Recognizing and Correcting Equipment Problems

All images suffer from a certain number of residual defects—streaks, black blobs, dark corners—the sorts of things that flat-fielding is supposed to eliminate, but does not always remove entirely. On astrophotographs, the same defects are often present, but they are small and hard to see. Because we inspect digital images so closely and apply powerful image-processing routines to enhance contrast, these same defects stand out with great clarity.

Common types of image defects are hot spots, field flooding, vignetting, and dust donuts. Field flooding and hot spots occur when light that should not reach the sensor falls on it anyway. Vignetting and dust donuts result when light that should fall on the sensor is blocked and does not reach it.

When reflections from focus tubes, flip mirrors, and focal-reducer lenses cause unwanted light to reach the focal plane, you get hot spots. You cure them with baffles and paint. An important variation on hot spots is field flooding: it occurs when the whole focal plane (rather than the center only) is awash in stray light. Internal baffling and a long "snoot"-type light shield on the front of the telescope tube greatly reduce or entirely cure field flooding.

Vignetting results from things that block light on its way to the focal plane; you deal with it by removing the obstructions. Too-small focusers are the most common cause. Dirt on the optics and dust on the CCD cause dust donuts; cleanliness is the fix. After removing light blockers and cleaning, what remains of vignetting and dust donuts can be cured by shooting good dark frames and flat fields so you can calibrate them out.

It is easy to get upset about image defects, but getting upset doesn't help much. Take an analytical approach to image defects: figure out what causes them so you can eliminate, greatly reduce, or calibrate them out of your images.

4.7.1 Hot Spots

Hot spots usually show as a bright region near the mechanical or optical axis of the telescope, which should be close to the center of the CCD chip. They may be large or small, sharp or diffuse, irregular or round. If your CCD is offset from the center of the camera body, the hot spot may be off-center in the image; but it probably lies close to the optical axis.

Hot spots result from excess light—non-image light—reaching the center of the sensor. Vignetting, which is caused by too little light reaching the edges of the sensor, looks a lot like a hot spot, but their causes are quite different.

If you were to look into the back end of an ideal optical system, you would see nothing but the objective. Everything else would be black. If you see hot spots in an image, it is very likely that when you look through the optics, you will see other sources of light. The best test for hot spots is to remove the camera and place your eye where the sensor normally goes—at the focal plane. As you look up the light path, any light not coming from the objective may be causing the problem.

In a system with a hot spot, light may reach the sensor by a variety of other routes. Once you have identified its source, you can probably eliminate it. Below are some likely hot-spot sources.

Cylinder Focusing. To check for cylinder focusing, place whatever adapter tubes you use with your camera in the eyepiece holder, but remove the camera body. With a deep-sky imaging system, the adapter may be a simple 1 ^-inch-to-T adapter; with eyepiece projection, the adapter may be a system of tubes and a projection eyepiece. In either case, set up the optical system as you would normally use it. Point the telescope at the bright daytime sky or at a bright, uniformly illuminated surface.

In most astronomical CCD cameras, the CCD is slightly behind the front flange of the camera. If possible, place your eye about the same distance behind the adapter tubes. You will see the objective filled with light; all else should be dark. If the shiny metal interior of the focuser tube lights up when you place your eye at the normal location of the CCD—bingo!—that's the cause of the hot spot. If it is difficult to place your eye at the focal plane, place your head behind the open tube and look in. Move your head from side to side and up and down, looking for reflections from the interior surfaces of the tube. These reflections direct light toward the CCD.

The cures for hot spots are baffles and black paint. Baffles are thin rings of metal shim stock, thin plastic painted black, or blackened cardboard lining the inside surfaces of the tubes. The length and size of the adapter tube determine how high the baffles should stand, and how many you will need. You can cut baffles with sharp scissors, a razor knife, or turn them on a lathe to fit snugly into the tube.

Slide the baffles into the tube and secure them with a dab of acetone-based solvent glue (such as Duco Cement). Paint all surfaces flat black. An inexpensive option to enhance ordinary black paint is to mix fine sawdust into Rustoleum flat black. Dab on this concoction with a cotton swab. It will dry to a dull, rough surface.

An alternative is to line the adaptor tube with flock paper (which has a velvet-like surface) or with black velvet. The nap of the flock paper and velvet acts like thousands of tiny baffles, breaking up and absorbing the reflections. Velvet linings are especially effective for short lengths of tubing; baffles are more effective for long ones.

