Chad,
Generally speaking, yes, dimmer shots require longer exposures, but it is not that simple. With any given telescope and camera there is a maximum exposure time after which the stars in the image start to “bloom” and are overexposed. To all who read this, please excuse the length of the post. I am going to expound in detail because it is something I wish I had known when I was starting at this sport.
I attended the astrophotography symposium at the Texas Star Party they had for a couple of years running but which they no longer offer. They brought in several professional and semi-professional astrophotographers (and paid them from our fees) to take us through the more advanced aspects of the subject. They were in consensus that experience is a good guide, but when in new territory, such as when I did the Wizard (Sh2-142) because it is so relatively dim, that there is no substitute for test shooting. My Celestron 1100 Edge HD with the Hyperstar and SX814 camera is generally limited to about 600 seconds before the brighter stars are oversaturated, so I do a 10 minute test shot in either Luminance or Hydrogen Alpha, whichever is going to be the lead/main channel. I then examine the brightest star in the image to determine if it exceeds the pixel capacity of the camera. Most capture software reads maximum exposure as 65535, even though every given chip has a maximum somewhat lower than that. Exceeding maximum exposure causes the flare and “bloom” to become excessive around he star and can create problems. The next step is to stretch the image to reveal the target in order to check if it is sufficiently exposed to get the target.
One of the things that Craig Stark teaches in his guide to Nebulosity is that the histogram is your friend. Ideally the left side of the chart should peak near the top, but then drop off in a cliff to the left and leave a gap between the left side of the graph window and the filled chart area. On the right side it should gradually decline in a concave slope and again hit zero just before the right edge. That sometimes takes some fine tuning.
A single shot, even using an ideal exposure, will be quite grainy. That is why we go for multiple exposures as the graininess of the exposure declines with the square root of the number of exposures that will be stacked. I commonly use six exposures of 600 seconds each with an f-2 lens/camera combination. I could get more detail and finer grained images with 9 or even 16 subs, but my best shooting skies are at star parties and I only have a limited number of nights there. The ideal would be to have a house in that subdivision at the base of the mountain where the MacDonald observatory is located and be able to spend lots of time with a permanently mounted scope on a single target, but, alas, that is not likely for me. Six exposures gets me a noise improvement of 2.45 (square root of six) while 9 exposures only takes me to 3, but with 50% more time and double the risk of losing frames to glitches. To go from about a 2.5x decrease in graininess (noise) to a 5x reduction would require 25 successful sub-frames, and probably far exceed the detail limits set by the atmosphere unless I was shooting from a high mountaintop on a very dark, very atmospherically stable night. That could be done from some observatories I have visited, but to get three channels of exposure would require about 17 hours of that ideal situation, not something one is likely to get at at star party.
With higher f-numbers, the exposure may go longer. At f-7 or f-8, the exposure can easily go 20 minutes and at f-10, 30 minute exposure are sometimes required. The problem with longer exposures is that the noise (hot and warm pixels) increases dramatically as do the probabilities of a wind gust, plane passing over, or some other glitch increase exponentially. Longer exposures generally require great telescope stability, hyper-cooled sensor chips, and the expectation that as many as half the exposures will be trashed by something. That is why I prefer shooting narrow-band, wide field images using a f-2.8 Canon lens and mid-field images using the f-2 Hyperstar lens set. Unless you are using a monster scope, shooting the smaller, faint galaxies and planetary nebulae will require a very stable platform, high f-numbers, a very dark sky, good seeing, and a lot of time.
As an interesting, but little published point, I can shoot an image using the Hyperstar f-2 at a 560mm focal length and get as much or more detail than using the 0.7 reducer at f-7 and 1,960mm focal length, but with much shorter exposures. The issue is that the “seeing” (atmospheric turbulence) limits the detail I can get, but the image improves with the number of exposures. Thus, I can shoot twice as many exposures in the same time with a shorter focal-length f-2 scope, so the detail is about the same and I lose fewer exposures to glitches. The problem I encounter at f-7 and f-10 is underexposure while the danger at f-2 is overexposure. There is also the counterintuitive effect that the lower the f-number the greater the detail that any shot will gather. Of course, if you didn’t already know, focus becomes more critical as the f-number decreases. Another issue is that with an 11” f-2 scope, the filters have to use a slightly different frequency window to capture a given color or narrow-band light frequency. It does get complicated.
Again, I apologize for the length, but as those who have been down this road have found, there are no simple answers to the issue of clarity and exposure when shooting images that are, in effect, invisible to the naked eye.
Jeff
Jeff,
These are amazing pieces of work. I haven't purchased nebulosity yet but will be happening soon. I appreciate you leaving the "technical card" info on your shots. How do you determine the length of your shots?
my understanding right now is that the dimmer the subject is, the longer the exposure. This is considering broadband frames. Narrow band requires sometimes twice the length of time?