×

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

Bi-monthly release with minor bug fixes and improvements

Meridian Flip

  • Posts: 910
  • Thank you received: 86

Replied by maxthebuilder on topic Meridian Flip

I was going to praise the meridian flip in Ekos as it didn't fail me... ever. I am praising it actually.
Last night, however, it did fail.
No updates to the system.
SM 1.5.2 on RPI4. Windows client. KStars 3.4.2.
Have been using this config since early March (in self-isolation).
So. Last night it failed. It started the flip but the slew never ended. The speed slowed down to almost zero and it was going and going and going very slowly...
I had to shut the system down and restart.
The night before was just fine (and many nights before too were fine).
Strange glitch.
-- Max S
ZWO AM5. RST-135. AZ-GTI. HEQ5. iOptron SkyTracker.
TPO RC6. FRA400. Rokinon 135 and other lenses.
ZWO ASI2600MC. D5500 modified with UVIR clip-in filter.
ZWO ASI120MM Mini x 2. ZWO 30F4 guider. Orion 50mm guider.
ZWO EAF x 2.
The following user(s) said Thank You: Jose Corazon
3 years 10 months ago #54582

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

  • Posts: 1009
  • Thank you received: 133

Replied by Peter Sütterlin on topic Meridian Flip

Hmm, doesn't sound like a software problem to me. Meridian flip only issues a slew command, and the slew is handled by the mount, with the set slew speed. So I'd suspect some mechanical issue - entangled cables, thickened grease due to temperature or alike causing steppers to lose steps. Strange indeed.

That said, the last nights MF worked just as flawless as the first one for me. Really very reassuring :)
The following user(s) said Thank You: Jasem Mutlaq
3 years 10 months ago #54585

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

Replied by Jasem Mutlaq on topic Meridian Flip

That's right, doesn't appear to be a software issue. I will do a full test tonight at the observatory to check everything including meridian flip.
3 years 10 months ago #54599

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

  • Posts: 1119
  • Thank you received: 182

Replied by Jose Corazon on topic Meridian Flip

I wish I could say the same! For the last 10 days, the meridian flip is no longer working for me. Did not have this problem before.

Please check out this thread: www.indilib.org/forum/ekos/7010-ekos-ent...r-parking.html#54620

Peter, I notice you have CEM60. I am using the same driver for my iOptron SmartEQPro+. I now get consistent failures of the flip. Check out the video and the log file in the link above.

Jo
3 years 10 months ago #54629

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

  • Posts: 554
  • Thank you received: 138

Replied by Chris Rowland on topic Meridian Flip

Jasem,
Can we please revert the changes I made to Ekos to try to improve the mount meridian flip process. It looks as if it is making things worse.

It looks as if the interaction between the mount, the capture module, the park process and the failings of individal mounts is so complex that trying to do ameridian flip process which inevitably needs much of the rest of the system to work properly isn't viable.

It's a shame but I thing it is better to have a syhstem with less functionaity that is reliabe is better than one which might have more functionailty but doesn't work.
3 years 10 months ago #54632

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

  • Posts: 1119
  • Thank you received: 182

Replied by Jose Corazon on topic Meridian Flip

Chris and Jasem,

Before doing that, could you have a brief look at my last post and see if that makes sense?

www.indilib.org/forum/ekos/7010-ekos-ent...r-parking.html#54631

If my hunch is correct the problem could be unrelated to your changes, but only you would know.

Jo
3 years 10 months ago #54633

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

  • Posts: 554
  • Thank you received: 138

Replied by Chris Rowland on topic Meridian Flip

Your log example, between the first and second orange text doesn not show a flip because the pier side does not change.

I haven't looked at the driver in detail but it probably checks the status that the mount reports to the driver to decide that a slew has finished. If the mount does not think it has finished slewing then trying to second guess it is asking for trouble.

I would have expected that slew problems would already have been flagged up before the meridian flip process was changed.

Reverting the meridian fip is the only way to be sure it isn't involved.
3 years 10 months ago #54635

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

  • Posts: 1119
  • Thank you received: 182

Replied by Jose Corazon on topic Meridian Flip


Chris,

The first sequence does not show a pier side change, because for space reasons I only posted the parts relevant for the case I was trying to make, i.e. that the mount maintained its stable position in DEC, whereas in the second example it did not, its position was unstable and alternating between two almost, but not quite, identical positions. I think that is what prevents the meridian flip from completing. In the first example, the meridian flip did obviously complete, and that is only possible if the pier side changed during the slew, otherwise there would be no meridian flip.

The reason I posted this was just to make sure that you don't throw out the baby with the bath tub. Meaning, the problem "could" be very simple and be rooted merely in the unnecessarily precise statement of the DEC position. Reducing the accuracy down to 2 decimal points, instead of the current 6, might have prevented the flip failure in my case.

Is there an easy way to change that decimal precision in Kstars or is this a value that is being processed in the mount itself and can only be changed by editing the firmware of the mount?

That's what I was asking.

I greatly appreciate the important improvement you made to Kstars for observers in the higher latitudes, I just want to make sure that that can be preserved if at all possible.

Jo
3 years 10 months ago #54636

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

  • Posts: 554
  • Thank you received: 138

Replied by Chris Rowland on topic Meridian Flip

Again, the mount position is not used to decide if the mount has finished a slew. The driver reads the mount status from the mount and when the mount status reports that the slew has finished it reports that the slew has finished. Nothing to do with the position. We trust the mount to know when a slew has finished.

The mount seems to think that the position has not been reached. It knows more about what is going on internally than we do and the change in the dec position, even thoug dmall, may be an indication that the slew process has not finished. This could be a side effect of a slew to a dec of 90.

If you feel that you could modify the driver so that it uses the mount position to manage the slew state then feel free.

As for the pier flip, you were suggesting that it was the cause of the problems with this driver. This is just the latest of a series of things that are questionable and I think that having something that is unreliable is worse than not having it. If you don't have it you know where you are, if it's unreliable then you have no idea.
The following user(s) said Thank You: Jose Corazon
3 years 10 months ago #54663

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

  • Posts: 1009
  • Thank you received: 133

Replied by Peter Sütterlin on topic Meridian Flip

So is the pier side info actually used in the MF code? Did I get it right that you assume different mounts give different answers to the :GEA command? Might be one reason why it's an undocumented command.... For the CEM60 it does work absolutely reliable so far....
3 years 10 months ago #54665

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

  • Posts: 554
  • Thank you received: 138

Replied by Chris Rowland on topic Meridian Flip

Yes the pier side info is used in the Ekos Meridian Flip code. It has to be correct for the meridian flip to be executed correctly.

No, the respone to the GEA command does not change depending on the mount type.

Reading the pier side from the mount should have nothing to do with the mount failing to report that the slew has finished, whch is the problem that Jo seems to be having.

But pehaps it does. Perhaps reading the per side from the mount does stop the mount reporting that the slew has finished.
We should remove that as well.
The following user(s) said Thank You: Jose Corazon, Peter Sütterlin
3 years 10 months ago #54666

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

  • Posts: 1119
  • Thank you received: 182

Replied by Jose Corazon on topic Meridian Flip



Hi Max,

I am having exactly the same problem, but only lately. Can you post your log file, there is something I want to check with regards to the very slow slew. Please also post a recent one where the flip worked. I would like to compare them with the log on my iOptron mount where the flip failed in a similar manner as what you are describing.

Thx, Jo
3 years 10 months ago #54674

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

Time to create page: 2.938 seconds