I have a Moonlite Nite Crawler which I’ve used with SGP for a while without too many problems. However, I have a second auto focus setup that I’m having issues with. It’s an Astromechanics focuser for Canon lens. I can control the focuser through SGP (i.e. move the focuser in and out), but when I try to use auto focus, it does not work as expected. If a focus run fails and the starting focus point is moved, it repeats the previous focus run (i.e. it starts at the same focus point). I assumed that SGP auto focus would move the starting point by a single step size (as defined on the auto focus setup panel), but it’s not behaving that way.
If the starting point is good, then auto focus works, although it’s very difficult to get anything like a V with the lens I’m using (Canon 200mm f/2.8 L).
If a focus run fails then it should always be returning back to the initial point. This is to prevent runaway cases where focus would continually fail and keep moving the focus point around.
If you have “Smart Focus” enabled then SGP will attempt to shift the focus point if there is a trend but if there are too many shifts then Smart Focus will eventually fail and it will return back to the initial start point.
I’m seeing multiple attempts to focus, but no shift when one attempt fails. It looks like “Smart Focus” is enabled by default, so I think that’s what’s happening, but no shift in the starting point. As a result, it’s just repeating focus with the same set of focus values and fails again.
I’ve copied a segment of the log file that should show an example of what I’m seeing. There’s a line in there “Focus cannot be determined…” that looks suspicious.
It doesn’t appear that I am able to upload a txt file, so I’ve uploaded it to my Google drive.
I have got interested in SGP AF from another post and so decided to take a look at your logfile.
I see two focus attempts. You are set for 7 data points with step size of 3 in the first attempt and 4 in the second.
AF Run 1: Focuser at 5006, Step size 3
Step
FocPos
#Stars
HFR
Quality %
1
5015
2
10.92
2
5012
48
3.92
3
5009
27
3.31
4
5006
34
3.28
96
5
5003
59
3.65
91
6
5000
26
4.83
86
7
4997
9
5.83
80
AF Run 2: Focuser at 5006, Step size 4
Step
FocPos
#Stars
HFR
Quality %
1
5018
1
12.81
2
5014
28
4.61
3
5010
31
3.65
4
5006
34
3.54
97
5
5002
28
4.78
93
6
4998
16
5.81
87
7
4994
6
8.01
83
I believe the reason that your AF runs are failing is that the the quality is below the default SGP target level. There is a way to change the target quality but I would need to do a search to find how this is done. {Smart Focus Lost - Frequently happening 3.1.0.313]}
I think the reason that you have such a low quality score is that on the first data point of each AF run you have a a suspiciously large HFR value .
As you say you seem to be starting at a pretty good focus position so therefore it is reasonable to expect the initial HFR value to be similar in value to the closing value i.e circa 6. This obviously is not the case!
If you have saved your AF packs (see Focus Control/Settings/Save Auto Focus packs) I suggest you take a good look at the opening image in each sequence to see if there are any clues. Perhaps you have a large backlash setting and your mount perhaps needs a longer time to settle.
I also have the Astromechanics EF focuser and have used it with the Canon 200mm f2.8L. I found that I had to find the sharpest focus manually, or close enough, before initiating autofocus. I used a step size of 3 and 9 steps, so AF first moves the focus out by 12 steps. I got a nice focus curve but it is important to start near the best focus. I’ve tried a couple of other lenses: the 400 f5.6L works well; the 100mm f2.8L macro does not work because the curve is very shallow and goes flat just 10 steps beyond best focus. So be prepared to have to spend some time determining the optimal AF parameters for each lens, and get the best focus you can manually before initiating AF.