×

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

Bi-monthly release with minor bug fixes and improvements

First Night with EKOS - Comments and Questions

  • Posts: 105
  • Thank you received: 30
I have been wanting to give Ekos/INDI a try for awhile and finally had a chance last night.

I have been imaging for over a decade with Windows/ASCOM/Maxim and more recently SGPro so I feel fairly comfortable with most imaging topics.

Overall I was pleased with the night but there are some issues which I hope I can get some help with understanding.

I even managed to capture some quality data:

astrob.in/full/301313/0/

Here are my comments/questions - most are likely either configuration settings I didn't see or understand properly.

GUIDING

The internal guider worked well with my SX Lodestar on my AP Mach1 mount. I also used PHD2 which did well.

Some questions:

- In the INDI lx200ap driver it seemed like the 'Use Pulse Cmd?' option didn't always show up and didn't seem to persist ON even when I saved the INDI configuration. Any input on how this option works would be helpful. I did finally get it enabled and PHD2 was able to use pulse guiding with the Mach1 so that was good.

- I could not find the option in Ekos or the internal guider to set the threshold for letting the mount settle after a dither operation - could someone please explain this to me?

ALIGNMENT MODULE

- Plate solving with the offline astronomy.net module worked GREAT out of the box. Well done!

- However I could never get a slew to work accurately. I would "Capture and Solve" - have it set to SYNC Mount (I set the lx200ap driver to use RECAL) - then I would slew to the target again in Kstars. The target was always offset slightly - maybe 10-15 arcminutes. It looked like a Jnow/J2000 problem. Would like some help here.

- How does the alignment module use all the historical plate solves with a pointing model? I do NOT want it to do anything with my slew requests besides pass them on to the mount. My polar alignment is accurate enough all I need so a plate solve/sync/reslew and targets in SGP are always centered. I'd like to replicate this setup under Ekos. I did not use the 'Model Editor' button on the Alignment screen to create a module.

- What is the easiest way to enter a J2000 RA/DEC and have the mount slew to it? I plan all my framing with aladin and have a RA/DEC in mind already. Also can I do with without using the scheduler - my preference is to just enter the coordinates manually and have it slew.

FITS VIEWER
- Is there anyway to manually stretch the image in the FITS viewer - I'd like to set a black point and gamma at least.

- When I saved an image from the FITS Viewer to a FITS file it was unreadable by Maxim and Pixinsight on my Windows machine - anyone seen this?
The following user(s) said Thank You: Jasem Mutlaq, Derek
6 years 8 months ago #17499

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

Beautiful results!
+ LX200 UsePulseCMD: I just added it to config file save/load, it should be there tomorrow.
+ PHD2 has settle, but Internal Guider doesn't. Will add this for next KStars release (in a week or two).
+ Best bet is to set it to "Slew to Target", this way it will iterate until it achieves good result governed by your accuracy requirements (set to 30 arcsecs by default)
+ For LX200AP driver, it doesn't use any history, it's just for the running session. It just depends on how well your AP mount updates its internal model after each sync. So when using Align module, it just solves, syncs, and then slews the mount. How the mount itself updates its internal model to account for this is not managed by Ekos.
+ There is currently no direct way to enter J2000. If you have JNow, you can use Mount Control Panel under Mount Tab.
+ No manual stretch yet... was hoping someone would submit a patch for that, but nothing so far.
+ Can you share link to this FITS file?
The following user(s) said Thank You: Michael Fulbright
6 years 8 months ago #17503

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

  • Posts: 105
  • Thank you received: 30
6 years 8 months ago #17507

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

1. Done, saved now.
2. The guide threshold is for guiding deviations during regular guide. It can be set so that capture is aborted when guiding deviations exceed a certain threshold.
3. If it oscillates a lot then either the internal pointing model got screwed up, or the accuracy is too demanding (if it set too low).
4. Looks like the FITS file itself got corrupted somehow, it cannot be opened in KStars.

For development, you can start here for KStars , it was made for Google students, but it's good for all. For INDI, start with INDI Developers Manual . Let me know if you face any trouble! You can download a Kubuntu 17.4 image and run it on VirtualBox.
The following user(s) said Thank You: Michael Fulbright
6 years 8 months ago #17509

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

  • Posts: 105
  • Thank you received: 30
I see you already committed some work with enable settling with the internal guide on the master branch - that was fast!

