I have been having much trouble obtaining rough focus (good enough for autofocus to be usable) with my ZWO-EAF autofocuser. Today I thought it might be a good idea to finally read the driver manual . It was instantly clear that I had not done my homework as described in the driver doc:

When the focuser is first connected and the ASI EAF driver used for the first time, it is recommended to configure and save a profile that synchronises the ASI EAF driver to the focuser, and protects the device and the telescope by limiting the range of motion of the focuser to the limits of the telescope's drawtube.

However this brought up as many questions as it answered, and I found I did not understand the documented procedure. Evidently, something has changed since the document was originally written.

Establish the furthest "in" position. This is the "0" position.
Move the telescope drawtube as far "in" as it goes, being careful not to force it past its limit. This can be done from the Main Control tab, by setting Direction -> Force In and set Relative Position to a small number and press Set. You will need to adjust the Relative Position number to a reasonable number through trial and error. Start small and increase. Repeat to move the focuser to the furthest "in" position.

The problem is that there is no Direction -> Force In: not in my version and not in the screen shot included in the document. Therre are Direction -> Focus In and Direction -> Focus Out buttons but they don't seem to do anything as far as moving the drawtube is concerned. They don't actually appear to do anything at all. Was the label changed with some earlier version, and what, if anything, doe these buttons do?

So my question is what is the right way to set the thing up? Should I manually move the tube as far in as it goes (counterclockwise in the case of my Celestron 8" SC telescope,), and then move it a few relative steps and call that 0? Or is there not actually any way to synchronise the ASI EAF driver to the focuser?

Also, the driver doc says, echoing the ZWO-EAF manual, that it supports a maximum of 5760 steps. This is quite a bit different from the default of 100,000. Should I set the maximum to some number slightly smaller than 5760, or are we talking about totally different units of measurement here?

Hi Steve,

The setup process syncs the range of travel of the focus mechanism with the EAF so it doesn't try and drive the mechanism beyond its limits, and allows absolute positioning, e.g. goto position 1234. By convention the furthest IN position is set to 0 and the furthest out is set to whatever your telescope supports. So you need to set the device to its furthest in and set this to 0 and then set the max limit for the furthest out. Its a while since I've done this so I can't exactly remember what the labels are called.

You don't technically need to allow the full range of motion, you can set 0 wherever you want and things would work fine providing the focus point is within the range you've allowed. However, on first setup you won't know where focus is so its best to allow the full range of motion.

As to what is IN and what is OUT, you want the last focuser movement (you said Anticlockise) to be IN and this is how the focus algorithm works.

The Direction->Focus In/Out buttons allow you to say move 5 ticks and the In/Out setting determines which way to move the 5 ticks. Once you've calibrated the focuser (0 and max as mentioned above) you can use absolute positioning, e.g. goto 1234 and don't need to use relative positioning e.g. move 5 ticks IN.

5760 is the number of ticks the EAF has in 1 complete revolution. How many revolutions you need to cover the range of motion of your focuser you'll have to determine yourself.

Thank you, John. You’ve explained what the focus in and focus out buttons do, but I think it’s even simpler. They set the direction for (or do they actually perform) the relative motion, moving in or out the relative motion number. This should clearly be explained in the documentation but isn’t. Whatever the old Force In/Force Out did, it no longer does. Also the 5760 steps per revolution ought to be in the documentation but isn’t. I know, the same problem is found in the ZWO-EAF manual.

One thing I notice here is that a lot of this functionality is duplicated in Ekos, which might be convenient but might also be a little confusing. More typical is doing configuration in the driver and executing actions in EKos.

Yesterday’s experimentation convinced me of something I have been avoiding considering for too long. There may be something mechanically wrong with my setup causing some all this diffuculty - serious or easily fixed I won’t know until I take it apart. I bought the EAF from Starizona and when I did, they also sold me a gizmo that allowed hand adjustment. I didn’t realize that using this plays hell with any previous calibration. Had I known that I would not have bought it. Too many moving parts to go wrong. Much better would be to have bought the ZWO-EAF hand controller, which I now have.

So I need to take it all apart and see what the problem is. I should have thought of this earlier but I didn’t want to go there. The EAF worked well when I first got it, until it didn’t.

Yeah, the EAF is pretty straight-forward, no encoders or anything to tell where it is. It just moves between 0 - max value and assumes the thing its moving... drawtube or focus mechanism moves appropriately. If there's any slippage of the drawtube or you disengage the EAF from the focus mechanism and move it independently then the EAF won't compensate for this movement the next time it's run.

I'm not sure why you want to disengage the EAF and move the focus independently, maybe manual focus? Anyway, that won't work very well (unless you're able to set the focus mechanism back to where it was before re-engaging the EAF). If you did it approximately it would probably be OK but first focus would be a bit problematic, which maybe is what your original issue is?

If you're able to just use the EAF / Ekos for Focus even if you're observing manually then that would avoid this issue but might bring other issues (not sure exactly what you're trying to do). If it is manual observing I might try an Optical Train with the CCD Simulator and EAF. That way you could move the focuser using Ekos whilst manually observing, e.g. in 100 ticks, out 20 ticks, etc etc. (You obviously won't be able to Autofocus with this setup as there's no camera to feed data to Focus).

You ask what I'm trying to do. Basically, I'm trying to get to a point where what for others is a relatively simple procedure, focusing, is no longer a source of intense aggravation for me. When I first got the EAF it wasn't too bad, but something happened and then it became difficult to impossible to use. I now suspect something mechanical. Even daylight hand-focus on local objects can be difficult to impossible for me. I recently purchased a 2" eyepiece which made it possible for me to manually focus on local objects. No hope with a 1.25. Then I started to realize that using the hand-focus addon that Starizona sold me disrupts the software tracking of INDI-based apps. The hardware has no mechanism to tell software "where it is now". Instead, the hand controller that EAF sells seems like a better choice for an EAA or Astrophotography setup.

I now think that something is wrong mechanically with the setup. I am going to rip it down and get back to basics, especially now that I understand the software side better. The thing worked all right when I first got it, then stopped.

As for my remarks on documentation they are for those who maintain it. They can be confusing, as I tried to show. If it is okay with the maintainers, I would like to have a try at reworking those, and those who maintain it can decide whether or not it gets published.