Telephotography of restricted areas

The basics....

The nature of a restricted area is you are not welcome there unless authorized. [If you are authorized, why are you reading this page?] For the restricted Groom Lake range, penalties start out with arrest and it goes down hill from there. To preserve your freedom,  you need some technology to see more than what is discernible by the naked eye. Should anyone inquire what you are doing with a high power telescope at the border of a secure area, tell them you are bird watching.

In most cases, your telescope is going to focus at infinity.  There are situations where you might want to observe something nearby (say 50ft), but in great detail, perhaps to read markings on it. Thus as you design your telephotography set up, you may want to consider close focusing. I suggest being capable of doing telephotography at a distance of fifteen to twenty feet. This will allow the setup to be tested in a garage, hallway, etc. I've tested hardware by photographing money, which is fine geometry target.

Before we examine using a telescope, consider stock camera lenses. Camera lenses, if sanely priced, tend to be no longer in focal length than 400mm to 500mm. [The focal length is the distance from the front of the lens to the focal plane, i.e. film or image sensor.]  The focal length of a good lens,  a prime (versus zoom), can be increased with a teleconverter. These are found in 1.4x and 2x magnifications. Note that if you use the 1.4x teleconverter, the exposure time needs to be increased by a factor of two, AKA one stop. Using the 2x teleconverter requires increasing the exposure time by a factor of 4, AKA two stops. Modern cameras do this time compensation automatically.  Regarding the teleconverter itself, you get what you pay for. Nikon and the later model Canon teleconverters are very good. Third party teleconverters are poor quality, which of course is why they are found on ebay.

Most people are familiar with viewing distant objects with binoculars. Binoculars are not specified in terms of focal length, but rather magnification. Binocular magnification ranges from about 10x to 18x.  You can investigate magnification of a camera by considering the lowly 50mm lens. On a 35mm camera, the 50mm lens is considered "normal." That is, the view is similar to what the eye sees. Double the focal length to 100mm, and the magnification is considered to be 2x. For 800mm, the magnification is 16x. 

If cameras max out at 800mm, why not just use a second teleconveter and get a second power of two? Well, it turns out as you increase magnification with teleconverters, the performance of the lens becomes more significant. Camera lenses tend to be complicated (many elements) to minimize geometric distortion in the image. Telescopes have a different goal in mind. They are designed to view stars, which are pinpoints of light. Even more demanding is to view two stars near each other, and resolve that they really are two distinct stars. The telescope should present the star as bright and small as possible. Telescopes tend to be simpler designs, generally two or three elements, while a prime lens such as the Canon EOS 400mm/f5.6 L series has 7 elements. Zoom lenses are even more complicated.  

You may have noticed I still haven't said how much magnification is possible. This isn't an exact science. The rule of thumb is you can achieve about 30x per inch of clear aperture, though you will find this number ranges from 20 to 60. (If you want a more rigorous approach to this problem, investigate "Dawes Limit.")  For a refractor telescope, the aperture is the diameter of the lens at the front of the telescope. Take a Takahashi Sky-90 for example. The aperture is 90mm, or about 3.54inches, which using 30x an inch would imply a magnification of 106.  The term "clear" comes into play with a reflector telescope. A reflector telescope has an obstruction in the middle of the scope, which reduces the aperture. Take a Celestron C-5, which has an aperture of  127mm and an obstruction of 47mm. Without showing all the math, subtract the area of the obstruction from the aperture of the scope, then compute the equivalent aperture required to achieve that area.  You should get 118mm.  

So on the surface, it looks like the refractor and reflector are on the same footing. However, the obstruction in the reflector telescope reduces the image contrast, and contrast is very significant for daylight terrestrial photography, especially over long distances. Examine the differences between nighttime astrophotography and daytime terrestrial telephotography. In the case of astronomy, you are viewing through about 2.5 miles of atmosphere if you are at sea level, and obviously less if you have some altitude. Now consider photographing Groom Lake from Tikaboo Peak. The distance is about 26 miles, so you are looking through 10 times as much atmosphere. Worse yet, this atmosphere is over ground, which has more air turbulence due to the sun heating the ground, which in turn heats the low level air. The atmosphere turns black to dark gray and white to light gray. It is actually worse, since it also adds a blue cast to the image, which will be discussed later. The image derived through 26 miles of atmosphere is what a photographer calls "thin", i.e. it lacks contrast.  Thus the refractor is the better choice since the image it produces is of higher contrast.

