This document will attempt to outline what is needed in order to create a system which can be used for Astrophotography, which both optimizes the electronic and optical systems, while keeping the price to something an individual can afford. I will attempt to cover the why's and how's of each choice for each device.

In order to successfully capture an image, using a CCD as the active device you must match both the size of the CCD array itself and the size of the individual pixels themselves to the image being projected on the CCD chip by the optical system. If you don't you may find that your image is either over or under sampled. This can cause you to perform extra work in actually taking the image, degrade the image, and waste light, something you dare not do in Astrophotography... First we must understand how a CCD Camera works in order to understand why a particular chip set is selected for use with a particular telescope.

A CCD camera is not just a single function device. It consists of a number of parts, each with a particular job to do, and each functioning together as a unit in order to bring a complete picture to the users computer. The cameras job is to take the image from the optical system and convert it into electronic data.

Figure 1
The Optical and CCD device

The above diagram represents both the Optical System and the CCD Device itself in the diagram below.

When you expose an image onto the CCD array the image is captured by the optics of the telescope and projected onto the CCD chip. The CCD Chip then converts the light into electrons, and stores them in small buckets called "pixels". Each pixel has a set location on the CCD chip, and can keep track of how much light fell on it during the exposure process. Look over figure 2 below. Each location can be considered a pixel of a small five by seven CCD chip.

The optical system is your telescope, or whatever piece of glass or plastic you decide to place in front of the CCD chip. The idea is for the image of what you are pointing your optical system at to be projected onto the CCD chip surface, much like a slide projector projects images onto a screen. Once the image is projected onto the CCD Chip, the chip will read the amount of light falling on each pixel and store it at the site in what is called a "well". Lets look at the chip closer...

The CCD Device itself

KAF-0400 Chip

The CCD chip takes the projected image from the Optical System and breaks the image up into discrete elements called pixels. Each pixel is a small engine that counts photons, (light), and turns the photon count into electrons, ann then deposits the electron into the well at the site. The brighter the light the more electrons are stored at any site. Look below, you will see in Figure 4. You will see the site address, (1,1), and a number after it, (1,1,5000), that last number is the electron count at the end of the exposure.

Figure 4
CCD Locations and electron count values

This portion of the camera takes the analog values, (number of electrons), from each photosite and converts them to digital values. If for instance the photo site located at chip coordinates 3,5 has an electron count of 3000, that would be converted to 10111000, or some other value representing the value of electrons in the CCD well. The data are then read out of the chip in a string of numbers representing the location, (1,5), and the value, (11011) into the computer.

This part takes the string of numbers the ADC (Analog to Digital Converter above) has read out from the CCD and converts them to Parallel or USB and then presents them in a way a computer can read.

ADUs

In order to understand ADU you have to know a bit about binary math and how a CCD works... A CCD camera needs to convert total charge per pixel, (number of electrons per pixel captured during the process of exposing the CCD chip to light), to some binary, (ones and zeros, like 1010001000100101110), number. Computers only read binary, so the A/D converter, (Analog to Digital Converter), converts the analog charges read off the CCD Chip to digital values of ones and zeros for use by the computer. Some cameras use 12 bit conversion, and others use 16 bit conversion. For a 12 bit camera the total number of possible states the A/D converter can represent is 4096, for 14 bit converter the possible number of states are 16384, or 2 raised to the 14 power. For a 16 bit camera it is 65536 states, or 2 raised to the 16 power. Lets look at a 12 bit camera. It has 4096 possible output states it can provide to the computer. Now a 16 bit camera, it has 65536 possible output states available to it. No matter how many states you have, 12 bit, 10 bit, 16 bit, or 128 bits, you MUST fit them into a Grey scale. 0 is black, and you move towards white as the number increases. White is ALWAYS the highest number available to you as a result of the A/D conversion. So for a 12 bit camera 4096 is white, for a 16 bit camera 65536 is white, 0 is always black. This means if you have a camera that has a resolution of 12 bits you have a Grey scale that runs from 0 (Black), to 4096, (white), so you can represent 4096 possible steps between black and white. If your camera is 16 bit, (65536 possible states), again zero is black, and this time 65536 is white. You can see that you can represent the same Grey scale in 65536 steps, this gives you more data to work with while Image processing. In essence more steps between black and white. The ADU is the number of steps between black and white that the camera can separate from the actual data being collected as presented to the computer. Each ADU can contain some number of electrons depending on the camera construction. Lets tale my ST7, the full well depth, (total number of electrons you can capture in one single photosite), of 100,000 electrons, or 100,000 possible output states, one state for each electron. Think of black as no electrons and 100,000 electrons as white. This is more than my 16 bit A/D converter can handle so it scales the total number of steps into 65535 steps, or 1.53, (100,000/65535=1.53), electrons per ADU, or 1.53 electrons for one output state change. So if I have 2 electrons in a photo site will get 3.06 ADU, this will read out as 3 ADU. Why 3 and not 3.06 like the math indicates... Rounding error!... Lets say I have 20 electrons stored in a single photosite. I do the math and I get 13.07 ADU's. I read them out using a 16 bit Analog to DIGITAL converter, I get a value of 13 ADU's. Why 13 and not 13.07 ADU because this is a digital unit representing an analog value, so things round to make it be an integer, and not a real number. Now back to the .07 that was dropped. Where did it go? It was lost in conversion from analog to digital, remember we have only 65536 possible states, and we are crushing 100,000 possible states to fit this. Back to your original question: Why would you want to know all of this? It's really cool to understand how all this stuff works, it helps you understand how best to use your camera, it allows you to do some really nice tests on your camera, and most important it lets you do actual science, because you have taken the first step in understanding exactly what the values your CCD camera puts out mean. Read the section on testing your camera in AIP4WIN, it is absolutely fascinating. You also might want to look at http://www.ccd.com/ccdu.html for more info on how CCD's work. They have a very well put together little course on CCD's there.

The computer is used both as a capture device, and as a digital darkroom. The computer takes the data stream from the Readout portion of the camera and saves it in some format. At this point the user has an image stored within his/her computer. From there the user can use Image Enhancment software to manipulate the image in various ways.

Now that you have a basic understand of how the CCD Camera functions, lets look more closely at how it works, and how it relates to the optics. We will next look into "sampling".