Bullet-proof auto focus with platesolving

The pointing model is within the mount control software. SGP does not access it directly. However when you do a platesolve followed by a SYNC the pointing model may get updated/refined depending on the Sync option you choose. Personally I do not use a pointing model so I am not the best person to ask this question. At session start from Home position I just do a slew to target followed by target centering. This gets me sufficiently well aligned for subsequent slewing/centering.

I have an Optec SWX30 focuser on a Skywatcher Esprit 100 ED. Is there some way to actually calculate the ideal step size? Using trial and error for the most part. Thanks for any info.

The basic process for doing this is set out in the Help File under the heading ‘Understanding Auto Focus’. I recollect it all seemed a bit daunting at first but it is actually quite straight-forward.

I assume your focusser is working and that you are able to get somewhere near a good focus using a Bhatinov mask or by monitoring the HFR stats for a series of images taken while using the In/Out buttons on the Focus Control.

I recommend a night of good steady viewing for doing this so you are not misled by fluctuating seeing.

Once you have established a good focus position (make a note of the value so you can easily get back to your start point if needs be) take a series of images as you gradually move out of focus . You are looking to get to the point where the measured HFR value is about x3 greater than the HFR** achieved at best focus. The Help File guidance suggests 3-5x smallest HFR but I suggest x5 is too much.

Knowing your focuser start point and focuser HFR x 3 position you can calculate a step size as described in the Help File.

Plug these values into the AF set-up , go back to your best focus position and you should now be in a position to try an AF run.

To start with I suggest you check the Save Auto Focus Packages option on the Focus Control / Settings screen so you have something to look at and diagnose from if your first settings do not work out too well.

The AF curve should be reasonably symetrical. If the right hand side of the curve is limp looking then you probably need to dial in some backlash compensation.

Once you have a start point you can then fine tune the step-size etc until you regularly get good quality curves and can set up for filters etc.

** If you are using SGP v4 and the ASTAP HFD metric then the metric is Half Flux Diameter not Half Flux Radius so you need 1.5 x your lowest HFD value rather than 3x HFR.

Thanks again for the great info. New issue (repeat of an old). I have an ASI 2600MC Pro connected directly to my laptop via a USB 3 port. SGP will not download an image whether I use Take One or a sequence. I had this problem a few weeks ago in SGP, Sharpcap and even ASICap, and I got advice on a forum to disable Selective Suspend in the Windows Power PLan / USB Settings and that seemed to fix the problem. I could then download to all three programs and even did a one hour sequence in SGP. Today I set up everything up - same configuration, same laptop, same ports on the laptop, Connected ALL and tried to do a Take One - hung up downloading… Tried a sequence hung up on downloading as well. Tried .01, .1, 1 and 10 sec exposure times with same problem - just reports “downloading…” and never actually downloads. But Sharpcap and ASICAP/ASIimg both will download with no issues. Any clues? Unbelievable frustrating - basically a month of technical challenges. Thanks for any feedback.

Try to update to the newer drivers, also try the new 64bit SGP version and see if that works better.

Thanks, but 4.x is not working for merbecause I have an Optec focuser (discussed elsewhere).
So I had the 2600 cable plugged directly into the USB 3 port - would not download. Closed SGP and switched the cable to USB2 and then everything worked… Had to leave the scope for a bit so closed the laptop cover - did not turn off - and when I came back, would not download,.Switched to USB 2 port and it worked again. What’s that all about??

Thanks, Mike. My scenario is as follows - I set up my rig during the day and choose a target several hundred yards away ( a niehgbor;s chimney, for example) Using the Optec SWX30 focuser, I unlock the focuser (turn knob counter-clockwise), do a continuous take on the target and carefully find a good focus point using the gross, then fine knobs on the OTA. Once I find the best point, I lock the focuser and wait until dark. So then it’s night time and I set up my sequence, and the autofocus fails. The first HFR values I see are around 7 - 8, and SGP can’t seem to find a spot that it’s happy with. I change the step size from, say 50 to 100 to 250 and start over, and there’s no change - Quality is 20, 30, maybe 40%, but never better - so it fails.

I don’t know what other tricks to try, but this is really a showstopper for imaging. Let me know if ther;sa anything you can think of. Thanks again.

Do you recollect the lowest HFR values you have seen with your scope? A starting HFR of 7-8 is in my experience is quite a long way from best focus. How many star detections do you see get at this HFR?

If you are seeing a handful of good star detections at this level then I would use the Focus Control/Coarse In/Out controls. In one or other direction after one or two increments (of say 50 steps) you should start quite rapidly to see a significant change in #stars and HFR. Once you start to see an improvement, continue this movement until you get to a minimum HFR / max star detections. Make a note of this HFR and focus position then continue until you are seeing an HFR of about 3x the minimum HFR. Your step size should be about 1/3 (for 7 step AF) the number of steps between the current focuser position and the minimum HFR position. Plug these values into the AF set-up. return to best focus then hopefully AF will start to return reasonable results.

Regarding your set-up routine there could be a significant temperature change between your initial focus and your imaging start time. It should not too difficult when stars are visible to get very close to best focus using the manual controls as described above.

When you are next experimenting with AF, if you still have issues, please post a copy of the SGP log file so there is some actual data to review.

Mike

Mike - thanks for your note, The best I have seen with this camera is down around 1.2 - I was astounded. But after running the AF routinely independent of a sequence on that occasion, the actual run before the sequence failed. So I don;t know if the gear slipped on the focuser or what happened. Yesterday I had a very good focus during the day using the knobs on the focuser, but after locking the focuser and not messing with the scope at all, the AF again failed prior to the actual sequence - that;s where it started out as 8 and never found its way to a decent HFR value, even with me changing the step value as noted.

Total white out again tonight but I’ll send you a log next time after running the entire process again. Let me know if you can think of anything else in the meantime and thanks again.
.d

Hi,
If you have obtained good AF results on a few occasions I would say your SGP settings are probably in reasonable shape. With clear skies and a more or less in-focus start position, if good AF is not regularly repeatable then I would start to suspect a mechanical issue - something either slipping or binding. Obviously focuser position changes need to be repeatable, accurately and regularly. And we are dealing with measurements at micron level!

Not sure if you are shooting mono or OSC. I found I needed to significantly increase exposure times for RGB filters (only 25-30% of photons captured compared with L) and even more with Ha.

I eventually reduced my step-size a small amount as I found with RGB filters ,at the extremes of the AF curve, I was getting too few star detections for reliable HFR measurements. Without decent detections at the extremes then the resulting AF curve is of poor qualiity leading to poor determination of best focal position. In my opinion SGP v4 is much more sensitive in this regard when using the ASTAP focus metrics option.

Hope you get some good results very soon