I’m having problems and some trust issues with the Auto focus. I spent around 4 hours a few nights ago manually refining my focus on each filter and the entering the counts in “The filter setup” window.
For an example Green is set at 1146, with a HFR of around 1.3, but when auto focus kicked in it moved it to around 30 steps away from this point and gave me they green X It then started to run the sequence. I checked the 1st sub and the stars were well out of focus with a HFR of around 2.4
I’m sure most of the above problems can be overcome with a little more education on auto focusing an so here are my questions…
1 Could someone please give me some tips on how to set the “Auto focus options” window, or point me to a help file.
2 Why does SGP give a green X, when it’s not at its optimum focus?
3 Does bahinov grabber work with SQP and how do you use it? I’ve checked youtube and there appears to be a good video but it’s not in English. Any links welcome.
Sorry if my questions are already on the forum somewhere and thanks for taking the time to read my post.
Darren,
It might be helpful to know your equipment. Different equipment can behave in different ways when it comes to autofocus. BTW if you briefly list your equipment in the “About Me” section of your profile, when people click on your avatar that info will show up in a tooltip.
Just to make sure we’re on the same page, when you enter the focus position numbers in the filter setup window, those numbers only come in to play when also using the “auto adjust focus per filter” in the focus tab on the control panel, and when choosing to focus with one filter (usually the LUM filter). So when autofocus runs it will use the LUM filter to focus, and then calculate the offset for whatever filter you want to image with. For example, lets say your LUM filter # in the filter setup list is 2000 and your Red filter is set at 2030. With the above options enabled when autofocus runs it will switch to the LUM filter, calculate focus (lets say it’s 1960), and then when switching to the Red filter it will take the current focus point for the LUM filter (1960) and then add 30 steps because thats the difference (offset) between the filters in the filter setup (LUM=2000, Red=2030). So the new focus position for the Red fitler will be 1990 (1960+30offset).
The reason I say all that is because it sounds to me like you are expecting the focus position of your green filter to always be at 1146 which won’t happen because of temperature differences etc. I could be way off in assuming this but I wasn’t clear from your description, so just making sure.
Let me know if I’m way off and we’ll go from there.
Here is a screen shot of my auto focus settings, just for reference.
I know I had the “Auto adjust per filter” ticked and I’m almost certain I had “Auto focus with filter Lum” ticked. However, when focusing didn’t play ball I did un-tick everything to do with Auto focusing, so I cannot be 100% sure. I’ve re-applied these settings this morning and so I should know on my next clear night if everything’s ok.
As per your suggestion, I have listed all of my equipment in “About me” and below.
Takahashi FSQ 85, With Reducer
Focal length 328mm
Focal ratio F/D 3,86.
Lakeside motor Focuser.
Atik 490EX H3380 x V2740 pixel size 3.69 x 3.69
Atik EFW 2 Filter wheel. LRGB, HA,Olll & Sll 1.25" Baader filters
OAG with Starlight Xpress Lodestar.
Avalon fast Reverse Mount.
I’ve also taken some screen shots of my focusing windows. Could you please give me some advice on the settings in general, especially the step size. In this screen shot it’s 50, but I’ve now reduced it to 10 (I have now reason to pick 10, I’m just playing).
The good news is that you’ve got great equipment and you should be able to get nice “V” curves for accurate focus. Here are some things I would try:
Please enable saving autofocus packs, so we can take a look at what you are seeing if problems arise:
As you have already noted, you may have your step size too large at 50 steps. Do you happen to know approximately how many steps there is in the full range of motion? When the focuser is racked all the way in (at 0 step) how many steps will the focuser report when extended all the way out? For example, my travel range is around 40,000 steps, and I have my step size set at 100. So for your focuser which I’m guessing has many less steps, 50 would probably be too large. This will take some experimenting.
Increase your data points to 9.
On the control panel, you do not have “Use auto focus” enabled. You’ll need to enable that…
Sorry for the late reply, but thanks for taking the time Joel to help me out. One thing I can assure you is that “use auto guide” was enabled. I just un checked it when it went a little pear shape. Could you please explain / expand on how the step size works.
I’m not exactly sure what you mean by “explain how the step size works” but here goes. If you have 9 “data points” and a step size of say 25, when a AF run is called for SGP will move the focuser out by 4 data points, in other words 100 steps of the focuser (4x25). SGP will then take an image (based on your exposure settings for filters found in the filter setup) and measure the Half Flux Radius (HFR) of the stars. Then SGP will move the focuser IN 1 data point or 25 steps of your focuser and take another image, which should hopefully be more in focus than the first data point. SGP will keep repeating this process (moving the focuser in by 25 steps at a time, take an image, measure HFR) until 9 data points have been reached. So in a perfect AF run you will get a “V” curve with 4 data points on each side of the bottom most point.
Here’s the rule of thumb we recommend: get as close to focus as you can before you start AF, using a bahtinov mask or just eyeballing. Set your step size so that your first data point (the right top of the “V”) is approximately 3x what your bottom most data point is (where good focus is). So let’s say that your in focus HFR measurement is 1. You want to adjust your step size so that when AF runs and the focuser is moved to the outermost position, the resulting HFR measurement for that data point is about 3 (3x1).
