SharpCap Forums

Hi,

I've been using SharpCap Pro for about nine months. I feel pretty comfortable with it and I've got a lot of use of the Polar Alignment, plate solving, and imaging settings. Recently I decided to get an autofocuser (ZWO EAF) having previously used the manual knob (Celestron 8" tube on a Celestron Advanced VX GOTO EQ).

I installed it using the dedicated bracket for this type of telescope. The initial checks show it seems to "work" - instructing the software to adjust focus produces a noise from the focuser, and the image (ASI 224MC) does alter.

However, it remains consistently very out of focus and I'm not sure how to get it "near" to focus well enough to precision tune it. I have tried to use the autofocus in Sharpcap (Contrast Edge Detection) on the moon and Jupiter. I get a very flat curve, but in a couple of cases the software identifies "best focus" and adjusts accordingly - but the image remains unusable.

A further complication is that the image seems pretty blown out (the black circle with the bright outer rim of a completely out-of-focus telescope) it cannot track the object, so I'm having to do a lot of manual readjustment to the tracking as I go. But when I make huge adjustments to the focus via the sharpcap interface the image changes exactly as I'd expect if you were changing the dial manually, so it seems properly attached to me.

What I'm mostly trying to figure out here is - what's the best way to get a "good enough" focus so that the tracking etc. can work while I fine-tune it? Or does this sound so bad that it might imply installation error?

Hi,

the first thing to note is that SharpCap's autofocus routines are designed to get you from 'decent focus' to 'best focus' - they are not going to help much if you are *way* out of focus, but from time to time there might be enough pattern in the graph that gets generated to make them think they are at best focus.

So, the trick is to use the Negative and Positive buttons in the SharpCap UI to get the initial focus done just like you would if you were focusing manually until you see a reasonably focused image of the planet. You can either set the 'Coarse' step to something large like 1000 and then keep pressing one button or the other, or you can just type a really big (or small) value into the box where the focuser position shows, press <ENTER> and watch SharpCap scan through 1000s of steps - press 'Stop' if you see a good image.

Since you are using an SCT, you have a harder job than someone using a Newt or refractor though...

1) Mirror shift - as you move the focuser one way or another, the mirror will tilt slightly and that moves the planet in the field of view (hopefully not out of it entirely) - best to limit movements to one direction by setting up 'Overshoot' backlash compensation in the SharpCap settings

2) Lots of backlash - you will need a fairly high value for backlash compensation - fortunately with 'overshoot' a too big value doesn't cause problems except everything taking a bit longer (as long as not massively too big)

3) A very large focus range, which means that it is easy to be lost far from focus. If you are way out of focus, you may see Jupiter (or any bright star) as a donut of light (maybe needing the display stretch in SharpCap turned on to see that). If you do see a donut, adjust the focus to make it smaller and you will be moving in the right direction.

If none of that helps, set the scope up in daytime, point it to a distant object, ideally on the horizon (tree, house roof, etc) and get focus on that - plenty of time and you can see what you are doing. Then leave the focus in that posuition as it will be pretty close (not exact) to what you need for the sky.

Hope this helps,

Robin

Thanks very much.

I tried this tonight using Jupiter, and I got to the point where it was planet-like on the camera, though with a tiny donut hole.

This was definitely the problem - having started with the autofocus function it found a "best focus" through, I guess, random noise. The actual roughly in focus point was very far away.

However, I wasn't able to get it completely in focus as I reached the outer limit of the autofocuser, so it wouldn't turn any more.

I assume that when I installed it, I somehow managed to connect it to the knob in such a way that the knob was turned to one extreme and so the autofocuser could not turn it all the way to proper focus? And if I reinstall it with the knob in a "central" position it ought to stop that?

Not sure if this is helpful but I made sure the physical focuser was in the middle of its range before attaching the EAF. In the ASI mount app, I set the focuser number of steps to 60,000, then set the position to 30,000 so the EAF is at the middle of its range. This gives me plenty of movement in both directions from within SharpCap. Hope this helps.

davemartin88 wrote: Wed Feb 26, 2025 1:13 pm Not sure if this is helpful but I made sure the physical focuser was in the middle of its range before attaching the EAF. In the ASI mount app, I set the focuser number of steps to 60,000, then set the position to 30,000 so the EAF is at the middle of its range. This gives me plenty of movement in both directions from within SharpCap. Hope this helps.

Yup same here.

Hi,

yes, good points there about the range of the autofocuser - it could be that you ran out of range in terms of numbers - 0 to 60000, rather than range in terms of how far the physical focuser knob will turn. I think you can use the 'Set Zero' button in the EAF focuser properties to change the current value and give yourself more movement range - you can also change the maximum limit value too.

cheers,

Robin

Thanks all.

I used the "Set Zero" button and it worked perfectly. I was able to get it focused properly - issue solved.

Hello.

Robin, I saw in one of your replies that you might need a large backlash compensation for an SCT. Can you quantify that for the EAF? Or can one of the forum members who has an EAF with an SCT share what they use?
I sometimes struggle with inconsistent results using the autofocus. I will get different results from the focus routine depending if I am moving forward or backward with the focuser. I do have a backlash number in there, but now I am thinking it is too small and wondering what it should be.

Thanks in advance everyone,

Keith

Hi Keith,

exact values are impossible to give, since they vary not only depending on the exact model of telescope, but also on the model of motor focus you are using (some have bigger steps than others, so require a smaller setting for the backlash allowance).

One way to determine is to run a focus scan in each direction (turn *off* any backlash settings first, and approach the start of the scan with a good, long movement in the direction you intend to scan). Maybe you will find best focus at 41200 when moving in the positive direction and at 40800 when moving in the negative direction - that then gives you a backlash figure of (41200-40800) = 1400. If you are using the recommended 'overshoot' correction for backlash then its worth setting the value a bit higher than the real amount of backlash, as there is no harm in the value being higher and it helps to stay on the safe side.

cheers,

Robin

kaymann wrote: Wed Feb 26, 2025 3:14 pm

davemartin88 wrote: Wed Feb 26, 2025 1:13 pm Not sure if this is helpful but I made sure the physical focuser was in the middle of its range before attaching the EAF. In the ASI mount app, I set the focuser number of steps to 60,000, then set the position to 30,000 so the EAF is at the middle of its range. This gives me plenty of movement in both directions from within SharpCap. Hope this helps.

Yup same here.

It looks like you guys need to upgrade the firmware on your ZWO focuser (how to do that is probably beyond the scope of this forum). 60,000 steps is the original number of my focuser from early 2024. The latest firmware goes to around 600,000. My current focus position on Jupiter is ~74,250 or so.

I am here to sort out autofocus of the planets which works so-so for me.