×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Meridian Flip fails

  • Posts: 66
  • Thank you received: 2
Last night, I was out trying to photograph galaxies in Makarians Chain. The object in question was crossing the meridian and Ekos appeared to send a signal to carry out a meridian flip. As can be seen on the attached log from a time of 2022-03-26T00:25:25.999 when this was initiated, it failed (log is the relevant part - whole log to large to attach). It subsequently retried and failed again. I tried to park the scope and execute a go to following several failure, but the mount kept going back to the west side of the pier when it was originally and although meriidan flip was initiated, it repeatedly failed. This is a real problem, as I cannot leave the mount tracking with the camera on the west side of the pier on this target as the dslr will hit the legs of the mount. Could somebody look at the logs and tell me if I have a setting wrong? Alternatively, is there a way to effectively do a manual meriidian flip to force the scope to the other side of the pier?
2 years 2 days ago #81755
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

No thoughts on this or any help?
2 years 16 hours ago #81805

Please Log in or Create an account to join the conversation.

  • Posts: 992
  • Thank you received: 155

Replied by Peter Kennett on topic Meridian Flip fails

MF failed for me last night too, with the latest Mac version. I didn't have a problem before.
I didn't have error logging turned on, so can't comment on what may have happened.
I simply did a GOTO back to the subject and it moved over to the other side, aligned and resumed guiding- so not the same issue you had.
2 years 15 hours ago #81808

Please Log in or Create an account to join the conversation.

  • Posts: 225
  • Thank you received: 16

Replied by Ron Clanton on topic Meridian Flip fails

I've had this problem as well... you'll find another thread related to this topic.

My current working theory is that while EKOS knows the object has crossed the meridian, the mount does not. This tells me that EKOS is not updating the mount as it should. Of course this is only a problem if there is a significant difference in their settings... i.e., your mount's RA is not exactly vertical. I confirmed that mine was about 3 degrees off... so I adjusted it a bit. I also changed the "Flip if HA >" parameter in the mount tab from 0.1 hours to 0.3 hours to allow a bit more "slack". So far, that seems to work... it's hard to prove a negative...

BTW... I figured out the error by looking at my initial alignment error when plate solving. I was alway off on the initial alignment by several thousand arc seconds. Curious as to your results.

Now, I'd really like to get a fix such that EKOS would truly update the mount.... but am okay for now... I think.

Ron
2 years 11 hours ago #81812

Please Log in or Create an account to join the conversation.

  • Posts: 593
  • Thank you received: 276

Replied by John on topic Meridian Flip fails

Looks like it tried at 00:25, 00:27 and 00:31 to do a MF but didn't move the mount.

I would suggest:
1. Try setting a HA > 0. You need to work out what works for your setup so that the camera doesn't crash into the tripod/pier. The most common reason I think is that the Mount and Ekos are slightly out of sync on timing or location. Its likely that your Mount will only flip when its past the Meridian (HA > 0) so if, for example, EKOS thinks its ready to flip but the Mount thinks there are a few minutes to go then when Ekos issues a MF, it sends a slew command to the mount for the exact position that the Mount is already pointing to. The Mount works out whether a MF is actually required or not, and if so, it performs the MF. If the Mount doesn't think it has reached the Meridian yet then it won't move. There is another Ekos bug that makes the situation worse. Ekos prints a log message saying its going to wait 4 minutes before trying again, but it doesn't wait 4 minutes, in your case it waited 2 minutes. I've seen it try again immediately on my system. I've also seen that if it tries multiple times it seems to get itself confused and needs manual intervention to sort things out.
2. You can practice all this in daylight with your Mount and Kstars. Pick a star that is a few minutes from traversing the Meridian and track it. Configure the MF in the Mount tab of Ekos playing around with HA until you get it working consistently. Obviously, if you can fix the problem in daylight it is sooo much easier.
1 year 11 months ago #81819

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

Peter,
I tried parking the mount and then executing a go to back to target. This did not work properly - it went to the same side (west) of the mount where it was previously.
1 year 11 months ago #81820

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

Ron,
Thank you for your thoughts. I do seem to find that my go to is off by a substantial amount. For example if I go to pleiades it seems about 10 degrees off. I wonder if this could be a factor? This is an improvement on a fortnight ago where if i did a go to the dec axis seemed to point somewhere in the region of 180 degrees off... It is frustrating because prior to have=ing to do a clean install of stellarmate it all seemed fine.

