Digiscoping with Catadioptric Scopes

The good and bad points of using optics with mirrors.


By Jay Turberville


June 23, 2004

Introduction

The best modern birding scopes are wonderful devices. They are waterproof, lightweight, relatively compact, ergonomically designed and provide beautiful images with wide views when the right eyepieces are used. These scopes use rather exotic and expensive special glass in their objectives to get exceptional color correction and high contrast. This class of scope is "the" class to look at when considering digiscoping. They have the high level of optical quality that is necessary for good digiscoped images. If your wallet can tolerate the hit, this is the category of scope to shop for. But for many people, there is just no reasonable way to fit one of these gems into the budget. And others may have the resources but they may want to try out the activity using something less costly before spending the larger sums. Perhaps with the exception of purchasing a used scope, the best low cost alternative is a catadioptric scope. This article explores the use of catadioptric scopes (often called "cats") for digiscoping and will hopefully give some insight into what you can expect from them.

TANSTAAFL (There Ain't No Such Thing As A Free Lunch)

You've probably heard that before. Its a favorite among economists. Purchasing a relatively inexpensive cat (there are expensive cats as well) means accepting some tradeoffs.

TANSTAAFL!

You can get digiscoping image quality from a catadioptric scope that rivals that from a top end refractor. And you can do it for less than what a top birding scope zoom eyepiece will set you back. That sounds great, and in many ways it is. Just remember TANSTAAFL. The catadioptric scope probably won't be as rugged, water resistant or as compact and light as one of the best refractors. At normal birding magnifications, most cats will have narrower fields of view and they often have focusing mechanisms that are not as easy to use as those found on many birding scopes.

Well, that's the short list of negatives, so here's a short list of positives. Even a low dollar cat will deliver pretty darned good views virtually free of chromatic aberrations. You can get a 90mm or 100mm aperture scope at a moderate price and these apertures along with the cat's typically longer focal length will make this class of scope a pretty good choice for high magnification views. They usually provide exceptional close focusing as well.

While there is no such thing as a "no compromise" design, the best birding scopes compromise far less than the inexpensive cats when it comes to birding. Understanding those tradeoffs will help you make the decision about whether a cat makes sense for your or not. So lets dive in.


Cat Guts

Catadioptric scopes perform well at a low cost because they are relatively simple to make and do not require exotic materials. They are fundamentally resistant to chromatic aberrations since their objective is reflective rather than refractive. Designers of refractors have to deal with the fact that when light is refracted (bent) through a lens, its component wavelengths are always bent at slightly different angles. You can see this effect in the rainbow pattern created by a prism. A mirror objective does not have this problem at all and the refractive elements in a cat only make very gentle bends in the light path to correct for other optical issues introduced by their mirrors. So a cat's refractive elements introduce very little of this prism effect.

But mirrors reflect light and this creates a logistical problem of getting the light from the scope entrance to the eyepiece. Cats accomplish this by reflecting the light from the main objective mirror located at the back of the scope towards the front of the scope where it hits a second smaller mirror which then reflects the light rearward and out through a hole in the main objective mirror where it finally reaches the eyepiece. The second small mirror and the hole in the mirror (and some interal baffles as well) create an obstruction. You can immediately recognize a cat because there will be either a mirror spot or a black mirror mount right smack in the middle of the scope's front corrector plate. Cats effectively have a donut shaped objective while refractors have a complete unobstructed disc.

A cat's obstruction limits its performance. Its sad but true. The mirror is a great thing, but the obstruction spoils the party a little bit. It tends to reduce contrast and light throughput. The reduction in contrast is the biggest issue since it affects apparent sharpness and how well subtle details are rendered. But you can compensate for this reduced contrast by using a cat with a significantly larger objective.

