Solar Astrophotography

   I used a Coronado SM-60 H-alpha filter on either my Takahashi 4" f/8 refractor or my TeleVue Pronto 70mm f/6.8 refractor in conjunction with a Philips ToUcam Pro webcam to obtain the images and video below. To be able to obtain sufficient contrast to see prominences, flares, filaments etc., the bandpass of the H-alpha filter must be narrow. The specifications of the SM-60 indicate the full filter bandwith to be less than 0.7 angstroms centered at 6562.8 angstroms (656.28nm). This wavelength of emission corresponds to the transition of an electron from level 3 to level 2 of the hydrogen atom. It is a member of the Balmer series, which consists of all possible transitions to electron shell 2. (The H-beta emission line, not isolated by this filter, is the electron transition from energy shell 4 to 2, occurring at a wavelength of 4861 angstroms). The H-alpha filter is attached to the T-Max tuner, which is a manual adjustment that enables one to tilt the filter such that it is not parallel to the optical plane of the telescope. Since it is an interference filter, the tilt shifts the central wavelength of the filter, thus enabling one to better see doppler shifted phenomena.
   The Philips ToUcam Pro is used to capture .avi video at prime focus. This webcam is USB1.1, and this affects the number of images per second that can be transferred to my laptop. My experience indicates that setting the capture rate to 2 frames per second results in minimal duplication of frames. The exposure is 1/50 second with zero gain for the 25-26 October 2003 imagery. Setting the gain above about 20-25% results in grainier images due to amplifier noise.
   The videos are processed using RegiStax, a powerful freeware product, in which each of the individual frames is co-located, and based on user selected criteria the best frames are retained, averaged together, and image processing routines can be applied. By averaging many frames the noise due to random distortions caused by atmospheric turbulence is reduced. It can be shown mathematically that the noise reduction is proportional to the square root of the number of frames averaged. For example, by averaging 100 frames the random noise is reduced by a factor of 10 compared to a single frame.
   2 August and 10 August 2003 (California time): These were my first attempts to obtain webcam imagery of the Sun. These were proof of concept tests using my TeleVue Pronto on a camera tripod. Since no motor drive was used the Sun quickly moved through the field of view. Even so, RegiStax was able to track the moving target. I learned that focusing is a challenge, and I still spend up to 20 minutes tweaking the focus in an effort to get sharp images. From the 2 August session I learned that I used too long of an exposure time, hence the overexposed image. The 10 August imagery was improved, but is a bit underexposed. I also learned that taking a darkframe video is essential. As seen in the 2 August solar image below, the background is bluish and the Sun is pinkish. The seams of the ToUcam let in stray light which indirectly illuminates the CCD chip. By covering the front of the telescope so that no light passes through the telescope, I take a darkframe video (1-2 minutes is adequate) which captures only the stray light leaking into the webcam. In RegiStax the darkframe video is processed to yield a single darkframe. This darkframe is subtracted from the frames of the solar videos, thus removing the stray light signal from the processed image. In the future I will seal the seams to reduce the stray light component at the source. Gert Gotschalk of TVS suggests using aluminum foil to cover the seams since only using tape may still permit infrared light to penetrate. The other reason to make a darkframe (even with the seams sealed) is to account for the thermal noise that builds up in the wells of the CCD chip.
   25 October 2003: This was my first attempt at solar webcam astrophotography using my Takahashi 4" f/8 refractor on a motor driven mount. A new video, consisting of approximately 120 frames, was generated each minute. I obtained approximately 90 videos of Sunspot group 484 and the prominences on the limb. Each of these videos was processed in RegiStax Thus, from each video I obtained one image. To make the movie of the prominence eruption, I used 47 images (from every other video) which cover the period 17:43-19:16UT. In Paint Shop Pro I manually co-located each of the individual images, and put on the various labels. Then I made the movie using the animation wizard. I estimate it took me about 12 hours to process all of the data and generate the movie. I wasn't entirely happy with resulting images, as they appear noisy, so for future webcam solar astrophotography I decided to obtain videos of 2 minutes duration each so that I had approximately 240 frames to process in RegiStax. Also, the videos for the first third of the movie were shot through high thin cloud that compromised the image quality.You will also notice that as the movie progresses a black band moves from left to right. This is because I did not have perfect polar alignment (difficult to do when setting up in the daytime!) and I did not enter any guide corrections over the 90 minute period. By the time I realized the Sun was drifting through the field of view it was too late to do anything about. When I manually co-located the processed images in Paint Shop Pro I had to increasingly shift the images to my chosen reference point, thus the progression of the black band.
   26 October 2003: I was very fortunate to capture a developing solar flare above Sunspot group 484. This was one of several large flares that occurred in late October and early November. So much activity over a period of a few weeks caught scientists by surprise since we are more than 2 years past the latest maximum in the 11-year sunspot cycle. The sky was crystal clear on this day, and in conjunction with using 2 minute videos the image quality improved compared to 25 October. However, the quality of frames in the latter portion of the Flare Eruption movie deteriorate. When taking images I noticed some drift of the sun through the field of view. In my haste to set-up to image the flare I forgot to balance the telescope tube, which I thought might be contributing to the drift, so I put a weight on the back near the eyepiece holder. This had the detrimental effect of causing the focus tube to progressively rack out of focus. As it turns out the tube imbalance had nothing to do with the drift. The drift occurred because one leg of the tripod sank into the damp grass. Haste does indeed make waste!!! Yet another lesson learned! I will try some image sharpening routines to improve the image quality of the latter frames of the movie.

Click on the pictures to see full size versions.

Prominence Eruption (42kb) Sunspot Group 484 25 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam			Prominence Eruption: The Movie (5.1mb) Sunspot Group 484 The prominences take magnification very well! The large prominence at the top-center exhibits plenty of material transfer, while the small vertical prominence at the top-right displays two lobes of material that pirouette about each other. The thin dark shading that arcs left to right in the upper third of the image is a filament. This is a prominence seen from above! 25 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam
Flare Eruption (32kb) Sunspot Group 484 26 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam			Flare Eruption: The Movie (4.9mb) Sunspot Group 484 26 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam
Prominence Eruption (37kb) Sunspot Group 486 25 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam			Prominence Eruption (40kb) Sunspot Group 486 26 October 2003 Takahashi 4" f/8, Coronado SM-60 H-Alpha filter with Philips ToUcam
11 August 2003 Pronto 70mm f/6.8, Coronado SM-60 H-Alpha filter with Philips ToUcam			2 August 2003 Pronto 70mm f/6.8, Coronado SM-60 H-Alpha filter with Philips ToUcam

All photos are copyright K. R. Sperber
Paint Shop Pro (c) Jasc Software Inc.
RegiStax (programming by C. Berrevoets)
Mention of copyrighted software does not imply endorsement.

Return to Ken's Astrophotography Homepage