Anyone have any idea why asiair starts off picking good stars in multi star guiding then the circles start drifting off the star centre after 20 minutes?
Here is a couple of screen shots where some stars are centered and others the circles have moved off center. Whenever this happens my guiding seems only marginal and the main star is off a bit too.
Calibration was great and orthagonal too.


    Kevin_A Anyone have any idea why asiair starts off picking good stars in multi star guiding then the circles start drifting off the star centre after 20 minutes?

    First guess is Field Rotation, Kevin.

    Field rotation (over time) will cause each guide star to move in a different vector, thus fighting one another, instead of cooperating and reducinng the variance.

    I would check your field rotation by using drift methods, instead of believeing what the ASIAIR tells your (hint:it lies).

    The other possibility is flexure.

    Chen

      w7ay
      I know it is most likely not flexture as my fma180 pro is very setscrew tightened tight. My polar alignment was checked twice and great.
      My question is…. How is it possible that within stars in the same close area… some are dead centre and some right beside them are on the edge of the circle? Why are some circles moving off target while one right beside it is centered. To me drift or flexture would affect all circles but maybe I am wrong. Could bad sky turbulence cause it? All my connections are tight and tonight guiding was worse and this is when I noticed this.

      • w7ay replied to this.

        Kevin_A My polar alignment was checked twice and great.

        Did you check it with some real tool, instead of the ASIAIR polar alignment? A couple of years back, when ASIAIR first implemented Polar Alignment, I posted how much error it has, based on measurements using my Mount simulator.

        https://bbs.zwoastro.com/d/11643-asiair-polar-alignment-without-visibility-of-the-pole

        When multi-star guiding became available on ASIAIR, I had written to them to warn them about the field rotation problem. The problem does not happen to single star, since the field actually rotates around the guide star (not the center of the FOV) in that case.

        One way to solve it is to allow field rotation to occur, but using a low-passed location (about 5 minutes time constant) for each guide star. It fell on deaf ears.

        Remember too that a non-flat image will also cause problems when the field rotates. That is why I do not recommend use of multi-star guiding unless you have a decent guide scope.

        What you can do is to record a guide image and then 20 minutes later, record a second guide image. Now go to PhotoShop to overlay one image transparently over the other. This will tell you what has actually moved (the stars or the circles).

        If the stars themselves have moved relative to one another, your guide scope did not produce a flat image.

        If the relative position of stars did not move, but the center of the circles did, ASIAIR has a centroid estimation bug.

        Chen

          Remember too that we are depending on sub-pixel movements to autoguide. Multi-star autoguiding is much more sensitive to field rotation than imaging, which can withstand a pixel or two of movement (depends on seeing and plate scale).

          Chen

          w7ay thanks Chen, that makes a lot of sense! Good food for thought.
          I have 3 setups and this one always has issues. My big and small setups run fine but this one just got the new fma180 pro guide scope and it still gives me grief!
          My FMA180 Pro has a field flattener built in so it should be flat when using my tiny 290mm mini guide scope.

          • w7ay replied to this.

            Kevin_A My big and small setups run fine but this one just got the new fma180 pro guide scope and it still gives me grief!

            If this is an outlier, I would suspect flexure.

            Remember that we are not talking about differential flexure between your main and sub OTAs here, but if you see the FOV change after 20 minutes (PhotoShop transparency), then I would check the usual suspects.

            Chen

              w7ay I have checked over all areas of possible flexture and cannot find any. If anything, everything is overly tight!

              • w7ay replied to this.

                Kevin_A I have checked over all areas of possible flexture and cannot find any.

                Repeat: please do an overlay in PhotoShop of two frames 20 miniutes apart.

                Chen

                  w7ay yes I will when I get time. Busy busy busy time of year. Tomorrow I travel, then Mexico then B.C. then Nova Scotia….
                  Did you say you have an Askar FMA180 Pro? I just got one and it is very sharp in all 4 corners on a fullframe. Using it for a guide scope now.
                  Would not all of them be angled in a similar or radiating direction?

                  • w7ay replied to this.

                    Kevin_A Did you say you have an Askar FMA180 Pro?

                    Yes, I have it, and also the non-Pro version. Same optical glass, but the "pro" is much more ergonomic.

                    Pretty reasonable guide scope for shorter focal length main OTA.

                    BTW, you need to check if the stars themselves have moved (relative to one another) from one frame and another frame 20 minutes later. This will tell you if you are fighting with the ASIAIR bad centroid algorithm.

                    Just switch from ASIAIR. Your problems with flakey ASIAIR algorithms will be over.

                    I have recently been indoor testing the usability of StellaMate Pro and INDIGO Sky running on Raspberry Pi 5 to see how much more "difficult" they are to use, compared with the ASIAIR. For someone with more than room temperature IQ and can actually read English documentation, it is pretty much a wash if you just need what is only available in ASIAIR (the INDIGO Web interface is actually quite nice for initial setup -- WiFi, drivers, etc.) StellarMate Pro might be more money than a lot of Easy as 1,2,3 types want to spend, but the Raspberry Pi 5 is dirt cheap ($100 -- even if you buy the complete INDIGO A1 software suite for $60, is still dirt cheap).

                    My All-Sky Camera has been running trouble free for about a year on a Raspberry Pi 4 INDIGO Sky, with my own macOS software as the GUI -- I may just swap the board for a RPi 5 to see if the image downloads are even faster.

                    If you have a Mac, the INDIGO A1 is a pretty complete turn key solution. To make it Easy as 1,2,3, just don't use the features that are not present in the ASIAIR.

                    Chen

                    Kevin_A Thank you for your feedback, we also received your log and checked that.
                    Does the problem happen every multi-star guiding? if so, can you please take a video and submit the log again?
                    We need the video and corresponding time log for further analysis, thank you very much.

                      Tech@ZWO I will verify if it happens on all sessions. I am using a am5 mount, asiair plus and asi290mm mini. Exactly what do you want in the video? Can you be a bit more specific on what you need regarding video.
                      The pictures I posted above were on March 4, 2024 between 9:59pm to 10:01pm at night on the log I submitted if that helps.


                        w7ay while I have not done an overlay yet I did notice after some info back from the tech dept that my polar alignment is moving all the time. I checked the log and after every 20 minutes is is off by quite a bit. I checked a few logs and my overall PA drift moves around all the time from 0.3 to 3.0 to 0.0 to 1,5 to 0.1 to 30.2 to 11.0 to 0.1 to 2.6 randomly on both axis after each guiding restart or new target etc… no matter what guide scope or imaging setup I have been using. Maybe my tripod is flexing or mount alt bolts or something loose. Interesting!

                        • w7ay replied to this.

                          Kevin_A Maybe my tripod is flexing or mount alt bolts or something loose.

                          Worth also checking if your mount's two axes are orthogonal. Plate solve after RA slew and DEC slew should tell you that.

                          Chen

                          5 days later

                          Kevin_A The video including the time and corresponding log will help us to confirm the drift direction, and what happened during that time. thank you.

                          Write a Reply...