Obstruction sizes are usually given as a percentage of aperture diameter. A 100mm aperture scope might have a 40% obstruction which would be 40mm in diameter. Subtract the obstruction diameter from the objective diameter and you get the equivalent refractor size that would theoretically give you about the same image contrast. This assumes that the two designs are otherwise of equal quality - which is probably seldom the case. So a 100mm scope with a 40% obstruction, should deliver contrast similar to that of a 60mm scope of otherwise identical quality. But all kinds of optical flaws contribute to lower image contrast. I think this rule of thumb might be a bit pessimistic in practice since the cats have such excellent inherent color performance. So if 60-80 is the normal range for a refractor, consider 90-100mm the normal range to look for in a cat. And all things being equal, we want a relatively small obstruction (40% or less). ( This article gives a more technical overview of this subject).

Cats usually have relatively long focal lengths. Well designed cats usually have f-ratios of f10 or more. So a 100mm aperture scope typically has a focal length of 1000mm or more. This longer focal length (refractor birding scopes are usually between 450mm and 500mm) has good and bad effects. Longer focal lengths require longer focal length eyepieces to deliver a given magnification. This works against us since long focal length eyepieces with 1.25" barrels have a fairly narrow field of view (35mm - 50mm and up). But it works for us in a different way. These same eyepieces tend to have very long eye reliefs. This can be a huge benefit when digiscoping with cameras that simply won't give decent results with the 17-20mm eye relief supplied by typical birding eyepieces.

But in general, we'd probably prefer a shorter focal length from our cats. This would allow us a greater variety of eyepieces to choose from. But remember the f10 or greater preference. There is a 500mm cat made in China by Synta that has a very short 500mm focal length. Reports are that it provides nice views, but poor digiscoping results. My guess is that its short design results in a large central obstruction. That means lower contrast that apparently overshadows the benefits of the short focal length.

Up or Down, Right or Left

Most cats are actually small astronomical scopes that are sold with different mounts and accessories to make them more suitable for terrestrial viewing. They typically use standard 1.25" eyepieces. This is a blessing and a curse. The blessing is in the huge choice of eyepieces that are available. You can get very good eyepieces for well under $100 and for just over $100, you can get truly excellent optics. The 32mm TeleVue Plossl that I am currently using is superb. The curse is that the erecting prisms sold with these scopes are usually of mediocre quality and they tend to increase the focal length of the scope to boot! Bummer.

As a general rule, astronomy folks don't use prisms. So its no wonder the cats are weak in this department. The astro crowd wouldn't dare degrade their images by introducing all those extra optical surfaces with their unnecessary refraction. Instead, they use 90 degree mirror finders. A mirror finder introduces only one additional optical surface and its reflective surface won't introduce any refractive issues. But a 90 degree finder only erects the image. It doesn't reverse it right to left. The astro crowd doesn't really care since issues of left and right aren't too important when viewing astronomical objects. These mirror erectors have a shorter light path and will usually yield an effective scope focal length close to that specified by the manufacturer. Good mirror finders have excellent optical characteristics.

The inexpensive prism that comes with most scopes is a roof prism. In its basic form, these are OK for viewing, but they will degrade the image some when digiscoping. Roof prisms actually split the image into two sections and recombine the image before it gets to the eyepiece. They are very compact and are the type of prism found in the small binoculars that have a "straight through" tube shape. But all of this cleverness introduces an optical phase shift that degrades the image. Really good roof prisms defeat this problem by using a special coating. But prisms like that can cost upwards of $200. There goes the budget. Ouch!

But you have another prism option. You could get a Porro prism that will also provide an upright and correctly oriented image. Many top end birding scopes use integral porro prisms. One for a cat will cost $100 or perhaps a bit less. They provide image quality that is superior to the standard roof prism and maybe even a little better than the really good phase corrected roof prisms. but porros will also increase effective focal length of the scope and they are much bulkier than the roof prism. The axis where the eyepiece is mounted will be offset a bit - much like you see in those large more bulky shaped binoculars that also use porro prisms.