Do you move things from the master branch to bleeding when they are ready for testing?

Another thing - looking at kstars-bleeding it looks like for PHD2 it uses these settle params which are currently hard coded:

(root/kstars/ekos/guide/externalguide/phd2.cpp)
    settle.insert("pixels", 1.5);
    settle.insert("time", 8);
    settle.insert("timeout", 45);
I guess I would need to recompile kstars if I wanted to change the PHD2 settle values at the moment?
Last edit: 6 years 8 months ago by Michael Fulbright. Reason: Needed to tag code segment for formatting
6 years 8 months ago #17510

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

  • Posts: 105
  • Thank you received: 30
I'm looking at the Alignment problem closer.

I did a 'Capture & Solve' and it returned the following in the log:
2017-06-28T22:18:24 Syncing to RA (16h 43m 28s) DEC ( 38° 51' 57")
2017-06-28T22:18:24 Solution coordinates: RA (16h 43m 28s) DEC ( 38° 51' 57") Telescope Coordinates: RA (16h 43m 00s) DEC ( 38° 51' 55")

If I look in the lx200ap log, however I see:
SCOPE   1924.023439 sec : <setObjectRA>
SCOPE   1924.023539 sec : CMD <:Sr 16:43.4#>
SCOPE   1924.038232 sec : CMD <:Sr 16:43.4#> successful.
SCOPE   1924.038312 sec : <setObjectDEC>
SCOPE   1924.038331 sec : CMD <:Sd +38*51#>
SCOPE   1924.052257 sec : CMD <:Sd +38*51#> successful.
SCOPE   1924.052353 sec : CMD (#:CMR#)
SCOPE   1924.094172 sec : RES (Coordinates     Matched.        )
DEBUG   1924.104388 sec : Synchronization successful Coordinates     Matched.        
INFO    1924.104496 sec : Synchronization successful.
SCOPE   1924.609301 sec : CMD <#:GR#>
SCOPE   1924.626986 sec : RES <16:43:00.0>
SCOPE   1924.627097 sec : VAL [16.7167]
SCOPE   1924.627116 sec : CMD <#:GD#>
SCOPE   1924.644018 sec : RES <+38*51:00>
SCOPE   1924.644153 sec : VAL [38.85]

Something is getting truncated - in the 'Alignment' tab it now reports the "Telescope Coordinates" as 16:43:00 +38:51:00.

So this is why it was oscillating when I did a "Slew to Target" since it was losing the arc seconds field.

I am running:

ii kstars-bleeding 5:17.04+201706100100~ubuntu16.04.1 amd64 desktop planetarium for KDE

ii indi-full 1.4~201706231909~ubuntu16.04.1 amd64 Instrument-Neutral Device Interface library - Full INDI
6 years 8 months ago #17512

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

  • Posts: 456
  • Thank you received: 76
Thats a very impressive image for your first run with a new system!!
The following user(s) said Thank You: Michael Fulbright
6 years 8 months ago #17516

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

For the PHD2, I updated it so it uses Options::ditherSettle() value. You can 'git pull' to update the code.
For the format. can you turn on DEBUG (and DEBUG LOGGING to FILE with everything selected under LOGGING) _before_ you press "Connect"? I want to see how it detects short/long formats for your case.
The following user(s) said Thank You: Michael Fulbright
6 years 8 months ago #17520

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

  • Posts: 105
  • Thank you received: 30
Thank you - I was pleased. The autofocus routine is a pleasant surprise - figured I'd have to spend time training it but just works!
6 years 8 months ago #17531

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

  • Posts: 105
  • Thank you received: 30

I'll try to get a build environment going this week and try it out.

I read the INDI Developer Manual and understood it - one question - is there a "best practices" for having a devel install on the same laptop I use for production? I don't want to destabilize my only imaging laptop I use! I guess I could just change the prefix used by the make process, for example?
6 years 8 months ago #17533

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

You could use a separate VM in VirtualBox and test everything there. On my machine, development is production as I need to know if anything breaks.
6 years 8 months ago #17534

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

Time to create page: 0.761 seconds