data on the computer than actually taking the images. The more you know - the more you can automate.
Some of the tasks are tedious and can be made bearable through scripting.
Windows PCs are the dominant platform for CCD systems (surprise). However, some Macintosh systems do exist (SBIG sells a
Macintosh adapter for most of their line of CCDs) and the new Macintosh OS X architecture is based on UNIX which
opens the door for UNIX-based CCD programs. Speaking of which, support for CCDs under Linux has grown measurably
in recent years. Almost any commercial CCD camera will have Linux drivers available (usually third party and free).
However, using Linux or a Macintosh to drive a CCD system should be attempted only by those who are thoroughly
familiar with their operating system.
Regarding monitors, if you have a very large CCD chip, make sure the monitor and video card can support the image resolution
your chip will supply. If it doesn't, you may find yourself scrolling madly or shrinking images.
Software
CCD software can be broken into two large categories: CCD Interfaces and Data Reduction. The first category
consists of software that actually controls the CCD during the observing session. The second category is made up of programs that
specialize in manipulating the data after it has been recorded. Most CCDs will come with a program that controls the camera well
but doesn't do much in the realm of data reduction or telescope control. You can either use this software to control the camera
and another software package to reduce the data or you can spend a little more and find a program that will do both.
Here are some links for some of the most popular software currently available:
(The AAVSO does not endorse any of these links or software packages)
Most professionals (and many advanced amateurs) use IRAF, a free program distributed by National Optical Astronomy Observatory (NOAO).
It is a very powerful and flexible collection of CCD control and
reduction packages. Right now it runs on most UNIX platforms, including Macintosh OS X. However, it can be
a challenge to learn and should only be attempted by those familiar with UNIX. If you have Perl, Python, Tcl, or
other kind of scripting skills then using IRAF will be much easier.
Choosing the software will be a little like choosing the telescope. You need to consider your hardware requirements
(does it support your CCD model and telescope mount? Does it run on your operating system?), cost, and ease of use.
For the latter, look for trial versions and, as with the telescope, ask friends about the software they prefer to use.
If possible, spend an observing session or two with the friend and let them teach you the ropes. Finally, lots of online
support exists for most programs (usually in the form of e-mail discussion groups).
Be Comfortable With Computers
One last thing to remember is that CCD observing is computer intensive. Computer skills are usually the prerequisite lacked
most often by new observers. Please make sure you are comfortable using computers before adding CCDs to your hobby. Many
observers e-mail us daunted and frustrated by the amount of tedious work they must put in to reduce and format their data. CCD
manufacturers don't advertise that! The highway of CCD observing is littered with debris from folks who got frustrated with the
technical requirements and threw their equipment out the window. Be patient and jump in when the time is right.
Here are a sample of some things you will need to know or learn:
- How to troubleshoot a bad connection to your camera. (i.e. is it a broken serial port? cable too long? temperature too
high?)
- The difference between various file formats? (especially RTF, TXT and other ASCII formats)
- Can you manipulate text data?
CCDs generates huge amounts of data. You need to be able to manipulate the data to get it into a format
acceptable by the AAVSO and the scientific community. Recently it has gotten easier as CCD
photometry packages have begun putting in data export features into their software. This formatted data can often
be uploaded to the AAVSO with little or know editing by the observer.
Back | Top | Forward