Internal Lens Reflections. The lenses in a focal reducer, focal extender, or projection eyepiece sometimes conspire to focus an image of the objective near the focal plane. Place your eye at the focus, but this time point the telescope at a dark target silhouetted against bright sky. The chimney on a nearby house usually works well for this purpose.

With your eye at the focus, the objective should appear fairly dark because you see it filled with light from the dark target. Move your eye around the focal plane; if the lens fills with light at the center of the field and becomes partially illuminated away from the axis, reflections are contributing to the hot spot. With focal reducers and extenders, you can switch to a different lens, change the spacing so the reflected light forms a larger, dimmer hot spot, or stop using the lens. With eyepiece projection systems, switch to another eyepiece.

CCD and Window Reflections. Your CCD reflects a significant amount of the light that strikes it; this light does not just disappear, but instead travels to the window of the camera where some of it is reflected again, and returns to the CCD. Another reflection comes from the glass window mounted on the CCD itself. It is rare for either reflection to cause serious problems, but when you are troubleshooting hot spots, you should keep it in mind lest you mistake it for a more serious problem. The easiest way to observe these reflections is to point the telescope at a bright star and make an integration of about 10 seconds. Upon processing the image, you should be able to find a large, faint, out-of-focus outline of the telescope objective.

A much smaller ghost image results from reflections between the CCD and the window covering it. The distance is shorter, so the spot will be brighter and smaller. This spot may be hidden by the glare of any star bright enough to make the reflection visible.

4.7.2 Field Flooding

When you carry out the hot-spot check, you may see sources of light that are continuously visible; these cause field flooding. In Newtonians, for example, a significant amount of light may reach the focus from the inside wall of the telescope tube opposite the focuser and from the open space around the mirror at the bottom end of the tube. Plug these leaks—field flooding increases noise, reduces contrast, whittles away at the limiting magnitude, and introduces systematic errors in flat-

Cylinder Reflection (Hot Spots)

Vignetting

Figure 4.8 Hot spots occur when stray light reflects from shiny interior surfaces of focus tubes to the center of the detector. Field flooding arises from stray light illuminating all or part of the detector. Vignetting occurs when focus tubes or filter holders clip the edge of the cone of light converging to focus.

Vignetting

Figure 4.8 Hot spots occur when stray light reflects from shiny interior surfaces of focus tubes to the center of the detector. Field flooding arises from stray light illuminating all or part of the detector. Vignetting occurs when focus tubes or filter holders clip the edge of the cone of light converging to focus.

fielding.

A ring-shaped baffle inside the bottom of the tube will stop light from coming around the mirror. You can also add baffles cut from thin cardboard or blackened brass shim stock to the front ends of tubes near the camera. When the opening in a baffle is between about 0.1 and 0.4 inches smaller than the inside diameter of the tube, light does not reach the inside surfaces. Install baffles temporarily with tape to check that you are not introducing any vignetting. Once you know they block reflections without vignetting, paint the baffles flat black, and attach them securely.

To stop field flooding from the upper end of the tube, extend it with a "superbaffle" twice the mirror diameter in length. This reduces the amount of stray light falling on the tube walls and diagonal and scattering into the CCD. The super-baffle should be larger in diameter than the rest of the tube, painted flat black on the inside, and lined with internal baffles of its own.

While it may seem like a hassle to eliminate field flooding and hot spots in your telescope, you cannot exploit the full power of your CCD as long as stray light reaches it. Proper calibration of images requires that the amount of light falling on each pixel of the CCD be proportional to the amount of light coming from the corresponding place in the sky.

Hot spots and field flooding disturb the linearity of the digital image: the anomalous light means that the amount of light reaching a pixel is not proportional to the intensity of the celestial source. When this happens, the flat-fielding magic does not work, and the observer wonders how he failed.

4.7.3 Vignetting

Vignetting results when some of the light from the objective cannot reach the CCD. The light loss is usually at the edges of the image. It can result from a too-small diagonal mirror, focus tube, focal reducer or focal extender, or from misalignment. All of these cut off light from the outer parts of the objective.

Diagnostic 1: View the Detector in the Objective. In this simple test, you view the reflection of the CCD in the telescope objective. Move your head back and forth until you have seen the reflection of the CCD in every part of the objective. You should always see the entire CCD reflected in the primary mirror. If you cannot see the whole CCD, light is getting lost. You will also be able to see what part of the telescope—be it an undersized diagonal mirror or the base of the focus-er—is causing the vignetting.

