Required Best Practices Article:
Public Viewing with a Low Light Video Camera
I'm a regular at ASEM's weekly Friday Night Open House (FNOH) public viewing sessions at Broemmelsiek Park. During a long string of poor weather during the winter and spring of 2011, I discovered and experimented with capturing deep sky objects (DSOs) with a Samsung SDC-435 low light security camera. Since then I've adapted three different low light security cameras to pretty much all the optics I can put on a mount including a C14, C8, f/6.1 Apogee refractor, 12" Dob, and a bag full of different camera lenses. At one time or another I've used all those combinations to work on various Astronomical League Observing Programs including the Outreach Program. I used a C14 and camera to produce images that I used to measure the position angles and separations for the Double Star Program. The combination gave me a 2x4 arcminute field of view. I've grown to love video and do think that GOTO and tracking are necessary parts of the puzzle. There are many difficulties finding, centering, and following DSOs manually via a video display. Consider that if you are trying to manually find and follow a DSO, you might have to wait for four to eight seconds for your frame to be updated. A lot of things can happen is that time, none of which will make your job easier and you temper less. After using video display of DSOs during the FNOHs, I've come to the conclusion that providing live video can be an fun, interesting and important part of public viewing session. Video could, but should notreplace traditional eye piece viewing. It can be very positive in a large group setting. I see video as a field experience while the audience is at the telescope and tripod out in the dark. I don't see an audience in a climate controlled theater setting looking at video streamed from outside in the dark.
Done well I see that video has the following benefits:
everyone in a party can see the same object at the same time
the party members can freely discuss the object on the screen...what they see, what they like, what impresses them
I can be sure everyone can see the image because I can point to it
young children don’t fumble at the eyepiece and have to hear “Do you see it? Do you see it? Do you see it?” If I were a young kid I’d say “See what?”
there is less dead time waiting compared to standing in a long line to view through an eyepiece
more objects of similar types can be seen for a better understanding when groups are large
fainter detail can be captured/seen because of extended integration/exposure times
objects can be seen around bright moon conditions
color is more obvious vs. eyepiece views
physically challenged individuals have a better opportunity to participate
it can be done with less expensive optics and smaller mounts
While capturing images with a video camera won't compare to those with a CCD or DSLR camera, it can give an eyepiece in a larger and more expensive scope and a run for the money. Some nights I feel I can see similar images with a video camera on a camera lens as I can see with my 12" SkyWatcher Dob.
I'll say that if you've never seen live video of DSOs you should probably take a visit to the Nightskynetwork at http://www.nightskiesnetwork.com/. On a dark night you may see up to twenty people broadcasting video of various DSOs at the same time. You'll either love it or hate it. If you love it, you'll probably take the plunge and get into video. If you have a way to connect to the Internet at your observing site, you'll probably broadcast at some point.
Public Outreach Possibilities
Going through various forums on Cloudy Nights will get all sorts of opinions about what kind of camera you'll need to get to do video. You'll see the latest discovery of the best camera ever made. Take it all in with a grain of salt. You don't need the biggest or most expensive gun on the block to get started since you will not be concerned with the dim wonders of the Herschel 400 or Flat Galaxy Programs. If those are your goal, then go for the Cadillac, the Mallincam and stop reading. Based on the three different low light security cameras I’ve used over the past two years, get a camera that is sensitive enough, won't break your budget, and you’ll be able to do the DSO biggies for less than $100. You will probably need to remove the red protective filter over the sensor to improve Hydrogen alpha red. Don't worry, it is easy. Doing video for public outreach means you will be concentrating on the brightest objects in the various deep sky classes. At first you'll limit galaxies to M31, M33, M51, M81 & M82, and the Sculptor galaxy. You'll pick the biggest globular clusters like M13, M15, and M20. It will be easy to see bright planetary nebula like M28 & M57. Diffuse nebula like M42, M8, and M17 will shine with color. You'll see open clusters like M6, M7, M41, or M45 under full moon conditions. Double stars will be easy too, assuming you have enough focal length to bring them in. Using a video camera with a 1/3" chip is like using a 6mm eyepiece. After you've had some practice with the camera controls take a shot at some of the fainter objects. A point to consider is that the cameras are not designed for what we are using them for and that the directions are few and not very helpful.You can do video by itself, but if you set up a side-by-side arrangement with a telescope and a piggybacked lens with video camera, you can give the crowd a preview of what they are going to see. I've found people are always a bit happier knowing what they need to find in the eyepiece. Video cameras usually have the ability to flip and mirror the image, so make sure your video view matches the eyepiece view. Be warned, it may be brighter and bigger on the screen than in the eyepiece so some people will have a bit of let down with the regular eyepiece view. The other option is to have them look through the eyepiece first and then show them what they saw. Put a piece of rubylith over your screen to preserve night vision unless you want to convince people how much better video is. The red filter works great unless the object is red like M8. Have the audience talk to each other about what they are seeing and encourage questions. There will be a whole lot more on topic talking than with the big long line method of the past.
