An interesting exchange with ChatGPT regarding a new onset problem with stars across my entire field of view being oval shaped, and all elongated in the same direction. Exchanges are summarized due to the wordy nature of ChatGPT.
What is interesting to me is how much of the solution came not from facts handed down by ChatGPT, but by the back and forth of ChatGPT offering suggestions, and me pushing back on things that I didn’t think made sense. Much like interaction with a person of slightly higher knowledge than myself. I haven’t tested the system out yet, but I think I found a solution. And it was one that I would probably not have gotten around to considering for several days/weeks if it were not thrown in my path.
📷 Screenshot 2026-02-13 145259.jpg
Me: I have a 150 aperture carbon fiber, F/4 Newtonian telescope. When doing astrophotography, the stars across my entire field of view are elongated in most but not all sub-exposures. All of the stars are elongated in the same direction. The extent of elongation is independent of exposure time from 1 to 30 seconds, independent of which camera I use, and independent of what coma corrector I use. Guiding is roughly 0.5 arc seconds. Images with and without elongation often appear within the same acquisition sequence. What are some common causes of astigmatism?
ChatGPT; It’s probably not optical astigmatism if you are sometimes getting subs without elongation. Consider, thermal currents in the tube, primary mirror pinch due to low temperatures, focuser sag, primary mirror flop, and high frequency vibration that is too rapid to be picked up in your guiding graph as possibilities (various tests for these hypotheses added)
Me: Wouldn’t the fact that the elongation is independent of exposure duration tend to rule out mechanical drivers like focuser sag?
ChatGPT: It doesn’t have to be a steady creep. It might be a fluctuation between two states
Me: Wouldn’t one of the two binary states be energetically favored and unlikely to flip during a 10 second sub?
ChatGPT: There are lots of energy sources that could induce a flip. Dithering, wind, autofocus, wind gusts… What happens when you rotate the camera 90 degrees? This will help you localize the problem.
Me: The elongation changes from parallel to the long side of the sensor to parallel to the short side of the sensor
ChatGPT: The elongation must arise in the optical tube or somewhere upstream of the focuser. (Returns to hypotheses of cold-induced mirror pinch and tube currents. Offers various tests to perform)
Me: Here’s another observation prompted by your suggestion of high frequency vibration. I’ve noticed some play in my RA axis, and when I adjust the focuser the image oscillates in a straight line for up to 10 seconds after I take my fingers off the focuser. But I’ve ignored it because the guiding is fine.
ChatGPT: A high frequency vibrational mode due to this play could readily explain your elongated stars. It won’t show up in your guiding if the frequency is high. The fact that the oscillation is in a straight line supports the hypothesis that your elongation is due to slop in the RA axis. (Suggests various discriminating experiments). You need to look at the RA axis bearing load and not at the worm gear assembly. If it were the belt you’d have guiding problems.
Me: I removed the polar alignment scope and its cap from the back of the scope and this is what I see (Photo). What do I adjust?
ChatGPT: (Explains what all the parts are, tells me to loosen the grub screws and turn the bearing retainer bolt in 1/64th of a turn)
Me: I can’t do that. The grub screws are stripped, probably due to a prior operator. But when I adjust these other two screws holding the retaining ring to the outer case, the play disappears. (We argue briefly about the appropriateness of my fix, and ChatGPT eventually concedes).
Tonight we’ll find out who was right.