So you say that you think that EKOS is not updating the mount? The thing is, I cannot work this out as I platesolve and sync results so ekos and the mount following platesolve appear to go to the right place.
1 year 11 months ago #81821

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

John,

I will give this a go. What puzzles me here though is the staement ' EKOS thinks its ready to flip but the Mount thinks there are a few minutes to go'. If I was using a synscan mount, then I would get it, but my mount is effectively a 'dumb' mount until connected to computer (see mount controller in my signature). Also 'The Mount works out whether a MF is actually required or not, and if so, it performs the MF. If the Mount doesn't think it has reached the Meridian yet then it won't move' - again the mount controller is essentially dumb without computer connection.

I shall give the HA settings a go in daylight when work etc allows. Many thanks.
1 year 11 months ago #81823

Please Log in or Create an account to join the conversation.

  • Posts: 225
  • Thank you received: 16

Replied by Ron Clanton on topic Meridian Flip fails

Richard,

As I understand it, EKOS does not send a MF command to the mount. It simply sends a "go-to" command to the mount. If the mount is past the meridian, then the mount will perform the MF to get there. If it's not past the meridian, then it won't perform a MF. At least, that's what I think is happening with my mount.

One difference in your situation is that if I perform a plate solve on the first object and it is off by a certain amount and it syncs/slews to correct correctly. However, if I go-to another object and perform a plate solve, it has an error of approximately the same amount as the first object. This is why I assume that it is not update the mount correct. It sounds like yours does not have this issue.

Ron
1 year 11 months ago #81826

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809

Replied by Rob Lancaster on topic Meridian Flip fails

Yes, I believe this is correct. There is no universal "Do a meridian flip" command. There are a couple of mounts/drivers that actually have a setting to do a Meridian Flip on the next slew command or a similar option, but as far as I know, there isn't a command to just do one. Software has to rely on the mount to want to do the Meridian Flip. Ekos issues a goto command at the right time, and as long as the mount thinks it is past the point where it needs to do a meridian flip, it will do a full flip even if it doesn't have to go very far to get to the position it is slewing to.
1 year 11 months ago #81828

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

So how would the mount 'know' it is past the point where it has to do a meridian flip? I can see how that would work with say a synscan mount of a mount with encoders. I thought that the process would be for the software to send a go to command but crucially, to have the scope do the go to on the other side of the pier? This seems not to be the case from what you have said.

I think I need to investigate this further. As per my other urrent thread, there do seem to be alignment problems with my mount following a reinstall of everyhing and I wonder if this is at the heart of the issue? Thread is here for reference:

indilib.org/forum/stellarmate/11462-furt....html?start=12#81830
1 year 11 months ago #81831

Please Log in or Create an account to join the conversation.

  • Posts: 66
  • Thank you received: 2

Replied by Richard Wesson on topic Meridian Flip fails

So how would the mount 'know' it is past the point where it has to do a meridian flip? I can see how that would work with say a synscan mount of a mount with encoders. I thought that the process would be for the software to send a go to command but crucially, to have the scope do the go to on the other side of the pier? This seems not to be the case from what you have said.

I think I need to investigate this further. As per my other urrent thread, there do seem to be alignment problems with my mount following a reinstall of everyhing and I wonder if this is at the heart of the issue? Thread is here for reference:

indilib.org/forum/stellarmate/11462-furt....html?start=12#81830
1 year 11 months ago #81832

Please Log in or Create an account to join the conversation.

Time to create page: 1.028 seconds