What can you expect?
It seems to me one of the more difficult objects is the Horse Head Nebula in Orion. On the left is one of my early attempts with an SDC-435 taken in early April of 2011 from the Bortle orange sky of Broemmelsiek Park. The bright star is Alnitak and the Horse Head can be seen as the dark patch towards the bottom left corner. The image on the computer screen was much more obvious than my screen capture displayed here. People, including experienced observers, ran to see it! Both the Horse Head and M8 above suffer from a number of issues including poor focus, hot pixels, exposure, and color balance. My technique has improved, but most of the time I simply display the image to an LCD TV rather than capture the image. Displaying the image to a small 400 line TV means you don't really need the best focus and resolution.The good news is that if you can see something as difficult as the Horse Head, many deep sky objects will be possible from your location. I'd venture to say that all the objects in the Urban Sky Observing Program are possible with video. One of the jobs on my to-do list is to see if I can do that from my Bortle red sky neighborhood.
During the past year Samsung has changed the name of the SDC-435 to SCB-2000. You can still find online sources for the SDC-435 but check for the new name too. I'd guess that most other low light security cameras would also work. Look for the lowest lux values you can find. There isn't any standardized test done to compare various makes and you might have to fiddle with settings to make it work. The last camera I got was a .0002 lux bullet camera. The bullet lets me push it into a lens for some interesting combinations and makes spacing a focal reducer easier.
Video works best with more light... the more, the better. In one post, the author suggested at least four times that you needed to be at f/3 or faster for DSOs. I don't know if it is absolutely necessary, since I've used my 90mm refractor @ f/6.1, but my C8 @ f/11 won't cut it without a .5x focal reducer. Power users say the SDC-435 works best at focal ratios less than f/4 and I agree.
Longer focal lengths and larger focal ratios should give acceptable results with the moon and larger planets, but if your optics are slower than f/6 you will need to add a focal reducer to get good results with DSOs. If splitting double stars is your goal, you'll be happy, with long focal lengths. As I've already said, I used a C14 to capture video for the all the stars in Double Star Program and routinely had a field of view smaller than 2x4 arc minutes.
Since the chip is so small, GOTOs must be spot on or you may have trouble locating the object on the sensor. Depending on which scope/lens/camera combination I use, I have resorted to swapping the camera with a long 40mm focal length parafocalized eyepiece, using a flip mirror, or using a zoom lens zoomed out to center the object in the field after slews. Each of these options has drawbacks: back focus problems with optics, inability to use filters or focal reducers, or spacing of optics. Nothing is every easy!
In my early days I used a Vicon 16-160mm f/1.8 lens that is a massive piece of glass I got on eBay for a song. As a telescope users you might think that with a focal length of 160mm would be much too short to use for any kind of astronomical work, after all a traditional 8" SCT has a focal length of over 2000mm. In fact, nothing could be further from the truth. The sensor in the SDC-435 is a 6mm chip. The small size of the chip means that you don't need a very long focal length lens to get a reasonably small field of view. The 160mm focal length of the Vicon gives a FOV of 77.5x103.4 arc minutes, which worked pretty well with the larger Messier objects like M42 or M31. Using a 2x teleconverter on the Vicon lens let me cut the FOV to half that, but added coma and make the objects dimmer. It seems to me that one certain disaster for a newbie is sticking one of these small chip cameras on a big SCT, Newt, or Dob. Too much magnification, gives too small a field of view, frequently at a too low a light level, and very poor results. Before you jump in with both feet and a credit card, use a program like CCD Calculator to give you an idea of what can be seen with your optics and a video camera. See:http://www.newastro.com/book_new/camera_app.phpIf you wish to enter the 435 in CCD Calculator it will help to know that the pixel size on the 435 is 6.35x7.4 and the array size is 768x494. The data for most 1/3” cameras will be about the same. You will soon discover that it doesn't take much focal length to fill the 435's sensor with a DSO.
Even a moderate telephoto lens from a DSLR or 35mm camera lens works well. A 200mm focal length lens will capture a 76x57 arc minute field with the SDC-435's small chip. In contrast, using a Celestron-8 with a Meade f/3.3 focal reducer captures a field of view of 25x19 arc minutes. I have a flip mirror that makes finding and centering the object easier with longer focal lengths. If you wish to use camera lenses rather than a telescope look for C-mount, CS-mount, or CCTV lenses and you will probably be able to avoid adapters although you will need to play with the lens flange to sensor spacing for focus. Camera stores like Adarama have adapters that will allow you to mount most popular brands of camera lenses to security cameras.
