A lot may depend on your focusing settings and this can depend on what focuser you are using and what focus motor you have attached to it.
For me, I think I use the Polynomial algorithm, and my intial step size / max step size and max travel are limited to quite low values, but big enough to notice a difference in HFR (by the time I am in focus and ready to start the scheduler I should not need to move the focus very much to correct focus for changes in temperature, sky position etc....)
I guess if some of these values are too big then the focuser could move way out of focus and back in again, and that, of course, is going to affect guiding.
"Steps" are different for different focusers and set ups, so it might make sense to check with others who are using the same type of equipment as you.
You focus using your imaging camera, and that focus is affected by the filter. The focus on the guide camera will therefore change when you focus your imaging camera (your focuser is in front of both cameras). If you don't have the guide camera well focused then it will move focus with an offset. This isn't always a problem, and some say guiding is better when slightly out of focus, but if you are not somewhat par-focal then with some filters the focus on your guide camera could be too far to guide properly when focused with your imaging camera.
I agree it is difficult to get the imaging and guide cameras using that grub screw, which is why I use the helical-focuser that ZWO supply as an accessory, it makes getting your guide camera par-focal very easy.
As per my review on the link I gave:
"If you've tried to adjust OAG focusing in the past with an alan key in the dark you will know how annoyingly tedious it can be and the worst thing that can happen is the camera and OAG fitting can just fall off to the ground if you don't support it while adjusting.
With the Helical Focuser getting parfocal is much easier and faster:
1. Use a Tablet with ASICAP, connect an OTG to both the Guide and imaging CCD.
2. Focus the Imaging CCD with the main focuser.
3. Switch ASICAP to the Guide camera, and focus the guide with the helical focuser."
With any given filter when the main imaging camera is focused correctly, is the guide camera also focused, or is it sometimes out of focus with some of your filters?
I use one of these on my OAG www.firstlightoptics.com/zwo-accessories...helical-focuser.html
It helps me easily adjust the focus.
I set it so that guide camera is completely in focus when we're using a filter that has an offset that's nearest the average of all the filters.
I have a similar set up, but all my filters are supposed to be "par-focal", and while the offsets are actually a little different from each other I don't lose guiding when re-focusing (I tend to use a low max-travel, not too high step size etc...).
It may also be that your OAG camera is at the limit of focus when your main imaging camera is focused, try checking that the guide camera is par-focal via the OAG to your main imaging camera?
I think the sestosenso2 driver supports the esatto.
With my Sestosenso 2 I find that I have to manually select the serial port for it to work correctly. Otherwise the driver connects to the wrong serial port and kind of hangs, even if I stop INDI completely the indi_sestosenso2 is still there running.
Perhaps it is the same case for the Esatto, I'd suggest setting the serial port manually, save the config. Reboot the system (to ensure that the driver isn't still running) and then see if you can restart INDI.
Ah right - I kind of didn't elaborate the full picture. I run GPSD, I connect INDI after polar alignment and it sync's time and location on the CGX, then I go to mount and check time/location is correct and choose last alignment.
After that I do the process of plate solving.
I use the latest Nexstar HC firmware that you can get teamcelestron.com/ - note this is the bleeding edge, and is usually ahead of what you can get from the normal Celestron site, but anyone can sign up and request access to the firmware files for their equipment.
Interesting, with my Celestron CGX, I do a polar-alignment with the polar scope, then just select "last-alignment", even though the telescope has moved since the last alignment. Then I got to plate-solving about three locations in the sky with the setting set to "Sync", after that I typically slew to my target with the setting changed from "Sync" to "Slew to Target" and it usually gets to within a few arc-seconds on the first slew.
I don't want to spend too much time out in the cold, doing a two star align with four calibration stars, as it is pretty time consuming and would require me to attach a finder scope, which I'm yet to ever use with my current scope.
While talking NUCs - I've noticed that iOptron have released a number of mount versions with a 19V supply and NUC mounting point:
You need to boot by a method that your current firmware supports in order to update to a firmware that supports a native SSD boot.
This means that you need and SD card, although you might possibly be able to Network PXE boot from a DHCP / NFS server. (The SD card is the method that would be cheapest and least technically challenging).
I agree, I feel I should have two images, one for testing the latest version, and another for trying to do astronomy with. There are far too many glitches / changes of operation with version updates for both INDI and Kstars that doing regular updates to something that already works isn't such a good idea. Especially when imaging time is limited.
The way the HDMI works on the Pi is that if it is already booted up and you then connect a monitor then you just see a blank screen.
I have found it necessary to have the screen connected and powered up, and then turn on the Pi - you should see something on the screen even before it finds the boot image.
Yes, I confirm I had the same issue with the SestoSenso2 driver, I'm not going to update my system to 1.9.0 until someone works out what went wrong, I think it might be as a result of changes to Ekos, as I have not found a specific commit with the driver that could cause this.