The advantage of this test is that the CCD is in place when you conduct it; the disadvantage is that the CCD appears small and distant. Since the image of the CCD appears at infinity, you can inspect it with a finder telescope or binoculars. Once you have identified the source of vignetting, you should fix it.

Diagnostic 2: Look for Blockage. Another way to find vignetting is just to look for clipping. Set up the optics for testing for internal reflections; that is, complete but with no camera. Place your eyeball exactly where the CCD would normally go and look at the objective. You should be able to see it in its entirety. As you move your eye away from the optical axis, you will see the edge of the objective blocked when vignetting is present.

Because CCDs are such small detectors, you might expect vignetting to be rare, but this is not the case. It often results from the simple mistake of using a tall 1 !/4-inch focuser on a telescope with a fast mirror, from an earnest attempt to prevent field flooding, or from an overzealous effort to keep the central obstruction small. Vignetting also occurs with focal reducers and focal extenders when the lens elements are too small to illuminate the whole CCD.

You may find vignetting and field-flooding in the same telescope. In Casseg-rain, Schmidt-Cassegrain, and Maksutov-Cassegrain systems especially, the baffles that should block light coming around the edge of the secondary mirror may be too small, resulting in field flooding; and the baffle that protrudes through the primary mirror may also be too small, resulting in vignetting.

The eyeball test is not easy to make because it is difficult to place your eye exactly where the CCD would be. If your eye is too far back, the objective may appear to be vignetted when it is really not. If you have difficulty placing your eye in the focal plane, put a diaphragm with a Vs-inch hole at the location of the CCD; you can then inspect the objective by peering through the hole.

Diagnostic 3: The Extrafocal Star Test. The idea in this test is to take two out-of-focus pictures of a field of reasonably bright stars, one on each side of focus. The stars should be far enough out of focus that the objective appears between 20 and 30 pixels across. With an integration time of 60 seconds, shoot several different fields so that you have star images in every part of the images.

If there is no vignetting, the star images will be complete donut images of the objective over the entire field of view in both images, with the diagonal mirror and spider vanes showing in Newtonians, the secondary showing in Cassegrains and SCTs, and complete unobstructed disks showing in refractors.

In severe cases of vignetting, half of the light from the objective may be clipped off. When vignetting is slight, it may be difficult to tell whether the image is complete. Newtonians built for visual observing are primary offenders when it comes to vignetting. Tall focusers, undersized diagonals, and misalignment are the main problems. Eliminating bad vignetting may require significant reworking of your telescope—think of it as "customizing" the instrument for CCD imaging and it won't hurt so much. With Cassegrains and SCTs, focal reducers often create severe vignetting—but they are compromises anyway. The best solution is to test several different focal reducers in hopes of finding one with better vignetting characteristics.

Unless vignetting is severe, though, there's no need to despair. Although losing 5 to 10 percent of the incident light at the edge of the field is hardly desirable, with proper flat-fielding, this loss can be calibrated out.

4.7.4 Dust Donuts

After a long night of imaging, you get up the next morning, turn on the computer, and there's an ugly dark donut in half the pictures. Relax—you're in good company. If you inspect the Voyager images of Jupiter, Saturn, Uranus, and Neptune, you'll see lots of dust donuts, shadows of dust particles in the optical system. The advantage you have over the Voyager Imaging Team is that you can clean your camera. They couldn't!

Dust donuts are shadows. It may seem counterintuitive that a small, round speck of dust will cast a ring-shaped shadow, but in Newtonians and Cassegraini-ans, where the light comes from a source with a central obstruction, that's what happens. One way to imagine it is to think of a dust particle as an anti-pinhole camera, with the anti-hole giving you an anti-image of the objective. In refractors and unobstructed systems, dust donuts have no holes.

Dust donuts are sneaky. A speck that is only 100 microns across (4 thou sandths of an inch) doesn't block much light when it is on the window 12 mm ahead of the CCD. Assuming an //10 optical system, the light blocked by the speck spreads to 1,200 microns diameter at the CCD, casting a shadow that is 99.4% as bright as the surrounding sky. The sneaky part is that you don't see this exceedingly pale shadow when you shoot the images. Only after processing and contrast stretching does that miniscule 0.6% drop in light stand out as a dark do-nut. Fortunately, they are easy to diagnose because the diameter of a dust donut is directly proportional to the distance, D, between the dust speck and the CCD:

where P is the width of the dust donut in pixels, d is the width of a single pixel in the units you used for D, and/is the focal ratio of the optical system.

