I know this is a few months old, but I'm having the exact same problem. I'm connected to a Celestron CGEM with the "SkyLink" WiFi adapter plugged into the Aux port and using the INDI "AUX" driver. Everything connects great and the mount moves in the right directions when using the virtual hand controller. I can sync to a star and it shows the correct RA/DEC values, but as soon as I tell it to goto a nearby star (only a short distance away), the mount shoots off in the opposite direction and I have to mash the STOP button to stop the camera crashing into the mount.
I've gone back to connecting through the hand controller with a USB/Serial ST4 cable which works great. Downside is you have to align with the HC first before using INDI.
One thing I did catch out of the corner of my eye, was when doing a goto, the virtual hand controller flashes "error" for a fraction of a second and then immediately changes to "slewing"
This is fixed in INDI 1.9.6 (or so I hope). The only missing configuration right now is Wedge. Alt-Az + Equatorial should in principle work OK. Might need to adjust tracking PID parameters (for Alt-Az mounts).
I've tried playing with this a couple times after updating to the latest... and there are still some issues.
I have a CGEM-II mount with the SkyPortal WiFi.
Raspberry Pi 4 with StellarMate OS and an additional wifi dongle (one connects to home network, one to the Celestron WiFi)
After working around a version issue with libgphoto2, I was able to run update_indi_core and update_indi_drivers.
What I saw was very similar to what Andrew described.
The mount connects and initializes with a seemingly random coordinate; so I sync to Polaris (roughly polar aligned).
The new coordinates look right; If I goto Polaris, it stays put... which is good.
But if I go to any other nearby star, it shoots off somewhere completely different (usually to a position that would damage equipment).
Also, unless I have a hand controller plugged in or go through alignment in SkySafari, there seems to be an artificial stop in one direction of the RA motor.
But, after any form of alignment (through hand controller or app) that artificial stop goes away.
However, I verified this happens when using the manual controls in SkySafari as well, if I have not run an alignment.
I also verified that if I run the alignment in SkySafari, it will then accurately slew the scope to selected targets within SkySafari (so, it can work).
Quick one (a problem I had a long time ago, which fixed itself.. but now picking things up again i'm having).
My KStars direction / movement is inverted when using ekos to control the mount. My guess is something to do with the southern hemisphere, mount is celestron wifi ALTAZ.
To simplify things in terms of logs, I did a quick connect, and a couple of seconds of *UPWARD* movement. The mount itself moves correctly upward. Logs attached.
Mount location seems to set correctly (KStars location also set to Sydney, Australia).
[2022-06-03T13:54:30.356 AEST INFO ][ org.kde.kstars.indi] - Celestron WiFi : "[INFO] Observer location updated: Latitude -33:54:59.0 (-33.92) Longitude 151:16:59.0 (151.28) "
However in diving into things, it seems the problem is the encoder direction might be inverted? As (see below) the AL output is decreasing rather than increasing.
Am I just completely down the wrong path here? The actual conversion from encoder numbers to AL seems to be correct from my calculation, so is it either that isNorthHemisphere() isn't firing correctly (I haven't found where that sits exactly yet), or does this model of mount have the encoder output reversed?
- From cold power on from a fresh install of the StellarMate beta stream, it slews to the correct target... sort of...
- If it tries to slew more than about 15-degrees clockwise of center on the RA, it abruptly stops motion on that axis.
- If I run alignment in an external app, then switch back to KStars/Ekos/StellarMate, it no long hits the stop... but it also does not slew to the correct location.
My SD card died, so I had to grab a new one and get it up and running again so I could get logs.
I got tired of setting my stuff up in the in the cold, so I'm testing by roughly polar aligning my mount and checking positions using SkySafari.
In order to get past the short artificial stop rotating east on the RA, I disconnected/stopped indi/ekos, and ran a scope alignment with SkySafari.
In the process, the StellarMate seemed to crash during that time, so I powered it off and on again before I could connected to it and start Ekos/Indi.
For good measure, I mash the 'sync gps' button in the StellarMate app; location in the CGEM-II looks almost right... except is show 240.xx degrees longitude instead of -119.yy (I'm not sure if that's normal).
Upon connecting, KStars shows the scope position somewhere in the south; way below the horizon (at 259P/Garradd).
So, I select Polaris and Sync the scope location.
In the _after_align.log...
I try to go to Merak (sort of near zenith), but it hits a hard stop (if I had my scope attached to the mount, it might have bumped the tripod).
So, then I send it to Pollux, and it kind of gets in the neighborhood.
It tell it to park, and it goes 90-degrees clockwise of the home position for both axes.
So, then I power the mount off and on again, and close/re-open kstars (_no_align.log)
It comes up with the cgem-ii pointing at Polaris this time, and correctly slews to Pollux and Vega... but hits an the artificial stop slewing to Merak (about 15-ish degrees past center on the RA).
Same issue as Rob. There's a phantom "hard stop" just past the meridian. There's no deceleration, just an instant stop (like it hit something) and even the virtual hand controller won't move past it.
If I align on a star in the west, the hard stop is just east of the meridian, and if I align on a star in the east, the hard stop is just west of the meridian. Wish I could figure out where the logs are, I'd upload them, lol.
My KStars direction / movement is inverted when using ekos to control the mount. Like Jack I'm in the southern hemisphere. My setup is a CPC925 on an equatorial wedge, using Astroberry to control the equipment. Mount location is set correctly (KStars location also set to Kyneton, Australia).
In addition to what Jack has noted in his post, I've found the following:
- after syncing to three stars I can successfully do a goto to those stars from Kstars.
- doing a goto to an un-synced star, after the above three star alignment, sees the telescope move in the wrong drection, as noted by Jack.
- after attempting to go to an 'un-synced' star, if I then do a goto back to a synced star, the location marker for the telescope in Kstars is no longer showing at the location of the synced star on Kstars even though the telescope has correctly returned to the poistion of the synced star.
- after syncing to a star, if I use plate solving to get more accurate positioning of the star, the initial plate solving result shows th ealignment is close, but then as it tries to move closer to the target star, it ends up moving further away.
I'll try and capture some of this behaviour in logs next clear night.
Any assistance in understanding and resolving this problem would be greatly appreciated.