There is a final erecting option and it is what I use and recommend. Nothing. Nothing is extremely compact. It takes no space at all. It introduces no additional optical surfaces, costs nothing, and has the side benefit of reducing the scope's effective focal length a bit (because the eyepiece is closer to the objective. Of course it does nothing about the upside-down and backwards image either. What do you want for nothing?

So you have to be willing to live with that if you choose a nothing erector. An erecting prism is definitely desirable for direct viewing. But for photography, I don't mind it at all. In fact, in some ways I prefer it when composing a shot. There is something about the wrongly oriented image that makes it easier for me to compose a shot.

Surprisingly, tracking birds isn't really complicated by the upside down and backwards image. In an unerected scope image, the bird (or whatever else we are looking at) is upside down and backwards. If we observe the bird moving up and to the right while looking through the scope, in reality it is moving down and to the left. But a scope on a tripod is a scope on a pivot. If we move the tripod handle up and to the right in the same direction that the bird seems to be moving, the scope actually pivots down and to the left - in the very same direction that the bird is actually moving. Nifty! Just move the handle in the same direction that the bird appears to be moving. This takes a very short time to get used to. In fact, now when I look through a scope that has an erecting prism, I tend to get disoriented and track backwards.

More on obstructions

The cat's obstruction makes the exit pupil (which is an image of the scope's aperture) donut shaped. It makes out of focus specular highlights of the same shape. Some people find these to be distracting.

The donut shaped exit pupil can also be a problem at very low magnifications where the exit pupil is larger. The exit pupil's "donut hole" gets larger as the exit pupil gets larger. If it gets large enough so that it is similar in size to the camera's aperture, you can get a darkening in the center of your digiscoped image. In fact, you can even get a large solid dark spot right in the center of your picture. This means that you may not be able to get a cat to deliver very low (for a spotting scope) magnifications of 15x and below. This is another reason to try to get a scope with a small central obstruction.

The out of focus areas of a cat are formed by a bunch of fuzzy donut shaped spots. On a refractor, these are generally uniform fuzzy disks. When there is a lot of slightly out of focus detail directly behind your subject, these discs can create a distracting pattern. The solid discs of refractors and conventional camera lenses have the reputation of giving a smoother look. Note how distracting the grass is in this image of two inca doves. Your only real recourse in situations like this is to retouch (gausian blur) the out of focus area in a program like Photoshop.

At one time I assumed that this was a very significant disadvantage when comparing a cat to a refractor. But I have found that from a practical standpoint, it seldom really is. In most of my images, there aren't background elements close enough to matter. And furthermore, I've found that refractors don't necessarily make nice even fuzzy dots in their out of focus areas. The Swarovski ATS80HD that I've also been using creates color fringing in its out of focus areas that can be very distracting. This fringing also must be retouched in an image processing program. Top quality refractors can be very good and even excellent. But they aren't perfect.

Ergonomics and other issues

A common criticism of cats is that they aren't weatherproof or waterproof. This is generally true, but there are at least two exceptions, the Jovian 90 by Park and the waterproof C90 by Celestron (about $200 with eyepieces finders etc.) Waterproof design may be a significant issue for some birding environments, but I think the issue is overrated for digiscoping. After all, there aren't too many waterproof digiscoping cameras out there. But you will have to judge the need for your own situation. I live near Phoenix, Arizona where it seldom rains. I just bring a pocket full of Saran QuickCovers that I can use to protect everything from my scope to my head in case I get caught in some rain. If you live where it is wet frequently, then you may need a waterproof scope and more extraordinary methods to protect your camera.

Cats are more sensitive to minor misalignments than refractors. Most cats with apertures greater than 90mm provide some means to make minor mirror alignment adjustments. Proper alignment or collimation assures you are getting the best quality possible from your scope. But this is a bit of maintenance that is not needed with refractors. I've been using cats for about a year and a half, but have yet to go to the trouble to align (collimate) one. Do a web search on "star test" to learn more about a common way to check scope collimation.

