Hey, I'm going to probably sound like a 5 year old but here goes. I recently installed a ZWO EAF on my AT102ED refractor. I'm running the latest version of SharpCap Pro and I'm trying to figure out how to get SharpCap to go to the best focus point. I've seen someone use NINA to do autofocusing and in that particular setup the software generated a graph with the various focus points plotted. I thought I was going to see something similar and perhaps I would if I knew which setting to set, perhaps not. I figured since my target was M33 the best method for focusing would be Multi-Star FWHM, but I admit, that is a guess. No experience here to back that up. I've attached 2 screen shots which were taken at different times.
I could use a little bit of a nudge in the proper direction. What am I doing wrong or what do I do next?
Trying to figure out focusing with ZWO EAF
Trying to figure out focusing with ZWO EAF
You do not have the required permissions to view the files attached to this post.
Re: Trying to figure out focusing with ZWO EAF
So I figured out that I needed to click the focus scan buttons. I clicked positive and let it run for awhile, then stopped it when i saw that it was getting too far away from the point where I had focused using a Batinov mask. I clicked negative and let it run completely in the other direction. This is what I got. I also displayed the Options to show what they were.
I tried clicking the Best Focus drop down, but nothing changed.
I just figured out how to manually focus the focuser.
I tried clicking the Best Focus drop down, but nothing changed.
I just figured out how to manually focus the focuser.
You do not have the required permissions to view the files attached to this post.
Re: Trying to figure out focusing with ZWO EAF
I toyed with the idea of deleting this entire thread because my learning curve has been steep and rapid, but then thought better of it since someone else may find it useful... I have to admit that I've been thrashing about trying to figure things out but they are starting to settle into my dull brain. Here's the two things I've learned so far that were causing me grief:
I can manually focus the EAF via the EQMOD HEQ 5/6 ASCOM control panel I have set up on my laptop. It is used by all of the astronomy programs I use, SharpCap Pro, PHD2, Cartes Du Ciel, etc. so the focusing is always available. The focusing section is located at the bottom of the control panel under Scope Controls.
A lot of the initial parms and the procedure to do an auto focus can be found here on Cloudy Nights. I was really stumbling around when I originally posted this thread as anyone who knows what they're doing can see.
https://www.cloudynights.com/topic/9074 ... uccessful/
I can manually focus the EAF via the EQMOD HEQ 5/6 ASCOM control panel I have set up on my laptop. It is used by all of the astronomy programs I use, SharpCap Pro, PHD2, Cartes Du Ciel, etc. so the focusing is always available. The focusing section is located at the bottom of the control panel under Scope Controls.
A lot of the initial parms and the procedure to do an auto focus can be found here on Cloudy Nights. I was really stumbling around when I originally posted this thread as anyone who knows what they're doing can see.
https://www.cloudynights.com/topic/9074 ... uccessful/
-
kaymann
- Posts: 188
- Joined: Thu Jun 16, 2022 2:59 pm
- Location: https://azkayfes.blogspot.com/
Re: Trying to figure out focusing with ZWO EAF
Hello Rich,
I am the the one who posted that thread on Cloudy Nights. The key to my thread is two points. One, you have to have a crystal clear idea of backlash and of how to get behind it so you are always pushing focus. I could say the same about pulling focus but why get confused with too many options. Two you have to start your autofocus journey just out of focus on the negative side so are pushing to get to perfect focus.
One of the hardest concepts some people miss is the auto focus routine will probably not actually hit perfect focus as it progresses through it'ss hunt for better focus points. But the graph produced by the all the misses and near misses will point to that perfect focus point. So why start out of focus on the negative side - so SC can be confident it is beyond any backlash you may have hanging around. Then it can start building that left hand side of the graph and then move onto building that right hand side... Being on the correct side of the graph and having sucked up the backlash are the key to success with autofocus.
So to summarize I get near eyeball focus I then shoot out on the negative side at least 3 times the number of backlash turns to get past that point and then push the focus in positive 1 times backlash to get the focuser resting tight on the gears. Then I conduct my auto focus with SC now pushing all focus with tight gears.
One other thing. SharpCap does have an auto focus routine based on the numbers it gets from a routine with your Bahtinov mask installed. You may have better luck with that routine.
I am the the one who posted that thread on Cloudy Nights. The key to my thread is two points. One, you have to have a crystal clear idea of backlash and of how to get behind it so you are always pushing focus. I could say the same about pulling focus but why get confused with too many options. Two you have to start your autofocus journey just out of focus on the negative side so are pushing to get to perfect focus.
One of the hardest concepts some people miss is the auto focus routine will probably not actually hit perfect focus as it progresses through it'ss hunt for better focus points. But the graph produced by the all the misses and near misses will point to that perfect focus point. So why start out of focus on the negative side - so SC can be confident it is beyond any backlash you may have hanging around. Then it can start building that left hand side of the graph and then move onto building that right hand side... Being on the correct side of the graph and having sucked up the backlash are the key to success with autofocus.
So to summarize I get near eyeball focus I then shoot out on the negative side at least 3 times the number of backlash turns to get past that point and then push the focus in positive 1 times backlash to get the focuser resting tight on the gears. Then I conduct my auto focus with SC now pushing all focus with tight gears.
One other thing. SharpCap does have an auto focus routine based on the numbers it gets from a routine with your Bahtinov mask installed. You may have better luck with that routine.
Last edited by kaymann on Mon Dec 02, 2024 7:28 pm, edited 2 times in total.
-
admin
- Site Admin
- Posts: 14836
- Joined: Sat Feb 11, 2017 3:52 pm
- Location: Vale of the White Horse, UK
- Contact:
Re: Trying to figure out focusing with ZWO EAF
Hi,
a good thread to keep, and great comments from @kaymann on the need to understand backlash before trying to autofocus.
My first experience of astro auto-focus was with another piece of software (I won't say which). It had just added an autofocus feature, so I tried it. It took some images, fiddled around with the focuser position a bit, took some more images and announced that it had succeded in autofocusing. It was miles out, and worse than just being wrong, there was no information that I could see that helped me understand why it had failed (no progress, no graph, etc). I'm fairly sure (with hindsight) that the failure was due to the movements only being in the backlash range, so there was actually no change to focus happening, but random variations in the FWHM number were being picked up as 'best' focus. That autofocus feature in the other software obviously improved over time, but that one experience helped me with the design of SharpCap's autofocus, giving me three key things:
* Graphs showing what was going on so you might have a chance of understanding when things go wrong
* Backlash compensation/handling code to aim to either have backlash compensated out or always work in one direction
* Checking the statistical robustness of any U or V curve to try to ensure SharpCap doesn't pick up random patterns in the FWHM measurement as a 'best fit curve'.
cheers,
Robin
a good thread to keep, and great comments from @kaymann on the need to understand backlash before trying to autofocus.
My first experience of astro auto-focus was with another piece of software (I won't say which). It had just added an autofocus feature, so I tried it. It took some images, fiddled around with the focuser position a bit, took some more images and announced that it had succeded in autofocusing. It was miles out, and worse than just being wrong, there was no information that I could see that helped me understand why it had failed (no progress, no graph, etc). I'm fairly sure (with hindsight) that the failure was due to the movements only being in the backlash range, so there was actually no change to focus happening, but random variations in the FWHM number were being picked up as 'best' focus. That autofocus feature in the other software obviously improved over time, but that one experience helped me with the design of SharpCap's autofocus, giving me three key things:
* Graphs showing what was going on so you might have a chance of understanding when things go wrong
* Backlash compensation/handling code to aim to either have backlash compensated out or always work in one direction
* Checking the statistical robustness of any U or V curve to try to ensure SharpCap doesn't pick up random patterns in the FWHM measurement as a 'best fit curve'.
cheers,
Robin
-
kaymann
- Posts: 188
- Joined: Thu Jun 16, 2022 2:59 pm
- Location: https://azkayfes.blogspot.com/
Re: Trying to figure out focusing with ZWO EAF
Here you go for a simple PH -
Assumptions (Once and Done):
1. SC > Multi-Star Auto Focus > Multi-Star FWHM Measurement > Options
a. Scan Step Size 50
b. Max Step Count 10
c. Samples to collect at each Step 1 or 2
d. Checkmark > Move to best focus position when scan complete
e. Checkmark > Use Smart Scanning
2. SC > Multi-Star Auto Focus > Multi-Star FWHM Measurement > Star Detection
a. Checkmark > Suppress Hot Pixels
b. Checkmark > Use Display Stretch
SC Auto Focus PHD
1. Using SC Live View aim for a an area of at least 15 or more moderate magnitude stars
2. Click on SC > Tools > Focus Assistant > Multi-Star FWHM Measurement
3. SC > Sidebar > ZWO Focuser > Checkmark (in Connected Box)
4. Using various SC focusing buttons, quickly focus stars based on size and or Scores Now close to Zero
5. Using SC focusing buttons click negative button until you have progressed at least 500 negative steps
6. Click SC > Focus Scan > Positive
7. Enjoy
Assumptions (Once and Done):
1. SC > Multi-Star Auto Focus > Multi-Star FWHM Measurement > Options
a. Scan Step Size 50
b. Max Step Count 10
c. Samples to collect at each Step 1 or 2
d. Checkmark > Move to best focus position when scan complete
e. Checkmark > Use Smart Scanning
2. SC > Multi-Star Auto Focus > Multi-Star FWHM Measurement > Star Detection
a. Checkmark > Suppress Hot Pixels
b. Checkmark > Use Display Stretch
SC Auto Focus PHD
1. Using SC Live View aim for a an area of at least 15 or more moderate magnitude stars
2. Click on SC > Tools > Focus Assistant > Multi-Star FWHM Measurement
3. SC > Sidebar > ZWO Focuser > Checkmark (in Connected Box)
4. Using various SC focusing buttons, quickly focus stars based on size and or Scores Now close to Zero
5. Using SC focusing buttons click negative button until you have progressed at least 500 negative steps
6. Click SC > Focus Scan > Positive
7. Enjoy