I also set GPS to resync every 600s in the INDI control panel

I wasn't aware that was possible. Gonna try it tonight...

I found it in the config settings of GPSD (not sure if this only applies to GPSD)

I did some testing last night, with GPS disabled and kstars set to update all devices, and Meridian flip failed again.
Here is what happend:

• Meridian flip triggers (still an exposure ongoing, it has to wait)
• Once exposure is finished, meridian flip starts
• Telescope moves extremly slow (as in Centering speed slow) during the meridian flip and moves in a wrong direction
• I manually abort the mount movements and trigger a manual GOTO
• Now the telescope moves fast, but misses the target slightly and overshoots (and does not stop moving further)
• I again abort the movement and trigger a new manual GOTO
• Telescope again moves fast (as it should) but misses target again and overshoots again
• I stop the movement again
• I stop the scheduler and delete / purge all the config of the mount in the Mount tab of EKOS
• I trigger a new manual GOTO
• Now it slews to the correct position and stops where it should
• I restart the Scheduler and now it works

Relevant bits from the logs are attached, can anybody make any sense of it? The logs show a very active communication between the mount and EKOS for some reason ongoing during the whole flip. Not sure why.

I do use a different (cheaper) mount than yours and my mount allows me to automatically trigger the MF. I turned that option off. Your mount can probably do that too, so check the settings if you haven't already done so. Not sure if that will help, but I think it's worth a try.

I have also recently had trouble with flips. Last night the meridian flip initiated during the last capture of a sequence. I waited until the capture was finished and successfully flipped. It did not focus even though the next sequence should have triggered a focus. It aligned and started guiding OK however. Then it said it was capturing and the log indicated it started capturing. However it did not have a sequence loaded and was doing nothing in the CCD panel. Nothing in the log to indicate a problem. I stopped it and restarted and it focused, align, guided and captured properly and continued for the rest of the night.

Here is a verbose log from a failed flip - its in the last part of the file. Flip initiated but it only slewed a short distance and stopped. It took three attempts restarting the job before it slewed all the way around.

The mount moving only a short distance sounds exactly like the issue I am having too.

Is there anyone we could ping maybe to take a look at the logs? (I am kind of lost in them)

I had a quick look at the log. The thing that looks slightly odd is that your mount doesn't report pier side. The manual says it does so it may be a limitation of the Indi driver or something else I'm not sure. The way Ekos checks if a Meridian Flip is done is wait for the slew to stop (tracking starts) and then check that the pier side after the MF is different to what it was before the MF. For mounts not reporting pier side, it expects the MF to take a minimum amount of time (looks to set to 20secs). Anyway, this is just something curious, its not the real issue.

The flip in the log looks to have only taken a couple of seconds which is consistent with the flip not happening.

So the question obviously is why didn't the mount flip? The mount will usually flip if its asked to slew to a target on the "other" side of the Meridian. If its asked to slew to a target on the same side of the Meridian it won't flip. Ekos does a flip by requesting a slew to the current target position. So if Ekos and the mount aren't lined up in terms of time and position there may be a problem. If Ekos thinks the target has transited the Meridian and requests a MF and if the mount thinks it hasn't quite transited then the mount will either not move at all or move a small distance (between where the mount thinks it is and the plate solved position). This could be the issue.

The easiest way to check this is to make sure the mount sync'd with Ekos for time and position. Make sure the mount isn't getting its time / position from elsewhere, e.g. GPS. There is an option in Ekos to sync the mount to Ekos for time and position data. This is the easiest way to do this.

If you've already done all this and still have the problem then the next thing will be the RST-135 Meridian Limit Setting. From looking at the manual (I don't have this mount) it seems that you can set this with the hand controller. This offers some more info...www.firstlightoptics.com/user/products/l...y%20Manual%20V16.pdf

It controls how far past the Meridian the mount is allowed to go. Once it reaches this point tracking is disabled. The other thing from googling your mount, it appears that the mount will only flip if it gets a slew beyond the Meridian Limit. So if Ekos starts a MF at 10 minutes past the Meridian and the RST-135 Meridian Limit is at 20 minutes then the mount won't flip. See this thread.... www.cloudynights.com/topic/666852-rainbo...maging-mount/page-53

So you'll probably need to work out a set of config that works for you. Something like:
1. Set time / position to be sync'd from Ekos to mount. Disable all other mount sources of time / position like GPS.
2. Set Ekos MF at say 10 minutes past Meridian.
3. Set RST-135 Meridian Limit to say 11 minutes past Meridian.

Note that the RST-135 Meridian Limit can't be before the Ekos MF because if it is the Mount will stop tracking and never reach the Ekos MF trigger point.

Ekos will trigger the MF at 10 minutes past the Meriidan. The RST-135 probably won't move. Ekos will keep trying. When the mount reaches 11 minutes past the Meridian tracking will be suspected and the mount will stop until the next Ekos MF slew is received. After that all should be good.

I would suggest setting the mount up indoors using the skymap to slew to targets near the Meridian. Configure your setup for a MF and let it run. Keep going until you get a set of config that reliably works for you.

Good luck!
HTH

Thanks for this very detailed feedback which gives me a couple of options to try. The mount doesn’t report pier side (making it impossible to reuse guiding calibration) but the driver has an option to simulate pier side so I will see if that helps. Previously flips were working and it was flipping when it was past the meridian but within the mount meridian limit so I don’t think that is the problem. I think the most likely culprit is time difference, I used to sync ekos from the mount but stopped because of problems with the time change - the mount doesn’t know about daylight savings time. As a result things could get confused by an hour. I will have to mess with this and see if there is a way to keep things in sync without confusion. At the moment I am setting mount location and time from GPS and ekos has manual location and time via internet.

Turns out I was already setup for kstars updates mount on time and location so it wasn't a timing issue. I have gotos that also stop quickly after moving only a few degrees so I am thinking this is not a meridian flip issue - it's a more general slew issue.

Might be worth checking what limits you have setup in the mount (nothing to do with Ekos). Even if you never explicitly set anything up, there may be defaults set. From the doco on your mount it seems you can setup general slew limits as well as a Meridian Flip limit. These would all need to be set self-consistently and consistently with Ekos for everything to work. If you were about to breach a limit I would expect the slew to be stopped by the mount whether its a MF or a general goto.

I just want to share a recently resolved problem, one manifestation of which was precisely the subject of this thread: Ekos-initiated meridian flips were failing. They would just move the telescope a tiny bit and settle there. I hope this can be helpful in some way.

Due to the system time being incorrect on the OS hosting INDI, the mount was receiving incorrect Local Sidereal Time and therefore incorrectly calculating the position of the meridian.

So when KStars/Ekos rightfully thought the target being tracked was behind the meridian and it was time to flip the mount, the mount had a vastly different opinion of where the meridian was. As I understand, all KStars/Ekos do to initiate a flip is issue a slew (go-to) command to the current coordinates where the telescope is already pointed, the expectation being that the mount will "understand" the target is now on the different side of the meridian than the current side-of-pier setting and therefore flip itself.

The current target coordinates given as the slew target to the mount with an incorrect "understanding" of the meridian location did not look to be on the other side of it, and no flip was being performed.

So yes, just as it is suggested in multiple self-diagnostic resources I've seen, the main problem was time not being synchronized between the devices, but it was not the mount driver/KStars mismatch that mattered, but rather the OS time on the system running the INDI mount driver.

Originally I was working through a different symptom of this issue: the mount would flip around lines that were not the meridian, but casually slew past the meridian itself. You can see how it is essentially the same problem. My full report on that is here: indilib.org/forum/ekos/13703-slewing-to-...the-mount.html#95506.