Chris,

there must be something wrong in my settings and I can't find what.

When I do the following:

with handcontroller:
boot the mount
search zero position
confirm zero position
Slew to Sun
The mount slew and the Sun is in the Solar ViewFinder. Just a few small corrections to get it centered.

Then Slew back to zero position with HC.
Connect Indi
From Kstars select Sun and goto Sun

The mount slews towards the Sun but is way off.

So Indi and mount disagree on position. When Indi thinks it is time for a flip, the mount in real is not over the meridian yet. So the slew does not move to the other pier side.

Must solve that first .... and generate a log

If you have ideas about the discrepancy ideas welcome.

Paul

(still trying to understand how this works coming from Skywatcher & Eqmod mount)

There is defenitally a timing issue between mount and indi. Something to do with UTC and time offset. Looks like there is 1 hour difference in RA.
The handcontrol shows the correct local time with HC. GPS is OK and time shows Daylight Savings Time UTC +120 minutes.

Slews with HC are correct.
Slews from Indi are incorrect

You can see the result:

This is what kstars shows after slew


Then bringing Sun with HC in center of viewfinder Kstars shows


Also log file attached

Paul

Don't have any problem with this with the Ascom interface, that seems to slew correctly with these HC settings.

Edit: but I remember having a similar problem witt EQAscom earlier this year before DST was in effect (Change to DST solved the problem, here it seems the opposite)

Found something: The mount has time settings with UTC offset and DST (Y or N)
For my timezone had set OFFSET 60 minutes DST Yes. And that gives correct info.

Then comes the Indi driver and overrules the time, setting OFFSET to 120 minutes but does not touch DST setting.
Result: time on the mount is off by 1 hour.

I will set DST to no on the handcontroller and redo all tests (and see if Ascom agrees, because for Solar observing the PI 's usb connection are toooooo slow, so I need my full speed PC for that)

I've not touched the time setting in this driver but as far as I can see the driver ignores the DST setting. You are probably best turning that off in the mount. I'm not sure if it will be best to set the time zone offset to +2 to compensate. One possibility may be to set the time zone and DST to zero and set the scope to UTC.

The log goes nowhere near the meridian so doesn't check what I'm looking for.

It was of no use testing meridian flip if mount and indi do not agree on time/position. I am now running a test with the Sun which should pass meridian in 20 minutes.

Had set offset to 120 and DST to No in the handcontroller, and that slewed correctly to the Sun from within Indi. So now waiting for the Sun. (and then I can also check how good the flip keeps position)
Then I will send the log file. (and also test if slewing directly from HC is ok. Suppose it is, but you never know)

Paul

Chris,

here is a full log of the session.

It starts with Slew to Sun
There will be few slews reported after the first one, these are from the handcontroller to center the Sun
then I did a sync in Ekos/kstart

and waited

13:42:49 Meridian Flip -> Success

And also again a few slews from the handcontroller to center after flip.


13:45 A slew to Beta Triangulum
13:50:03 Meridian Flip -> Success

13:51 Slew to Vega (low in west) Looks like pier side wrong?


13:54 Slew to Casror (low Eas) Doubts about pier side

13:57 slow to M82 again pier side incorrect?


All picures taken direction South. Looks like when scope is pointing north something inverse? (never imaged north, the house is in the way, so have not really an idea how that is handled)

The log is long (you can skip al lot between first slew and meridian flip)

Hope this helps,

Paul

Make sure that KStars uses the correct location, time and DST (click on the globe in the KStars icon bar). Then also make sure that the correct time setting is chosen for the connection between KStars and the mount. Open te KStars settings and then the INDI tab:




HTH, Wouter

Just what I wanted, thanks.

The movement past the meridian is happening and communication remains good, which is what I was worried about. There are a number of flips and the mount is reporting the axis hour angle, pole angle and pointing state as I would expect.

I know what you mean about the pier side looking wrong when looking North. The name Pier Side is only valid when the mount is looking South, this is why I wish we could change it for Pointing State which is what it actually is. Calling it Pier Side confuses people.

I've already pushed a new version to my repository which I hope will cure the PS not being shown problem and shows the hour ange axis correctly in the debug log. Both the ieqpro.cpp and ieqdriverbase.cpp files need to be updated.

This is pretty close to being ready to escape, thanks for your help,

Chris

At least that one is covered in my log (tracking from 2min before to 2 min past meridian, then a slew to the same target that flips)

Thanks, i'd missed that. Can see it now.

I feel a PR coming on..

Wow, seems that went at speed of light. Just did a git pull in the official indi tree, and it's already in!
Compiling now (I assume I can use the main tree there now?). Thanks <strong>so</strong> much for this!

Edit: I'ts running the new (official) version now. Indeed, it now shows pier side already when initially connecting with the telescope in home position / pointing at NCP.

Good to hear that it apears to be working. i'd like to hear how people are getting on with it.

If people are reporting things that you think I ought to fix then a debug log is essential, with a good description of what you see. Remember that I'm not looking over your shoulder.

Posts saying that it's working perfectly and no changes are neede are welcome of course.

Chris

As the OP, I appreciate all the work Chris Rowland did on the code and DerPit on testing. I look forward to seeing more confirmation this issue is fixed, and I can confidently keep the CEM60 as a prime candidate on my wish list.