The atmosphere contributes to other problems, the worse being a time varying geometric distortion, AKA heat wiggles. In cinema, the boiling image conveys to the audience that the scene is hot. Fine, but our task isn't art, but rather espionage. While the camera shutter is open, it is "integrating" the light presented to the film or sensor. If the image moves, the integration process creates a blur. The frequency of these distortions tend to be low, generally 10 cycles per second or less. Certainly, you need the shutter speed higher than 1/10th of a second not to capture more than one thermal distortion. From experience, 1/60th or so is about as low as you should go for this "poor seeing" condition.  [Poor seeing is an actual astronomical term, not a medical diagnosis.]  For reasons (begin hand waving) beyond the scope of this document, smaller aperture telescopes perform better in poor seeing conditions.]

Given a lower limit on the shutter speed, the ISO rating of the film or camera, and the aperture of the lens, we can compute how much magnification can be achieved.  This is better than pulling a mystery number between 20 and 60 from one's arse. Just one more rule needs to be introduced, that being the "sunny 16." To use the sunny 16 rule, set the shutter speed to the ISO of the film or sensor. The focal ratio of the lens should be 16. No problem for a camera lens, since you can just dial the focal ratio (f-stop). For a telescope, you need to determine the focal ratio. Let's take the Sky-90 example. The  focal length is 500mm and the aperture is 90mm, so the ratio is 500/90=5.6. Since the focal ratio is fixed, we need to either alter the ISO (not a good idea) or the shutter speed to achieve a situation similar to the sunny 16 rule. We could do some math, but for this example, let's count the F stops to determine how much light we need to lose. Counting in full stops, we have f5.6 f8 f11 f16. Thus we have three F stops, which means 1/8 the light, so for ISO 100, set the shutter speed to 1/800th. [One stop is a factor of two, two stops is a factor of 4, etc.) 

Now that calculation is interesting, but what we really want to find out is how much we can change the effective focal length to get more magnification. It was already determined there is an extra factor of 8 of light. If we doubled the focal length, the light would be reduced by a factor of 4. If we scaled the focal length by the square root of 8, then the light would be reduced by a factor of 8. The square root of 8 is 2.83, so the effective focal length should be increased to 500mm*2.83 =  1451. If we drop the shutter speed to 1/50th, then the effective focal length can be increased by a factor of two, or 2830.  So we can insert a 4x to 6x magnification in the optical path. For a smaller telescope, say 3 inches aperture, the magnification should be closer to 4 than 6. None of this is cast in stone, but rather a guideline. 

To summarize the set up thus far, we have a refractor telescope, somewhere between 3 to 4 inches of aperture, a device to increase the focal length by a factor of 4, and the camera. Needless to say, a tripod will be required. As says, "But wait, there's more!" Not any camera will do. To minimize shaking the set up, the camera must have a "mirror lock-up" feature.  This function is a bit different depending on the camera used, but the common feature is that the mirror is up long before the shutter is opened. Also required is a remote shutter release. This allows you to trigger the camera without touching it. Touching the camera would blur the image, even on a tripod. On older cameras, the shutter release is a flexible cable. On more modern cameras, it is an electronic device. Finally, some way to magnify the image during focusing is required. On older Nikon cameras, you can remove the pentaprism and put a magnifier right on the focusing screen. For other cameras, a magnifier is placed at the back of the viewfinder. Focusing a high magnification telescopic system with yet another magnifier on the viewfinder can be frustrating. This is because the mere act of touching the telescope moves the image, making it hard to focus on the scene. [There are aftermarket motors you can put on a telescope to get around this problem. but I'm not sure you want to carry one up Tikaboo.]

"But wait, there's more!"  Pointing the telescope at such high magnification requires a gear head on the tripod. These range in price from expensive to freakin' expensive. The gear head let you pan the camera to create a series of images that can be stitched together to create a panorama.

Not really essential but nice to have is an image rotator in the telescope. The rotator lens you rotate the telescope to level the image. In theory, you can do this in a number of way, but the rotator does this smoothly. As far as I know, only Takahashi has this option. This is because with Takahashi gear, it is possible to have all the components in the optical path screw together. This is much more stable than using components employing set screws. However, with everything firmly screwed together, the only way to rotate the image would be to rotate the telescope tube, hence the need for the image rotator.

We're almost done spending your hard earned money. As I mentioned earlier, viewing through all the atmosphere puts a cast (color bias) on the image. The cast tends toward blue. To some degree, you can rebalance the color in a photo editor (Photoshop, the GIMP, etc.) but it is best to provide some filtering before the image is recorded.

Whew!