I have just been trying to autofocus on the Globular Cluster M15 and have run into problems. The routine appears to be seeing very large stars in the core of the Cluster when it is out of focus and this is creating false readings in the V curve. Here are screenshots of the focus routine running:
As you can see there are peaks where there shouldn’t be. I tried imaging the globular M2 last year, but binned the image as it appeared out of focus. I’m pretty sure the same thing was going on with the autofocus, but I didn’t really realise at the time.
Has anyone else experienced problems running autofocus on globulars? Or have I got a setting wrong somewhere?!
Yes, they can be tricky. I usually tried to get close manually, using frame and focus and then just fine tune with smaller step sizes. The alternative is (though I have never tried it and I have never seen much forum traffic on its use) is to try the alternative focus target (it is one of the dialog boxes in SGP)
Thank you for your response Buzz. From reading the manual it appears that the different Focus Position is for manual focusing only - otherwise that would have been perfect!
Not sure what to do re. Autofocus in this instance… Ken, Jared, any thoughts please??!
It might be helpful if you tell us your equipment like brand, aperture and focal ratio of your scope, brand and model of your camera and brand of motorized focuser. Do you remember if the seeing condition was good or bad or were there any high clouds?
I imaged M13 last month and it was one of the best auto focus runs I’ve seen. My scope is TEC 140 APO f/5 (705mm focal length) using Astro-Physics Quad TCC, camera is QSI660wsg and motorized focuser is Optec Focus Boss II with Handy Stepper Motor. This is my latest M13:
My Auto Focus run uses step size of 100 and total of 9 steps and binned 2x2.
Love the image. My M13 was slightly deeper but my resolution using a RC10 cannot match your image. I particularly like the rendition of the small galaxy.
My system is: Celestron EdgeHD 8” (2032mm @ f10) with Lakeside motor focuser and QSI 683-WSG8. The system has worked perfectly on other targets so far, but is struggling with globulars…
Which globular cluster are you attempting to image right now? M13 or M15? M13 is bigger than M15 and you may have better chance focusing with M13 than M15.
I have been trying to focus on both M15 and M2, but they are both relatively small and things are going wrong. I will give M13 a try just to see if it is any better.
Nigel, that’s exactly what I tried last night. The problem is that the Focus Target Module is a manual process, autofocus ignores the settings in it. It would have been the perfect solution and I wonder if it could be activated in the code (Ken, Jared?). In the short term, I think I am going to have to select alternative targets or go manual…
I’m having similar problems though I’m not trying to image a glob, just a target which has a couple of groups of 3-4 stars towards the center of the frame, with SGP detecting each group as one big star at different points of the focusing run which seems to throw off the curve. I’ve tried lots of variations of step size, number of steps, min star size and can’t quite seem to get there.
I can get close to focus, but not good enough on this target. Very frustrating! I’ve basically given up on trying to image the target I’ve chosen after 3-4 nights of failing to get it to focus properly.
If we could define a subframe for use with AF, that would be a seemingly simple solution to the problem. Right now we can define a single crop region, but it’s centered, which doesn’t help when the problematic stars / features are in the center of the frame.
SGP’s AF routine has been a continual source of problems for me. It’s enough of a problem that as above, I can’t seem to image some targets and I’m starting to wonder whether other software solutions might be worth considering. That’s not meant as a threat and I’m not trying to be a douche but it really is a big problem for me and I’m just trying to be honest in conveying the possible implications.
I’m sorry to hear that you are having focus issues, but I’m also relieved to hear that you are having focus issues… it’s not just me! I think I am going to adopt a semi-manual approach with this target, using the Focus Reminder feature along with Focus Target & Image Target, slewing to the Focus Target, running AutoFocus and then back to Image Target to continue imaging. It’s a hassle as not fully auto, go to sleep and let it run style, which I have got used to! Hopefully it will work though and at least let me acquire good sharp data of this globular. Good luck in your quest and perhaps Ken & Jared will like your definable area for AF idea - that should work…
Hi Photogav, did you get any resolution to this, as I have been experiencing exactly this problem imaging M2 with my C14 + QSI583 mono CCD. I want to be able to auto focus on filter change without slewing to another part of the sky and certainly don’t want the manual intervention. Looking back at images I’ve taken previously of M13 and M15, I note that focus was off on some of the filters, so i think this an ongoing problem. Other capture software that I have used runs their focus routine on a single star, but that option doesn’t seem to be available with SGP and globular clusters are definitely a problem. I’ve tried adjusting exposure and min star size, with limited success.Just observing the downloaded images from each focus step it is clear that the best focus point (smallest stars/lowest HFD per star) is not what SGP calculates as best focus position. I didn’t see any response from Ken or Jared, so I’m hoping that you found a working automated resolution.
Cheers, Geof
Hi Geof, it’s bad news I’m afraid - I never sussed out an automatic solution to this issue. I ended up doing it manually, setting a reminder when the sequence wanted to focus and paused waiting for my intervention, then manually slewing to an area away from the glob, running autofocus, centring back on the target and continuing the sequence. It was a right hassle, especially when I’m used to sequences looking after themselves while I’m tucked up in bed! Consequently globs are not high on my target list!!
If you come up with an automatic solution, please share it…!
Gav.
