SGP Autofocusing and bright nebula cores

[that_guy]

Hi Gang, just a bit of a frustrating thing I've been dealing with regarding autofocusing routine and bright nebula such as tarantula. The autofocus star detection detects the bright core of tarantula nebula as several blobby stars and really skews the mean hfr value and makes autofocusing impossible. I keep having to slew slightly off frame every time i need to autofocus and centre back. Is there any "magic" setting I can use in SGP?

[The_bluester]

I have not found it to be an issue myself, and have imaged the tarantula using SGP as well. It does throw the average HFR value of the frame off but at least in my experience it did not really make a difference. The detected HFR of the "Star" that clumps of nebulosity were detected as went up and down with the rest of the frame, so I found that while it would result in an bad looking final HFR compared to a more sparse star field, the actual focus run was still good.

[astrod]

I sometimes see this problem. One thing I do, for some targets, is increase "Auto Focus Data Points" to a higher value, at least 7. Then the algorithm better estimates the linear trend on each side of focus. Of course more points makes auto focus more time consuming.
(And sometimes I intervene manually which defeats the purpose.)

[codemonkey]

I get this on galaxy cores and globular clusters as well as nebula. It's a common cause for me choosing not to image a given target. I'm really hoping that SGP implements outlier rejection soon.

[that_guy]

I know that you can "crop" a field by a certain percentage to speed up autofocusing but it would be great to be able to choose what part of the field to focus on so we can avoid bright cores. I'll try increasing the focus points. I've always stopped the autofocus routine if the 'V' curve wasn't perfect. Maybe I'll just let it run and see what happens. Maybe it'll still be usable. Thanks guys

[ChrisV]

I know it doesn't directly address your issue. But you could try the beta version that has parabolic fitting. I've found it works better than the old straight line fitting - and seems more robust