Consider the following example: You have taken a night's worth of lunar images that show dust donuts, and you know that you need to clean something to get rid of them—but what should you clean? One particularly clear donut is 13 pixels in width by 15 lines high. The images were taken with a Cookbook 245 in 252-wide mode using eyepiece projection with an effective focal ratio of//44.4. Since the images were taken in 252-wide mode, the pixels are 0.0255 mm wide by 0.01975 mm high. You compute their location:

Given the difficulty of estimating the size of the dust donuts, the agreement is pretty good. The estimates place the dust roughly 14 mm ahead of the CCD chip, which pretty clearly puts the offending speck on the window of the camera. With eyepiece projection, small shadows are caused by dust on the cover glass of the CCD itself.

In deep-sky images taken at/75, dust donuts on the camera window cast such dilute shadows that they are invisible. However, dust on the front of the CCD cover glass is a candidate for causing problems. The glass cover on a typical CCD is 1 mm thick, located about 1 mm ahead of the chip, and has an index of refraction of 1.53. Refraction decreases the effective air thickness of the glass to 0.65 mm, so the total distance from the front of the cover glass to the CCD is 1.65 mm. Solving the distance equation for the diameter of the shadow:

df where again P is the width of the dust donut in pixels, D is the distance from the focal plane, d is the width of a single pixel,/ the focal ratio of the optical system.

In this example, an image taken with an//5 optical system using a Cookbook 245

Figure 4.9 Select imaging tools to produce the results you want. To obtain a wide field of view, use a short focal length and/or a large detector. Today's digital SLRs combine quality short-focus lenses with large CCD or CMOS sensors, and they provide constellation-spanning fields of view.
Figure 4.10 The right imaging tools will give you the results you want. To hold the Moon comfortably, the focal length of the optical system should be 100 times the size of the detector. This focal length comfortably accomodates Messier objects and most of the better-known NGC and IC objects.
Figure 4.11 Webcams and small-chip astronomical CCDs are great for capturing diffraction-limited detail on the planets and Moon. For such objects, a field of view no more than a few arcminutes wide is all that's needed. The key to successful imaging is to match the tools you use to the imaging task.

in its 252-wide mode (i.e., 25.5 micron pixels), dust on the CCD cover glass will show up as dust donuts 13 pixels wide. When you are confronted with dust donuts, measure their size, calculate their location, and then clean the appropriate surface with alcohol on a swab.

4.8 Reaping the Benefits

Shooting digital images is easy, but shooting great images takes effort. Great images reflect not only careful attention to hardware and techniques, but also identification of and devotion to specific imaging goals.

Begin by determining your imaging goals. On a clean sheet of paper, write down what you expect to accomplish with your digital images. Reduce your goals to basic facts and figures—fields of view, limiting magnitudes, star-image size— and then select equipment that can and will meet those goals. Every minute that you invest in learning what equipment can meet your goals will be time well spent.

Be prepared to discover that you cannot attain your initial set of goals. It may be necessary to tailor your choice of sensor to an existing optical system—or to tailor the optics to a specific sensor system. You must decide whether field-of-view trumps resolution, or whether resolution trumps field-of-view. This depends on your goals and the celestial bodies that you are imaging.

The three F's of imaging—finding, focusing, and following—are prerequisites that your equipment must satisfy for you to achieve success. Select equipment that will make it easy and efficient to locate, sharply focus, and accurately track the objects that you wish to image. It does not much matter how you accomplish these goals, it only matters that you do accomplish them.

Once you have selected and installed your imaging equipment, it is still important to remain goal-oriented. It is easy and tempting to expect "the right equipment" to solve all of your imaging problems. If you see hot spots or dark corners in your images, the equipment is trying to tell you that something is not right. This does not mean that the equipment is bad, defective, or broken; it may simply be that you have put the components together incorrectly. Do not blame equipment for problems, but rather look into the matter, identify the source of the problem, and with the aid of the insight gained, cure it. Your reward will be better images.

Finally, learn all you can about your equipment. In digital imaging, your camera, your telescope, the weather, and astronomical objects interact with one another. The more your learn about esoteric subjects such as the spectra of galaxies and the wavelength sensitivity of your camera's sensor, the more things will make sense to you, and the more focused and better directed your activities will become. Setting broad goals for your imaging and paying attention to the nitpicky details pays off handsomely in results that make you feel proud.

Was this article helpful?

0 0
Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment