Back to forums

ZWO EAF picks the wrong focus point

ZWO EAF Auto-Focusing Troubleshooting
20 replies•934 views
Ferenc Szabo avatar
Ferenc Szabo avatar
I asked this on the cloudy nights forum, but people are clueless. 

I never had an issue with autofocus for 3 years, but now suddenly I do. 

Here is the problem. This new scope - the SVBONY SV555  when it's out of focus it looks like a small central dot and a large halo around it. 

The size of the halo depend how far I am out of focus.   This is when I turned the focus too far in from the actual real focus. 

If you ever used a DSLR camera with lens, you know what I'm talking about. 

People with camera lens understands what I'm talking about? 





Asiair picks up on that central dot as "focused".  It doesn't matter how big the halo around it, it uses the central dot for measurement, it doesn't see the large halo around. 


Here is the real focus when I do it manually. Is there any technique to get around this issue or should I just go back to mini PC and NINA instead of Asiair? 

mikejwheatley avatar
mikejwheatley avatar
I would drop your exposure down, I expect 3s is too long and the halo is being read as star brightness. Try 1s. If that fails, try reducing the gain as well.
Helpful Concise
TKall avatar
TKall avatar
I generally focus manually to get close before letting the asiair take over.  In my experience if the focus is way out, the air will have trouble.  If I start with a fairly sharp focus, the air handles it well.
Helpful Concise
noon avatar
noon avatar
Check to make sure the grub screws that hold the adapter to the output shaft of the EAF are tight. They can work loose over time. I had the same issue and I found that the grub screws had worked slightly loose, generating enough backlash to affect the focus accuracy.

Once I tightened them up, the backlash went away and my focus got tight again. Much like you it seemed to happen suddenly after about 18 months of perfect operation.

If this is what is happening, another symptom that you might notice is that your "V" curve will get flatter. Once you tighten the grub screws your "V" curve should get more "V" like and less "U" like… if that makes sense
Helpful Respectful Engaging Supportive
Doug Crowe avatar
Doug Crowe avatar
You say that this started with the new scope. I had no problem with my scopes focusing with the asiair and eaf, but when I switched to a camera lens for wide field it would not focus. I lowered my step count from 70 to 30 and that solved the problem.
Helpful Concise
Ferenc Szabo avatar
Ferenc Szabo avatar
Doug Crowe:
You say that this started with the new scope. I had no problem with my scopes focusing with the asiair and eaf, but when I switched to a camera lens for wide field it would not focus. I lowered my step count from 70 to 30 and that solved the problem.

  This sounds like a good idea!  I only tried increasing the step size, but that didn't work , I guess I should try to decrease it. 

For those people saying that my grub screws loose, I just put this assembly together literally before trying it out. All components are brand new. 

The only difference to my other scopes that the SV555 scope has a helical focuser and the EAF is belt driven, from an EAF bracket under the scope. 



I never used a system like this before, but I do know I made sure that everything was tight and the focuser would move the telescope helical focuser around.
TKall avatar
TKall avatar
Thinking about it,  when stars are out of focus in any of my scopes they are big, circular orbs of relatively uniform brightness.  I have never seen a halo around a central disk like you have shown.  My guess, based on this, is that the out of focus star shape is a function of the scope and the air doesn't know how to deal with it.  I wonder if that new Askar55 does the same thing.  Maybe try picking a smaller star to focus on by clicking on it?
Helpful Insightful Respectful Concise Engaging Supportive
noon avatar
noon avatar
Thinking about it,  when stars are out of focus in any of my scopes they are big, circular orbs of relatively uniform brightness.  I have never seen a halo around a central disk like you have shown.  My guess, based on this, is that the out of focus star shape is a function of the scope and the air doesn't know how to deal with it.  I wonder if that new Askar55 does the same thing.  Maybe try picking a smaller star to focus on by clicking on it?

...Or shortening your autofocus exposure length if you have an extremely bright star in the frame perhaps?
Ferenc Szabo avatar
Ferenc Szabo avatar
Thinking about it,  when stars are out of focus in any of my scopes they are big, circular orbs of relatively uniform brightness.  I have never seen a halo around a central disk like you have shown.  My guess, based on this, is that the out of focus star shape is a function of the scope and the air doesn't know how to deal with it.  I wonder if that new Askar55 does the same thing.  Maybe try picking a smaller star to focus on by clicking on it?

I'm not sure, but since I was using a 533mmpro camera  with RGB HaO filters  (5 filter wheel)  and I know for sure that those filters (Optolong filters) work good on all my other telescopes and with Asiair.  So the camera+filters is definitely a tried and working combo, so it comes down to the lens itself. 

This is my first experience with a helical  - Petzval telescope and I'm getting suggestions from many people - one recommending to drop the step size and drop the exposure time, others saying the opposite. 

I also want to try out using an OSC camera and see what happens.  Anyway, good thing that this is my secondary setup, so I actually had my main imaging rig running while I was tinkering with this new rig. 
Also I tried to blame the atmospheric conditions, but my other rig with the OSC camera and RC scope shown no sign of star halos or high clouds running and imaging at the same time when this was running.
TKall avatar
TKall avatar
I have three scopes with helical focusers, FMA 230, FMA 180 and FMA 135….Never again.  My scopes don't do what yours is doing but I don't like the belt driven focuser because it gets so cold here and I just don't trust it.  I was really interested in looking at the new 55's but the helical focuser made me reconsider.  Oh yeah, I have two EvoGuide scopes with helical focusers and those focusers are kind of awful.

I hope you get it figured out.
Merry Christmas
Vroobel avatar
Vroobel avatar
I also have Askar FMA230 and the EAF fastened using DeepSkyDad's solution. It works well, no problems at all, but maybe my low temperatures aren't enough low? The only strange I think is the 'V' curve which isn't 'V' or "U', it looks rather like 'L', the right side is much flatter than the left one. Anyway, I'm getting an expected star size.
Rob avatar
Rob avatar
It sounds like the ASIAir software is selecting the centroid (similar to autoguiding) instead of determining the half flux diameter (HFD) of the stars. 🤔  Is this all that's available for adjustable settings?



Edit:  If it's stuck using the centroid, I would definitely lower the step size so it's not skipping past the actual focus while searching.
Well Written Insightful Respectful Concise Engaging
Chris Ashford avatar
Chris Ashford avatar
I had similar problems when I bought a new scope - my ASIair/ZWO EAF, which had never failed before, would fail to find the focus point. I fixed it by adjusting the autofocus coarse step size using trial and error, concentrating on the shape of the AF curve. Make sure that curve is a smooth U shape. If the coarse step size is too small, the samples will just go up and down randomly due to seeing variations but the ASIair will calculate a fit curve anyway, and give you an erroneous result. If too large, the ASIair cannot accurately calculate the correct fit curve resulting in a significant error. I hope this helps!

Clear skies,

Chris
Helpful Insightful Respectful Concise Supportive
John Favalessa avatar
John Favalessa avatar
in addition I would calculate the backlash and enter that into the EAF settings.  Search YouTube for  Backlash Calibration and Tuning.  -john
NeilM avatar
NeilM avatar
Deleted
Chris Ashford avatar
Chris Ashford avatar
John Favalessa:
in addition I would calculate the backlash and enter that into the EAF settings.  Search YouTube for  Backlash Calibration and Tuning.  -john

That’s a good idea. I have noticed that there is quite a lot of backlash on one of my scopes and the focus point certainly changes depending on which direction I am moving the focuser. If focusing manually I always make sure that I am approaching the focus point from high to low.
Well Written Concise
Wim van Berlo avatar
Wim van Berlo avatar
I think that what you're seeing is the diffraction near focus. Have a look at this article. https://www.telescope-optics.net/diffraction_pattern_and_aberrations.htm
Figure 94 shows the issue. The longitudinal cross-section shows the star profile when focus is changed. The centre of the figure is at best focus. At this point, the star profile is somewhat wider than the central part of the defocused star image. This is because light goes into the side lobes (a ring in the cross section star profile). The overall star profile is of course wider when defocused. In essence, this is a software issue, or rather, a flaw in the implementation of how the star profile is to be interpreted. The reason why it doesn't show up in all imaging systems has to do with additional aberrations in the optics, and probably imaging conditions.

Wim
Well Written Helpful Insightful
Chris Ashford avatar
Chris Ashford avatar
Wim van Berlo:
I think that what you're seeing is the diffraction near focus. Have a look at this article. https://www.telescope-optics.net/diffraction_pattern_and_aberrations.htm
Figure 94 shows the issue. The longitudinal cross-section shows the star profile when focus is changed. The centre of the figure is at best focus. At this point, the star profile is somewhat wider than the central part of the defocused star image. This is because light goes into the side lobes (a ring in the cross section star profile). The overall star profile is of course wider when defocused. In essence, this is a software issue, or rather, a flaw in the implementation of how the star profile is to be interpreted. The reason why it doesn't show up in all imaging systems has to do with additional aberrations in the optics, and probably imaging conditions.

Wim

Thanks for sharing that interesting article - I learned something
Monty Chandler avatar
Monty Chandler avatar
Undoubtedly go back to the mini PC and try APT.  Fully supported functional software.  Then you can use a better auto focuser that doesn't require a bracket or a belt.  InvPwr is a game changer for auto focus - background versus a star.  Cheers!
Ferenc Szabo avatar
Ferenc Szabo avatar
I tried using NINA and hocus focus. 

To me, this is start to look like a scope problem, not a focuser or Asiair problem. 

I'm not even sure what's up, to me the stars appears to be bloating at focus point anyway. 






Anyone interested in a test fits file with the best focus via NINA then see the attachment and let me know what you think: 

https://drive.google.com/file/d/1xwoPPsQloW4IjU1YtB9AOQIevnmOf09T/view?usp=sharing
NeilM avatar
NeilM avatar
x
Related discussions
Lessons learned & tips from building an observatory (remote or not)
Hi all, I wanted to start a topic specifically for those of us who either have an observatory or are thinking of building one be that remote robotic or not. Whether you're in the planning stages or already running a fully automated system, I beli...
Jun 9, 2025
Both posts are incomplete, with the first cutting off mid-sentence while describing a telescope focus issue and the second ending abruptly in the middle of introducing an observatory discussion topic.