Questions on First Attempt at AutoFocus


Hi Group,
Target M27 Dumbbell, shooting at 2182mm focal length
Atik 484L+mono at 5.4 pixels
Step setting calculated at 18 steps.
Autofocus data points 9.
Bin 2x2
Shooting in OIII with 25sec for autofocus exposure.
There was a bit more than 3C temp drop during the session.
Autofocus was set before sequence start, at every 6 images (equals 60 minutes) and every 1C temp change.

  1. The focuser started out (or finished) around 29650 at the 8:49:09pm initial autofocus. It appeared to finish at 29828 at 12:50:31 this morning. Step increases are consistent with falling temps.
  2. Being my first trial, there is a bit of starting and stopping at the beginning of the log file (my fault).
  3. I did a manual focus (with B-mask) prior to beginning the sequence.

a) Can the “V” Curve generated during focus run be redone for review?
b) My autofocus runs did not create the victory “V” curve. What should I take from that?
c) Log makes reference to “old autofocus method”.What is the “old autofocus” method?
d) In spite of what may be unsuccessful focus attemps, my images remained reasonably focused (and very similar) throughout the almost 4hr session. Any thoughts on that?

Here are a couple of screen shots of the initial focus and one later in the evening:

Here is the DropBox link to the SGPro log file, the PHD2 log file, and the last autofocus folder:

Thanks for your patience and help.


Not by you…but if you turn on saving of the Auto Focus frames then we can inspect.

[quote]b) My autofocus runs did not create the victory “V” curve. What should I take from that?
Looking at your graph your step size is much too small or you didn’t take a long enough exposure.

It’s the algorithm that SGP previously used to find focus. It is used when 2 convergent slopes cannot be found in your sample space.

Luck? Or it could be that you never moved out of critical focus and the temp was stable. Fairly difficult to tell. But generally focus moves in when temp drops, not out.

Personally I wouldn’t focus through a Narrowband filter. Setting up your filter positions and using a single filter (lum) for focus is much better and faster.