I mentioned focusing mechanisms earlier. I have a number of cats and each has a somewhat different focusing mechanism. These cats with their astro pedigrees vary significantly in how easy they are to focus. My LOMO 70 has a knob about the diameter of a quarter mounted on the side. It focuses smoothly and quickly. Its the best of the bunch and I like it a lot. My LOMO 95 is very slow at focusing, but well suited for fine focusing. The focus on my Rubinar 100 is just like the focusing ring on a large lens (because the Rubinar is a large lens) and is fairly quick but stiff and isn't well suited to fine adjustments. By comparison, the focus ring on the Swarovski ATS80HD is just excellent. Its smooth and quick. And since it is a collar around the middle of the scope, either hand works equally well. So when picking out a cat, consider how it focuses. You want a scope that will let you move quickly through its focusing range. Fine focus is not usually a critical issue since most people let the camera handle the fine focus chores.

Cats are usually heavier and bulkier than standard birding scopes, but not by a huge margin. And their lengths are pretty short (because of the folded optical path). My 70mm LOMO is very light and compact and is about the same size and weight as the very lightweight ATS80HD (though the LOMO 70 has a smaller aperture). But the larger aperture scopes between 90mm and around 100mm will probably weigh about a pound or so more than the more advanced lightweight refractors like the ATS80HD. Thats not a huge penalty but its still an additional pound or so that nobody really wants.

All of my cats give a yellowish cast to the images that they create. I think this is from the anti-reflection coating on the front correcting scope element. Most scopes impart a cast of some kind (though the ATS80HD and a few other top scopes are pretty neutral). But the solution to this when digiscoping is pretty simple. Just do a white balance preset with your camera while imaging a white card through the scope. Bingo. Nice neutral colors.

All of my cats are Russian made. Russia has a very good reputation in optics and they produce scopes at a moderate cost. You can find cheaper scopes though. These will probably be made in China. As a general rule, they won't be quite as good optically as those from Russia and quality control may very well be poorer. But those are huge generalizations. Buy from a reputable dealer and do a star test as soon as you receive the scope or even before you buy the scope. If you think you've got a dud, exchange it.

The Pudding

If the proof is in the pudding, then I guess this must be the pudding. People talk a lot about image quality and then make their point with web-sized images that are taken in very uncontrolled circumstances. Taking good test images is a lot of trouble. So I understand why it is seldom done. The following are pretty decent tests. My methods have been developed from many previous tests sequences that helped me learn some of the testing pitfalls. The tests show a resolution target and Phred the Wonder Pheasant. The resolution target attempts to quantify the image quality with a resolution measurment. The picture of Phred tries to give you some idea of how the optics image a bird (Phred has real feathers). The tests include images from a Nikon Coolpix 5000 without a scope, a LOMO Astele 70, a 1000mm f10 Rubinar lens, and a Swarovski ATS80HD. All of the cats used a 32mm TeleVue Plossl. The Swarovski used the Swarovsk 20-60x zoom at 20x.

Testing methods

These tests were taken in my backyard from a distance of about 50 feet. I established a white balance preset for each scope/eyepiece combination before testing and took eight pictures for each scope test.

Four pictures were taken of a resolution target and four more were taken of Phred the Wonder Pheasant. For each group of four, two images are taken with the camera aperture wide open and the other two are made with the camera stopped down to around f8 or as far as the camera can be stopped down before encountering a central area blackout. The 50 foot distance means that these images are taken at a relatively low digiscoping power which makes the blackout issue with a cat more likely. It also makes it impossible to eliminate vignetting with the Rubinar. The Rubinar has the longest focal length and therefore the greatest magnification. So to match image sizes with the other scopes, it requires a lower camera zoom which gives us a camera angle and camera "eye relief" distance that causes vignetting.