Adapting a zoom lens to the camera is an adventure. Spacing between the lens flange and camera sensor is very critical if focus is to be held throughout the entire zoom range. The spacing is adjusted by a combination of adding C-mount spacers between the camera and lens AND/OR screwing the camera in/out from the lens. Adjust the spacing to focus on a distant object at the shortest focal length with the lens racked in/set on infinity. Slowly zoom to longer focal lengths and check the focus. If the focus does not hold, return to the widest focal length and readjust. This is DIFFICULT but rewarding because afterwards you have a digital zoom finder. It makes finding and centering the object on the chip much easier. This is an important concern when a line of people is forming during a public viewing session!As I said at the beginning, providing live video can be an important part of public viewing sessions. Video will not replace eye piece viewing, but it can be very positive in a large group setting. A beginning video setup is much less money than a good eyepiece, although I'd guess that it won't last nearly as long. Video is still in it's infancy and as time passes technology will improve even more. In a few years we should see cheaper, bigger, more sensitive chips that can record even fainter objects and detail. Someday I suspect that new telescopes will come with video cameras and not eyepieces. Someday we'll be able to look at a magnitude 14 galaxy and see some structure. There are interesting threads at Cloudy Nights predicting the future of telescopes, viewing, video and such over the next 30 years. This is one of the predicted trends.
If you are in the St. Louis area and interested in live video demo maybe we can meet at Broemmelsiek sometime. Drop me a line and I'll see you there.
Video FAQ:
Can the SDC-435 or other low light security camera be used for astrophotography?
Yes, but as you can see, a CCD or DSLR will produce much better results. The actual pixel size of the image is very low, typically 720x480 or less. Using 2x or 3x drizzle processing can help increase resolution. It is possible to take flats and darks. However AGC may do strange things to the darks. Remember, the purpose is to augment visual and not astrophotography.
Grabbing individual frames is a very easy way to record an image for later review or proof of object.
You will notice lens aberrations if you use a focal reducer and the spacing isn't correct.
Are there any idiosyncrasies?
When using long integration times (exposure lengths) it takes a while for the screen to update. When slewing from one object to another bright area in the first image take a while to disappear from the screen, sometimes obscuring the second object for a while. It may take a minute or so for the second object to be fully revealed. You could erase the old image faster by going into the menu and reducing the exposure time to about x16 and then stepping back up to the longer integration time, but it takes almost as long as letting it sit and the "dissolve" to new image is kind of fun to watch.Sometime changing the shutter speed too quickly "locks" the camera. Change the shutter speed to a much faster setting and slowly change the speed to slower usually corrects the problem.
Hot pixels will appear as the camera gets warmer or older. Stacking software may think that they are stars and produce strange results.
Amp glow will slowly build as the night progresses. You probably won't notice it on the computer screen, but if you record video, you'll probably have to deal with it. Taking darks should help. Amp glow produces the lighter upper left corner in the picture to the right.
Unless your mount is much more solid than mine, pressing the controls on the camera will cause the image to shake for some time. There is a hardware hack, but I’m not brave enough to solder on the camera’s board.
I've mentioned that the 6mm chip is pretty small. Unless your GOTOs are spot on, you may have a bit of difficulty locating the DSO in the camera's field. Swapping the camera for a parafocalized longer focal length eyepiece or using a flip mirror can really help locate and center the object. This is one of the main frustrations of doing video.
While on the subject of small chip, if your mount suffers drift from poor alignment or periodic error, you'll notice the object wandering in the field.
Looking at the computer screen all night does kill night vision pretty well even when you cover it with rubylith. Curb your excitement when you see a particularly difficult object like the Horse Head Nebula, the visual observers will want to see, but then complain about you destroying their night vision.
In an unfair test I set up by a friend with an 8” dob. We both displayed the same objects and the crowd bounced back and forth between the two views while we did our astronomer talk in the background. Things were bigger and brighter with my video setup AND I killed the night vision for the observers. They loved my images and couldn’t see his.
What are all the parts?
A low light security camera.
Power supply--AC adapter to provide 12v 1000ma or DC plug and battery
BNC to RCA adapter and RCA cable (RadioShack) to connect the camera to the display device
OTA/Lenses (Any will work if they are fast enough)
Adapter to connect the optics to the camera... types might include: (depends on camera and your chosen optics--zoom lenses can be a problem)
1.25" (or 2") to C-mount lens adapter to use OTAs with focuser
1.25" to M12 board camera adapter to use OTAs with focuser
camera lens to C-mount adapter to use camera lens
home made camera to optics adapter
T-mount to C-mount adapter
Alt/Az or EQ mount with drive
TV/monitor or computer for video display and capture
Frame grabber to get video to the computer
Frame/video capture software
IR/UV filter if you remove the stock filter
Focal Reducer: f/6.3, 0.5x, or Meade f/3.3 if your optics are too slow
Light pollution filter
Flip mirror
various home made adapters to hold filters & focal reducers
The bottom line is that it will probably run you about $200 to get into video via this route.