Aug 242011
 
http://micro.magnet.fsu.edu/primer/photomicrography/images/exposure/characteristicfeatures.jpg

Astropotamus has been absent this summer from the Time Machine.  This is due, in part, to being busy, but also because it’s somewhat frustrating to take image of the heavens with the camera that I put at the end of the Time Machine.  Luckily, I just bought a new one.

Cameras are deviously simple complex devices.  Or is that fiendishly complex simple devices?  At any rate, it used to be, back when film was all we had, that cameras were a light-tight box with a hole in them.  The hole was different sized based on the lens you put in front of it, and the hole was opened up for different amounts of time based on the film you put in it, and how much light you wanted to have reach the film.  That all changes with digital.

Digital sensors are linear – twice as much light hitting them creates twice as much signal.  Film is not.  But then, neither are your eyes.  Your eyes see in multiplication, not addition.  A different way of saying that is to say that you see in a logarithmic fashion instead of a linear one.  If I asked you to dim/brighten a set of lights so that their brightnesses were distributed equally from “low” to “high” you would probably set them to 1x, 2x, 4x, 8x, and 16x.  Or maybe 1x, 3x, 9x, 27x, etc.  This is because your eyes see in a manner where levels of 1, 2, 4, and 8 all look like they are equally more bright than the dimmer value.  This is logarithmic.

Film is the same way.  If you google “film response curves” you’ll see talk about toes, shoulders, and gammas.  Basically the toe is the beginning, non-logarithmic part, the gamma is the middle, mostly logarithmic part, and the shoulder is the top non-logarithmic part.  It’s that toe and shoulder that causes problems for film.  If you expose film over time, you move from toe to the gamma portion to the shoulder portion.  In the toe and shoulder portions, the response of the film doesn’t behave very well.  Exposing for long exposures that are required in astroimaging cause the film to max out the overall exposure – great for preventing overexposed family portraits, not so great for grabbing pictures of deep sky objects.

http://micro.magnet.fsu.edu/primer/photomicrography/images/exposure/characteristicfeatures.jpg

Toe, Gamma, Shoulder

Enter digital light sensors (CCDs or CMOS sensors).

Digital sensors are linear.  If you arranged digital sensors so that you could adjust their sensitivity and lined them up so that you had equal levels of recorded brightness between them, you’d set them at 1x, 2x, 3x, 4x, and 5x.  So unlike your eye and film, the linear sensors don’t have complicated response curves and simply put:  the more light you give them, the more exposed the resulting image will be.

The tradeoff comes in noise.  The longer you leave an electric circuit running (especially a light sensitive CMOS sensor) the more random noise you’ll get.  This is evident as random colored dots in a supposedly “dark” part of a digital picture enlarged to 100% or more.  Again, it doesn’t matter for family portraits, but it sucks when the noise pixels are just as big as your star pixels while doing astroimaging.

So the balance is long exposures versus lots of noise.  Some cameras are better at low noise than others.  Some also have remote control software, intervalometers (taking a picture every X number of seconds), mirror lockup (which minimizes camera shake when the mirror is moved to take the picture), AC adapters (batteries run out quickly with long exposures), and “live view” to see what you’re focusing on so you can make minute adjustments (sometimes with computer controlled focusing).  Couple this all with the idea that it should be EASY to take pictures, and you have Astropotamus’s new astroimaging camera, the Canon Rebel XSi.

It’s not the latest Canon Rebel digital SLR, nor is it the biggest resolution.  But it does have very low noise, it’s easily modified to remove some filters that block out light you can’t see but the sensors can (astroimaging wants this light, so we tend to rip the filters out), and it’s fully controlled by computer.  This is the best part – Astropotamus can hook the laptop up to the camera, hook the camera up to the Time Machine, plug all three things into AC power, and then go inside with warm fuzzy slippers on and remote control the camera and Time Machine all by connecting to the laptop over wireless.

So soon, I hope to have my first round of newest astroimages posted.  Until then, you’ll have to wait for me to finish processing my last round of images taken with my previous camera.