I take two images at each setting because I have found that even at the relatively short distance of 50 feet, there can be noticeable variations in image quality - presumably due to air turbulence. The best example from each group of four is shown. The idea behind these tests is to show the best image quality that I was able to obtain from each combination. The images taken with the cats are rotated 180 degrees so they are correctly oriented. There are links to the unaltered original test images at the end of the article. All pictures were taken with a 5MP Nikon Coolpix 5000. The test target is a customization of the test target found on Norman Koren's excellent website. Norman's site is a must read if you want to learn more about the fundamentals of digital imaging and related topics.

You can also see pictures taken with various cats in my gallery of bird pictures. Information about the optics that were used is in the right hand margin of most of the pictures. Of course these images are all reduced in size from the originals and carefully sharpened and retouched as needed. They were also taken at different distances and under varying conditions. So they really aren't an effective way of critically judging the optical performance of the scopes used. But they can give you some idea of what kind of images are possible. Many of these images are sharp enough to make a very nice quality 8x10 print.

Phred's Head

These are unmodified 100% crops of Phred's head. Starting from the upper left and moving clockwise: Taken with the camera by itself from a few feet away. Taken with the Swarovski ATS80HD with 20-60x zoom at 20x. Taken with a Lomo Astele 70 with a Televue 32mm Plossl. Taken with the Rubinar 1000mm F10 lens with a TeleVue 32mm plossl. All subsequent four panel images are in the same order

Phred's Head Sharpened

The same images as above but I have applied sharpening to the digiscoped images to varying degrees to try to match the sharpness of the camera image.

Phred's Feathers

Feathers on Phred's back. 100% crops with no modifications.

Phred's Feathers Sharp

The same images as above but I have applied sharpening to the digiscoped images to varying degrees to try to match the sharpness of the camera image.

Resolution Closeup

These are unmodified crops of the resolution shots except that they have been enlarged 200% to make it easier to compare the point where resolution breaks down. Note the vertical color color banding near the limit of resolution in the camera only image and how it is barely present in the Swarovski image and slightly less apparent in the Rubinar image. This occurs because the Bayer mask on the sensor is recording detail less than two sensor cells wide. At this level of detail, the demosaicing algorithm starts to have difficulty figuring out what color the lines are supposed to be. This is the limit of resolution of the CCD. Note how the colors are absent in the LOMO 70 image that resolves fewer lines.

Resolution Sharpened

These are 100% crops of the same images shown above. But these have been sharpened and the levels adjusted in an attempt to match the camera only image.

Conclusion

I'm not going to bother describing in detail what I see in the pictures or how significant the effects of a little sharpening are. I'll leave that to the reader. But I do think that these tests show that it is possible to get very good digiscoping results with fairly modest equipment. They also show that at low magnifications and moderate distances, a good digiscoping rig can deliver an image that is very close to the quality of an image created by the camera alone.

Original Test Images

These are the original test images. Most are about 1.6MB and may take a little while to download.

Phred Camera Only 1/482 @ f3.6
Resolution Chart Camera Only 1/158 @ f7.2

Phred Swarovski ATS80HD 20-60x zoom @ 20x 1/145 @ f7.5
Resolution Chart Swarovski ATS80HD 20-60x zoom @ 20x 1/129 @ f7.5

Phred Rubinar 1000mm f10 with TeleVue 32mm Plossl 1/290 @ f3.3
Resolution Chart Rubinar 1000mm f10 with TeleVue 32mm Plossl 1/212 @ f3.3

Phred LOMO Astele 70 w/ TeleVue 32mm Plossl 1/243 @ f4.0
Resolution Chart LOMO Astele 70 w/ TeleVue 32mm Plossl 1/178 @ f4.0


You are visitor # since 07-23-2004


If you have questions, comments, suggestions etc. please feel free to write me at:
jay@studio522.com


Copyright © 2004 by Jay Turberville