Converting an 8-Inch Schmidt Camera to Digital Imaging

Prehistory
In the early 1970s, the American company Celestron started building Schmidt cameras for amateur use. Devices with apertures of 5 ½, 8 and 14 inches were sold. Even now these remain the stuff of legend. When film emulsion provided the only means to record deepsky images, Celestron's cameras offered amateur astronomers the best means of capturing wide fields. Thus, Hans Vehrenberg (among others) published breath-taking celestial photos, exciting and inspiring many amateur astronomers with his Schmidt cameras in Germany and Namibia.

But starting with the advent of CCDs, and certainly since the decline of commercial film production about ten years ago, Bernhard Schmidt's ingenious system seemed at last to have finished its time. This is because a Schmidt camera has a serious problem: its images are formed on a curved surface. This isn’t an insurmountable problem with photographic film and even thin glass plates, since it’s possible to overcome by bending them over curved mandrels (or carriers) during exposures. Afterward they can spring back to flatness, providing a large, well-corrected field of images. Digital detectors, on the other hand, cannot be so bent. They come ready-made on flat substrates. And so, many small Schmidt cameras disappeared into their storages cases, and took up bleak residence in dark corners of basements. Professional observatories with large Schmidts, on the other hand, (as for example the Thuringian State Observatory in Tautenburg) of necessity devised a different solution -- namely, use of a corrective lens (field flattener), placed in front of their digital detectors. This allowed large Schmidts to adapt themselves to a new life, with a new detector type. And so such Schmidts continued to be used.

But the same solution can be employed on small cameras too! That’s what the present article describes: adapting a small Schmidt camera to digital imaging. My adventure in digitizing an 8-inch Celestron Schmidt started last year with an e-mail from Walter Stephani. He is a researcher who has studied Bernhard Schmidt for decades at the Hamburg Observatory, ii in association with Roger Ceragioli (University of Arizona). After reading (on the Münsterschwarzach monastery’s website) about my conversion of a 5 ½-inch Schmidt camera to digital imaging, he called me. He had an old and long disused 8-inch. Though somewhat damaged, would I like to repair it and convert it for use once again? Since Dr. Ceragioli (an optics expert) was on board, I assented cautiously, but with inner enthusiasm.

The ca. 50-year-old coating on the Schmidt camera's mirror was no longer in good condition. So Walter had it recoated before sending the device on to me. My initial, gloomy impression of its condition soon brightened once the corrector plate was carefully cleaned. Then I recognized that before me was (in principle) an optically flawless system -- albeit with an old-fashioned carrier for photographic film. Now came the task of replacing that with a digital detector, to bring this classic device into the modern age!

Test Phase
With the conversion complete, testing began. The first target was a daytime wind turbine. I could already find the overall focus position relatively well. But small errors, as expected, could be seen throughout the field: the left side was slightly sharper than the right, and the corners were not sharp. Further tests followed using a C8 as collimator.i Employing a special procedure I found on the Web (see the links below), the entire field could be adjusted and corrected. A crosshair eyepiece in the C8 was focused on infinity and directed to a bright background. The Schmidt camera now looked directly into the C8 and the crosshairs of the latter were imaged onto the CMOS detector. By positioning the crosshair first on the detector's center and then its four corners, image sharpness across the field could be accurately adjusted.

First Light
On the first clear night, the converted Schmidt camera was turned to the sky. The target selected was Comet C/2022 E3, near NGC 1647. There were many bright stars in a large field, in addition to the comet. Even the raw images (after a final adjustment) gave convincing proof: all the stars were sharp. Success! The new focal length of the Schmidt camera with the addition of the field flattener and filter, as well as the altered position of the corrector plate, was determined as 290 mm, the focal ratio as f/1.45, and the image diagonal as 4.6° (3.8° x 2.6°). One pixel equals 3.3 arcsec.

During image stacking, a curved streak was noticed beyond the comet: a ghost of a bright star lying outside the field had surreptitiously found its way in! And something else became clear during adjustment of the light histogram: flat-fielding had to be done very subtly. Otherwise, large dark and light rings formed around the image periphery.

A New Paint Job
Another building block in the improvement of the imaging was to provide the inside of the tube with a new, ultra-black paint job. Initially, I had only planned on recoating the outside of the tube white. The irregularity of the original black aluminum had to go! So it was via synergy that I recoated the inside of the tube as well. Surprisingly, reflections and attendant ghosting were greatly reduced in this way.

But then another – so far not totally understood – phenomenon came to attention: the oil-spacing of UV/IR blocking filter and the flattener lens produced rainbow appearances at certain angles of light incidence. Apparently, this is an interference phenomenon, which probably occurs in the boundary layer between the filter and the plano surface of the flattener. It would seem that this should not be possible due to the oil-spacing. One possibility, however, is that we have here white-light Newton’s fringes: the oil may not match the indices of refraction in the glasses exactly. Nevertheless, the fringes can be markedly ameliorated during image processing by subtracting out a very accurate flat-field exposure (a so-called “sky flat”), where the aperture of the camera is covered with black cloth. Subtracting sky flats has yielded the best results.

Optimization
After the first series of exposures, the images were analyzed carefully. A slight coma was observed in stars, especially those in the corners of the frame. To eliminate this, the tube was shortened again by 7 mm (in total, the corrector plate now sits 65 mm closer to the mirror than originally). The next set of images evinced the success of this measure. Only a slight chromatic aberration remained in the corners of the field. This is an unavoidable product of the non-achromatic field flattener lens. It can, however, easily be corrected in image processing. The red channel must be reduced by 0.05%, and the blue channel enlarged by 0.05%. With 4144 pixels in the detector length, this is only 2 pixels, but it improves the image noticeably. The limiting magnitude is about 20.5 for exposures of two hours duration.

A dew cap helps to reduce stray light. Images taken after the crescent phase of the Moon show a noticeable gradient. And cirrus clouds quickly envelope brighter stars in a halo. Thus, we see the high sensitivity of the Schmidt camera thru such unanticipated disturbances!

In a later step, the oil filling was replaced by an air gap. Small spacers of 40 µm thin paper were inserted between the filter and the flattening lens. The rainbow phenomena disappeared completely and there were no new disturbances, such as "ghost images". Unfortunately, however, the collimation of the Schmidt camera had to be readjusted!

Practical Use
Because of the fast focal ratio, exposures are taken without active guiding. In the case of individual exposures, times greater than one minute are unnecessary. Long series of exposures are merely a challenge to the hardware and stacking software during post-processing. There is very little need of correction of the stacked images. Colors appear natural and balanced. Thermal noise in the uncooled camera is surprisingly low and not much worse than with a cooled model.

The conversion of this Schmidt camera was rewarding. Not only did it result in an outwardly attractive recording device, it also brought Bernhard Schmidt’s invention into the digital age, yielding high resolution over a large field, coupled with the highest light intensity!